SPECIAL SECTION ON CLOUD - FOG - EDGE COMPUTING

IN CYBER-PHYSICAL-SOCIAL SYSTEMS (CPSS)

Received February 26, 2020, accepted March 13, 2020, date of publication April 7, 2020, date of current version April 22, 2020.
Digital Object Identifier 10.1109/ACCESS.2020.2986369

Smart Routing Management Framework

Exploiting Dynamic Data Resources
of Cross-Layer Design and Machine
Learning Approaches for
Mobile Cognitive Radio
Networks: A Survey
QUSAY MEDHAT SALIH 1 , MD. ARAFATUR RAHMAN 1,2 , (Senior Member, IEEE),
FADI AL-TURJMAN 3,4 , AND ZAFRIL RIZAL M. AZMI1
1 Facultyof Computing, University Malaysia Pahang, Gambang 26300, Malaysia
2 IBM CoE and ERAS, University Malaysia Pahang, Gambang 26300, Malaysia
3 Artificial
Intelligence Engineering Department, Near East University, 99138 Nicosia, Turkey
4 Research Center for AI and IoT, Near East University, 99138 Nicosia, Turkey

Corresponding authors: Qusay Medhat Salih ([email protected]) and Md. Arafatur Rahman ([email protected])
This work was supported in part by the Ministry of Higher Education Malaysia through the Fundamental Research Grant Scheme (FRGS)
under Grant FRGS/1/2019/ICT03/UMP/02/2, and in part by the Universiti Malaysia Pahang (Flagship Grant) under Grant RDU182202-2.

ABSTRACT The concept of Cognitive Radio (CR) has emerged as a practical solution to solve the issue of
the fixed spectrum and bandwidth scarcity in wireless communication. However, the nature of dynamic
Mobile Cognitive Radio Networks (MCRNs) drives to the emergence of new challenges, especially
concerning the routing protocol operations. Applying a cross-layer design is considered a sufficient remedy
to overcome routing protocol challenges such (e.g. channel diversity, integration route discovery with
spectrum decision, mobility, etc.). Consequently, the cross-layer design has a magic solution to overwhelm
routing challenges in MCRNs due to the ability to be free from the strict boundary and share the information
and services with other layers in a manner that contributes to enhancing routing performance. Thus, the scope
of this survey is to review and taxonomy numerous routing protocols in MCRNs according to methods of
design to highlight the strength and weakness points. Also, machine learning has acquired much interest
in this literature. A cross-layer framework for smart routing protocol in MCRNs has been proposed by
exploiting machine learning mechanisms. Finally, the open research issues of routing protocol in MCRNs
are summed up.

INDEX TERMS Mobile cognitive radio network, cross-layer design, non-cross-layer design, machine
learning, smart routing protocol.

I. INTRODUCTION technology known as cognitive radio. CR technology allows

The growing demand for wireless applications, coupled with an unlicensed user namely Cognitive User (CU) or Secondary
the ineffectiveness in spectrum usage, invite for the appear- User (SU) to compose links of communication over licensed
ance of a new wireless communication paradigm that based spectrum bands that used by the Primary User (PU) on
on the dynamic spectrum participation rather than a fixed an opportunistic foundation to increments the manage-
spectrum. To implement the concept of dynamic spectrum ment spectrum efficiency [1]. However, CUs have imposed
in wireless technology, this required the development of new serious dilemmas because of the dynamic nature of the
The associate editor coordinating the review of this manuscript and channel, mobility nodes, and the interference probability with
approving it for publication was Xiaokang Wang. the PU [2].

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
VOLUME 8, 2020 67835
 Q. M. Salih et al.: Smart Routing Management Framework Exploiting Dynamic Data Resources

Under these circumstances, the routing protocols had • The prominent routing methods or techniques in
a wide range of problems in MCRNs, especially channel MCRNs are reviewed and highlighted their strengths and
selection, path stability, QoS, PU interference, and weaknesses.
others [3], [4], [6]. In addition, MCNRs can significantly • A generic cross-layer framework for intelligent routing
effect on characteristics of the stack-layer protocols due to protocol in MCRNs has been proposed. The frame-
the dynamic network resources and PU activities [3]. Hence, work will offer a quantum leap in principle working of
there has been increasingly focused on getting rid of routing routing protocols in MCRNs through engaging machine
problems in MCRNs. According to this, numerous encour- learning functions with layer stack services.
aging routing protocols have been proposed in MCRNs, • The open research challenges for routing protocol in
which followed different techniques [2], [5]. For instance, MCRNs are outlined.
joining path and spectrum diversity can assist MCRNs in The rest of the paper is structured as follows. Section II
avoiding the activity of PUs and improving routing perfor- explorers the fundamentals of routing protocols in MCRNs.
mance. Moreover,The cross-layer design is a premium solu- In Section III, it describes the classification of routing proto-
tion through sharing services and information in the effective cols in MCRNs. Subsequently, a critical review of existing
way between non-adjacent layers to accommodate changes methods or techniques for the routing protocol in MCRNs
CRNs topology [7]. is done, where their strengths and weaknesses are needed In
Today, many new applications in Cognitive Radio Section IV. After then, Section V provides a description of
Networks (CRNs) can be highlighted. For instance, the smart cross-layer framework routing protocol in MCRNs.
(CR)-based internet of things [8], military applications [9], Next, Section VI presents suggestions for the directions
health monitoring [10], 5G technology [11], etc. However, for research challenges that seeking to improve the routing
all these technologies require a robust and flexible routing performance in MCRNs. In the end, the paper is concluded
protocol that can transfer data packet applications between in Section VII.
the nodes reliably. In contrast, the routing protocols in CRNs
still are facing more hurdles due to new emerging applica- II. FUNDAMENTALS OF CROSS-LAYER ROUTING
tions, which needs to find more intelligent routing protocols. Cross-layer design has been broadly investigated and
Learning from the CRNs environment is a funda- developed to enhance the decision resource allocation in
mental demand for providing intelligent communication MCRNs [103], [107]. Recently, especially in the field of
services [100]. In fact, routing protocols in MCRNs can routing protocol in MCRNs, cross-layer design theory has
learn and reconfigure its services to adapt with the dynamical been the focus of attention by many researchers concerning
network resource [99]. In other words, machine learning can improving route decision and link-channel selection [3].
play to optimise routing performance and utilise network Fundamentals of cross-layer routing in MCRNs require
spectrum resources [98], [99]. acknowledging the cross-layer resources and limitations
There are existing surveys for a routing protocol in MCRNs considering the routing challenges [85]. The cross-layer facil-
that try to highlight routing protocol in MCRNs from different ities can present an ideal model that can trade off guid-
aspects, e.g., routing and channel selection, routing metric, ance and resource performance across layers. From the
etc. However, there are still no serious attempts to study same perspective, diagnosing the problem of routing proto-
routing protocols in MCRNs from a perspective of employing cols in MCRNs can guide to knowledge of the required
machine learning beside CRNs protocol to signify a smart services and information from other layers [87]. For example,
routing protocol in MCRNs by using a cross-layer design. the path is defined based on physical layer sensing infor-
Also, It explorers routing protocols in MCRNs from a mation expressed by the probability of channel availability.
perspective of the cross and non-cross layer design to identify By setting up a routing algorithm to adapt to new features and
and distinguish the difference between each model. Hence, implementing them across the layered framework can lead to
a critical review has been done to recapitulation and evalua- create a valid routing protocol in CRNs [5], [106], [107].
tion of routing protocol in MCRNs to highlight the strength There must be harmony between traditional routing func-
and weakness points. Besides, the cross-layer design and tions and the extended functions to create a reliable routing
machine learning methods have analysed and synthesised protocol.
to design intelligent routing framework in MCRNs. Conse- For more details, the next subsections will discuss the basis
quently, we have proposed an intelligent routing framework of fundamental cross-layer routing.
that has features of cross-layer and machine learning to make
the right routing decision. A. BASIC ROUTING FRAMEWORK FOR COGNITIVE
The foremost contributions of this paper are summarized RADIO NETWORKS
as follows.: The general routing framework in CRNs has described
• This study explorers the fundamentals of routing in Fig. 1, which consists of many various blocks such
protocol in MCRNs. as (QoS) evaluation, routing information, learning decision
• It has been proposed a new taxonomy for routing and route establishment block [12]. The block of routing
protocol in MCRNs. decision executes base on the knowledge of the interaction

67836 VOLUME 8, 2020

Q. M. Salih et al.: Smart Routing Management Framework Exploiting Dynamic Data Resources

outcome of preceding block information can be signifi-

cant (according to a selected metric), to establish a new
route to achieve the better router performance.

B. CROSS-LAYER-ROUTING CONCEPT
The concept of the cross-layer design has introduced to
establish a link between different protocols in different
layers and to minimize the overhead in the layer stack [17].
In another way, the cross-layer design breakdowns the
traditional network stack in which each protocol in the
network layer’s stack operates independently [18]. In CRNs,
a cross-layer methodology attempts to intensify the accom-
plishment of routing by combined the lower layers. Conse-
quently, the overall performance of the wireless systems in
FIGURE 1. Basic routing framework for CRNs. term of data rate, error, and radio resource utilization is
improved [19]. Hence, a cross-layer routing strategy in CRNs
between all these blocks, whose descriptions clarity in the is a must, not a choice [20].
ensuing: In order to address the satisfactory network performance,
• Block of Routing Information: This block includes apply QoS requirements, and reduce interference, the tradi-
information such as next-hop, the availability and tional network layer stack and stratified protocol reference
quality of the channel, modulation, transmission rate, models are limited and not able to solve the wireless network
and different other parameters which are specific to challenges in CRNs [21]. For that, the static layer model
each link. Besides, the process of channel switching cannot address all the challenges of CRNs and provide an effi-
includes a finite delay, which diminishes the rate of cient routing performance. The cross-layer design depends on
throughput and increases the latency in the end-to-end the fact that traditional static models can be revised in many
connection. [13]. In the routing protocol, the channel ways to achieve QoS, network performance, and interference
selection is considered the primary role in order to get a mitigation [17]. Also, spectrum management in CRNs needs
high performance and stability [1]. For that, the channel a cross-layer design mechanism where different layers can
select strategy has to regard choice the channel with less be considered in attaining the optimal design and imple-
occupied by PU to reduce the total interference, high mentation results [22]. Indeed, the routing performance can
availability and high connectivity with other neighbours. take advantage of the stack-layer information and make them
• Block of QoS Evaluation: The QoS rate in CRNs useful tools to develop routing operations [7].
is related to the effectiveness of the routing algori-
thm [14], [15]. In other words, application layer C. A FUNCTION KEY OF THE STACK LAYERS IN THE
requirements receive by this QoS block and gauge to COGNITIVE RADIO NETWORK
discover how the performance of current routing to these The cross-layer is defined as the design that violation
requirements. of the communication architecture of a reference protocol
• Block of Learning: The network paradigm is moving concerning the architecture of the particular layered [22],
towards learning from observation and experience to [23]. In general, routing protocol in MCRNs is considered the
enhance performance over time [16]. It is imperative to main issue because of the fluctuation availability of spectrum
incorporate the routing framework into the fabric of the resource and PU activity [6]. Therefore, it is worth noting
learning block. In more, the networks develop on the that the objective is to find appropriate solutions for routing
way to the individual-learning and environment-aware issues through suggested a cross-layer design as a significant
paradigm. As a result, the learning block became neces- solution.
sary to take part in the routing framework. By following In more specific, mobility CU and/or PU impacts on
up for the prior history of the channel, This block traces routing protocol to recognise the best idle channel and
the routing operations layer over time. It serves the block the stable path between a pair of CUs [24]. For this
of the decision to determine a more reliable idle channel reason, SU must be able to accommodate the dynamic
and a stable route. change in the spectrum utilisation by PU [5], respectively.
• Block of Decision: According to the output of sensing Routing protocol has many challenges [2] such as (channel
knowledge, the upshot of the block of QoS evaluation, switching, CR mobility, PU interference, etc.), which gener-
learning, and decision leads to make a decision that ally affect the routing performance in CRNs [25]. In partic-
might modify the current path or to exchange a channel ular, applying a cross-layer method with a multi-channel
or might keep continuing to the current route. and the multi-path concept it draws a new communica-
• Block of Route establishment: Lastly, this block erects a tion environment in principle working of routing protocol
route from the source node to the target node, when the in CRNs [26].

VOLUME 8, 2020 67837

 Q. M. Salih et al.: Smart Routing Management Framework Exploiting Dynamic Data Resources

FIGURE 2. Stack-layers’ functions in cognitive radio technology [12], [27], [30], [101].

For instance, the cooperation of physical and MAC layers • Network Layer: The network layer can forward a data
can feed the routing algorithm very prominent information packet from the sender node to the target node. More-
about spectrum sensing, which utilized by routing deci- over, maintenance and update of routes that it use for
sions [5], [13]. Shortly, the routing can get more optimization delivering a packet over it [7], [29]. It also acts frag-
by allowing the layer of physical and MAC for collabo- mentation and reassembly of packets, if required [12].
ration [7]. In more details, every layer in the stack-layer In contrast, the traditional routing protocol at the
can provide different information and services to adapt with network layer has many challenges in MCRNs. In other
MCRNs [12], [27], [28] as explained in Fig. 2. Hence, every words, the routing protocol in CRNs is not like conven-
layer can serve different functions by creating new interfaces tional self-organising wireless ad hoc networks [2].
to the other layers in order to facilitate information exchange, It has not designed to work with a dynamic spec-
flexible control, and protocol optimization. In more details, trum frequency band and PU activity [31]. In a word,
every layer function has been explained as mentioned the limitation of information about channel availability,
below: SU mobility, probability PU interference, and other
• Application layer: The services of application layer factors that can affect the routing performance [32].
introduce information about the availability of the • Link-layer: This layer is capable of handling with
resource, synchronizing communication, and identi- multiplexing/multiple access of data across one or more
fying the communicating devices [29]. Furthermore, physical links. Besides, it is responsible for error correc-
it can introduce methods of data dissemination, data tion operations. From the same perspective, the link
aggregation, and fusion [30]. For that, the application layer, in MCRNs, is accountable for accessing and
layer plays a crucial role in the definition of QoS. capable of using the available spectrum opportuni-
In contrast, the efficient QoS in CR networks is consid- ties [29]. More functions are provided by the MAC,
ered highly challenging due to increasing dynamic of which is a sub-layer of the link-layer. MAC layer can
network conditions that it cannot always be guaranteed allow for multiple users to be simultaneously sharing the
the availability of the required resources [7]. resource of the channel within the same network [7]. The
• Transport layer: It can provide different services that MCRNs has imposed on link-layer some new functions
including congestion control, flow control, and end- compare to the services are offered traditional wireless
to-end error recovery. Also, it is a title from many vulner- networks. It needs to explore PU activity to avoid inter-
abilities subjects that infect the MCRNs [29]. ference with it [7]. Besides, due to spectrum mobility,

67838 VOLUME 8, 2020

Q. M. Salih et al.: Smart Routing Management Framework Exploiting Dynamic Data Resources

the Common Control Channel CCC) might be not useful demand for each node in the network to deliver a data bit
to use to transfer the routing control packet to coordinate to the target node safely [102]. As noted, the cross-layer
the users [2]. method could reduce energy consumption in the term (per
• Physical-layer: Physical layer manages the connection bit), through combining lower layer parameters.
between the data link layer and the physical wire-
less medium. The physical layer interest that how to D. WHY CROSS-LAYER
transfer the bits of information from a sender to a The classical network layer stack model has not provided
receiver [33]. The processes that occur within this layer quick accessibility for services and information with the other
are modulation/demodulation, coding/ decoding, and layers. In general, The process of exchange control informa-
signal processing for transmission and reception [7]. tion carries out only between the neighbor’s layers protocol
In more, the difference in this, that in the case of by applying the service access point (SAP) concept [103].
CRNs, the physical layer has to able to adapt rapidly The function of SAP is to enable the ability to access and
due to dynamically of the spectrum, which brings more choose the functionalities of subset protocol by precisely
issues [7]. The function of the sensing spectrum, at the defined primitive operations [34]. In more details, since the
physical layer, is responsible for discovering spectrum different layers are ordered in the network structure, so each
opportunities over a frequency band and estimating the layer has to operate sequentially (e.g., a lower layer has to
probability of interference with PU [29]. wait until a higher layer has finished processing) [103]. This
As mentioned above, the layers in network stack need to sequential operation results in computational overhead and
include the advanced functions and services at each layer latency. With the layers being is isolated, components in one
to allow MCRNs devices to operate. Therefore, the tradi- layer may not be able to access information in other layers.
tional network is unable to meet the challenges of MCRN The wireless-link issues and opportunist conduct of CR are
routing due to resource constraints. To emphasize, the routing the initial impulses for cross-layer design in MCRNs. There
protocol in MCRNs needs to update its information about the are several problems in MCRNs cannot be solved by handling
free channels to avoid the interference with the PU, and that it by the traditional wireless communication [19], [107].
can create a stable path routing. Consequently, the cross-layer In MCRNs, the limitation of discovering channel diversity
design is emerging as a method for interaction and sharing the at the network and rapid change in spectrum resource, that
layer’s parameters to adapt the layer’s protocols operations it makes participation the physical layer parameters about
in MCRNs environment. Thus, through applying cross-layer spectrum sensing a legal requirement [69]. For instance,
design, routing protocol algorithm at the network, layers can sharing spectrum sensing with MAC protocol might supply to
be sharing different parameters, i.e., (channel availability), select the best channel, and that can contribute to increasing
for more efficiently and effectively routing decision to select data rate is an assignment for a particular link [34]. The
the path and the channel [33]. Thus, the routing algorithm decisions at the network layers’ stack must be considered and
performance can be more significant by: accounted for the parameters of the lower-layers [101]. Thus,
• Choose best an idle channel, More link life. the link-layer optimisation services will affect positively on
• Avoid the interference probability with PU. other layers performance such a network layer [101].
• Higher throughput, less delay. Consequently, the collaboration among the elements in
• Stability the route between the source node and the different layers is poor [3]. Thus, the performance of
destination. the entire system may be not enhanced due to the lack of
• The highest overall network performance. global information exchange, and that is contradicted with
Hence, the cross-layer design can offer a lot of active the requirements of MCRNs. In short, MCRNs need to share
elements for routing algorithm, especially in case of mobility spectrum sensing the information between the layers in the
and time diversity [27], [107]. In more specific, there are fast and effective way and that due to the dynamically of
remarkable differences between the traditional network and spectrum in MCRNs to identify the spectrum hole and a avoid
the cross-layer network [7], [30]. In the traditional network, PU appearance.
the nodes cannot share and enjoy the parameters of neighbour
layers, because every layer cares about its neighbours [7]. E. CHALLENGES OF CROSS-LAYER IN COGNITIVE
In contrast, the cross-layer provides a bridge to take part RADIO ROUTING
and utilise to serve the interests of other layers [7]. Network Generally, the cross-layer design also has some challenges
performance always needs to be enhanced with expansive and issues which are inevitable due to the nature of the
applications [3]. The tradition network, however, cannot characteristics of these design [33]. As the following:
provide high performance for the limited resource. Thereby, • Coexistence of cross-layer designs: It is a challenge
the inclusion of the available resources of each layer with to coexistence different wireless communication [33].
other layers can achieve higher network effectiveness [30]. In other words, for wireless networking using different
The energy consumption sector also plays as a critical key technologies, it needs to find a common language for
in term of network resources [102]. In more, Higher energy communication [23]. The cross-layer design can be
consumption in the term (per bit) refers to the total energy the universal language to standardization and adapt the

VOLUME 8, 2020 67839

 Q. M. Salih et al.: Smart Routing Management Framework Exploiting Dynamic Data Resources

characteristic heterogeneous of wireless networks [23]. time-variant impact and the limited routing tables information
On the other side, the idea of coexistence different that save only the next-hop. Hence, cross-layer architecture is
cross-layer design also can execute through finding a needed to assist routing protocol for right channel selection
chance for standardization of the mechanism of building decision. On the other side, there is not much research that
the cross-layer. Therefore, To create a state of coexis- has highlighted the routing protocol from the perspective
tence between different cross-layer design, that requires of routing design methods. As shown in 3, we have been
to set the standardization rules [33]. Thus, to enable taxonomy the routing protocol based on design methods,
different cross-layer environment technology to coexis- namely, the Non-Cross-layer Routing Protocol (NCLRP) for
tence needs to take into account to put the standardiza- Proactive protocol, Reactive protocol, and Hybrid protocol,
tion for the interface of communication between them. respectively. Another taxonomy is called Cross-layer Routing
• Cross-layer signalling: To interchange the services and Protocol (CLRP), which combines the networks layer with
information of cross-layer between the nodes in wireless other layers. Meanwhile, machine-learning in this survey has
communication, the signalling operations need to be gotten much respect through exploring smart routing protocol
considered [95]. The signalling inside cross-layer has in MCRNs. This taxonomy can contribute to an overview
to control and the manner of exchanges network infor- of the routing protocol in MCRNs and the knowledge of
mation inside the cross-layer model [30], [95]. Thus, weaknesses and strengths. Moreover, it provides a chance to
cross-layer has to set the control to operation exchange discover new research areas for MCRNs routing protocol that
information in a wireless network. Consequently, one of it needs more contributions.
the most critical issues that cross-design has to solve is
signalling. A. NON-CROSS-LAYER-ROUTING PROTOCOL (N-CLRP)
• Universal cross-layer design: Numerous applications Routing is a transfer packet across the network from source to
have inspired the cross-layer that emphasis on a partic- destination [2]. Cognitive radio routing is various from tradi-
ular application, e.g., audio, video, protected connec- tional routing in wireless Ad-hoc [22]. In MCRNs, the routing
tions [7], [30]. Nevertheless, the cross-layer model that algorithm poses several from critical challenges due to spec-
has designed for a single application or a group is not trum diversity and PU activities. Differently, in the tradi-
necessary to be suitable for different applications. One tional wireless network, e.g., Ad-hoc network, the routing
of the principal problems in the cross-layer design is the algorithm operates with a fixed licensed spectrum frequency,
absence of cross-layer design that automatically accli- e.g., 2.4 GHz and 5 GHz. On the other hand, the routing
mate with various applications. The number of appli- algorithm, in MCRNs, operates with a license and unlicensed
cations is continuously increasing, which call to design spectrum [8]. In non-cross-layer design, the network stack
the universal cross-layer [95]. Creating a universal cross- layers cannot violate the others layer. From that, the routing
layer, which can deal with multiple applications, is a real algorithm strove to find innovative solutions to overcome
dilemma. cognitive routing challenges. Hence, based on type classifi-
In more specific, all these problems that are referred to in cation, routing protocols are categorized into proactive, reac-
the cross-layer can be negatively affected in routing perfor- tive, and hybrid [8], [82].
mance. For that, it is necessary to address these challenges
during the design of the cross-layer model [22]. The signif- 1) PROACTIVE ON N-CLRP
icant optimizations of cross-layer could cause struggles in a In the proactive routing protocol, such as OLSR and DSDV,
layer. Thus, when the layered design is a breakdown, some per node recurrently changes the link-information with other
interactions are not quickly expected [35]. In short, most of nodes [5]. In more, the proactive protocol always maintains
these challenges were taken seriously during the design of our information about paths with other neighbours’ nodes [5].
cross-layer routing framework. Thus, the proactive protocol always keeps ongoing updating
to eschew the stale paths. For that, SU can always obtain
III. TAXONOMY OF ROUTING PROTOCOLS IN MOBILE path information meanwhile demanded by solely looking in
COGNITIVE RADIO NETWORKS the routing table [36]. In contrast, the higher overhead and
The cognitive radio technology presents an efficient design bandwidth waste are considered one of the main flaws and
for utilising the available spectrum. However, it also intro- sins in the proactive routing protocol. In this case, the large
duces new challenging problems which are not present network size and high mobility of SU can lead to an increase
in traditional wireless networks, especially the changing in routing overhead.
channel availability over time. In more, the routing protocol The author in [37] presented a path-centric spec-
in MCRNs is a complex mission because of the dynamic trum assignment framework (Cog-Net). This scheme was
spectrum access, node mobility, and PU interference. There- addressed the challenge of how to incorporate path discovery
fore, one of the critical routing design is how SU should take with channel resolution to reduce delay the channel
decisions about which channel they will use and at which switching. The aim of channel switching in a wireless
time to enable SU communication while avoiding harm to multi-channel network is to minimise the collision of data-
PU. This problem becomes much more complicated under the packet with the neighbours’ nodes by using a similar channel,

67840 VOLUME 8, 2020

Q. M. Salih et al.: Smart Routing Management Framework Exploiting Dynamic Data Resources

FIGURE 3. Routing protocol taxonomy in MCRNs.

and that can increase the rate of throughput. More in details, strategies for solving these challenges into two stages. First,
the Cog-Net model is also considered the structure of a it specifies a weight in the term of estimate the delay of trans-
multi-layer graph network at each CU. In more, each layer mission for every link that regards the channel bandwidth,
refers to one channel, and a provided CU can show in all availability of channel, and the quantity of channel avail-
the layer stack as a sub-SU or a peak point. In more, vertical ability for every link. Second, it estimates the minimum rate
edges amongst the sub-nodes of SU, that connected through delay for the end-to-end connection. The results explained
the exact SU, oppose for the ability to forward the data that the proposed routing algorithm for CR could find a
amongst the various channels for the node. Likewise, other lighter path delay compared to classical routing algorithms.
SUs, which are accessible from an offered node using a In more, this algorithm assumes a CU is static, whereas
channel, are joined by the matching horizontal edges layers it recognizes the band of the free spectrum as a perpetual
in frequency. Hence, the vertical edges might have given a resource accessible during its activity. It is noted that it
weight that equivalent to the switching period on the spec- confirms such a presumption for traditional multi-channel
trum. In contrast, it sets the horizontal edge to provide a networks. In contrast, this work has not considered the time
weighted for spectrum access delay. However, this work has of spectrum availability. The time of channel switching and
dropped the overhead effect of channel reassignment once the transmission has a decisive impact on routing and network
channels prove to be unavailable because of the preemption connectivity.
by PU.
The author in [38] proposed a Smallest Delay Cognitive 2) REACTIVE ON N-CLRP
Routing (SDCR) that aims to reduce the delay of the end- The reactive (or on-demand) routing approach, such as
to-end connection. The SDCR can handle the challenges of ADOV, DSR, each SU source node establishes the route when
the dynamics of channel availability. They have divided the it needs to send packets to the target node. [5]. In more,

VOLUME 8, 2020 67841

 Q. M. Salih et al.: Smart Routing Management Framework Exploiting Dynamic Data Resources

SU sends a spate of Route REQuest (RREQ) packets for the to embrace all the spectrum probability areas. In more, spec-
neighbour’s nodes. After receiving that RREQ, The desti- trum management operation has been done by using various
nation node will send the Route REPly (RREP) packet [5]. slots for spectrum management for a gather of spectrum
The reactive routing profit is to decrease the consumption sensing information and delivery slots for forwarding the
of bandwidth and network overhead. Plus, it saves the node data packet. Plus, SU explores the paths through sharing an
memory. On the other hand, the drawback of reactive routing idle frequency and next-hop election in the sending periods.
is incurred a higher delay in route discovery [36]. In other words, SU transmits its RREQ to the forwarding area
The author in [39] proposed a routing scheme on-demand with regards to SU with the highest priority to receive this
for the Multi-Hop Single-transceiver Cognitive Radio control packet, while the irrelevant neighbours drop it. In this
network routing Protocol (MSCRP). This routing algorithm way, the routing overhead reduced, with increasing the range
focuses on the challenges of the dynamic spectrum resource, of spectrum coverage. On the other hand, although selection
absent of the common control channel, and reduce the delay next-hop depends on neighbor nodes’ location as a metric
of channel switching. A technique is intended to combine rather than using a dedicated control channel to exchange
the control information of protocol amongst SUs without spectrum status. In this case, this technique is inappropriate
the framework of routing. That it can increase the rate of with the nature of CR dynamicity.
throughput of every flow across attaining the optimal channel The author in [80] proposed Stable Routing (PSR) that
selection and improve the convergence rate of selecting an chooses the most suitable channel from the list of channels
optimal path. Thereby, the throughput wins enhance due to available for a pair of CUs. In more, each CU computes the
the proposition of the assignment channel can harmony a steadiness of its free channels relative to the activity of PU
load of the channel. Thus, depending on the result of delay to contact its relay nodes. Then, it attaches channels that
outline, channel utilization improvement is achieved. On the have a higher stability possibility to the neighbour table. The
other hand, permitting CUs to switch channels can appear of framework of channel assignment and routing was construct
deafness problem, where a pair of CUs are unable to connect based on the multilevel graph. Moreover, to identify the
due to they are sensing on different channels. surrounding neighbors, CU in advance exchanges a hello
The author in [40] introduced Cognitive Ad-hoc packet with them. Then, it gathers information about all chan-
On-demand Distance Vector (CAODV) routing based on nels to its every neighbor, and periodically calculate channel
modifying the AODV protocol to evade PU activity and availability in advance by a CU and saves in a table. On the
sharing path and spectrum. At the time of route establishment, downside, the complexity and inflexibility of the multilevel
CAODV excluded the channels which are exploited by PUs graph model make it not efficient for CRNs. Additionally,
from the process of discovery route. However, at the time of using hello-packet periodically to identify surrounding neigh-
forwarding, when the channels become occupied by the PUs, bors increases network overhead.
the CU neighbors invalidate all the routes that it used by that The author in [81] proposed Spectrum-Aware Anypath
channels through broadcast PU-RERR control message. The Routing (SAAR) for Multi-Hop networks. The uncertainty
shortest path was used as a nominee to choose among the spectrum and capricious transmission characteristics are both
fresh routes that tag lower hop-numbers. The drawback of metrics to predict and assess the quality of any route in
CAODV protocol has selected the channel randomly without MCRNs. Besides, extensively simulations have done to
considering the best idle channel availability results in poor compute the packet dropping ratio, delay of connection and
routing performance and PU interference. throughput. In a word, the SAAR algorithm can achieve the
The author in [41] presented Dual Diversity Cognitive objective of improving the multi-hop performance in MCRNs
Ad-hoc Routing Protocol (D2 CARP) is the modified version and efficient spectrum utilization. However, the network
of CAODV. The path and spectrum are individually exploited performance is suffering from high overhead due to the statis-
in CAODV, whereas D2 CARP join utilization of path and tical information size of the packet and the routing table
spectrum diversity for more effectively using spectrum in packet. Also, when the channel status alteration, the prede-
MCRNs. This method allows for CUs to move and commu- termined candidate and channels might be inaccessible and
nicate with other CUs node in dynamically way over several non-perfect solutions, that fetches an eloquent challenge.
paths and channels. Notwithstanding the above, D2 CARP The author in [31] suggested the Cognitive radio Routing
selected an idle channel arbitrarily without considering the Protocol (CROP), which focused on select a route that offers
channel estimation model to detect the best idle channel. high throughput between the two end nodes. The methods
Also, D2 CARP is suffering from a phenomenon of higher of Smart Spectrum Selection (SSS) and Succeeding Hop
network overhead. Selection (SHS) were promoted to allow for CUs in a single
The author in [79] presented the Dynamic Spectrum process to choose the unoccupied spectrum by the relay
Aware Routing (DSAR). The process of packet forwarding node. That it will make the path configuration procedure is
depends on gleaned monitor spectrum mobility and spectrum a simple process as well as decrease overhead of routing.
sensing information for CR network topology. The method However, the suggested solutions may not be able to find the
of spectrum selection and the next nodes coordination at shorter path, and this leads to an increased delay and energy-
network layer together with constrained geographical routing consuming. Besides, the author assumed that the MAC to be

67842 VOLUME 8, 2020

Q. M. Salih et al.: Smart Routing Management Framework Exploiting Dynamic Data Resources

ideal without a wrong alarm and missed detection and that the related nodes with the zone are governed and construed
not always possible in case of CU mobility and spectrum about the availability of a path.
diversity. The author [43] proposed a Spectrum-Tree based
The author in [82] proffered merge channel selection and On-Demand routing protocol (STOD-RP) to construction a
routing protocol name as (CSRP) in CRNs that is built tree routing protocol for every channel availability to facili-
based on AODV routing technicality. The mission of this tate a channel and path selection. This tree routing protocol
algorithm is to guarantee the stability of routing, channel diagnoses several problems, such as the fluctuation of channel
accessibility, and switching delay are used as the election availability, absence of a fixed common control channel,
metric. On a broader scope, the central control did not use and union of channel decision and route discovery. In more
as a unit to govern the spectrum distributed information for detail, this routing scheme is a hybrid between a proactive
overall CRNs. In this case, the knowledge of a free channel and reactive routing algorithm. Proactive routing is needed
depended on channel history to permit CU nodes to take part to keep an intra-channel routing tree that is recognised by
license spectrum hole. In more, when the target node receives adopting a single channel. Also, SU might be set in cases of
more than one RREQ packet over different channels and overlapping. Thus, SU can arrive at various channels which
paths, then the routing algorithm filters these routes through might become a portion of multiple routing trees.
measures the rate of delivery of the entire link agreeing to The author [44] introduced a joint venture between
available channel probability in each RREQ packet. As a combine routing and channel chosen scheme known as (MPP)
result, this process has a clear footprint in terms of reducing to meet bandwidth requirements of flows, and to overcome on
interference rate with PUs during a transmission time and the challenge of the dynamic spectrum and how to integrate of
also to obtain higher data delivery rates and lower time-delay. channel decision and route discovery. Base on the probability
However, the procedure of the channel selection based on of PU-CU interference, the bandwidth for each channel is
channel history is not an efficient way, especially for dynamic estimated. In more, every CU chooses a path according to
spectrum and PU/ CU mobility. the probability of matching the bandwidth requirements of its
The author [83] introduced a shared stability-based flows. Then, the CU has to approve the selected route. Mean-
routing, link scheduling, and channel assignment (SRLC) while, in the case of the path has no corresponding the band-
algorithm. Also, the author has been taken into consid- width requirements, so the CU requirements to append more
eration the impacts of SUs’ mobility and PUs’ vitality of channels to diminish the bottleneck links. On the downside,
on the link-span. In more detail, the principal goal in the channel selects based on the probability (i.e., history) of
the first part is to compute the link lifetime because the channel in a randomly way rather than regarding best
of the PUs’ vitality, SUs’ mobility, and relative speeds. history according to information to be transmitted.
The next hope selection also has been studied in terms The author [45] proposed a hybrid protocol knows as
of balancing the consumption of energy in the network. On-demand Cluster-based Hybrid Routing (OCHR), where
Lastly, based on the above mechanism, the SLR algorithm a proactive technique is used for intra-cluster routing,
selects the better neighbor with the increase transmission and reactive technique is used for inter-cluster routing.
ability and the better free channel with the reduce channel In more details, the first of all insert the mechanism of
switching cost. On the other hand, it has not to take into spectrum-aware clustering, that distributed CUs into clusters
account the effect of the probability distribution of the route rely on the availability of spectrum, the level of power, and
lifespan in MCRNs, which rely on many parameters like stability. Additionally, they have promoted a routing protocol
the density of nodes, PUs’ activities model, and connectivity to reduce the delay and fulfil an adequate rate for the delivery
of path. ratio, and that has a positive impact to supply steady and trust-
worthy routing in CRNs. In contrast, the author has not been
3) HYBRID ON N-CLRP considering the concept of dynamic spectrum availability,
The hybrid strategy blends the features of both proactive which has a meaningful impact on routing performance.
and interactive orientation schemes. It earns a stable achieve-
ment trade-off between proactive and interactive routing B. CROSS-LAYER ROUTING PROTOCOL
schemes in various network scenarios with various require- The standardisation of layered protocol stacks has permitted
ments [5], [6]. In more, the advantage of this strategy the rapid development of interoperable systems, but at the
can decrease the routing overhead, also boost routing same time limited the overall structure performance, due to
performance in a case adjacent node are more inclined the lack of coordination between the layers. In more specific,
to collaborate [6]. An example of a routing protocol of this problem is especially relevant for routing protocol in
the hybrid style is named Zone Routing Protocol (ZRP) CRNs due to the dynamic spectrum access, time-varying
which has a pre-defined zone centred at itself regard to the behaviour of channel, severe interference with PU, etc.,
hop-numbers [42]. In more details, when the nodes inside the which called for the modification of the layering paradigm.
zone, it can use the proactive routing protocols to save the For that, the cross-layer design has been found as a method
routing information. On the other hand, for the nodes are out for a changing of the layering paradigm. The main idea of
of the zone, it takes on reactive routing methods. In contrast, cross-layer design is to preserve the functions associated

VOLUME 8, 2020 67843

 Q. M. Salih et al.: Smart Routing Management Framework Exploiting Dynamic Data Resources

with the original layers but to allow coordination, interac-

tion and joint improvement of protocols that cross different
layers. Consequently, cross-layer optimisation approaches
attempt to dynamically match the requirements of routing
protocol in CRNs to maxima network performance. For that,
the cross-layer routing protocol is classifications by focusing
on the combination of two or more layers through join routing
algorithm at a network layer with other layers, e.g., physical,
MAC, etc., and to highlight the strengths and weakness.
Our classification also takes into account adopting of the
FIGURE 4. Messages interchange between network and application layer.
machine-learning method for the routing protocol in CRNs
to design the smart routing framework.
Further details for routing protocol taxonomy has these layers, to provide these layers with all necessary
discussed in the next subsections. information that assists routing to make the right deci-
sion to serve the main objectives concerning operate to
1) TWO-CROSS-LAYERS ROUTING PROTOCOL these layers.
In this section, the previous studies for routing protocol In contrast, the application layer can adapt its application
in CRNs are reviewing based on joining two layers such packet to meet the route resource [35]. For that, the term
as (combine MAC layer with the network layer, etc.), and of QoS routing algorithm concept can apply not only
discuss in details. In contrast, there are still some cross-layer from metric selection (e.g., bandwidth, delay, etc.), path
designs that have not addressed, such as (network layer computation, QoS state propagation, maintenance, scal-
with application layer) in any previous studies. Accordingly, ability [20], but also to answer the requirements other
we try to draw a relation map between these layers’ parame- layers. In a word, this design carries within it a mutual
ters with expected future direction for it. benefit for both segments.
1.1 Application and Network Cross-layers: Over and above that, the routing algorithm also needs
In general, the application layer functions need to to study the impact PUs activities impact the network
support different user’s applications such as (QoS, file layer and application layer [7]. The proposed framework
transfer, internet browsing, multimedia applications, will have an introduction to the white spectrum avail-
intra-refreshing rate, etc.) [46]. However, the adequate ability information in the network to meet the applica-
provision of QoS at the application layer is an extremely tion imperatives. The application layer calls information
challenging issue in CRNs, primarily due to the raise about the available resource and other services to adapt
dynamism of the networking conditions which are not its’ needs according to that [47]. In CRNs, the network
able to continually secure the supply of the required layer has acknowledgment about network resource that
resources [7]. For instance, the selection of a stable path it inherits it from the lower layers, which can help the
and higher lifetime channel availability can minimize upper layers. In that case, the application layer can
the distortion at the application. reconfigure its configuration base on that [35]. The
In more specific, the observed decrease of the QoS at network characteristics can provide information such as
the application layer by SUs may restrict the chance mobility, power, and network lifetime.
of success CR technologies. Again, a new cognitive Really, the responsibility is not lies solely with the layer
radio module can be designed that can interface to the of the directive, but also the application layer has a key
other layers’ protocols. Hence, a right cross-layer design role to play in improving the functioning of the routing
can create a new interface between the lower layers protocol. Thus, cross-layer design can hold a new part-
and the upper layers variables to optimization the target nership between the network layer and application layer
applications. A formulation cross-layer model that joins to serve the work of each layer.
the application layer with the network layer has to be 1.2 Transport and Network Cross-layer:
taking into account various factors which have a real In general, the transport layer deals with TCP and
impact such as scalable, path selection, channel selec- UDP protocols, also provides different functions and
tion, throughput, and other elements. Fig 4 shows a services such as round-trip time, receiver window, most
general cross-layer framework via cooperation between magnificent transmission unit, congestion window, rate
the network layers with the application layer of packets lost and throughput [48]. On the other hand,
According to our acknowledgements, although many the transport layer is suffering from adaptive with CR
research papers have tried to meet the user require- technique. In more detail, as an example of TCP, TCP
ments at the application layer in CRNs, still this gap of is not suitable meanwhile applied in mobile networks,
co-operating the application layer with the network layer multi-hop in CRNs due to the fluctuated channels with
is a challenge. For the sake of clarity, Fig. 4 demonstrates variable characteristics [9]. In a word, the TCP is
the types of services that can be exchanged between slow to adapt to the rapid frequency change, interim

67844 VOLUME 8, 2020

Q. M. Salih et al.: Smart Routing Management Framework Exploiting Dynamic Data Resources

disconnections because of PU activity and mobility of The author [84] introduced a cross-layer to join between
spectrum, and that may result in packet loss and delay. routing protocol (Network layer) with spectrum infor-
In fact, few research papers have addressed the transport mation (Mac layer) for ad-hoc mobile in CRNs. The aim
layer issues but shyly [9]. is to mitigate the frequency channel switching, because
The author [48] called to design The Cognitive Radio a re-routing process in CRNs is expensive, regarding
Transport Layer (CTRL) that aims to provide the chart energy, delay, and throughput. Therefore, to reduce
between the QoS at the application layer and down channel switching, the investigator presented a smart
layers stack and to breed a connection of transport selection way by using cross-layer-based on cognitive
layer. In more, the CRTL is in charge of treating with radio routing (CLC-routing) protocol. The CLC-routing
various packet data context and the requirements of QoS can obtain information about channel availability from
between applications and control layer management in the MAC layer and chose the next-hop node. Thus,
cognitive radio. the network layer can select the next hop according to
Also, the CRTL executes QoS filtering to aid with the the list idle channels. The advantage of this way, it can be
routing of application data with diverse QoS demands reducing the rerouting frequency, data packet collisions,
to pre-define contexts, administer the flow of packets to a maximum average of throughput, and increase the
lower layers, and to supply utilization to context creation lifetime of the network. However, the author has not
and modification. Anyway, to enable CRTL services that analyzed the delay of channel access, which defines as
it needs to implement the cross-layer design to offer it the average time spent by a data packet in the MAC
with the necessary data to activate the services that can queue, respectively. It can affect the channel selection
adapt with CR. process.
Indeed, there are no real attempts to use cross-layer The author [19] introduced cross-layer-based routing
that join between network-layer with the transport layer layer solution that intends to reduce channel switching
to serve the functions of them. In more, the problems frequently. The routing algorithm has been taken into
at the transport layer in CRNs are related to sensing account that it is a better way to save the costs of energy,
operations [12]. At the time a middle node on the path delay, and throughput through select a route in such
is employed in spectrum sensing, it will not be able to a way that requires less channel switching. In more,
forward packets [75]. In more specific, the routing path the CU requires specifying a link-failure result in
cut-off while sensing is virtual, i.e., the route itself will spectrum mobility from a node failure for better perfor-
resume as soon as the sensing is complete. Hence, the SU mance. Also, the author has design to channel selec-
source has to reduce the transmission rate to an ideal tion methods that called smart- selection and random
rate to block the overflow of buffer in the intermediate selection to explain that the rate of channel switching
nodes, rather than ultimately stop the path connection process can reduce considerably where insert channel
as usual in classical ad hoc networks [12], [49]. For selection parameters in routing. In short, implemented
that, the integration between the network layer param- smart-selection in cross-layer routing has a particular
eters with transport with CR routing protocol can create resonance on routing performance rather than random
a successful model and enhance the transport layer selection. On the downside, this work did not consider
protocol. the channel access delay due to lower channel access
1.3 Network and Link (MAC) Cross-layers: time, it is better for MAC and routing performance.
The function of MAC in CRNs is to define the available The author [57] adopted a cross-layer approach to link
spectrum resource during spectrum sensing, determine the control of power, the assignment of the channel,
the optimum sensing and transmission times, and orga- and route selection in a bounded optimisation frame-
nize spectrum access, and channel assignment with the work. The goal of this way is to reduce interference
other users [29], [50]. between SU and PU, and for proving a routing session
The author [56] proposed a cross-layer design between in CR based on the constraint an average data rate of
data-link (MAC) and network layer. The aim is to reduce the end-to-end connection for SU. Further, the routing
wastage resources used by the packet in their previously approach does not only depend on the reducing of the
hops. For that, the resource allocation, at the link-layer, interference probability between SU and PU but also
is accounting by the insert of hop-count information the spectrum sensing values. The resulting explained
that obtains from the network layer. So, in cased of a significant decrease in mean interference and higher
a packet is missing at the maximal hop-count, this performance. Nevertheless, the spectrum sensing model
outcome has represented the wastage of resources in the is not practical since it did not examine the fading nature
earlier hops. Thus, to surmount those obstacles, after of the spectrum sensing channels.
channel reservations to transfer the packets, he power is The author [58] introduced a cross-layer routing based
distributed between packet and transmission through the on joining Rate adaptation, Channel Assignment, and
most suitable available channel without any regression Routing, which called (J-RCR). The J-RCR protocol
throughput. aims to increase social welfare through the exploitation

VOLUME 8, 2020 67845

 Q. M. Salih et al.: Smart Routing Management Framework Exploiting Dynamic Data Resources

of spectrum resources in multi-channel multi-hop 1.4 Network and Physical Cross-layers:

CRAHNs. At link-layer, the routing algorithm combines In CRNs, SU is responsible for discovering the trans-
the data rate average for each time-slot and specifies mission range of PU and trying to avoid interference
for all link-channels along the route in case it arrives a with it [56]. For that, SU has to be intelligent for sensing
new flow or any PU activities. For the guarantee rate the spectrum to prevent interference with PU transmis-
requirements, the virtual queue model was applied in sion. Furthermore, the process of spectrum sensing can
the routing algorithm. Furthermore, the J-RCR routing provide more spectrum access opportunities. The phys-
algorithm not only has been considering for the level ical layer in CRNs can supply these requirements. Thus,
of workload and spacing amidst relay and target CU the physical layer manages the operation of spectrum
nodes but also has implemented the interference due to sensing, detecting the PUs, and assessing the quality of
co-channel. Also, a new routing metric was proposed available channels [7].
that it relies on service price per packet, which combine The author [62] proposed a cross-layer Routing Protocol
queue backlog and distance to the target node. The (CLRP). The purpose of CLRP is to increase the
output result shows that the J-RCR reveals improve- throughput and to build a stable path and increment the
ments in throughput rate, path stability, and reduced possibility of detecting an end-to-end route. In another
delay. Although the routing algorithm has succeeded in way, through channel sensing, each SU preserves a list
joining the channel and routing decision and reducing of observing channels that specify the probability of
the probability of interference with PU, it did not availability with 0 and 1. A common control channel
consider the channel/link characteristics. In more, did the signaling of the control routine. The author
the operation of selecting an idle channel based is supposed that the relay nodes (i.e., in-between nodes)
unlearned, without also taking into consideration any might observe and sense a group of selection the ideal
factor to measure the quality of the channel. channel after setting the path. On the downside, the other
The author [59] presented a cross-layer, that is to say, channels that are not sensing are considered as always
joint routing and channel assignment (J-SRCA) algo- available with a probability.
rithm for mobile cognitive radio Ad hoc networks. The author [63] suggested a cross-layer design to inte-
This schema is involving node mobility, PU and CU grate the the transmission-guide information of the phys-
co-channel interference, the burden of dealing with a ical layer and process coding at the network layer. This
channel and measure the space between the relay and schema is aim to assist the connection between multiple
target node. At the results level, the J-SRCA signif- a pair of cognitive users, that permits to use of the
icantly enhance the performances, especially in the spectrum resources. Also, the interference rate is less
case of higher PU density and mobility. In any case, with primary licensees. Therefore, the author employs
the J-SRCA algorithm has not been taken into account network coding at between nodes to associate the
the factor of a lifetime to find a more stable link, packet received prior re-sending, which gains reliability,
especially in node mobility. Nonetheless, it has not increase in throughput, and more bandwidth utilisation.
considered the influence of node density on the routing In contrast, underlay mode habitually runs with lower
performance. power which leads to reduce data rate transmission.
The author [60] proposed the distributed joint dynamic the author [64] offered undercover, a cross-layer routing
Routing and Channel Allocation (RCA) algorithm by protocol that connects the physical layer services with
using a cross-layer design. RCA algorithm has been routing algorithm at the network layer. In more, under-
designed based on a join the link-layer (MAC) and cover employs collaborative groups with applies beam-
a network as one optimization challenge. In practical, forming to transmit data, typically albeit PU actives
the MAC services such as (dynamic spectrum allo- on the same channel. In that case, this property can
cation, scheduling, and transmit power control) has improve the packet delivery rate for undercover than
joined with the routing algorithm to achieve the aims other protocols. Expressively, the capability to transfer
of the routing and to meet the QoS requirements. packets concurrently with PU over the same channel
In addition, this scheme aspires to build a state of can provide a new level of freedom that was never
balance between the functionalities of MAC and routing. available before. In order to evaluate the link-quality,
Hence, the utility function was called to increase the more than one metric has been used between the
link-capacity between CU nodes. In more, the process next hop and relay node. Thus, the diverse values for
of path selection from the source to the target node the routing metric have been computed, according to
relies on the Traveling Salesman Problem (TSP) to find different groups, which can facilitate the carriage from
the shortest path to the destination node. The output the neighbor’s node to the next hop. The output results
results of the RCA protocol observed attains a higher clarify enhancement of goodput rate of up to 250%
throughput and packet delivery ratio. On the other hand, contrast to other routing protocols, with the lowest over-
RCA is suffering from an increase in rates of high end head. Even so, allowing SUs to use the licensed chan-
to end latency. nels with PUs at the same time that it enforces to use

67846 VOLUME 8, 2020

Q. M. Salih et al.: Smart Routing Management Framework Exploiting Dynamic Data Resources

low-power transmissions, which leads to a decline data resource [18]. Thus, every layer of these three layers,
rates. Application, Transport, and Network, has different
parameters, and combining these layers or parameters
2) THREE-CROSS-LAYERS ROUTING PROTOCOL can be more practical [7]. In more, the cross-layer
The integration between three layers have been explained in mechanism joins the transport-layer with the applica-
this part, with presented many previous studies to show what tion and network layer to solve a particular issue. The
is the features and disadvantages of participating between transport layer can handle various data-packets contexts
three layers in different ways in CRNs. and perform operations of QoS. It can be filtering
2.1 Application-Transport and Network Cross-layers: the requirements of the application according to the
The QoS is regarded as an essential and sensitive point QoS-routing resources. Also, it manages the stream of
at the application layer than at other network layers. IP packets to lower layers and supplies help for setting
Besides, CRNs-based services for SU would have a making and alteration [30]. Thus, it is the fact that the
severely lower QoS than radio services that enjoy secure application layer, transport layer, and network layer are
spectrum access [47]. At the application layer, if QoS all interdependent, and the combination among these
is not prudently considered in MCRNs, the perceived three layers can enhancement the QoS, and help the
decrease in QoS related to the MCRNs could obstruct the routing be more efficient to support the application
success of MCRNs technology. Therefore, to improve layer.
the quality of service, the upper layers of the stack layer 2.2 Transport-Network and MAC Cross-layers
need to share and aware information [60]. Using an efficient transport protocol in CR is essential
The transport layer is accountable for controlling the to send the data in a reliable manner [63]. Conges-
transmission data rate of per link to ensure the QoS tion control and reliability are provided by a transport
requirement and to prohibit a rapid row rate from the protocol when the congestion control is imperative to
restrained channel rate [61]. In more, there are many avert congestion by arrangement, the rate of transmis-
factors in a wireless network that can be an effect on sion, or by employing other strategies [30]. Reliability
QoS [60]. For instance, the transmission error average is anxious about providing steady and mistake-free data
(e.g., package loss once using a TCP) is considered a transmission. Once a data transmission incident takes
severe problem that can solve by using the cross-layer place, the data sender should be able to affirm that the
design environment [33]. Cross-layer design can signif- receiver has to receive the data correctly.
icantly achieve the QoS requirements on multi-hop The protocols in the network layers’ stack need to
networks and improve system performance [18]. enhancement to accommodate the extra functionalities
Choosing a high-quality path for real-time applications of MCRNs. The phenomenon of switching, congestion,
is considered one of the critical rules for QoS at the and queuing products a delay at each of the relay nodes
network layer [62]. By the same token, delivery paths for along with a communication influx, also, the interfer-
streams are defined by using the information resource of ence due to the spectrum fluctuating in the surrounding
network availability as well the QoS necessities of iden- environment [64]. For that, these problems are called
tical flows [18]. In more specialized, real-time applica- the cross-layer design in order to combine MAC layer
tions for SU are considered as a natural judge to test the scheduling with the network layer [64]. For this reason,
routing performance for MCRNs. In wireless networks, the joint between MAC parameters such as (coordi-
it has developed a traditional approach to support QoS nate dynamic spectrum access, cooperate in sensing the
in CRNs [20]. Pragmatically, new multi-aim metrics spectrum, etc.) and network layer by cross-layer design
have been joined route fastness with other QoS metrics can ease off the routing issues such as switching the
such as (delay, bandwidth, channel selection, channel channel [7], [65].
quality, etc.), candidate as essential metrics for QoS Undoubtedly, combining the transport layer with the
in MCRNs. In more detail and point out, cooperation MAC and network layer can be more imperative
these metrics with multipath routing metrics for SU is to according to essential services that provide by every
choose backup routes or route simultaneously on various layer. In more detail, join the transport layer with the
routes. Also, it can supply a valuable framework for MAC layer and network in MCRNs can be more signif-
real-time applications. icant due to it can improve the higher layer functions
As far as we know, this combination between these through the incorporation of lower-layer services such
layers yet is not addressed in any previous research as spectrum access, spectrum mobility, routing deci-
practically. Without a doubt, a mix between these sion, and congestion-free end-to-end reliability [50].
layers’ parameters via a cross-layer intends to enhance In contrast, these attributes can be located in the
the performance of a system by combining the layers infrastructure networks at the centre of CRNs control.
protocol [7], [49]. In the same way, the outputs are flexi- However, the CRNs distributed systems network needs
bility in sharing the information between network layers to smart collaboration between different protocols in the
and provide better QoS for network dynamic and limited network layers’ stack.

VOLUME 8, 2020 67847

 Q. M. Salih et al.: Smart Routing Management Framework Exploiting Dynamic Data Resources

2.3 Network-Link (MAC) and Physical Cross-layers sequential Bayesian estimation scheme where each SU
It is attracting the attention of some researchers estimates the probability of the channel availability for
processes of combining the lower layer with the network each frequency band.
layer due to its positive results on the routing perfor- Moreover, designing a reward metric corresponding to
mance in MCRNs. On the other hand, CR technique the channel availability probability and their capacity,
imposes the lower layers also to update its structure respectively. A reward metric is used by the MAC layer
to adapt with new features offered by CR such as to select the channel that the link between two SUs,
dynamic spectrum access, multi-channel. Definitely, which has the most significant reward metric. At the
joining lower layers’ parameters with routing algorithm, network layer, its candidates the first-best-seek routing
especial in routing decision, can represent compelling protocol that chooses the relay node in the area of
solutions. the forwarding packet to the destination node. As a
The author [85] realized a cross-layer routing and result, the numerical results explained the trade-off PUs
dynamic spectrum strategy for CRNs, which knows as collection and increased the successful packet transmis-
Routing and Spectrum Allocation Algorithm (ROSA). sions rate. In contrast, the Best-first-search routing algo-
For the routing metric, it is designed based on spectrum rithm consumes a large amount of memory and network
utility and spectrum holes. In more, the ROSA algo- resources, since it has to keep the tree of backtracking in
rithm aims to increase the network throughput and that its memory.
through sharing routing, customize the dynamic spec- The author [87] Cross Layered Opportunistic Routing
trum, scheduling, and control on transmit power, respec- Protocol (CLORP) that joins between physical layers
tively.It can deal with the dynamical spectrum resources spectrum sensing, MAC layers opportunistic link
to maximizing the link-capacity without harmful inter- discovery, and network layers opportunistic data trans-
ference to other users. Moreover, it can exploit the mission. In more detail, the energy detection (ED)
weighted sum of difference backlogs to steady the method at the physical layer has used to detect the
system by granting precedence to excellent link-capacity idle channels that allow the CUs to transmit the data
with a high differential backlog. However, the ROSA over it. Moreover, in order to gauge the link-quality for
algorithm has a drawback in its schema. Equally, selecting, the Error vector magnitude (EVM) method,
the routing algorithm is suffering by default from the at the MAC layer, was considered. The CLORP has
problem of competes for the low economical cost route extended the route request structure, which is called
to the destination node, and the policies of model cognitive radio route request (CRRREQ), that included
attempt to employ links that maximize queue differential the available free channels. Meanwhile, the metric of
backlog. Thus, these two procedures could probably be path selection was designed based on the minimum
in inconsistency in the calculated paths that point for length CRRREQ to the target node, including selecting
directing the packets from source SU to destination. As a the spectrum occasion at each SU node.
matter of fact, this can lead to an outcome in routing More clearly, the CLOPR algorithm has been focused on
loops that have a terrible effect on the end-to-end delay discovering an optimal route across opportunistic link
of delivered packets and on the end-to-end delay of discovery with employs the most significant spectrum
delivered packets and at a critical state, the situation opportunity (SOP) toward each hop, including the prob-
could have a negative influence on network throughput ability of delivery in a secure way. The SOP obtains
due to continued interference by the extreme looping its information through the messages exchanged of a
effect. Over and above, the author also suggested that cross-layer model based on the layer spectrum sensing
routing and scheduling can be accomplished through function (PSSF). By looking deeply, the author has
a single interface with the presumption of CSMA-CA kept on reducing the number of control packets through
MAC design at the lower layer. Notwithstanding, this merging SOP information inside the RREQ packet,
presumption is not considered the deafness problem. which called CRRREQ. For that, the phenomenon of
The author [86] used cross-layer design for a routing link failure has been avoided by using the informa-
protocol in CRNs, through join sensing with MAC and tion of the best link (BL) and the SOP. This infor-
network layer. The physical layer function is to process mation has already included inside CRRREQ rather
the data and control frames into identical bit-streams than using PU-RERR message as in CAODV. Even so,
and directs them to the data (if the MAC allocates using the CRRREQ as a rescue letter to avoid PU inter-
the channel) and the control channels. The cross-layer fering is not a practical solution. In more, the mobility
combine between MAC and network, where the sensing mode of SU always exposed to the risk of PU inter-
process already joins from the physical layer to the fering, which leads to more rescue letters. That means
MAC layer. This approach aims to evaluate performance more CRRREQ messages that including more than
by measure the average of packet sending successfully one control packet, which considers an increase in the
between the sender-receiver node and the average rate of volume of consumption of network sources and more
clash with the PUs. In particular, the author presented a overhead.

67848 VOLUME 8, 2020

Q. M. Salih et al.: Smart Routing Management Framework Exploiting Dynamic Data Resources

2.4 Application-Network and Physical Cross-layers: has the aim to consider the quality of service require-
Joining application parameters with the lower layers can ments and channel availability by presented a new
affect positively on acceptable for variation of delay, scheme jointly. From that, the mechanism of integra-
required throughput, delay tolerance, and fair packet loss tion of allocates channels and determines a path that
rate, etc. [22]. Similarly, the MCRNs has authorised can keep the QoS claims for the real and non-real-time
the dynamic control of necessary the parameters of a applications. These schemes offer to reduce the packet
physical layer such as (transmission power, modulation, loss rate for the applications of real-time and throughput
constellation size, etc.) [66]. For that, CR adaptives the for the applications of non-real-time, which is consid-
physical layer information to make an optimal deci- ered as evidence for the effectiveness of cross-layer
sion to obtain on maximise performance for QoS at the design-related to the QoS challenges in CRNs. In a word,
application layer.In more, application layer QoS can be the cross-layer will have a crucial role to play in solving
amelioration significantly if the intra-refreshing-rate is problems of QoS and meeting the growing demands of
modified together with parameters at low layers, such services.
as spectrum sensing at the physical layer [47]. There- 2.6 Transport-Network and Physical Cross-layers:
fore, enhancement of the QoS by the cross-layer design The parameter selection map between the layers must
amongst the application layer and physical has a sound. be according to the objectives to be achieved and the
Typically, one of the aims of the routing protocol is to challenges that must be overcome. Actuality, the deep
provide QoS that meet QoS at the application layer in understanding of the nature operating of layers can
MCRNs [67]. For that, to enable a QoS-aware routing provide refresh solutions. For instance, the transport
protocol in MCRNs that it has to take into consideration layer parameters, e.g., congestion control, adaptive
the delay-sensitive and bandwidth resource for hungry control, flow control when combining truly with lower
applications [68]. Likewise, combine, at the network layers, can be as an antibiotic for CR challenges. On the
layer, to its routing agent information about, for instance, contrary, the wrong building cross-layer design can
channel state information, PU activity, etc., whereas give fruitless results. In MCRNs, during the process
the application layer, e.g., define the type of applica- of sensing, SU cannot forward the packet [28]. Thus,
tions such as (audio, video, text. etc.). This information the sensing time needs to value and share with the trans-
pushes the routing protocol to make the right decision port layer to evade the retransmission error, which can be
about channel/link selection, which serving application a reason to influence overall routing performance [28].
layer requirements [7]. This area still needs more effort As well, at the transport layer, the congestion control
to discover the positive results that can be gained through of TCP can be enhanced through obtaining information
this design and expected problems. for the physical layer into consideration. For instance,
2.5 Application-Network and Link (MAC) Cross-layers: the knowledge of the physical layer can use to distin-
The synergy between the layers of the network must guish packet loss because of the PU interference or the
be consistent with the goals it needs to achieve [23]. bad link quality [70].
For instance, the service that it provides by the phys- In more details, much research about the transport layer
ical layer has a different effect, in contrast, when the has been focused on the enhancement of TCP protocol,
cross-layer design passes through the MAC layer [7]. increased throughput, reduce end-end delay [71]. For
For the QoS in CR, the application layer has to more simplicity, integration congestion control can
account and implement in the cross-layer design [47]. decrease the rate of packet loss and improve perfor-
Thus, to preserve the end-to-end QoS, it is compul- mance [71]. Hence, meanwhile creating transport proto-
sory to add conduct at the wireless link [55]. At the cols for CRNs, it should consider some factors such
MAC sub-layer, the parameters can facilitate to support as congestion control, lower-layer parameters, and loss
the routing protocol. For example, the MAC sublayer recovery into account for more spectrum-efficiency and
information from the channel location and channel optimization performance. However, still, the existing
assignment can use to promote the routing algorithm cross-layer has limitations due to many factors such
in MCRNs [33]. From that, it is evident that the as PU behavior, spectrum dynamicity. Nevertheless,
cross-layer between MAC and network can achieve the the challenges of design window-size that it has to take
routing process, whereas multiple license channels are features of CRNs and activity of PU/CU are open chal-
scheduling to the multi CU in a quest to meet QoS lenges [71]. In effect, improvements in the operating of
specifications [55]. the network layers will not only affect the target layer but
There is no much research that touches this type of also will take part in improving the process of the other
design. For the sake of simplicity, the study by [69] has layers. In fact, the cross-layer system is a series of related
assumed just as an illustrative example. The cross-layer events and actions, where the exchange of necessary
protocol has been designed based on the channel alloca- information will be a rescue bridge to overcome the
tion and routing with QoS requirements. This approach obstacles of CR challenges.

VOLUME 8, 2020 67849

 Q. M. Salih et al.: Smart Routing Management Framework Exploiting Dynamic Data Resources

3) FOUR-CROSS-LAYER ROUTING PROTOCOL are needed and change in channel quality [19]. The
Combining between different four-layer explained in this flexibility of cross-layer design could invest in service
section, with more reviewing and information about the to implement this design that could be included, for
design mechanism and how these cross-layer methods can instance, QoS, congestion control, scheduling, resource
boost the operations of network layers and to be aware of the allocation, routing, and the list goes on as long as these
CR technology requirements. layers can provide different services. This design is still
3.1 Application-Transport-Network and Link (MAC) unexplored in the field of research.
Cross-layers: 3.2 Transport-Network-Link (MAC) and Physical
The single-layer process, e.g., where the improvement Cross-layers:
of the performance is the target of the only a particular The author [74] devoted a distributed cognitive
layer, whereas the different layers operate as an assistant cross-layer design algorithm for multi-hop wireless
to supply the necessary information parameters, has not networks enhancement through merging the upper layer
enough high flexibility to adequately addressed to meet (transport layer) with the lower layers at (network, data-
the MCRNs demands [72]. At the same time, the steps link, physical). In more, the author explained that the
of the solution need to define explicit targets to build traditional network layer stack could not be a source
a model that matches the requirements of the current of inspiration for optimal performance for the wire-
stage. To enable unlicensed devices to send and receive less network. In other words, this method tried to
the data effectively that the cross-layer design routing solve the predicament of Network Utility Maximization
must be adapting for more efficient dynamic spectrum (NUM). In this, per CU source independently setting
access [74]. In more specific, integrated the data-link its rates, according to the current capacity of the link
and the network layer can reduce re-routing initiate, settled by the physical layer. The previous data from its
which increases the routing stability and minimises the neighbors define the link-layer scheduling, and routing
routing delay [19]. In contrast, the delay is considered has done by using the AODV protocol. In that case,
an essential metric for the QoS in MCRNs [20]. The The concurrent enhancement for the control through
QoS at a higher layer in MCRNs is a hard job to multi-protocol layers can motivate the cross-layer design
provide a significant QoS due to there is increasing in to gain higher throughput, support the capacity and
the average of the dynamism of the networking, and utilization of network, minimize power-consuming, and
that makes difficult to supply enough stable resource interference.
for QoS requirements [7]. Again, the QoS operations at In a word, cross-layer design can assure the aims of
the application layer can promote significantly by the the end-to-end data stream through wholly utilising
exploitation of management services that it provides by the CRNs resources. On the downside, this design has
the link layer [7]. inherited complexity. That means the higher network
A little research has been touch this type of design [30]. equipment resources are needed to allow this model.
However, it was implemented to solve the challenges Thus, replacing the layer with another layer can lead
of traditional wireless networks. Today, with the emer- to change in scenery of the process cross-layer design
gence of the MCRN challenges, that it requires applying and the output results. It is reasonable since every
the cross-layer design to overcome new challenges [7]. stack-layer has its parameters which affect negatively on
Pooled data between the transport-MAC layer is needful the performance.
because of the sensing process effects on transport-layer 3.3 Application-Network-Link (MAC) and Physical
protocols. [28]. In that case, the sensing-period has to Cross-layers:
be an accuracy account at the transport layer to save The relation between the data link layer and the physical
any unnecessary re-transmissions and packet loss on the layer has a close coupling to attain the performance and
route. For that, this process will call to join between to be ambidextrous to dealing with the dynamic spec-
the MAC and transport layer to reduce packet drop trum in MCRNs. From another perspective, the phys-
and re-transmissions process, which also to avoid the ical layer factors such as signal processing, modulation,
buffer overflow at an intermediate node [12]. In more, coding, can be imperative the network performance in
a cross-layer design is needed to interchange informa- the term of delay, throughput, packet loss rate, and inter-
tion between the network layer stack in a motivating ference. Thus, optimisation is achieved by cross-design
way and thus allowing to diminish the overhead [7]. combining physical layer and MAC layer.. Besides, link
A cross-layer environment between application, trans- error-rate and MAC protocol factors also can influ-
port, network, and data link layer, have to include the ence on the link’s usability, and that leads to re-routing
adaptive protocol, which should know the divergence in operations. For further clarification, the example below
the cognitive radio environment [73]. In particular, they meets the same cross-layer design as mentioned here.
should consider the traffic activity of PU in the adap- However, this work has not focused precisely on routing
tive protocol. Besides, SU transmission characteristics performance.

67850 VOLUME 8, 2020

Q. M. Salih et al.: Smart Routing Management Framework Exploiting Dynamic Data Resources

The author [105] proposed a cross-layer optimisation sharing also touches the physical layer. Meanwhile,
architecture to minimise the bit error rate, out-of-band the higher layers also share some tasks [12]. That means
interference, power usage, and to maximise the overall the pre-existing common elements between existing
throughput. Besides, the genetic algorithm has also layers will contribute to increasing the efficiency of
applied to the optimisation of distributed problems, layer protocols [76]. In the same way, the process of
where it depends on the methods of frequency allo- conjunction the upper layer (Application, Transport)
cation for distributed cognitive. In essence, the tradi- with the lower layer (Network, Physical) can improve
tional genetic algorithm is changed, which is a namely in several aspects, such as reduce the routing delay
serial sub-carrier-wise genetic algorithm that uses and increase the throughput [3], [47]. Besides, join
multi-objective fitness function. To further minimise the application layer and transport have an impact
out-of-band interference, partial quantisation within one on QoS [3].
transmission can be used. From that, there are four
network system parameters that it selected to optimise 4) FIVE-CROSS-LAYERS ROUTING PROTOCOL
through the recommended method, where the results This classification explains the cross-layer design between
are shown that the proposed method can simultaneously the fives layer. It is still this classification is not vastly
decrease the bit error rate and the out-of-band interfer- employed, but from the previous studies can obtain some
ence while maximising the overall throughput. For the evidence to support this suggestion.
routing protocol, the author called that routing stability, 4.1 Application-Transport-Network-Link (MAC) and Phys-
and link availability relies on the BER and locations of ical Cross-layers:
nodes, and the number of hops and congestion over a To exploit interaction bonds between five layers’ param-
route located by the delay value for the transmission eters, that requires finding a language of dialogue among
application. layers functions to accommodate the new tasks entrusted
3.4 Application-Transport-Network and physical to it [30]. Admittedly, designing a model that coop-
Cross-layers: erate with five layers of services needs to drastic the
For stack-layers, each layer carries on different critical changes inside each stack-layer [33]. That means that
parameters inside it, which can be harnessed to serve and the process of synthesis between layers’ protocols leads
optimise other layers’ operations [30]. The transform to receives new inputs [102]. For instance, joining
from the traditional wireless network to cognitive radio channel availability with the routing decision requires
has called for the birth of some new concepts [75]. For the routing algorithm to adapt and coexistence with new
instance, spectrum sensing is an old concept which has facts. Indeed, coupling two layers or more need always
already applied with the traditional wireless network. to prepare that target layer to digest new inputs and
Nevertheless, with the appearance of MCRNs, that it deals smoothly with it. Undoubtedly, the expansion of
invites to develop the spectrum sensing operation to cross-layer to join different layers will increase the capa-
be able to discover the spectrum hole and to void PUs bilities of layer protocols to deal with more challenges
activity [12]. At the same, the cross-layer design also has in a practical way [3]. However, it will contribute to
been developed by building new relationships between growing the complexity of the cross-layer model and
layers parameters. consumption of network sources.
Practically speaking, the process of direct connect
among the parameters of each layer allows for those
parameters at any layer to be noticeable to other layers
at run-time. The five cross-layer design has been inves-
tigated, which proposed by [3]. It is mentioned just
as an example for clarifying the concept design of
a cross-layer over five-layers. In contrast, the routing
problem is considered a sub-problem. In more infor-
mation, the vertical calibration framework was used
as a model for five cross-layer design, which took
care of the network objectives and constraints imposed
FIGURE 5. Spectrum management process. the lower layers to the higher layers. Accordingly,
the author has translated the generalised utility optimi-
For more understanding, the function of layers joining sation problem (GUOP) mathematically in CRNs for
can be done in different stages. For instance, Fig. 5 covering various layers starting from the physical layer
elucidates that although spectrum sensing is considered to the application layer.
as the primary task for a physical layer, the link-layer These constraints have been represented in the access
also has been sharing this process, whereas the spectrum management components from the power of access

VOLUME 8, 2020 67851

 Q. M. Salih et al.: Smart Routing Management Framework Exploiting Dynamic Data Resources

control, frequency functions, choose the routing algo-

rithm, flow control to ensure QoS through incorrect
mixed non-linear programming. For that, the optimisa-
tion cross-layer design has been used that be applied by
the vertical decomposition (COVD) pattern to separate
objectives and constraints. For more simplicity, CRN
multi-hop networking problems have decomposed into
several subs-problems. In other words, the optimisa-
tion process can pass through the optimal spectrum,
scheduling with dual-factor (OPT2), multiple (OPT3)
protocols at the network layer, and engineering traffic
with QoS assurance methods (OPT4). Furthermore,
the complexity of computation operations of optimi-
sation a cross-layer design was reduced by using the
proposed heuristic algorithm (COHA). Actually, this
work has been focused on the problem of improving
FIGURE 6. Cross-layer cooperative parameters over five layers.
the utility optimisation of the network. However, there
are various restrictions, from the lower layers to
upper-layers functions, should take into account.
In [3] has introduced different cross-layer models In the same context, the recently proposed protocol was
that it partially satisfies the concept of the design in introduced by [76], which is called Mobility Adaptive
this section, which based on computing in a repeti- Cross-layer Routing (MACRO) protocol. The MACRO
tive loop among the layers with flowing-back services depended on the concept of cross-layer intercommunica-
and information and forth execution different tasks. tion strategy across five-layers from the physical layer
In essence, the map of interactions divided into network- to the application layer. MACRO seeks to provide the
physical, application- network, and transport-physical. dependability of a path by considering the variations
With a simple scenario, based on the possible next-hop topology and channel conditions like node congestion,
map, the resource spectrum has been allocated by failures and to harnesses knowledge like the average
the PHY-MAC algorithm, and after that, the routing mobility and RSSI over multiple-layers. Similar to the
table has founded at each CU user benefiting from AODV protocol, it applied the route discovery tech-
combining the network-physical layers resource. With nique to decrease the number of deluge by trapping
more evidence, the author mentioned that further knowl- the group of CUs mobile ready to answer the requests.
edge underpass requires between application-network Also, it selected the most secure paths for forwarding
that to make it possible for the routing algorithm can packets based on mobility of CU and the link-quality.
satisfy the QoS requirement. In the end, a new road Simulation outcomes revealed that the MACRO affords
within the transport and the physical layer are demanded significant enhancements in terms of the packet delivery
to achieve the traffic balance and power regulation ratio and delay of end-to-end connection. However,
process to save energy. All these evident statements are the discovering process takes part in reduces the amount
considered an attractive point for more researches in this of flooding, but it raises the delay of end-to-end connec-
domain. tion and the consumption energy rate.
In more clarifying, Fig. 6 shows a sample for cooper-
ating cross-layer parameters that are used as a critical 5) SMART CROSS-LAYER ROUTING BASED ON
solution for facing the CR challenges. The wise from MACHINE LEARNING METHOD
this example is to explain how the interaction methods Today, machine learning is widely used in various applica-
between layer can play as a critical role to deal with tions and technologies [89]. The routing protocol in CRNs is
different challenges through setting up a new cross-layer considered one of this application that beneficial interest from
relationship for parameters to win on CR challenges the features of machine learning [89], [90]. In more, the CRNs
such as route selection, spectrum diversity, channel routing protocol at the network layer has various factors
selection, QoS, traffic control, and others. In more that can be trained by machine learning for the true route
specific, this method can give official status to the decision, respectively [109]. The cross-layer routing protocol
idea of using cross-layer design for routing challenges is considered more suitable for using machine learning due to
in CRNs. Consequently, the routing protocol, in the its ability to provide various useful services and information
cross-layer architecture of GUOP, chooses stable paths from other layers [91]. In practical terms, the algorithms of
instead of finding the shortest-hop routing, respectively. machine learning are trained according to the routing protocol
The proliferation of endless hops is prohibited by aiming factors such as hop count, node speed, node energy, CAP, CQ,
to reduce the delay of routing. etc., and that constitutes a smart routing protocol [89], [109].

67852 VOLUME 8, 2020

Q. M. Salih et al.: Smart Routing Management Framework Exploiting Dynamic Data Resources

In fact, the development of the principle work of machine in CRNs [3], [7]. Accordingly, there has been a great
learning algorithms has made them useful solutions in routing diversity of practical ideas to present mimic this type
protocols [94]. This is reasonable due to the machine learning of application. An example of this point is the achieve-
algorithms has the flexibility and ability to integrate and ment of cross-layer techniques for allocating resources. The
understand the challenges of routing algorithms in CRNs. cross-layering approach to resource allocation has been
On the other side, the number of research in this field is known to drive to better solutions to enhance routing perfor-
still shy, which make it an attractive research area. For that, mance. Although cross-layer available resource resilience,
we have introduced in this survey a new smart framework for it is still limited due to need central control for resource allo-
cross-layer routing protocol in CRNs in order to explore the cation management. On the other side, the machine-learning
mechanism of integrating machine learning and cross-layer technique is considered a key solution to manage and train the
routing protocol in CRNs. cross-layer resources. In more, machine-learning proposes an
engaging opportunity to manage resource allocation, which
IV. A CRITICAL REVIEW OF EXISTING WORKS FOR THE considers one of the main shortcomings of the cross-layer
ROUTING PROTOCOLS IN MOBILE COGNITIVE routing protocol algorithm in CRNs [77]. For instance,
RADIO NETWORKS the supervised algorithm is one of the machine learning tech-
Numerous techniques for routing protocol in CRNs have nique that aims to forecast the value of an outcome measure
been proposed. All appropriate methods or techniques are based on some input featuers, where learning is done base
reviewed, and their strengths and weaknesses are highlighted on a set of training samples. In more, Table 2 illustrates a
as mention in Table 1. In more detail, based on critical reviews sample of the input resources for services and information
tabulated in Table 1, most of the work in the early stage that have been shared from stack-layer to the network layer.
of CRNs discussed in non-Cross-layer approach. However, Hence, It has become necessary to find a smart cross-layer
recent related work focuses on cross-layer issues because of routing framework to manage cross-layer routing resource
its positive impact on CRNs. As the CRNs is very much and overcome routing challenges in CRNs [77]. For that,
dynamic due to its characteristics, a smart routing approach is Fig. 7 describes the proposed cross-layer routing framework
required by analysing environmental behaviour. Fewer works in CRNs, whereas machine learning has employed its services
have been addressed in the literature. and functions for more intelligent routing decisions. The
Hence, machine learning could be a prominent approach details of the framework are in the following:
to analysing the environmental behaviour of CRNs. In the
same context, smart routing protocols in CRNs will permit
A. ATTRIBUTES SERVICES AND INFORMATION OF
a robust modern spirit of a smart adaptive at the level of
THE HIGHER AND LOWER LAYER IN THE
CRNs which will expedite useful improvement and will be
CROSS-LAYER FRAMEWORK
an outstanding aid In the general framework of the principle
he successful cross-layer routing protocol in MCRNs
work of routing protocol. In this paper, we weave the idea of
depends on to adapt the layer’s services and information to
proposing a framework for smart cross-layer routing protocol
make a correct routing decision for next-hop [92]. In fact,
in CRNs by using machine-learning techniques. Applying
the smart routing protocol in MCRNs needs to define the
machine-learning for cross-layer routing protocols in CRNs
right metrics, which affects positively to optimize routing
can be optimisation routing performance at the level of
performance [93]. From that, we need to explorer the influ-
multi-path/multi-channel selection.
ence of taking part in the information and services of the
In practical terms, cross-layer features can provide the
higher and lower layer on routing algorithms. In more details,
machine-learning engine insufficient information (such as
the attributes of the higher and lower layer explain as the
CAP, CQ, link-interference, etc.), which can be serviceable
following:
by the smart routing protocol particularly for the jobs of
routing in CRNs. In more, machine-learning can play as the
boss that manages the cross-layer resource at network-layer 1) INPUT SERVICES AND INFORMATION PROVIDE BY
and trains the routing protocol to predicate the future route APPLICATION LAYER IN CROSS-LAYER FRAMEWORK
decision. To our knowledge, this is the first survey article Each layer in a stack can provide different services and share
that concentrates on issues of design smart routing protocol in it with its neighbours according to the layer stack policy [7].
CRNs through presenting a framework for smart cross-layer When the CU prepares its data packet to send to the target
routing protocol in CRNs by employing machine-learning node, the application layer will make the data and share it
technology. with the network layer through cross-layer design, as illus-
trated in Fig. 7. The advantage of this information is that
V. SMART CROSS-LAYER MANAGEMENT FRAMEWORK the routing protocol can define the data packet (i.e., text,
FOR ROUTING PROTOCOL IN MOBILE COGNITIVE audio, video, etc.) to provide adequate resources [3]. Also,
RADIO NETWORKS the routing algorithm can meet the QoS requirements
Many cross-layer designs have been proposed and simu- at the application layer by sharing the characterised of
lated various challenges and requirements routing protocol data-packet [14].

VOLUME 8, 2020 67853

 Q. M. Salih et al.: Smart Routing Management Framework Exploiting Dynamic Data Resources

TABLE 1. A critical review of techniques in MCRNs routing protocol based on the cross and non-cross layer design.

67854 VOLUME 8, 2020

Q. M. Salih et al.: Smart Routing Management Framework Exploiting Dynamic Data Resources

TABLE 1. (Continued.) A critical review of techniques in MCRNs routing protocol based on the cross and non-cross layer design.

VOLUME 8, 2020 67855

 Q. M. Salih et al.: Smart Routing Management Framework Exploiting Dynamic Data Resources

TABLE 1. (Continued.) A critical review of techniques in MCRNs routing protocol based on the cross and non-cross layer design.

67856 VOLUME 8, 2020

Q. M. Salih et al.: Smart Routing Management Framework Exploiting Dynamic Data Resources

TABLE 1. (Continued.) A critical review of techniques in MCRNs routing protocol based on the cross and non-cross layer design.

VOLUME 8, 2020 67857

 Q. M. Salih et al.: Smart Routing Management Framework Exploiting Dynamic Data Resources

TABLE 1. (Continued.) A critical review of techniques in MCRNs routing protocol based on the cross and non-cross layer design.

67858 VOLUME 8, 2020

Q. M. Salih et al.: Smart Routing Management Framework Exploiting Dynamic Data Resources

TABLE 1. (Continued.) A critical review of techniques in MCRNs routing protocol based on the cross and non-cross layer design.

VOLUME 8, 2020 67859

 Q. M. Salih et al.: Smart Routing Management Framework Exploiting Dynamic Data Resources

FIGURE 7. A generic cross-layer framework for an intelligent routing protocol in MCRNs based on machine learning.

TABLE 2. Sample of cross-layer input attributes.

2) INPUT SERVICES AND INFORMATION PROVIDE BY size of the routing protocol can take part in increasing the
TRANSPORT LAYER IN CROSS-LAYER FRAMEWORK PLP. The CU sends the data-packet at random times, and
The transport layer can directly share its services and infor- that can lead the queue to build up at a router [93]. For
mation with the network layer. It can contribute effectively that, if the routing has not enough properly buffer sized,
to refining the performance of the routing protocol in CRNs, the route will force to drop packets [7]. On the other hand,
especially in the QoS and throughout [7]. Packet Loss Proba- Enabling routing protocol to select a better idle channel with
bility (PLP) is related to the transport layer [93]. In CRNs, more bandwidth resource can reduce the probability of packet
many reasons can lead to an increase in the rate of PLP, loss [2]. In contrast, the increasing rate of the PLP can lead to
such as (Delay-sensitive, PU interference. etc.). The buffer more packet drops, which harmful the routing performance.

67860 VOLUME 8, 2020

Q. M. Salih et al.: Smart Routing Management Framework Exploiting Dynamic Data Resources

That means the buffer of routing should always be willing 2) ROUTING TABLE WITH EXTENDED INFORMATION
to receive CU’s data packet and monitoring the average of The routing operations performed through using a routing
PLP. Moreover, selecting the higher channel availability with table [12]. This table assists routing protocols to determine
more bandwidth availability can play a crucial role to reduce the best path to the destination based on a pre-defined scale
the PLP [30]. (i.e., routing metric) [12]. However, the routing table in
the traditional infrastructure-less wireless networks main-
3) INPUT SERVICES AND INFORMATION PROVIDE BY tains only limited information (i.e., next-hope informa-
MAC LAYER IN CROSS-LAYER FRAMEWORK tion) [59]. In contrast, spectrum mobility, PU interference,
The MAC layer also plays through giving its services/ channel switching, have paralysed the traditional routing
information with the network layer [55]. It considered as a table to meet new challenges in MCRNs [12]. In Fig. 7,
significant player through participation routing with channel it explains an extended routing table to include extra infor-
decision for next hope. Moreover, enabling routing protocol mation (e.g., CAP, CQ) for the link-channel of next hope.
to select the channel with higher availability that can increase For instance, CAP can reduce the probability of channel
routing stability, reducing channel switching, PU interfer- switching and provide a stable channel [59]. Thus, successful
ence and others, especially in the mobility scenario [45], [46]. routing protocol in MCRNs needs to extend its routing table
Fig. 7 illustrates the MAC layer services, namely scheduling with full information.
transmission on a free ideal channel (STFC), which has coop-
erated with the network layer. This service can guarantee 3) LEARNING ROUTING DECISION ENGINE
to route the proper to the better idle channel (i.e., higher The Learning Routing Decision Engine (LRDE) is the
CAP and CQ) [90]. Also, it can avoid CU the probability of lusty part that is attached to machine learning. Once the
interference with the PU [48]. link-connection is established between a pair of CU users and
the data packet is waiting inside the buffer of routing protocol
4) INPUT SERVICES AND INFORMATION PROVIDE BY
for sending, then the CU will forward data-packet to next CU
PHYSICAL LAYER IN CROSS-LAYER FRAMEWORK
node based on the LRDE feedback. In more detail, LRDE is
accountable for handling services and information gathered
The dynamic nature of CRNs has been imposed on the phys-
from other layers and making a true routing protocol decision.
ical layer to extend its functions. In a traditional wireless
Fig. 7 reveals that the LRDE has the different processing
system, there is no PU and CU. However, the physical layer
inside the training phase and the regression (i.e., estimation)
in CRNs has to differentiate between PU and CU. Conse-
phase.
quently, physical layer develops its mechanism to sensing
the idle channel and provide this information and service
4) TRAINING PHASE
to other layers such as (MAC layer, network layer, etc.).
Training is an iterative process in which the data gradually
Fig. 7 explains that the physical layer has different functions.
help to improve the quality of the prediction [92]. For more
Channel availability probability, one of the physical layer
details, we are going to explain the functionalities of every
functions, refers to the availability of a licence channel to
process, as shown below:
an unlicensed user. However, higher channel availability is 1) Feature Extraction:
not always represented as the best idle channel. The quality To learn the LRDE to produce curate estimate labels,
of the channel is also interference in decision making to it needs to train the LREAD firstly. The first step in the
select a better idle channel. In a wireless network, a link training phase is to extract the (features, labels) from
refers to a wireless connection between two nodes. By the the input of the cross-layer mechanism [93]. However,
same token in CRNs, a link is referred to as connecting the cross-layer mechanism can serve different services
between two CU nodes. However, when one or a pair of and information and that it needs to prepare this infor-
CUs are inside the interference range of PU that mention to mation [7]. Therefore, before applying the data for
link-interference. The link-interference can also take part in the learning phase, it has to go into a pre-processing
estimating link-throughput and link-delay. phase to clean the data and papering it for extracting
phase [89]. This phase plays an essential key because
B. INTELLIGENT ROUTING PROTOCOL WITH THE of in MCRNs due to these metrics can be used to predict
SUPERVISED MACHINE AT THE NETWORK such link-delay, link-throughput, and others.
LAYER IN CROSS-LAYER FRAMEWORK 2) Sample Collection:
1) CROSS-LAYER MECHANISM It is referred to select a sample from the features
It performs the cooperation of cross-layer information and and pre-filtration before sent it to regression
services. In addition, the cross-layer mechanism embraces the algorithm [91], [99]. It is an essential process because
services of other layers and shares them with the network the layer stack can provide various features. Therefore,
layer. Fig. 7 also clarifies that this mechanism has a vital the regression algorithm needs to filter these features
communication with other parts to serve the interests of the and prepare its configured according to routing opera-
functions of these parts in the network layer. tions such as regression link-delay, link-throughput.

VOLUME 8, 2020 67861

 Q. M. Salih et al.: Smart Routing Management Framework Exploiting Dynamic Data Resources

3) Supervised regression learning: such as (IoT, multimedia application, etc.) lead to

Supervised regression learning aims to predicate a emerge significant challenges [3], [7]. Mobility of
value based on features input [89], [98]. The output of spectrum-quality availability and application-mobility
regression could be numerical or categorical according may be a necessity to face the flood of QoS chal-
to the input natural features [99]. The learning process lenges of CR user’s requirements in the heteroge-
always needs to find a relation between the features neous network [71]. The mobility of CU and/or PU
to increase the learning regression accuracy [99]. is a real crisis due to various spectrum quality and
For instance, using features of a physical layer, availability in the region. The QoS is also suffering
i.e., CAP, CQ, and LI can train it to estimate the from application-mobility, which refers to the variety
link-throughput [88]. Hence, supervised first must and difference of applications [9]. Hence, to meet
train the algorithm on this data to learn how to esti- the heterogeneous QoS in MCRNs that need to think
mate link-throughput for the smart routing protocol seriously concerning the mobility of spectrum-quality
and that it can make the CU smarter to select the and applications. Movement of CR user is not mean
higher throughput link and can increase the routing only geographic change but also including change the
efficiency. spectrum-quality characteristics to match QoS require-
ments. Besides, application-mobility such as ( voice,
5) REGRESSION PHASE multimedia, IoT, 5G, etc.) need to find what satisfies
After training the supervised regression to estimate the link its greedy from the spectrum. Different parameters and
throughput based on the input metrics from the cross-layer services can contribute to addressing challenges QoS
mechanism, the learning regression algorithm now has the at the application layer. For instance, the priorities of
experience start to fill the routing table with regression CR user application, delay sensitivity, loss tolerance,
value [94]. In the same way, new input services and infor- traffic type and other services can play a pivotal role
mation will pass through the extraction phase and transfer in containing a crisis of providing the best QoS for
to a regression algorithm [92]. Fig. 7 shown that the output MCRNs.
estimation values has been sharing with a routing protocol
to update the routing table. On the other side, the LRDE B. TRANSPORT LAYER CHALLENGE IN MOBILE
also has an explicit link with routing operations to learn COGNITIVE RADIO NETWORK
the CU node to select the most appropriate route decision. • Congestion control and PU mobility: PU mobility leads
On the other hand, there are different supervised regres- to grow the phenomena of congestion and causes in
sion algorithms such (linear regression, logistic regression, service interruption [71]. The transport layer protocol
decision tree, etc.), which has to select according to input needs to take into account the tradeoff spectrum
metrics. sensing, time-variant of spectrum availability, and PU
activity [3], [7], [54], [75]. In a traditional network,
6) ROUTING OPERATIONS BY MCRNs there is no PU and SU, which makes the reasons
Routing operations is a sequence of steps that it targets to for the interruption of service belongs to traditional
reach the data-packet to the destination node [2]. Different reasons such as buffer occupies, window size, network
from traditional routing operation, this phase includes congestion and others [3]. In contrast, the protocol of
PU-RERR for handling PU activity. In more, routing services connection-oriented and connectionless is unsuitable for
such (RREQ, RREP, etc.) has to extend its structure to include adapting to the MCRNs. That means when the inter-
channel information. That it is reasonable due to CU node ruption of service occurs, the protocols at the transport
needs to discover the best idle channel through injection this layer will refer to the traditional reasons. Consequently,
information inside the control packet. the process of re-connection will cost the MCRNs
losing to its resource. Therefore, sharing the appropriate
VI. RESEARCH CHALLENGES spectrum sensing with the transport layer can beat the
Several studies have been presented the routing challenges transport layer protocol challenges in MCRNs. Thus,
in MCRNs [2], [5], [12]. However, with the increasing and cross-layer optimizations considering spectrum sensing
varied CR applications, it can create new routing challenges and the mechanisms of congestion control will be highly
and issues. Thus, this section tries to review and discuss these desirable.
new challenges.
C. NETWORK LAYER CHALLENGE IN MOBILE
A. APPLICATION LAYER CHALLENGE IN MOBILE COGNITIVE RADIO NETWORK
COGNITIVE RADIO NETWORK The routing challenges at the network layer in MCRNs are
• QoS-aware: The QoS-aware at the application layer is abstracted below:
a fateful matter to success the CR technology [52]. • Routing scalability: Scalability is one of the main funda-
Increasing the numbers and types of applications mental concerns in CRNs due to it can be considered

67862 VOLUME 8, 2020

Q. M. Salih et al.: Smart Routing Management Framework Exploiting Dynamic Data Resources

as the principal performance measure for routing [5]. D. LINK LAYER (MAC) CHALLENGE IN MOBILE
In general, scalability could be mentioned as the routing COGNITIVE RADIO NETWORK
ability to achieve valuable work with the rise in network Control sensing and transmission: Mobility CU and PU will
density size, i.e., network load. For instance, the scal- increase the complexity of the duties of the MAC protocol in
ability factor can be useful to determine if the perfor- CRNs. MAC protocol has to update its strategies to deal with
mance of CRNs increases with the increase in the mobility challenges concerning sensing control [7], [101].
number of CUs, and also if the routing protocol perfor- Sensing control is responsible for facilitating MCRNs to
mance increase with the increase in the network size, etc. do its sensing process in adaptively way with mobility of
From that, scalability plays a critical issue routing design spectrum resource [12]. In more, spectrum sensing aims to
in CRNs due to it defines the routing algorithm’s ability avoid interfering with PU during to explore opportunities for
to provide reliability, fairness, and ingest the network spectrum access [12], [55]. Indeed, the MAC protocol needs
load [78]. In a word, routing scalability requires devel- to know-how to avoid Interference with PU mobile and the
oping routing algorithm operations through the integra- fast-discovery for spectrum resource for MCRNs. The MAC
tion of new parameters and routing metric to win more protocol has to impose its control on session operations by
robust routing. defining a mechanism for a sensing period. At the link-layer,
• Limitation routing algorithm resource with upper layers: the MAC protocol demands to identify the extended period of
In the traditional network, Ad hoc routing algorithm spectrum sensing [55]. By the same fashion, the speed diver-
saves in its routing table only the neighbour nodes sity will demand to the fast-discover process for spectrum.
information. In contrast, the routing table in CRNs Longer and shorter time can play on the tune of spectrum
needs to extend its rows to include more parame- accuracy discover [55]. On the other hand, transmission-time
ters such as network characteristics, list free channel, also needs to take into account. It is considered as another
etc. because routing in CRNs demands to a dynamic problem that needs to take into consideration. The spectrum
environment and PU activity [7], [12]. In fact, many sensing and transmission are one part complete the other.
studies, as mention previously, have been referred to this Thus, the MAC has to know how to determine the appropriate
challenge. However, all these efforts nearly have been times of sensing and transmission. In other words, it is a
focused on lower layer parameters. According to our fundamental problem in controlling the spectrum sensing and
classification in this research, it has revealed an urgent transmission, especially in CU and PU mobility.
need for taking into account the higher network layer.
In more specific, not only the lower layer parameters
can share in achieving routing performance, but also E. PHYSICAL LAYER CHALLENGE IN MOBILE
the upper layer [73]. For instance, the upper layer can COGNITIVE RADIO NETWORK
provide routing algorithm guideline information about • Physical layer services and functions: The services and
the nature of the applications such as the nature of functions at the physical layer have already not been
traffic, e.g., audio, video, frame size, and other param- specified for CR [7]. For that, the physical layer must
eters [7]. Exchange upper layer service with routing develop its services and features to adapt to CR tech-
algorithm is not only reflected on the performance nology [87]. One of the critical function that needs
routing algorithm but also the network performance in to explore at the physical layer is the fast-adapt with
general. the time-varying channel. The physical layer must be
• Time-variant for relative distance between PU and CU: vigilant to any varying in licensed channel availability.
Not only the resource location factor but also the Because of the mobility of CU and PU, physical layer
time-variant spectrum availability should be consid- jobs to detect the PU signal on the licensed channel
ered [95]–[97]. In more, when the PU and/or CU is a become a daunting task [88]. Numerous studies have
mobile, the time of channel availability is variant due focused on spectrum sensing to identify PU signal [12].
to mobility relative distance between SU and PU [41]. However, it is still considered a chronic disease. The
As a result, the time-variant for network resource makes physical layer has a great responsibility, which is
the path and channel instability. Thereupon, time-variant restricted to monitoring the varying period of channel
plays a decisive factor for successfully avoiding PUs availability [88]. Also, the physical layer must improve
activity and determining the correct link communica- its ability to detect the PU signal early. The rapid
tion, especially when PU/CU or together mobility. So, change in mobility and channel availability will make
the routing protocol, due to time-variant for channel PU fishing more difficult. Therefore, the physical layer
availability, needs every period to update its information can not deal with the CU node as an Ad- hoc node.
about the list of channel availability. Hence, time-variant In MCRNs, the temporal diversity of availability of the
imposes on routing a new challenge that demands to licensed channel distance mobility between CU and PU
develop a new mechanism for channel selection. needs to be radically addressed.

VOLUME 8, 2020 67863

 Q. M. Salih et al.: Smart Routing Management Framework Exploiting Dynamic Data Resources

VII. CONCLUSION [17] L. Senthilkumar and M. Meenakshi, ‘‘Cross-layer based asymmetric

The technical routing protocol details are explained resource allocation in relay-aided cognitive radio networks,’’ Wireless
Pers. Commun., vol. 97, no. 4, pp. 5543–5571, Dec. 2017.
throughout this paper, which including explored a cross-layer [18] Q. Zhang and Y.-Q. Zhang, ‘‘Cross-layer design for QoS support in
framework for the smart routing protocol in MCRNs. In more, multihop wireless networks,’’ Proc. IEEE, vol. 96, no. 1, pp. 64–76,
machine learning plays as a critical interaction factor by Jan. 2008.
[19] G. P. Joshi, S. W. Kim, and S. Y. Nam, ‘‘Routing layer solution for miti-
proposed machine learning features with the routing algo- gating frequent channel switching in ad hoc cognitive radio networks,’’
rithm, so that MCRNs can communicate reliably, over a IEEE Commun. Lett., vol. 19, no. 11, pp. 1917–1920, Nov. 2015.
multi-path/multi-spectrum environment. It also classifies the [20] M. Youssef, M. Ibrahim, M. Abdelatif, L. Chen, and A. V. Vasilakos,
routing algorithm based on methods of design, which belongs ‘‘Routing metrics of cognitive radio networks: A survey,’’ IEEE Commun.
Surveys Tuts., vol. 16, no. 1, pp. 92–109, 1st Quart., 2014.
to cross and non-cross-layer design. Next, critical reviewing [21] M. Cesana, F. Cuomo, and E. Ekici, ‘‘Routing in cognitive radio networks:
has done for related work to explorer the strengths and Challenges and solutions,’’ Ad Hoc Netw., vol. 9, no. 3, pp. 228–248,
weakness. In the end, this paper vigorously discussed the May 2011.
[22] V. Karyotis, E. Anifantis, and S. Papavassiliou, ‘‘Cross-layer based
open issues related to the routing protocol in CRNs. resource management frameworks for mobile cognitive radio networks,’’
in Resource Management in Mobile Computing Environments. Cham,
REFERENCES Switzerland: Springer, May 2014, pp. 285–322, doi: 10.1007/978-3-319-
06704-9_13.
[1] G. Baldini, O. Holland, V. Stavroulaki, K. Tsagkaris, P. Demestichas, [23] V. Srivastava and M. Motani, ‘‘Cross-layer design: A survey and the road
A. Polydoros, S. Karanasios, and D. Allen, ‘‘The evolution of cognitive ahead,’’ IEEE Commun. Mag., vol. 43, no. 12, pp. 112–119, Dec. 2005.
radio technology in Europe: Regulatory and standardization aspects,’’ [24] I. Das, D. K. Lobiyal, and C. P. Katti, ‘‘Multipath routing in mobile ad hoc
Telecommun. Policy, vol. 37, nos. 2–3, pp. 96–107, Mar. 2013. network with probabilistic splitting of traffic,’’ Wireless Netw., vol. 22,
[2] Y. Saleem, F. Salim, and M. H. Rehmani, ‘‘Routing and channel selection no. 7, pp. 2287–2298, Oct. 2016.
from cognitive radio network’s perspective: A survey,’’ Comput. Electr. [25] H. Su and X. Zhang, ‘‘Cross-layer based opportunistic MAC protocols
Eng., vol. 42, pp. 117–134, Feb. 2015. for QoS provisionings over cognitive radio wireless networks,’’ IEEE J.
[3] Y. Teng and M. Song, ‘‘Cross-layer optimization and protocol anal- Sel. Areas Commun., vol. 26, no. 1, pp. 118–129, Jan. 2008.
ysis for cognitive ad hoc communications,’’ IEEE Access, vol. 5, [26] M. Zhan, P. Ren, C. Zhang, and F. Li, ‘‘Throughput-optimized cross-
pp. 18692–18706, 2017. layer routing for cognitive radio networks,’’ in Proc. 1st IEEE Int. Conf.
[4] S. AlQahtani and A. Alotaibi, ‘‘A route stability-based multipath QoS Commun. China (ICCC), Aug. 2012, pp. 745–750.
routing protocol in cognitive radio ad hoc networks,’’ Wireless Netw., [27] Y.-C. Liang, K.-C. Chen, G. Ye Li, and P. Mahonen, ‘‘Cognitive
vol. 25, no. 5, pp. 2931–2951, Jul. 2019, doi: 10.1007/s11276-019- radio networking and communications: An overview,’’ IEEE Trans. Veh.
02014-6. Technol., vol. 60, no. 7, pp. 3386–3407, Sep. 2011.
[5] A. S. Cacciapuoti, M. Caleffi, L. Paura, and M. A. Rahman, ‘‘Channel [28] J. Marinho and E. Monteiro, ‘‘Cognitive radio: Survey on communica-
availability for mobile cognitive radio networks,’’ J. Netw. Comput. Appl., tion protocols, spectrum decision issues, and future research directions,’’
vol. 47, pp. 131–136, Jan. 2015. Wireless Netw., vol. 18, no. 2, pp. 147–164, Feb. 2012.
[6] H. A. A. Al-Rawi and K.-L.-A. Yau, ‘‘Routing in distributed cognitive [29] D. Hlavacek and J. M. Chang, ‘‘A layered approach to cognitive radio
radio networks: A survey,’’ Wireless Pers. Commun., vol. 69, no. 4, network security: A survey,’’ Comput. Netw., vol. 75, pp. 414–436,
pp. 1983–2020, Apr. 2013. Dec. 2014.
[7] A. Kliks, D. Triantafyllopoulou, L. De Nardis, O. Holland, [30] R. Edirisinghe and A. Zaslavsky, ‘‘Cross-layer contextual interactions
L. Gavrilovska, and A. Bantouna, ‘‘Cross-layer analysis in cognitive in wireless networks,’’ IEEE Commun. Surveys Tuts., vol. 16, no. 2,
radio—Context identification and decision making aspects,’’ IEEE pp. 1114–1134, 2nd Quart., 2014.
Trans. Cognit. Commun. Netw., vol. 1, no. 4, pp. 450–463, Dec. 2015. [31] J. S. P. Singh and M. K. Rai, ‘‘CROP: Cognitive radio ROuting protocol
[8] F. A. Awin, Y. M. Alginahi, E. Abdel-Raheem, and K. Tepe, ‘‘Technical for link quality channel diverse cognitive networks,’’ J. Netw. Comput.
issues on cognitive radio-based Internet of Things systems: A survey,’’ Appl., vol. 104, pp. 48–60, Feb. 2018.
IEEE Access, vol. 7, pp. 97887–97908, 2019. [32] A. Guirguis, F. Digham, K. G. Seddik, M. Ibrahim, K. A. Harras, and
[9] P. Rawat, K. D. Singh, and J. M. Bonnin, ‘‘Cognitive radio for M2M M. Youssef, ‘‘Primary user-aware optimal discovery routing for cogni-
and Internet of Things: A survey,’’ Comput. Commun., vol. 94, pp. 1–29, tive radio networks,’’ IEEE Trans. Mobile Comput., vol. 18, no. 1,
Nov. 2016. pp. 193–206, Jan. 2019.
[10] R. M. Aileni, G. Suciu, V. Suciu, S. Pasca, and R. Strungaru, Health [33] B. Fu, Y. Xiao, H. J. Deng, and H. Zeng, ‘‘A survey of cross-layer de-
Monitoring Using Wearable Technologies and Cognitive Radio for IoT. signs in wireless networks,’’ IEEE Commun. Surveys Tuts., vol. 16, no. 1,
Cham, Switzerland: Springer, Jul. 2019, pp. 143–165. pp. 110–126, 1st Quart., 2013.
[11] A. Morgado, K. M. S. Huq, S. Mumtaz, and J. Rodriguez, ‘‘A survey of 5G [34] D. Singhal and R. M. Garimella, ‘‘Cognitive cross-layer multipath prob-
technologies: Regulatory, standardization and industrial perspectives,’’ abilistic routing for cognitive networks,’’ Wireless Netw., vol. 21, no. 4,
Digit. Commun. Netw., vol. 4, no. 2, pp. 87–97, Apr. 2018. pp. 1181–1192, May 2015.
[12] I. F. Akyildiz, W.-Y. Lee, and K. R. Chowdhury, ‘‘CRAHNs: Cognitive [35] R. Ranjan and S. Varma, ‘‘Challenges and implementation on cross layer
radio ad hoc networks,’’ Ad Hoc Netw., vol. 7, no. 5, pp. 810–836, design for wireless sensor networks,’’ Wireless Pers. Commun., vol. 86,
Jul. 2009. no. 2, pp. 1037–1060, Jan. 2016.
[13] E. Ahmed, A. Gani, S. Abolfazli, L. J. Yao, and S. U. Khan, ‘‘Channel [36] A. S. Cacciapuoti, M. Caleffi, F. Marino, and L. Paura, ‘‘On the route
assignment algorithms in cognitive radio networks: Taxonomy, open priority for cognitive radio networks,’’ IEEE Trans. Commun., vol. 63,
issues, and challenges,’’ IEEE Commun. Surveys Tuts., vol. 18, no. 1, no. 9, pp. 3103–3117, Jul. 2015.
pp. 795–823, 1st Quart., 2016. [37] C. Xin, L. Ma, and C.-C. Shen, ‘‘A path-centric channel assignment
[14] K. C. How, M. Ma, and Y. Qin, ‘‘Routing and QoS provisioning in framework for cognitive radio wireless networks,’’ Mobile Netw. Appl.,
cognitive radio networks,’’ Comput. Netw., vol. 55, no. 1, pp. 330–342, vol. 13, no. 5, pp. 463–476, Oct. 2008.
Jan. 2011. [38] L. Yun, Q. Fengxie, L. Zhanjun, and Z. Hongcheng, ‘‘Cognitive radio
[15] A. Sevin, C. Bayilmis, and I. Kirbas, ‘‘Design and implementation of routing algorithm based on the smallest transmission delay,’’ in Proc. 2nd
a new quality of service-aware cross-layer medium access protocol for Int. Conf. Future Comput. Commun., vol. 2, May 2010, p. V2-306.
wireless body area networks,’’ Comput. Electr. Eng., vol. 56, pp. 145–156, [39] H. Ma, L. Zheng, X. Ma, and Y. Luo, ‘‘Spectrum aware routing for multi-
Nov. 2016. hop cognitive radio networks with a single transceiver,’’ in Proc. 3rd
[16] J. Qadir, ‘‘Artificial intelligence based cognitive routing for cognitive Int. Conf. Cognit. Radio Oriented Wireless Netw. Commun. (CrownCom),
radio networks,’’ Artif. Intell. Rev., vol. 45, no. 1, pp. 25–96, Jan. 2016. May 2008, pp. 1–6.

67864 VOLUME 8, 2020

Q. M. Salih et al.: Smart Routing Management Framework Exploiting Dynamic Data Resources

[40] A. S. Cacciapuoti, M. Caleffi, and L. Paura, ‘‘Reactive routing for [63] Y. Yang and S. Aissa, ‘‘Cross-layer combining of information-guided
mobile cognitive radio ad hoc networks,’’ Ad Hoc Netw., vol. 10, no. 5, transmission with network coding relaying for multiuser cognitive radio
pp. 803–815, Jul. 2012. systems,’’ IEEE Wireless Commun. Lett., vol. 2, no. 1, pp. 26–29,
[41] M. A. Rahman, M. Caleffi, and L. Paura, ‘‘Joint path and spectrum diver- Feb. 2013.
sity in cognitive radio ad-hoc networks,’’ EURASIP J. Wireless Commun. [64] A. Guirguis, M. Karmoose, K. Habak, M. El-Nainay, and M. Youssef,
Netw., vol. 2012, no. 1, p. 235, Dec. 2012. ‘‘Cooperation-based multi-hop routing protocol for cognitive radio
[42] O. Abedi and R. Berangi, ‘‘Beaconless dynamic spectrum-aware routing networks,’’ J. Netw. Comput. Appl., vol. 110, pp. 27–42, May 2018.
protocol for cognitive radio ad hoc networks,’’ Arabian J. Sci. Eng., [65] F. Foukalas, V. Gazis, and N. Alonistioti, ‘‘Cross-layer design proposals
vol. 39, no. 5, pp. 3941–3952, May 2014. for wireless mobile networks: A survey and taxonomy,’’ IEEE Commun.
[43] N. Dutta and H. K. D. Sarma, ‘‘A probability based stable routing Surveys Tuts., vol. 10, no. 1, pp. 70–85, 1st Quart., 2008.
for cognitive radio adhoc networks,’’ Wireless Netw., vol. 23, no. 1, [66] S. Yin, Y. Xiong, Q. Zhang, and X. Lin, ‘‘Traffic-aware routing for
pp. 65–78, Jan. 2017, doi: 10.1007/s11276-015-1138-2. real-time communications in wireless multi-hop networks,’’ Wireless
[44] J. Wang, H. Yue, L. Hai, and Y. Fang, ‘‘Spectrum-aware anypath routing Commun. Mobile Comput., vol. 6, no. 6, pp. 825–843, Sep. 2006.
in multi-hop cognitive radio networks,’’ IEEE Trans. Mobile Comput., [67] S. Salim and S. Moh, ‘‘A robust and energy-efficient transport
vol. 16, no. 4, pp. 1176–1187, Apr. 2017. protocol for cognitive radio sensor networks,’’ Sensors, vol. 14, no. 10,
[45] Y. Wang, G. Zheng, H. Ma, Y. Li, and J. Li, ‘‘A joint channel selection and pp. 19533–19550, 2014.
routing protocol for cognitive radio network,’’ Wireless Commun. Mobile [68] S. Sengupta and K. P. Subbalakshmi, ‘‘Open research issues in multi-
Comput., vol. 2018, pp. 1–7, Mar. 2018. hop cognitive radio networks,’’ IEEE Commun. Mag., vol. 51, no. 4,
[46] S. Amiri-Doomari, G. Mirjalily, and J. Abouei, ‘‘Stability-based routing, pp. 168–176, Apr. 2013.
link scheduling and channel assignment in cognitive radio mobile ad-hoc [69] N. Mansoor, A. K. M. M. Islam, M. Zareei, and C. Vargas-Rosales,
networks,’’ Wireless Netw., vol. 25, no. 4, pp. 2013–2026, May 2019. ‘‘RARE: A spectrum aware cross-layer MAC protocol for cognitive radio
[47] Z. J. Haas and M. R. Pearlman, ‘‘ZRP: A hybrid framework for ad-hoc networks,’’ IEEE Access, vol. 6, pp. 22210–22227, 2018.
routing in ad hoc networks,’’ in Ad Hoc Networking. Boston, MA, [70] G. S. Let and G. J. Bala, ‘‘A review of cross-layer design in dynamic
USA: Addison-Wesley, Jan. 2001, pp. 221–253. [Online]. Available: spectrum access for cognitive radio networks,’’ J. Comput. Inf. Technol.,
http://dl.acm.org/citation.cfm?id=374547.374554 vol. 22, no. 1, pp. 21–29, 2014.
[48] G.-M. Zhu, I. F. Akyildiz, and G.-S. Kuo, ‘‘STOD-RP: A spectrum- [71] Z. Shu, Y. Qian, Y. Yang, and H. Sharif, ‘‘A cross-layer study
tree based on-demand routing protocol for multi-hop cognitive radio for application-aware multi-hop cognitive radio networks,’’ Wireless
networks,’’ in Proc. IEEE Global Telecommun. Conf. (GLOBECOM), Commun. Mobile Comput., vol. 16, no. 5, pp. 607–619, Apr. 2016.
Nov. 2008, pp. 1–5. [72] M. Amjad, M. H. Rehmani, and S. Mao, ‘‘Wireless multimedia cognitive
[49] H. Khalife, S. Ahuja, N. Malouch, and M. Krunz, ‘‘Probabilistic path radio networks: A comprehensive survey,’’ IEEE Commun. Surveys Tuts.,
selection in opportunistic cognitive radio networks,’’ in Proc. IEEE vol. 20, no. 2, pp. 1056–1103, 2nd Quart., 2018.
Global Telecommun. Conf. (GLOBECOM), Nov. 2008, pp. 1–5. [73] J. Jagannath, S. Furman, T. Melodia, and A. Drozd, ‘‘Design and experi-
mental evaluation of a cross-layer deadline-based joint routing and spec-
[50] M. Zareei, E. M. Mohamed, M. H. Anisi, C. V. Rosales, K. Tsukamoto,
trum allocation algorithm,’’ IEEE Trans. Mobile Comput., vol. 18, no. 8,
and M. K. Khan, ‘‘On-demand hybrid routing for cognitive radio ad-hoc
pp. 1774–1788, Aug. 2019.
network,’’ IEEE Access, vol. 4, pp. 8294–8302, 2016.
[74] Y. Liu and Q. Zhou, ‘‘State of the art in cross-layer design for cognitive
[51] S. Shakkottai, T. S. Rappaport, and P. C. Karlsson, ‘‘Cross-layer design
radio wireless networks,’’ in Proc. Int. Symp. Intell. Ubiquitous Comput.
for wireless networks,’’ IEEE Commun. Mag., vol. 41, no. 10, pp. 74–80,
Educ., May 2009, pp. 366–369.
Oct. 2003.
[75] X. Zhong, Y. Qin, and L. Li, ‘‘Transport protocols in cognitive radio
[52] F. R. Yu, B. Sun, V. Krishnamurthy, and S. Ali, ‘‘Application layer QoS
networks: A survey,’’ 2015, arXiv:1501.02869. [Online]. Available:
optimization for multimedia transmission over cognitive radio networks,’’
http://arxiv.org/abs/1501.02869
Wireless Netw., vol. 17, no. 2, pp. 371–383, Feb. 2011.
[76] C. Dong, Y. Qu, H. Dai, S. Guo, and Q. Wu, ‘‘Multicast in multi-channel
[53] Z. Damljanovic, ‘‘Mobility management strategies in heterogeneous cognitive radio ad hoc networks: Challenges and research aspects,’’
cognitive radio networks,’’ J. Netw. Syst. Manage., vol. 18, no. 1, Comput. Commun., vol. 132, pp. 10–16, Nov. 2018.
pp. 4–22, Mar. 2010.
[77] J. L. Mauri, K. Z. Ghafoor, D. B. Rawat, and J. M. A. Perez, Cogni-
[54] D. Sarkar and H. Narayan, ‘‘Transport layer protocols for cogni- tive Networks: Applications and Deployments. Boca Raton, FL, USA:
tive networks,’’ in Proc. IEEE Conf. Comput. Commun. Workshops CRC Press, Dec. 2014.
(INFOCOM), Mar. 2010, pp. 1–6. [78] F. Yao, J. Zhang, H. Zhao, and Y. Liu, ‘‘Practical cross-layer routing
[55] C. Cormio and K. R. Chowdhury, ‘‘A survey on MAC protocols for and channel assignment in cognitive radio ad hoc networks,’’ Wireless
cognitive radio networks,’’ Ad Hoc Netw., vol. 7, no. 7, pp. 1315–1329, Commun. Mobile Comput., vol. 15, no. 10, pp. 1444–1455, Jul. 2015.
Sep. 2009. [79] G.-Q. Chen, S. Mei, Y. Zhang, and J.-D. Song, ‘‘Enhanced performance
[56] S. C. Jha, U. Phuyal, and V. K. Bhargava, ‘‘Cross-layer resource allocation based on cross-layer design from physical to transport layers,’’ Arabian
approach for multi-hop distributed cognitive radio network,’’ in Proc. J. Sci. Eng., vol. 39, no. 5, pp. 3941–3952, May 2014.
12th Can. Workshop Inf. Theory, May 2011, pp. 211–215. [80] K. Bian, J. M. Park, and B. Gao, Cognitive Radio Networks:
[57] S. Basak and T. Acharya, ‘‘Route selection for interference minimization Medium Access Control for Coexistence of Wireless Systems.
to primary users in cognitive radio ad hoc networks: A cross layer Cham, Switzerland: Springer, Jul. 2014. [Online]. Available:
approach,’’ Phys. Commun., vol. 19, pp. 118–132, Jun. 2016. https://books.google.com.my/books?id=r88lBAAAQBAJ
[58] F. Tang and J. Li, ‘‘Joint rate adaptation, channel assignment and routing [81] S. Cakici, I. Erturk, S. Atmaca, and A. Karahan, ‘‘A novel cross-layer
to maximize social welfare in multi-hop cognitive radio networks,’’ IEEE routing protocol for increasing packet transfer reliability in mobile sensor
Trans. Wireless Commun., vol. 16, no. 4, pp. 2097–2110, Apr. 2017. networks,’’ Wireless Pers. Commun., vol. 77, no. 3, pp. 2235–2254,
[59] F. Tang, M. Guo, S. Guo, and C.-Z. Xu, ‘‘Mobility prediction based Aug. 2014.
joint stable routing and channel assignment for mobile ad hoc cogni- [82] D. G. Narayan and U. Mudenagudi, ‘‘A cross-layer framework for joint
tive networks,’’ IEEE Trans. Parallel Distrib. Syst., vol. 27, no. 3, routing and rate adaptation in infrastructure wireless mesh networks,’’
pp. 789–802, Mar. 2016. Comput. Electr. Eng., vol. 56, pp. 113–129, Nov. 2016.
[60] K. Moyaba, M. Velempini, and S. Dlamini, ‘‘A joint MAC and network [83] M. Musavi, K.-L.-A. Yau, A. R. Syed, H. Mohamad, and N. Ramli,
layer QoS scheme which mitigates primary and secondary Users’ inter- ‘‘Route selection over clustered cognitive radio networks: An experi-
ference,’’ in Proc. Int. Conf. Adv. Big Data, Comput. Data Commun. Syst. mental evaluation,’’ Comput. Commun., vol. 129, pp. 138–151, Sep. 2018.
(icABCD), Aug. 2018, pp. 1–8. [84] G. P. Joshi, S. Y. Nam, C.-S. Kim, and S. W. Kim, ‘‘A cross-layer-based
[61] Z.-S. Li and Q. Zhu, ‘‘A novel cross layer routing protocol for CR Ad routing protocol for ad hoc cognitive radio networks,’’ Int. J. Distrib.
hoc network,’’ in Proc. 6th Int. Conf. Wireless Commun. Netw. Mobile Sensor Netw., vol. 2015, p. 4, Oct. 2015.
Comput. (WiCOM), Sep. 2010, pp. 1–4. [85] L. Ding, T. Melodia, S. N. Batalama, J. D. Matyjas, and M. J. Medley,
[62] R. Saifan, A. E. Kamal, and Y. Guan, ‘‘A cross-layer routing protocol ‘‘Cross-layer routing and dynamic spectrum allocation in cognitive
(CLRP) for cognitive radio network,’’ in Proc. IEEE Global Commun. radio ad hoc networks,’’ IEEE Trans. Veh. Technol., vol. 59, no. 4,
Conf. (GLOBECOM), Dec. 2013, pp. 896–901. pp. 1969–1979, May 2010.

VOLUME 8, 2020 67865

 Q. M. Salih et al.: Smart Routing Management Framework Exploiting Dynamic Data Resources

[86] S. Gogineni, O. Ozdemir, E. Masazade, C. K. Mohan, and P. K. Varshney, [108] F. Tang, C. Tang, Y. Yang, L. T. Yang, T. Zhou, J. Li, and M. Guo,
‘‘A cross layer routing protocol for cognitive radio networks using ‘‘Delay-minimized routing in mobile cognitive networks for time-critical
channel activity tracking,’’ in Proc. Conf. Rec. 46th Asilomar Conf. applications,’’ IEEE Trans. Ind. Informat., vol. 13, no. 3, pp. 1398–1409,
Signals, Syst. Comput. (ASILOMAR), Nov. 2012, pp. 1079–1083. Jun. 2017.
[87] S. Lavanya and M. A. Bhagyaveni, ‘‘Design of SOP based cross-layered [109] D. K. Sharma, S. K. Dhurandher, I. Woungang, R. K. Srivastava,
opportunistic routing protocol for CR ad-hoc networks,’’ Wireless Pers. A. Mohananey, and J. J. P. C. Rodrigues, ‘‘A machine learning-based
Commun., vol. 96, no. 4, pp. 6543–6556, Oct. 2017. protocol for efficient routing in opportunistic networks,’’ IEEE Syst. J.,
[88] M. A. Rahman, A. T. Asyhari, M. Z. A. Bhuiyan, Q. M. Salih, vol. 12, no. 3, pp. 2207–2213, Sep. 2018.
and K. Z. B. Zamli, ‘‘L-CAQ: Joint link-oriented channel-availability
and channel-quality based channel selection for mobile cognitive radio
networks,’’ J. Netw. Comput. Appl., vol. 113, pp. 26–35, Jul. 2018.
[89] R. Boutaba, M. A. Salahuddin, N. Limam, S. Ayoubi, N. Shahriar,
F. Estrada-Solano, and O. M. Caicedo, ‘‘A comprehensive survey on
machine learning for networking: Evolution, applications and research
opportunities,’’ J. Internet Services Appl., vol. 9, no. 1, p. 16, Dec. 2018.
[90] M. Bkassiny, Y. Li, and S. K. Jayaweera, ‘‘A survey on machine-learning
techniques in cognitive radios,’’ IEEE Commun. Surveys Tuts., vol. 15, QUSAY MEDHAT SALIH received the master’s
no. 3, pp. 1136–1159, 3rd Quart., 2013. degree in computer systems and networking from
[91] P. V. Klaine, M. A. Imran, O. Onireti, and R. D. Souza, ‘‘A survey the Faculty of Computer Systems and Soft-
of machine learning techniques applied to self-organizing cellular ware Engineering, University Malaysia Pahang,
networks,’’ IEEE Commun. Surveys Tuts., vol. 19, no. 4, pp. 2392–2431, in 2015, where he is currently pursuing the
4th Quart., 2017. Ph.D. degree. His research interests include a
[92] K. Singh and S. Moh, ‘‘Routing protocols in cognitive radio ad hoc computer networks, handoff, wireless commu-
networks: A comprehensive review,’’ J. Netw. Comput. Appl., vol. 72, nication, cognitive radio networks, and routing
pp. 28–37, Sep. 2016. protocol.
[93] D. R. Bhadra, C. A. Joshi, P. R. Soni, N. P. Vyas, and R. H. Jhaveri,
‘‘Packet loss probability in wireless networks: A survey,’’ in Proc. Int.
Conf. Commun. Signal Process. (ICCSP), Apr. 2015, pp. 1348–1354.
[94] Y. Wang, Z. Ye, P. Wan, and J. Zhao, ‘‘A survey of dynamic spectrum
allocation based on reinforcement learning algorithms in cognitive radio
networks,’’ Artif. Intell. Rev., vol. 51, no. 3, pp. 493–506, Mar. 2019.
[95] Q. Wang and M. A. Abu-Rgheff, ‘‘Cross-layer signalling for next-
generation wireless systems,’’ in Proc. IEEE Wireless Commun. Netw.,
vol. 2, Mar. 2003, pp. 1084–1089.
[96] J. Lu, Z. Cai, X. Wang, L. Zhang, P. Li, and Z. He, ‘‘User social activity- MD. ARAFATUR RAHMAN (Senior Member,
based routing for cognitive radio networks,’’ Pers. Ubiquitous Comput., IEEE) received the Ph.D. degree in electronic and
vol. 22, no. 3, pp. 471–487, Jun. 2018. telecommunications engineering from the Univer-
[97] S. Kim, O. Lee, S. Choi, and S.-J. Lee, ‘‘Comparative analysis of link sity of Naples Federico II, Naples, Italy, in 2013.
quality metrics and routing protocols for optimal route construction in He has more than ten years of experience in
wireless mesh networks,’’ Ad Hoc Netw., vol. 9, no. 7, pp. 1343–1358, research and teaching, in the domain of computer
Sep. 2011. and communications engineering. He worked as a
[98] R. N. Raj, A. Nayak, and M. S. Kumar, ‘‘A survey and performance Postdoctoral Research Fellow with the University
evaluation of reinforcement learning based spectrum aware routing in of Naples Federico II, in 2014, and a Visiting
cognitive radio ad hoc networks,’’ Int. J. Wireless Inf. Netw., vol. 27, no. 1, Researcher with the Sapienza University of Rome,
pp. 144–163, Mar. 2020. in 2016. He is currently a Senior Lecturer (equivalent to Assistant Professor)
[99] B. Pourpeighambar, M. Dehghan, and M. Sabaei, ‘‘Joint routing and with the Faculty of Computing, University Malaysia Pahang. His research
channel assignment using online learning in cognitive radio networks,’’
interests include the Internet of Things (IoT), wireless communication
Wireless Netw., vol. 25, no. 5, pp. 2407–2421, Jul. 2019.
networks, cognitive radio networks, 5G, vehicular communication, cyber
[100] Y. Du, C. Chen, P. Ma, and L. Xue, ‘‘A cross-layer routing protocol based
on quasi-cooperative multi-agent learning for multi-hop cognitive radio physical systems, big data, cloud-fog-edge computing, machine learning,
networks,’’ Sensors, vol. 19, no. 1, p. 151, Jan. 2019. and security. He has developed excellent track record of academic leadership
[101] A. Kumar and K. Kumar, ‘‘Multiple access schemes for cognitive as well as management and execution of international ICT projects that are
radio networks: A survey,’’ Phys. Commun., vol. 38, Feb. 2020, supported by agencies in Italy, EU, and Malaysia. He is a Fellow of the
Art. no. 100953. IBM Center of Excellence and Earth Resources and Sustainability Center,
[102] J. Zuo, C. Dong, S. X. Ng, L.-L. Yang, and L. Hanzo, ‘‘Cross-layer aided Malaysia. He has received number of prestigious international research
energy-efficient routing design for ad hoc networks,’’ IEEE Commun. awards, notably the Best Paper Award at ICNS’15, Italy, the IC0902 Grant,
Surveys Tuts., vol. 17, no. 3, pp. 1214–1238, 3rd Quart., 2015. France, the Italian Government Ph.D. Research Scholarship, the IIUM Best
[103] F. Foukalas, V. Gazis, and N. Alonistioti, ‘‘Cross-layer design proposals Masters Student Award, and the Best Supervisor Award at UMP, awards in
for wireless mobile networks: A survey and taxonomy,’’ IEEE Commun. international exhibitions, including the Diamond and Gold in BiS’17 U.K.,
Surveys Tuts., vol. 10, no. 1, pp. 70–85, 1st Quart., 2008. the Best of the Best Innovation Award, and the Most Commercial IT
[104] N. Gupta and S. K. Dhurandher, ‘‘Cross-layer perspective for channel Innovation Award, Malaysia, and the Gold and Silver medals in iENA’17,
assignment in cognitive radio networks: A survey,’’ Int. J. Commun. Syst., Germany. He has coauthored of over 80 prestigious IEEE and Elsevier
vol. 33, no. 5, p. e4261, Mar. 2020. journals (e.g., the IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS, the IEEE
[105] S. Chen and A. M. Wyglinski, ‘‘Efficient spectrum utilization via cross-
TRANSACTIONS ON SERVICES COMPUTING, the IEEE Communications Maga-
layer optimization in distributed cognitive radio networks,’’ Comput.
zine, Journal of Network and Computer Applications (Elsevier), and Future
Commun., vol. 32, no. 18, pp. 1931–1943, Dec. 2009.
Generation Computing Systems (Elsevier)) and conference publications
[106] Y. Mhaidat, M. Alsmirat, O. S. Badarneh, Y. Jararweh, and
H. A. B. Salameh, ‘‘A cross-layer video multicasting routing protocol (e.g., IEEE Globecom and IEEE DASC) and has served as an Advisory Board
for cognitive radio networks,’’ in Proc. IEEE 10th Int. Conf. Wireless Member, an Editor of Computers (MDPI), a Lead Guest Editor of IEEE
Mobile Comput., Netw. Commun. (WiMob), Oct. 2014, pp. 384–389. ACCESS and Computers, an Associate Editor of IEEE ACCESS, an Organizing
[107] M. Zareei, A. K. M. M. Islam, N. Mansoor, S. Baharun, E. M. Mohamed, Committee, the Chair, the Publicity Chair, the Session Chair, a Programme
and S. Sampei, ‘‘CMCS: A cross-layer mobility-aware MAC protocol for Committee and a member of Technical Programme Committee (TPC) in
cognitive radio sensor networks,’’ EURASIP J. Wireless Commun. Netw., numerous leading conferences worldwide (e.g., IEEE Globecom, IEEE
vol. 2016, no. 1, pp. 1–15, Dec. 2016. DASC, and IEEE iSCI) and journals.

67866 VOLUME 8, 2020

Q. M. Salih et al.: Smart Routing Management Framework Exploiting Dynamic Data Resources

FADI AL-TURJMAN received the Ph.D. degree ZAFRIL RIZAL M. AZMI received the B.Sc.,
in computer science from Queen’s University, M.Sc., and Ph.D. degrees in computer science
Kingston, ON, Canada, in 2011. He is currently a from Universiti Teknologi Malaysia. He is
Full Professor with Near East University, Nicosia, currently a Senior Lecturer with the Faculty of
Cyprus, where he is also the Director of the Computing, Universiti Malaysia Pahang, and a
Research Center. He is a leading authority in the member of the Systems, Network and Security
areas of smart/intelligent, wireless, and mobile (SysNetS) Research Group. His research interests
networks’ architectures, protocols, deployments, include mobile computing and distributed systems.
and performance evaluation. His publication
history spans over 250 publications in journals,
conferences, patents, books, and book chapters, in addition to numerous
keynotes and plenary talks at flagship venues. He has authored and edited
more than 25 books about cognition, security, and wireless sensor networks’
deployments in smart environments, published by Taylor & Francis, Elsevier,
and Springer. He has received several recognitions and best papers’ awards at
top international conferences. He also received the prestigious Best Research
Paper Award from Computer Communications (Elsevier) journal for the
period 2015–2018, in addition to the Top Researcher Award for 2018 at
Antalya Bilim University, Turkey. He has led a number of international
symposia and workshops in flagship communication society conferences.
He currently serves as an Associate Editor and the Lead Guest/Associate
Editor for several well reputed journals, including the IEEE COMMUNICATIONS
SURVEYS AND TUTORIALS (IF 22.9) and Sustainable Cities and Society (Else-
vier) (IF 4.7).

VOLUME 8, 2020 67867