Received September 21, 2021, accepted September 30, 2021, date of publication October 4, 2021, date of current version

October 12, 2021.

Digital Object Identifier 10.1109/ACCESS.2021.3117836

Restricted Access Window-Based Resource

Allocation Scheme for Performance
Enhancement of IEEE 802.11ah
Multi-Rate IoT Networks
SRI PAVAN BADARLA , (Associate Member, IEEE),
AND V. P. HARIGOVINDAN , (Senior Member, IEEE)
Department of Electronics and Communication Engineering, National Institute of Technology Puducherry, Karaikal, Puducherry 609609, India
Corresponding author: Sri Pavan Badarla ([email protected])

ABSTRACT IEEE 802.11ah (Wi-Fi HaLow) is introduced to support Internet of things (IoT) applications.
In this paper, we consider the IEEE 802.11ah multi-rate IoT network where stations (STAs) contend for
channel access using group-based restricted access window (RAW) mechanism with the default uniform
grouping scheme. RAW divides the channel time into various slots, STAs into different groups and assigns
a dedicated RAW slot to each group of STAs. As uniform grouping scheme performs grouping on random
basis, each group will comprise of distinct data rate STAs, which leads to severe degradation of throughput
because of performance anomaly. We propose a novel RAW based resource allocation (RAW-RA) scheme
which groups the STAs in a multi-rate IoT network based on data rates and allocates the RAW slots to
the groups proportional to their data rates. We also develop an accurate analytical model with unsaturated
traffic conditions for calculating the throughput and energy efficiency of the IEEE 802.11ah multi-rate IoT
network with the RAW mechanism. From the analytical results, it is evident that the proposed RAW-RA
scheme can resolve the performance anomaly as well as significantly improves the aggregate throughput
and energy efficiency of the IoT network. Finally, all the analytical results are validated with extensive
simulation studies.

INDEX TERMS Internet of Things, IEEE 802.11ah, performance anomaly, restricted access window based
resource allocation, Wi-Fi HaLow.

I. INTRODUCTION users or STAs which make the network denser with challeng-
The technological advancements in the networking field are ing tasks [2].
evolving rapidly; as a consequence, stations (STAs) are able Wireless local area networks (WLANs) have gained a lot
to access the Internet everywhere at high speed. This con- of attention in recent years for providing Internet services
tributes to the rise in number of connected users to the Inter- compared to cellular networks. Despite WLANs are grow-
net. The Internet of things (IoT) is a powerful paradigm in ing more popular in license-free bands, due to the widely
the modern world for interconnecting physical objects/things deployed Wi-Fi access points (APs) for Internet access [3].
which are mostly energy constrained, for the applications The IEEE 802.11 WLAN provides specifications for physical
such as smart grid, smart city, smart agriculture, smart home, and medium access control (MAC) layers. The STAs at the
and multimedia, etc. [1]. IoT enables objects with intelligence physical layer of IEEE 802.11 WLAN exhibit multiple data
for sensing, computation, as well as information sharing rates over the AP’s coverage area. The STAs closer to the
through the Internet at anytime, anywhere, and for any ser- AP attain higher signal strength and the STAs far from the
vice. The applications like e-Learning, e-Health, and smart AP exhibit poor signal strength depending on the distance
environment lead to an increase in the number of connected from the AP. Thus, based on the signal-to-noise ratio (SNR),
the STAs experience higher data rates when they are closer
The associate editor coordinating the review of this manuscript and to the AP and lower data rates when they are far from the
approving it for publication was Stefano Scanzio . AP [4]. Using rate adaptation, the STAs can transmit the data

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
VOLUME 9, 2021 136507
 S. P. Badarla, V. P. Harigovindan: RAW-RA Scheme

at various transmission rates to the AP. Besides, the MAC window (RAW) mechanism is introduced in the IEEE
layer protocols play an important role in determining net- 802.11ah standard [7]. Here, a specified group of STAs are
work performance. In the MAC layer, the STAs will contend only permitted for medium access within the allocated time.
using the fundamental channel access mechanism known as The RAW mechanism of IEEE 802.11ah is shown in Fig. 1.
distributed coordination function (DCF) which works based In this, AP periodically transmits multiple beacon frames to
on the carrier sense multiple access with collision avoidance every STA for network synchronization. In a beacon interval,
(CSMA-CA). When STAs encounter collision, they undergo there can be one or more RAW periods could be present.
exponential back-off (BO) process [5]. Here, DCF guarantees Under the RAW mechanism, the RAW period is divided into
equal channel access to all the STAs irrespective of their several RAW slots and the STAs are classified into various
data rates to achieve fairness. Initially, Wi-Fi networks were groups. Further, the RAW slots are assigned to the intended
designed for short distance communication with limited num- group of STAs. The group of STAs are allowed for channel
ber of STAs for Internet access to ensure high throughput. access within the allocated RAW slot, which result in reduced
Moreover, several traditional IEEE 802.11 standards operate collisions among the STAs. In this, the STAs can choose the
around ISM bands of 2.4 GHz and 5 GHz which make the deputed RAW slot using the expression φslot = (AIDn +
spectrum denser [6]. Therefore, the IEEE 802 working group Foffset ) mod K , where AIDn is the nth STA’s AID, Foffset is
introduced IEEE 802.11ah, as an amendment to the IEEE for STA fairness, and K represents number of RAW slots [7].
802.11 to support dense IoT networks [7]. It features low Here, beacon frames carry RAW parameter set-information
power operation, massive number of STAs, and long trans- element (RPS-IE) to all the associated STAs which consists
mission range. of RAW duration, each STA’s AID, initial and end time of the
corresponding RAW slot [10]. As shown in Fig. 1, the RAW
A. BRIEF DISCUSSION ON IEEE 802.11ah STANDARD slot is categorized into two parts. (1) Contention-based access
IEEE 802.11ah standard is primarily developed for dense IoT period (2) holding period. The STAs identify the deputed
networks. The standard operates in sub-1 GHz ISM bands RAW slot with the help of RPS-IE. Here, the RPS-IE consists
which are unlicensed excluding the white space bands of tele- of a cross slot boundary (CSB) field. If the CSB field is
vision (TV). Therefore, a critical issue of the crowded spec- equal to zero, then the STAs should originate and finish their
trum that arises in other amendments of IEEE 802.11 WLAN transmission within the RAW slot referred to as no-crossing
could be minimized in IEEE 802.11ah [8]. In addition, case (NCR). Whereas, if the CSB field is set to one, the STAs
the standard provides various enhancements to the physical can initiate the transmission within the RAW slot and can be
and MAC layers, which are described as follows. permitted to continue their transmission after the boundary of
the RAW slot known as crossing case [11].
1) IEEE 802.11ah PHYSICAL LAYER CHARACTERISTICS
The physical layer of the IEEE 802.11ah standard is inherited
from the IEEE 802.11ac and is ten times down-clocked to
provide channel bandwidths of 2, 4, 8, and 16 MHz. Besides,
1 MHz channel is additionally introduced in IEEE 802.11ah
for the extended range [7]. The standard exploits the orthogo-
nal frequency division multiplexing (OFDM) technique at the
physical layer. To support multiple data rates, the standard
affords different modulation and coding schemes (MCSs)
based on the channel conditions. Here, every STA can trans-
mit at a particular data rate with the help of different MCSs
(MCS0-MCS9) using rate adaptation. For extended coverage, FIGURE 1. RAW mechanism of IEEE 802.11ah.
MCS10 is introduced in 1 MHz channel with the BPSK
modulation [9].
B. PERFORMANCE ANOMALY IN IEEE 802.11 MULTI-RATE
2) IEEE 802.11ah MAC LAYER CHARACTERISTICS NETWORKS
Several MAC layer enhancements are introduced in IEEE In IEEE 802.11 multi-rate networks, the STAs which are
802.11ah such as improved channel access mechanism, closer to the AP attain a higher rate due to the good chan-
power-save (PS) methods, and scalability to support large nel conditions. These STAs use the minimum channel time
number of STAs. To coordinate the large number of STAs, for transmission of information once they acquire channel
IEEE 802.11ah uses a hierarchical structure of association access. Despite, due to the poor channel conditions, the STAs
identifiers (AIDs), where the AP assigns a unique AID to at the fringes of the AP’s coverage area exhibit lower data
every STA (typically 8192) [6]. In dense IoT networks, rates. To ensure fairness, IEEE 802.11 DCF guarantees equal
a high contention process is involved among the STAs for opportunity for all the STAs in the network irrespective of
channel access. As a consequence, the network performance their data rates [5]. As a result, the lower data rate STAs take
will be degraded. Hence, a group-based restricted access longer transmission time once they acquire channel access.

136508 VOLUME 9, 2021

S. P. Badarla, V. P. Harigovindan: RAW-RA Scheme

Thus, the higher data rate STAs are unable to achieve high suggested two grouping based MAC protocols that outper-
throughput because the medium is occupied by the lower data form the legacy DCF in dense networks. In their study,
rate STAs for the majority of the time. Therefore, it penalizes the authors modified the Markov-chain model by adding
the throughput performance of higher rate STAs and degrades busy channel probability. In [13], the authors proposed a
the throughput down to that of lower rate STAs. This leads service differentiation scheme using GS-DCF to provide dif-
to performance anomaly in the network [12]. Earlier, IoT ferentiated services. The authors in [17] proposed a model
networks were designed for sending small data packets with to improve the throughput of IEEE 802.11ah with the UG
infrequent transmissions. Nowadays, new emerging data- scheme. The authors of [18] presented a DTMC model for
intensive IoT applications such as acoustic sensing, telep- IEEE 802.11ah network to evaluate the average aggregate
resence robots, wireless capsule endoscopy, and unmanned throughput with network simulations. In [19], the authors
aerial vehicles require higher data rates. At present, the IoT developed an analytical model based on a novel BO algorithm
scenario is a multi-rate scenario where different STAs has to compute the IEEE 802.11ah saturation throughput. The
different data rate requirements [13]. In IEEE 802.11ah multi- authors of [20] implemented an analytical model to improve
rate network, the STAs contend for channel access using the throughput of IEEE 802.11ah WLAN. In [21], the authors
enhanced distributed channel access (EDCA) mechanism developed a mathematical model which is suitable for the IoT
which is based on the DCF. Besides, the standard draft specif- scenario according to the RAW mechanism. Here, the system
ically does not mention any scheme for grouping. Hence, performance was optimized by finding the optimal RAW
by default, a uniform grouping (UG) scheme is used [7]. parameters. The authors of [22] implemented a DTMC model
UG scheme is a group-based scheme that randomly forms with reset and renewal of BO to assess the performance
groups of equal size. In this paper, we propose a novel scheme of IEEE 802.11ah MAC layer for differentiated quality-of-
to resolve the performance anomaly in IEEE 802.11ah multi- service (QoS), without degrading the network performance.
rate IoT networks because of the distinct data rate STAs In [23], the authors proposed a QoS-aware priority grouping
present in the network, due to varying channel conditions. for real-time traffic conditions in IEEE 802.11ah network
The remaining portions of the paper are structured as fol- using a RAW scheduling scheme. It mitigates the collisions
lows: Section II presents the related research work. Section III and provided the required bandwidth rare but priority to the
describes the architecture of IEEE 802.11ah multi-rate IoT critical traffic STAs.
network and achievable data rate analysis. Section IV pro- As the IEEE 802.11ah standard promoted for low power
vides a detailed discussion of analytical model developed applications, the authors of [24]–[27] evaluated the energy
which is based on the discrete-time Markov-chain (DTMC) efficiency of the IEEE 802.11ah network. In [24], the authors
to evaluate the throughput and energy efficiency of the IEEE proposed an optimal traffic sensor grouping scheme for the
802.11ah multi-rate IoT network. Section V shows the exis- IEEE 802.11ah network. They developed a heuristic traffic
tence of performance anomaly in IEEE 802.11ah multi- sensor mapping algorithm to achieve sub-optimal energy
rate IoT network. Section VI presents the proposed RAW efficiency with max-min fairness among groups. Using the
based resource allocation (RAW-RA) scheme to resolve DTMC model of [5], the authors in [25] provided a simple
the anomaly and to improve the performance of the IEEE analytical model where the RAW mechanism gave better
802.11ah multi-rate IoT network. Section VII provides the throughput and energy efficiency compared to legacy DCF.
analytical and simulation results. Section VIII describes The authors of [26] proposed a novel medium access mech-
conclusion. anism to mitigate the energy consumption in IEEE 802.11ah
network. Here, it estimates the service interval and schedule
II. RELATED WORK the subsequent frames by eliminating the contention. In [27],
The performance analysis of IEEE 802.11ah had been exten- the authors presented a novel directional MAC scheme using
sively studied in the literature [10], [11], [13]–[27]. In [10], PS mechanism to improve the throughput and reduce the
[14], [15], the authors provided the various features of IEEE energy consumption for the IEEE 802.11ah network.
802.11ah for different IoT scenarios. In [5], the author pro- From the comprehensive survey of IEEE 802.11 multi-rate
posed a DTMC model to find the saturation throughput WLANs, a critical problem known as performance anomaly
of IEEE 802.11 WLAN using DCF. After his pioneering was identified in multi-rate networks that affects the through-
work, several authors had extended the DTMC model for put performance of high rate STAs [12]. Several studies
different versions of IEEE 802.11 WLAN supporting various were reported in the literature based on MAC parameters to
applications. The authors of [11], [13], [16]–[27] evaluated mitigate the performance anomaly in IEEE 802.11 multi-rate
the performance of IEEE 802.11ah network. A completely WLANs [28]–[31]. The authors of [28] proposed a differen-
new approach referred to as mean value analysis was used tiated reservation algorithm to minimize the collisions and
by the authors of [11] for the throughput performance of reduce the performance anomaly by sending multiple packets
IEEE 802.11ah network. Here, the average throughput, col- for one transmission opportunity. In [29], the authors used a
lision, and transmission probabilities were obtained by con- proportional fairness approach to overcome the anomaly. The
sidering group-synchronized DCF (GS-DCF). The authors authors of [30] proposed a solution to anomaly by control-
of [16] designed a smart metering network, in which they had ling the parameters such as frame size, initial and maximum

VOLUME 9, 2021 136509

 S. P. Badarla, V. P. Harigovindan: RAW-RA Scheme

BO window. To resolve the anomaly, the authors in [31] in conformance with IEEE 802.11ah standard and hence
proposed a MAC layer protocol for air-time fairness, where accurate.
each STA runs numerous instances of conventional DCF (2) We propose a novel RAW-RA scheme to resolve the
BO process. Targeting IEEE 802.11ah multi-rate networks performance anomaly as well as to improve the aggregate
[32], [33], the authors of [32] proposed an analytical model utility and energy efficiency of the IEEE 802.11ah multi-rate
where the STA after counting down to unity, checks for suffi- IoT network with proportional resource allocation.
cient residual time for transmission of the frame and in case The contributions are described as follows.
of insufficient residual time, STA enters the defer state. Also, • We develop an accurate analytical model where resid-
the authors proposed a data rate grouping (DRG) scheme to ual time is checked at each state of the DTMC with
solve the anomaly. In [33], the authors provided an analytical unsaturated traffic conditions to evaluate the throughput
model for IEEE 802.11ah multi-rate network and proposed and energy efficiency of the IEEE 802.11ah multi-rate
a novel grouping scheme to overcome the anomaly in dense IoT network which uses DCF with RAW mechanism for
networks. medium sharing [5], [20], [22], [34]. This model takes
In the RAW mechanism, each group of STAs assigned with into account the distinct features of the IEEE 802.11ah
a RAW slot. The STAs performing the BO decrement process RAW mechanism in multi-rate IoT networks where the
in the assigned RAW slot, has to verify at each state for suf- STAs communicate with different data rates.
ficient time to complete the frame transmission in the RAW • Using this model, we analyze the aggregate throughput
slot [7]. If the residual time is insufficient, then the STAs will of IEEE 802.11ah multi-rate IoT network with default
defer their attempt and wait for the next designated RAW UG scheme. From the results, we observe the signifi-
slot by entering into a defer state. In such a case, the STAs cant degradation of throughput because of performance
must reset the BO counter and restart from the start of the anomaly.
next allocated RAW slot. The analytical models developed by • We propose a novel RAW-RA scheme which groups
the authors of [11], [13], [16]–[20], [23]–[27], [33] did not the STAs in a multi-rate IoT network based on data
consider this important aspect of RAW mechanism, which rates and assigns RAW slots to the groups proportional
is the reset and renewal of BO process at the beginning to their data rates. We also analyze the performance
of RAW slot as per the IEEE 802.11ah standard [7]. Also, of RAW-RA scheme using the analytical model devel-
in [21] and [22], the authors proposed the analytical models oped with unsaturated traffic conditions. From the
for single-rate IEEE 802.11ah network; but neither of these results, we observe that RAW-RA scheme can resolve
two works provided a closed-form expression of throughput. the performance anomaly and significantly improves
In [32], the authors developed an analytical model which does the aggregate throughput as well as energy efficiency
not consider the checking of BO at each state in the DTMC of multi-rate IoT network compared to the DRG and
model of the IEEE 802.11ah RAW mechanism. Hence, these default UG schemes. The proposed scheme can be
models lack accuracy. implemented with lesser additional overhead because
In this paper, differently from the previous works, grouping is an imperative part of the IEEE 802.11ah
we develop an accurate analytical model using DTMC analy- RAW mechanism.
sis [5], [20], [22], [34] with unsaturated traffic conditions for • We have conducted extensive simulation studies using
IEEE 802.11ah, by considering the reset and renewal of BO network simulator (ns-3) which follows specifications of
counter at the RAW slot beginning as per the IEEE 802.11ah IEEE 802.11ah physical and MAC layers to corroborate
RAW mechanism [7]. The proposed model takes into account the analytical findings.
the reset of BO counter values at the end of the RAW slot
irrespective of their BO stage. The model checks the residual
time at each state and in case of insufficient residual time
for a frame transmission, the STA enters into a defer state.
After that, the STA will reset their BO counter values and
renew them from the initial BO at the beginning of the next
RAW slot. Thus the previous RAW slot counter values will
not be carried to the next RAW slot according to the RAW
mechanism. Hence, this model is accurate and completely in
accordance with the IEEE 802.11ah standard [7]. The novel
aspects of this research work are as follows.
(1) We develop an analytical model according to the IEEE FIGURE 2. Network architecture with different zones. Zone 1, which is
near to the AP consists of higher data rate STAs, while the zone c which is
802.11ah RAW mechanism using DTMC analysis with unsat- far from the AP comprises of lower data rate STAs.
urated traffic conditions. Unlike the existing models, here
residual time is checked at each state of the DTMC and III. NETWORK ARCHITECTURE
STA enters into a defer state in case of insufficient resid- In this article, we consider an IEEE 802.11ah multi-rate IoT
ual time for frame transmission. This model is completely network with an AP at the center of the coverage area and

136510 VOLUME 9, 2021

S. P. Badarla, V. P. Harigovindan: RAW-RA Scheme

N number of STAs which are uniformly deployed around the From the Eq. (2) and Eq. (4), the achievable data rate of the
AP as shown in Fig. 2. The STAs can communicate with the receiver STA in zone-x is computed as,
AP for uplink transmissions. At the IEEE 802.11ah physical  
Eb
layer, we consider c distinct data rates based on the average C(R(C
x
x ) ) = P +A +A −L(R(Cx ) )−(N )
t t r x 0 dB −F M − .
distance from the AP. The total coverage area of AP is divided N0 dB
into c zones with each of length wx . Each STA is at a distance (5)
(C )
Rx x |x ∈ [1, c] from the AP. Hence, the supported data
rates are 3c = {Cx |x ∈ [1, c]}. The zones are such that The achievable data rates for the corresponding zone
zone 1 is near the AP, whereas zone c is at the fringes of the lengths are shown in Table 1.
AP. Further, the STAs near the AP exhibit a higher data rate,
while the STAs at the fringes of the AP support lower data IV. THROUGHPUT ANALYSIS OF IEEE 802.11ah
rate [4]. As the IEEE 802.11ah physical layer affords multiple MULTI-RATE IoT NETWORK
data rates, each STA in a specific zone can transmit with the In this section, we develop an analytical model with unsat-
appropriate MCS using rate adaptation technique. urated traffic conditions to determine the throughput and
energy efficiency of the IEEE 802.11ah multi-rate IoT net-
A. COMPUTATION OF ACHIEVABLE DATA RATE ANALYSIS work, which uses DCF with RAW mechanism for medium
sharing. We assume that no hidden STAs present in the net-
The analysis of achievable data rate is provided in this sub-
work. From the network architecture discussed in Section III,
section for IEEE 802.11ah multi-rate network with macro
the STAs areP divided into c groups where the size of every
deployment scenario [33], [35]. Here, the link budget for
group is nx ( cx=1 nx = N ). Here, we consider each group
power received at the receiver STA of data rate Cx |x ∈ [1, c]
comprises an equal number of STAs and the STAs in each
can be expressed as,
group are of the same data rate. Further, the RAW slots are
x ) = Pt + At − L(Rx ) + Ar ,
Pr (R(Cx) (Cx )
(1) formed by dividing the RAW period into various slots [11],
with TRAW representing the RAW period and Ts,k gives the
where At denotes transmitter gain, Ar is receiver gain, length of the k th RAW slot. The duration of all the RAW slots
(C )
Pt is transmitter power, and path loss L(Rx x ) = 8 + is equal.
(Cx )
37.6 log10 (Rx ). The power loss component known as fade
margin (FM ) exists in the network due to multipath fading. A. MARKOV-CHAIN MODEL
The power obtained in dB at the receiver STA of data rate
This subsection presents an accurate analytical model which
Cx |x ∈ [1, c] in zone-x considering Rayleigh-fading channel
is developed with unsaturated traffic conditions to evaluate
between the AP and STAs is given by,
the transmission probability of the x th group STA of data rate
x ) = Pt + At − L(Rx ) + Ar − FM .
Pr (R(Cx) (Cx )
(2) Cx |x ∈ [1, c] using DCF with the RAW mechanism [5], [20],
[22], [34]. Here, the channel time is divided into mini-slots of
duration σm . When the STA is ready to transmit a frame, it lis-
TABLE 1. Achievable data rates with zone lengths. tens to the channel for distributed inter-frame space (DIFS)
duration and contends for channel access by invoking the
exponential BO procedure in the deputed RAW slot. During
the BO, the STAs can select contention window (CW) in the
range [0, Wk,0 − 1], where Wk,0 is the minimum CW. The
(C ) (C )
two-dimensional Markov-chain model {ak x (t), bk x (t)} is
The fade margin with respect to link outage is expressed as, shown in Fig. 3 with a queue empty state I of probability
FM = −10 log10 (− ln(1 − Pout )), where Pout is the outage q0,k which follows Poisson process with packet arrival rate
probability. For the required level of link-outage Pout = of 100 packets/second [34]. Let the stochastic process of
0.1 [35]. Furthermore, the SNR at the receiver STA is given (C ) (C )
BO stage is ak x (t), and bk x (t) be the stochastic process
by, corresponds to the BO counter that is evenly selected over
S Eb 1 CEb the interval [0, Wk,0 − 1]. Here, the discrete mini-slot times t
= × = , (3) and t + 1 represent the beginning of two consecutive BO of
N N 0 B tb kT0 FB
the STA. The duration of a mini-slot forms a random variable
where S is the power obtained at the receiver. According to (Cx )
that could be an ideal/successful/collision slot. Let pcn,k is the
Eq. (3), Eb represents energy consumed per bit, N0 is the noise (Cx )
spectral density, tb is bit duration, B is bandwidth, C is the probability of collision and τk is the transmission probabil-
data rate, k represents the Boltzmann constant, F is the noise ity of the x th group STA with data rate Cx |x ∈ [1, c]. These
figure, and receiver temperature in Kelvin is T0 . The Eq. (3) probabilities can be represented by Markov-chain model. Let
can be expressed in dB as, the transmission duration of a frame for the STA with data
      rate Cx |x ∈ [1, c] is given by,
S Eb C
= + . (4) (C )
N dB N0 dB B dB T (Cx ) = TE[P]
x
+ TSIFS + Tack + σp , (6)

VOLUME 9, 2021 136511

 S. P. Badarla, V. P. Harigovindan: RAW-RA Scheme

FIGURE 3. Markov-Chain model for IEEE 802.11ah RAW mechanism.

x (C )
where payload duration is TE[P] , TSIFS represents the length During the BO decrement process, at each state, the STA
of short IFS frame, Tack is the acknowledgement (ACK) checks the residual time in the deputed RAW slot. When
frame duration, and the propagation delay is σp . As shown the residual time is not adequate for frame transmission dur-
in Fig. 1, AP specifies holding period at the end of every ing j ∈ [2, Wk,i − 1], then the x th group STA with data
(Cx )
RAW slot for NCR case [11]. The holding period is given by rate Cx |x ∈ [1, c] enters into the deferred state is pde,k .
Th = T (Cx ) −σm +σg , where σg is guard band interval. Within Here, the residual time is distributed evenly over the interval
the RAW slot, the initiation of any transmission should end (0, Ts,k ). The probability of x th group STA with data rate
before the holding period. As illustrated in Fig. 1, the effective Cx |x ∈ [1, c] entering into deferred state is given by,
RAW slot duration for contention is Ts0 ,k = Ts,k − Th . Here,  (Cx )
the group of STAs access the channel in the deputed RAW T ; T (Cx ) ≤ Ts,k
(Cx )
slot using EDCA which is based on DCF. As the RAW slot pde,k = Ts,k (8)
is organized in contention-based access period and holding T (Cx ) > Ts,k

1;
period, the total duration of RAW slot can be expressed as
Ts,k = Ts0 ,k + Th . Here, the duration of a RAW slot is Similarly, when the BO counter reaches one, it decrements
chosen such that, it must be at least as long as the transmission to zero, if the holding period in the deputed RAW slot is
duration. sufficient for the frame transmission. In case of the residual
During the BO process, the STA decrements the BO time is insufficient for successful frame transmission, then
counter in the deputed RAW slot when the channel is the STA moves into the deferred state as depicted in Fig. 3.
idle; otherwise, it freezes the counter. The x th group STA’s Further, the STA will reset their BO functions and renew
BO counter would be frozen due to the following aspects. them from the initial BO for the next deputed RAW slot
(1) When the channel is sensed busy due to the transmission arrival. When the BO counter reaches one, let the probability
of another STA of the same group. (2) The contention-based of x th group STA with data rate Cx |x ∈ [1, c] entering
(Cx )
access period of the deputed RAW slot is expired. Thus, into deferred state be pdh,k . We assume that the residual time
the probability of BO counter will be frozen is given by, in the contention-based access period of the deputed RAW
slot is distributed evenly over the interval (0, Ts0 ,k ) [32]. It is
(C ) Ts0 ,k  (Cx ) (n(Cx ) −1)
 given by,
pfn,kx = 1 − (1 − τk )
k
TRAW 
(Cx )
T
 
; T (Cx ) ≤ Ts0 ,k
c

Y (Cp ) (Cp ) (TRAW − Ts0 ,k ) (Cx )
Ts0 ,k
× (1 − τk ) nk + . (7) pdh,k = (9)
TRAW 
 1; T (Cx ) > Ts0 ,k
p=1,p6=x

136512 VOLUME 9, 2021

S. P. Badarla, V. P. Harigovindan: RAW-RA Scheme

The probability of x th group STA transit from the deferred of STA moves into the deferred state from the BO decrement
(C ) Ts,k
state be pr,kx = TRAW , and the probability of STA being in state for j ∈ [2, Wk,i − 1], i ∈ [0, m]; and Eq. (10i) signifies
(C ) s,k T the probability of STA moves into the deferred state from the
the deferred state is pd,kx = 1 − TRAW . Furthermore, a frame
transmission occurs when the BO counter approaches zero BO decrement state for j = 1, i ∈ [0, m]; Eq. (10j) gives the
irrespective of their BO stage. For every successful trans- probability of STA being in the queue empty state; Eq. (10k)
mission, the STAs are being acknowledged with the ACK represents the probability of STA moves to queue empty state
frame, and the BO counter reset to Wk,0 . Whenever a packet from the deferred state when the MAC layer buffer is empty;
collision occurs, the BO counter is doubled till the maximum Eq. (10l) accounts for the probability of STA moves into
CW. Here, we consider the maximum BO stage (m) is same initial BO stage when the MAC layer buffer is ready with a
as the retry limit.1 Hence, the CW for ith BO stage is given by frame.
Wk,i = 2i Wk,0 , i ∈ [0, m]. From the Markov-chain model, Stationary distribution of the Markov-chain model corre-
the probabilities corresponding to x th group STA with data sponds to the x th group STA with data rate Cx |x ∈ [1, c] is
rate Cx |x ∈ [1, c] can be expressed as, given by,
(C )
P(i, j | i, j) = pfn,kx ; j ∈ [1, Wk,i −1], i ∈ [0, m] (10a) (Cx ) (Cx ) (Cx )
( (C ) (Cx )
bk (i, j) = lim P{ak (t) = i, bk (t) = j};
1−pfn,kx −pde,k ; j ∈ [2, Wk,i −1] t→∞
P(i, j−1 | i, j) = (Cx ) (Cx ) (10b) j ∈ [0, Wk,i − 1], i ∈ [0, m] (11)
1−pfn,k −pdh,k ; j = 1, i ∈ [0, m]
(C )
x
pcn,k Under steady-state conditions, the following relations
P(i, j | i−1, 0) = ; j ∈ [0, Wk,i −1], i ∈ [0, m] (10c)
Wk,i obtained by solving the Markov-chain model.
x (C )
(1−pcn,k )q0,k
P(0, j | i, 0) = ; j ∈ [0, Wk,0 −1], (Cx ) (C )
x i (Cx )
Wk,0 bk (i, 0) = (pcn,k ) bk (0, 0); i ∈ [0, m] (12)
i ∈ [0, m−1] (10d) x i (C )
x (C )
q0,k (Cx ) Wk,i − j (pcn,k ) bk (0, 0)
P(0, j | m, 0) = ; j ∈ [0, Wk,0 −1], i ∈ [0, m] (10e) bk (i, j) = ;
Wk,0 Wk,i 1 − p(Cx ) − p(Cx )
fn,k de,k
x (C )
P(I | i, 0) = (1−q0,k )(1−pcn,k ); i ∈ [0, m] (10f) j ∈ [2, Wk,i − 1], i ∈ [1, m] (13)
(C )
P(Di,j | Di,j ) = pd,kx ; j ∈ [1, Wk,i −1], i ∈ [0, m] (10g) (Cx ) i (Cx )
(pcn,k ) bk (0, 0)
(Cx ) Wk,i − 1
x(C )
P(Di,j | i, j) = pde,k ; j ∈ [2, Wk,i −1], i ∈ [0, m] (10h) bk (i, 1) = (C ) (Cx )
; i ∈ [1, m] (14)
Wk,i 1 − pfn,kx − pdh,k
x (C )
P(Di,1 | i, 1) = pdh,k ; j = 1, i ∈ [0, m] (10i) x i (C ) (C ) !
x
Wk,i − j (pcn,k ) bk (0, 0) pde,k
P(I /I ) = (1−q0,k ); (10j) (C )
bk x (Di,j ) = ;
(C ) Wk,i 1 − p(Cx ) − p(Cx ) (C )
pr,kx
P(I | Di,j ) = pr,kx (1−q0,k ); i ∈ [0, m] (10k) fn,k de,k
q0,k j ∈ [2, Wk,i − 1], i ∈ [0, m] (15)
P(0, j | I ) = ; j ∈ [0, Wk,0 −1] (10l)
Wk,0 (Cx ) i (Cx ) (Cx ) !
(Cx ) Wk,i − 1 (pcn,k ) bk (0, 0) pdh,k
Here, Eq. (10a) represents the freezing probability of bk (Di,1 ) = (C ) (Cx ) (C )
;
the x th group STA when the channel is busy; Eq. (10b) Wk,i 1 − pfn,kx − pdh,k pr,kx
signifies the probability of BO counter j decrements for j = 1, i ∈ [0, m] (16)
(C ) (Cx )
j ∈ [2, Wk,i − 1] with probability 1 − pfn,kx − pde,k , while (C ) Wk,i − j
the BO counter reaches to zero by means of probability bk x (0, j) = (C ) (Cx )
× G; j ∈ [2, Wk,i − 1],
(C ) (Cx ) 1 − pfn,kx − pde,k
1 − pfn,kx − pdh,k ; Eq. (10c) represents the probability of
STA entering the subsequent BO stage when it experiences (17)
collision; Eq. (10d) depicts the probability of STA reaches (Cx ) Wk,i − 1
the BO stage zero when it successfully transmits, on a con- bk (0, 1) = (C ) (C )
× G; j=1 (18)
1 − pfn,kx − pdh,k
x
dition that the frame is ready in MAC layer buffer; Eq. (10e)
accounts for the probability that, after the maximum num-
ber of transmission attempts, the STA moves into BO stage where
zero under the condition of MAC layer buffer nonempty;
Eq. (10f) gives the probability of STA moves into queue
 (C )

x X
q0,k (1−pcn,k ) m−1 (C ) q0,k (Cx )
empty state after the successful transmission when buffer is b x (i, 0)+ b (m, 0) 
i=0 k Wk,0 k

empty; Eq. (10g) represents the probability of STA remains
 Wk,0 

G= pde,k q0,k Xm XWk,i −1 (Cx ) .

in the deferred state; Eq. (10h) accounts for the probability  + bk (i, j) 
 Wk,0 X i=0 j=2 
1 Here, the number of BO stages m = 7 is the retry limit and minimum
 pdh,k q0,k m (C ) q 0,k 
+ bk x (i, 1) + bk (I )
CW=32. Wk,0 i=0 Wk,0

VOLUME 9, 2021 136513

 S. P. Badarla, V. P. Harigovindan: RAW-RA Scheme

Under the steady-state probabilities, the normalized condi- where packet error rate er = 0.1. The probability that the
tion is given by, x th group STA with data rate Cx |x ∈ [1, c] succeeds its
k,i −1
m WX m m
transmission in the mini-slot of the deputed k th RAW slot is
X
bk
(Cx )
(i, j) +
X
bk
(Cx )
(i, 1) +
X
bk
(Cx )
(i, 0) given by,
i=0 j=2 i=0 i=0 (C ) 1 
(C ) (C ) (C ) (Cx )

ps,kx = (C ) × nk x τk x (1 − τk x )nk −1
k,i −1
m WX m ptr,kx
(Cx ) (Cx )
X X
+ bk (Di,j ) + bk (Di,1 ) + bk (I ) = 1 (19) 
c

(Cp )
Y (C )
i=0 j=2 i=0
× (1 − τk p )nk (1 − er ) , (23)
After substituting and simplifying the equations from (12) p=1,p6=x
(C )
to (18) in the Eq. (19), the expression for bk x (0, 0) is given where the probability of at least one STA transmits with data
by, rate Cx |x ∈ [1, c] in the mini-slot of the deputed k th RAW
(Cx ) 1 slot is expressed as,
bk (0, 0) = 1−q0,k ! ,
1 + Y1 + Y2 + Y3 + Y4 + c
q0,k + (C )
Y (Cp ) n(Cp )
1−q0,k ptr,kx = 1 − (1 − τk k ) . (24)
(Y5 + Y6 )(α + β + q0,k ) p=1
(20)
The sum of a T (Cx ) , TDIFS , and the number of BO slots
where spent prior to the transmission would account for a transac-
x(C )
α pcn,k x m
(1 − (2pcn,k ) )Wk,0
(C ) tion. Let λb,ω implies the number of mini-slots spent before
Y1 =
x (C )
the ωth transmission and δω is the length of ω-transactions.
1 − 2pcn,k Thus, the probability of δω in the deputed RAW slot that
(C ) (C ) utilizes the λ mini-slots [11] is expressed as,
x
pcn,k x m
(1 − (pcn,k ) )(1 − 1.5α + β)
Y2 = (C )
 0
λ − ωT − 1
 0
(ptr,kx )ω (1 − ptr,kx )λ−ωT −ω ,
x
1 − pcn,k (C ) (C )
Pδω (λ) =
λ − ωT − ω
0
(C ) (C )
x
pcn,k (2m − (pcn,k
x m
) )(α − β) (25)
Y3 = (C ) 0
2m (2 − pcn,k
x
)Wk,0 where T = T (Cx ) + TDIFS . Let the random variable H be
the total number of transactions during Ts0 ,k . The probabil-
((Wk,0 )2 − 3Wk,0 + 2)(α + 2β)
Y4 = ity of ω-transactions in the deputed k th RAW slot can be
Wk,0
expressed as,
(C ) (C )
x m+1
0.5γ (1 − (2pcn,k ) ) x m+1
(1 − (pcn,k ) )(1.5γ − η) T 0 −(ω−1)T
0
−TDIFS −σm ( ω
s ,k
)
Y5 = (Cx )
− (Cx )
X X
(1 − 2pcn,k ) (1 − pcn,k )Wk,0 PH,k (ω) = Prob λb,k = x .
(Cx ) m+1 x=ω k=1
(2m+1 − (pcn,k ) )(γ − η)
Y6 = (26)
m 2 (Cx )
2 (Wk,0 ) (2 − pcn,k )
The maximum number of transmissions during Ts0 ,k can be
(C )
x /p x ) (C ) (C )
x /p x ) (C ) T 0
1+(pde,k 1+(pdh,k s ,k
also, α = r,k
, β = r,k
, γ = expressed as NT = b 0 c. Therefore, the average value of
(C ) (Cx )
1−pfn,kx −pde,k
(C ) (Cx )
1−pfn,kx −pdh,k T +σm
(Cx ) (Cx ) H during the deputed k th RAW slot can be expressed as,
pde,k pdh,k
(Cx ) (Cx ) , and η = (Cx ) (Cx ) . NT (Ts,k −Th )
1−pfn,k −pde,k 1−pfn,k −pdh,k X
The transmission probability of the x th group STA with Ek [H] = ωPH,k (ω). (27)
data rate Cx |x ∈ [1, c] in the mini-slot of the deputed ω=1
k th RAW slot can be computed as, Therefore, the average throughput of the x th group STA
m x m+1
1 − (pcn,k )
(C ) with data rate Cx |x ∈ [1, c] in the deputed k th RAW slot is
(Cx ) (Cx ) (Cx )
X
τk = bk (i, 0) = (C )
bk (0, 0). (21) given by,
x
i=0 1 − pcn,k (C )
(C ) c E[P] Ek [H] ps,kx
The probability of the x th
group STA with data rate UTk x = . (28)
TRAW
Cx |x ∈ [1, c] which suffers from the collision in the mini-
slot of the deputed k th RAW slot is expressed as, By considering the UG scheme, the aggregate through-
put of a group having distinct data rate STAs of data rate
(C ) (Cx ) n(Cx ) −1
Cx |x ∈ [1, c] in the deputed k th RAW slot is given by,
x
pcn,k = (1 − (1 − τk k) )
c c
(Cp ) n(Cp ) (Cx )
X
.
Y
× (1 − τk )
k (1 − er ), (22) UTk = UTk (29)
p=1,p6=x x=1

136514 VOLUME 9, 2021

S. P. Badarla, V. P. Harigovindan: RAW-RA Scheme

The data transferred by the x th group STA with data rate Using the equations from (32), (35), (36), and (38),
Cx |x ∈ [1, c] can be computed as, the energy efficiency of the network can be computed based
(C ) on Eq. (31).
(Cx ) c E[P] Ek [H] ps,kx  
Dk = Ts0 ,k + Th . (30)
TRAW V. PERFORMANCE ANOMALY IN IEEE 802.11ah
MULTI-RATE IoT NETWORK
B. ENERGY EFFICIENCY ANALYSIS
In this section, we consider an AP whose coverage area is
The energy efficiency is the amount of energy required to
classified into five different zones. The zones are positioned
successfully transmit the data bits [36], [37]. During the DCF
in such a way that, zone 5 is at the fringes of the AP having
mechanism, the STA can be in one of the following states such
STAs of lowest data rate. While, zone 1 is closer to the AP
as decrementation, freezing of BO counter, successful trans-
with the STAs of highest data rate as depicted in Table 1 of
mission or collision. The STA with data rate Cx |x ∈ [1, c]
(C ) Section III. To observe the existence of performance anomaly,
that can consume energy in the BO procedure is EB,kx , freez-
(C ) (C ) we have examined two cases. In the case-1, we consider
ing state is EF,kx , successful transmission is Es,kx , and the the UG scheme of group size nx = 200 which comprises
(Cx )
collision process is Ecn,k . Hence, the energy efficiency can of 40 distinct data rate STAs from each of the five zones.
be expressed as, Further, we have computed the aggregate throughput using
(C ) (C )
EB,kx + EF,kx + Es,kx + Ecn,k
x (C ) (C ) Eq. (29). In case-2, we consider the STAs with a group size
(Cx )
ET = . (31) of nx = 200 of lowest data rate alone (0.3 Mbps), and we
E[P] compute the throughput using Eq. (28). The results of case-1
During the BO counter decremented process, the energy and case-2 have been tabulated in Table 2. From Table 2,
consumed is given by, we have observed the throughput of the group of distinct
m r
Wk,u − 1 data rate STAs (case-1) is degraded close to the lowest data
(C ) (Cx ) (Cx ) r
X X
EB,kx = (1 − pcn,k )(pcn,k ) σm Pid , (32) rate group STAs (case-2). This is because of performance
2
r=0 u=0 anomaly. Since all the distinct data rate STAs in the UG
here Pid is the idle power. In the deputed RAW slot, the prob- scheme have an equal chance of accessing the channel, which
(C ) result in lowest data rate STAs use the channel most of the
ability of one of the (nk x −1) STAs transmit successfully can
be expressed as, time. This penalizes the channel time of highest rate STAs
(Cx ) (Cx ) (Cx ) (n(Cx ) −2)
and the throughput performance degrades because of inap-
(C ) (nk − 1)τk (1 − τk ) k
propriate usage of channel resources.
ps0 ,kx = . (33)
(C ) (Cx )
1 − (1 − τk x )(nk −1)
TABLE 2. Throughput for case-1 and case-2.
In the BO process, the number of times that the transmis-
sions overhear by the other STAs is given by,
(Cx ) Pm (Cx ) (Cx ) r Pr Wk,u −1
pcn,k r=0 (1 − pcn,k )(pcn,k ) u=0 2
Q= (Cx )
. (34)
1 − pcn,k
The energy consumed by a STA in the BO process due to
overhear from the other STAs is expressed as, VI. RESOLVING PERFORMANCE ANOMALY AND
(C ) (C ) (C ) RESOURCE ALLOCATION: RAW-RA SCHEME
EF,kx = Q[1cn (1 − ps0 ,kx ) + σm ps0 ,kx ]Pid , (35)
In this section, we propose a RAW-RA scheme to resolve
where 1cn = Tpoll + TDIFS + σp is duration of collision the performance anomaly as well as to improve the aggre-
process. Here, Tpoll is PS-Poll frame duration. The energy gate utility of the IEEE 802.11ah multi-rate IoT network.
consumed during the successful transmission is given by, The proposed scheme groups the STAs based on the data
(C ) (C ) (C ) rates and allows them to contend in the deputed RAW slot.
Es,kx = (Tpoll + TE[P]
x x
)Pt + (1s − Tpoll − TE[P] )Pr , (36)
As illustrated in Algorithm 1, the proposed RAW-RA scheme
(C )
where successful transmission duration 1s = Tpoll + TE[P] x
+ is described as follows. Initially, at regular intervals, AP trans-
2Tack + 3TSIFS + TDIFS + 3σp . The average number of mits multiple beacons to every STA in the network. After
transmission attempts due to the collisions prior to the success hearing the beacon frame, every STA computes its data rate
can be computed as, as stated in Table 1 of Section III before the association
m process. The achievable data rate of each STA is shared with
(C ) (C ) the AP through the rate bits of the physical layer convergence
X
Qcn = x
r(1 − pcn,k x r
)(pcn,k ) . (37)
r=0
protocol (PLCP) header during the association process. The
AP then classifies the STAs based on data rates and form a
The energy consumed because of the collision process is
RAW-RA matrix. Every row in the RAW-RA matrix represent
given by,
a group of the same data rate STAs which are formed into
(C )
x
Ecn,k = Qcn [(Tpoll )Pt + (TDIFS + σp )Pr ]. (38) c groups, and every STA is assigned with a unique AID by

VOLUME 9, 2021 136515

 S. P. Badarla, V. P. Harigovindan: RAW-RA Scheme

Algorithm 1 Proposed RAW-RA Scheme scheme. Here, the proposed scheme can be implemented very
1: Each STA initially share its achievable data rate with the easily without any significant modifications to the existing
AP; standard, because grouping is an imperative part of the RAW
2: A RAW-RA matrix is formed and initialized as null
matrix; mechanism in IEEE 802.11ah.
3: p ← 1, q ← 1;
4: while p ≤ 3c do
5: while q ≤ N do VII. RESULTS AND DISCUSSION
6: if nq transmits with data rate Cx |x ∈ [1, c] then This section presents analytical results that are validated
7: M [p][q] ← nq ; with extensive simulation studies. Simulations are performed
8: end if
9: q←q+1 using ns-3.29 [38]. We consider an IEEE 802.11ah multi-rate
10: end while IoT network with uniformly deployed 1000 number of STAs
11: p←p+1
12: end while in the AP’s coverage area as shown in Fig. 2. The AP’s cov-
13: p ← 1, q ← 1; erage area is divided into five different zones corresponding
14: while p ≤ 3c do
15: while q ≤ N do to the data rates as provided in Table 1. The zone 1, which is
16: Each RAW-RA matrix’s row is formed as a group by near to the AP has highest data rate STAs, while the zone 5
the AP based on the same data rates and assigns a which is at the fringes of the AP has lowest data rate STAs.
group ID to each group;
17: q←q+1 The simulations are conducted for 100 seconds, 30 times with
18: end while a confidence interval of 95% and averaged to evaluate the
19: p←p+1 throughput. We have selected the system parameters based
20: end while/∗ Slot allocation procedure∗/
21: u ← 1; on the IEEE 802.11ah standard. The system parameters are
22: while u ≤ 3c do provided in Table 4.
23: φSlot−initial ← (AIDn + Foffset ) mod K ;
24: Slot_num ← φSlot−initial ;
25: y ← c; /∗ Temporary variable ∗/ TABLE 4. System parameters.
26: z ← Gid ;
27: while z < c do
28: Slot_num ← Slot_num + y;
29: y ← (y − 1);
30: A[Slot_num] ← Slot_num;
31: z←z+1
32: end while
33: u←u+1
34: end while
35: STAs in the respective group can choose any RAW slot of
A[Slot_num] & φSlot−initial ;

the AP. Here, the LSB bits of all STAs belong to a group are
the same. Therefore, AP can assign a group ID (Gid ) to each
group. Using RPS-IE, the details of assigned AIDs, RAW
initial, end time, and the number of groups are shared with
the STAs.

TABLE 3. RAW slots allocation based on data rates.

Furthermore, we allocate the channel resources to every Figure 4 shows the throughput performance of the pro-
group proportional to their data rates. We consider the entire posed RAW-RA scheme compared to the DRG scheme [32]
RAW duration has 15 RAW slots numbered from RAW slot 0 and default UG scheme. In the RAW-RA scheme, all the STAs
to RAW slot 14. The slot allocation procedure is provided in are divided into five groups based on the data rates using
Algorithm 1. Based on the slot allocation procedure, the STAs Algorithm 1. Here, we provide more number of RAW slots
in the group 1, which has the highest data rate can choose to high rate STAs. To illustrate this, the total RAW duration
the RAW slots of 0, 5, 9, 12, and 14 as shown in Table. 3. is divided into 15 RAW slots, and the RAW slots are allocated
Similarly, the STAs in the group 5 of lowest data rate can only proportional to their data rates as provided in Table. 3. Here,
choose the RAW slot 4. As shown in Table. 3, the throughput group 1 (zone 1) comprises of highest data rate STAs assigned
of the group of same data rate STAs is computed based with five RAW slots, whereas group 5 (zone 5) consists of
on the number of RAW slots assigned using the RAW-RA lowest data rate STAs allocated with one RAW slot. Similarly,

136516 VOLUME 9, 2021

S. P. Badarla, V. P. Harigovindan: RAW-RA Scheme

FIGURE 4. Throughput per station versus different zones. FIGURE 6. Data transferred per station versus different zones.

FIGURE 5. Aggregate throughput per zone versus different zones. FIGURE 7. Total data transferred per zone versus different zones.

that the data transferred per STA is considerably improved

in the UG scheme, the network is classified into five different compared to DRG and UG schemes. Figure 7 shows the
groups with an equal number of distinct data rate STAs from total data transferred (Kb) per zone for IEEE 802.11ah
each of the five zones. In the UG scheme, the slot allocation multi-rate IoT network. From Fig. 7, we can observe
procedure is uniform as it gives equal chance to every STA the proposed RAW-RA scheme substantially improves the
irrespective of their data rates. Hence, we provide every group total data transferred per zone over the DRG and UG
with one RAW slot of duration 100 ms. From Fig. 4, it is schemes.
evident that the proposed RAW-RA scheme resolves the per- Figure 8 shows the energy efficiency analysis with respect
formance anomaly and significantly improves the throughput to the zones. From Fig. 8, we can observe less energy con-
performance of the IEEE 802.11ah multi-rate IoT network. sumption using the proposed scheme when compared to the
In addition, we also compute the aggregate throughput per DRG and UG schemes. This is due to the STAs which are
zone. Figure 5 gives the aggregate throughput per zone for far from the AP have lower data rates. Thus, these STAs
the RAW-RA, DRG, and UG schemes. The findings show utilize the channel for longer time and consume more energy.
that a significant improvement in the aggregate throughput While, the STAs near the AP has a higher data rate, eventually
per zone can be observed with the proposed RAW-RA scheme these STAs use minimum channel time for transmission and
compared to the DRG and UG schemes. consume less energy. The energy efficiency is significantly
Figure 6 shows the data transferred (Kb) per STA for the better in the proposed scheme when compared to the DRG
IEEE 802.11ah multi-rate IoT network. The data transferred and UG schemes. By examining all the results, it is evi-
per STA is computed using Eq. (30). After grouping the STAs dent that the proposed RAW-RA scheme outperforms the
based on their data rates, the proposed RAW-RA scheme DRG and UG schemes in terms of throughput performance
allocates more channel resources to higher rate STAs (zone 1) and energy efficiency of the IEEE 802.11ah multi-rate
than lower rate STAs (zone 5). From Fig. 6, we can observe IoT network.

VOLUME 9, 2021 136517

 S. P. Badarla, V. P. Harigovindan: RAW-RA Scheme

[7] IEEE Standard for Information Technology—Telecommunications and

Information Exchange Between Systems—Local and Metropolitan Area
Networks—Specific Requirements—Part 11: Wireless LAN Medium Access
Control (MAC) and Physical Layer (PHY) Specifications Amendment 2:
Sub 1 GHz License Exempt Operation, Standard 802.11ah-2016, IEEE,
May 2017, pp. 1–594.
[8] B. S. Pavan, M. Mahesh, and V. P. Harigovindan, IEEE 802.11ah for
Internet Vehicles: Design Issues Challenges. Cham, Switzerland: Springer,
2021, pp. 41–61.
[9] F. Tramarin, S. Vitturi, and M. Luvisotto, ‘‘A dynamic rate selection
algorithm for IEEE 802.11 industrial wireless LAN,’’ IEEE Trans. Ind.
Informat., vol. 13, no. 2, pp. 846–855, Apr. 2017.
[10] E. Khorov, A. Lyakhov, A. Krotov, and A. Guschin, ‘‘A survey on IEEE
802.11ah: An enabling networking technology for smart cities,’’ Comput.
Commun., vol. 58, pp. 53–69, Mar. 2015.
[11] L. Zheng, M. Ni, L. Cai, J. Pan, C. Ghosh, and K. Doppler, ‘‘Performance
analysis of group-synchronized DCF for dense IEEE 802.11 networks,’’
IEEE Trans. Wireless Commun., vol. 13, no. 11, pp. 6180–6192, Nov. 2014.
[12] M. Heusse, F. Rousseau, G. Berger-Sabbatel, and A. Duda, ‘‘Performance
FIGURE 8. Energy efficiency versus different zones. anomaly of 802.11b,’’ in Proc. IEEE INFOCOM, vol. 2, Mar. 2003,
pp. 836–843.
[13] M. Mahesh and V. P. Harigovindan, ‘‘Restricted access window-based
novel service differentiation scheme for group-synchronized DCF,’’ IEEE
VIII. CONCLUSION Commun. Lett., vol. 23, no. 5, pp. 900–903, May 2019.
In this paper, we have developed an accurate analytical model [14] H. Wang and A. O. Fapojuwo, ‘‘A survey of enabling technologies of
low power and long range machine-to-machine communications,’’ IEEE
with unsaturated traffic conditions to compute the throughput Commun. Surveys Tuts., vol. 19, no. 4, pp. 2621–2639, 4th Quart., 2017.
and energy efficiency of the IEEE 802.11ah multi-rate IoT [15] S. Aust, R. V. Prasad, and I. G. M. M. Niemegeers, ‘‘Outdoor long-range
network under the RAW mechanism, considering the STAs WLANs: A lesson for IEEE 802.11ah,’’ IEEE Commun. Surveys Tuts.,
vol. 17, no. 3, pp. 1761–1775, 3rd Quart., 2015.
operate at distinct data rates. We have evaluated the aggregate
[16] Y. Yang and S. Roy, ‘‘Grouping-based MAC protocols for EV charging data
throughput using the default UG scheme and observed that transmission in smart metering network,’’ IEEE J. Sel. Areas Commun.,
the throughput has been degraded greatly because of perfor- vol. 32, no. 7, pp. 1328–1343, Jul. 2014.
mance anomaly. This is due to the inappropriate usage of [17] N. Nawaz, M. Hafeez, S. A. R. Zaidi, D. C. Mclernon, and M. Ghogho,
‘‘Throughput enhancement of restricted access window for uniform group-
channel resources by the lower rate STAs. Hence, we have ing scheme in IEEE 802.11ah,’’ in Proc. IEEE Int. Conf. Commun. (ICC),
proposed a novel RAW based resource allocation (RAW-RA) May 2017, pp. 1–7.
scheme for the IEEE 802.11ah multi-rate IoT network. With [18] S. M. Soares and M. M. Carvalho, ‘‘Throughput analytical modeling of
IEEE 802.11ah wireless networks,’’ in Proc. 16th IEEE Annu. Consum.
the proposed scheme, the STAs in a multi-rate IoT network Commun. Netw. Conf. (CCNC), Jan. 2019, pp. 1–4.
are grouped based on the data rates and the RAW slots are [19] A. J. Gopinath and B. Nithya, ‘‘Mathematical and simulation analysis of
allocated to each group proportional to their data rates. It is contention resolution mechanism for IEEE 802.11ah networks,’’ Comput.
Commun., vol. 124, pp. 87–100, Jun. 2018.
observed that the proposed RAW-RA scheme can simulta- [20] U. Sangeetha and A. Babu, ‘‘Performance analysis of IEEE 802.11ah
neously resolve the performance anomaly and improved the wireless local area network under the restricted access window-based
aggregate throughput as well as energy efficiency of the IEEE mechanism,’’ Int. J. Commun. Syst., vol. 32, no. 4, p. e3888, Mar. 2019.
[21] E. Khorov, A. Krotov, A. Lyakhov, R. Yusupov, M. Condoluci, M. Dohler,
802.11ah multi-rate IoT network when compared to the DRG
and I. Akyildiz, ‘‘Enabling the Internet of Things with Wi-Fi halow—
and UG schemes. We have used the network simulator (ns-3) Performance evaluation of the restricted access window,’’ IEEE Access,
which follows IEEE 802.11ah physical and MAC layer spec- vol. 7, pp. 127402–127415, 2019.
ifications to validate the analytical results. [22] Z. Ali, J. Misic, and V. B. Misic, ‘‘Performance evaluation of het-
erogeneous IoT nodes with differentiated QoS in IEEE 802.11ah raw
mechanism,’’ IEEE Trans. Veh. Technol., vol. 68, no. 4, pp. 3905–3918,
REFERENCES Apr. 2019.
[1] A. Nauman, Y. A. Qadri, M. Amjad, Y. B. Zikria, M. K. Afzal, and [23] N. Ahmed, D. De, and M. I. Hussain, ‘‘A QoS-aware MAC protocol for
S. W. Kim, ‘‘Multimedia Internet of Things: A comprehensive survey,’’ IEEE 802.11ah-based Internet of Things,’’ in Proc. 15th Int. Conf. Wireless
IEEE Access, vol. 8, pp. 8202–8250, 2020. Opt. Commun. Netw. (WOCN), Feb. 2018, pp. 1–5.
[2] V. Baños-Gonzalez, M. Afaqui, E. Lopez-Aguilera, and E. Garcia-Villegas, [24] C. Kai, J. Zhang, X. Zhang, and W. Huang, ‘‘Energy-efficient sensor
‘‘IEEE 802.11ah: A technology to face the IoT challenge,’’ Sensors, grouping for IEEE 802.11ah networks with max-min fairness guarantees,’’
vol. 16, no. 11, p. 1960, Nov. 2016. IEEE Access, vol. 7, pp. 102284–102294, 2019.
[3] B. Bellalta, L. Bononi, R. Bruno, and A. Kassler, ‘‘Next generation IEEE [25] O. Raeesi, J. Pirskanen, A. Hazmi, J. Talvitie, and M. Valkama, ‘‘Perfor-
802.11 wireless local area networks: Current status, future directions and mance enhancement and evaluation of IEEE 802.11 ah multi-access point
open challenges,’’ Comp. Commun., vol. 75, pp. 1–25, Feb. 2016. network using restricted access window mechanism,’’ in Proc. IEEE Int.
[4] V. P. Harigovindan, A. V. Babu, and L. Jacob, ‘‘Proportional fair resource Conf. Distrib. Comput. Sensor Syst., May 2014, pp. 287–293.
allocation in vehicle-to-infrastructure networks for drive-thru internet [26] N. Ahmed and M. I. Hussain, ‘‘Periodic traffic scheduling for IEEE
applications,’’ Comput. Commun., vol. 40, pp. 33–50, Mar. 2014. 802.11ah networks,’’ IEEE Commun. Lett., vol. 24, no. 7, pp. 1510–1513,
[5] G. Bianchi, ‘‘Performance analysis of the IEEE 802.11 distributed coordi- Jul. 2020.
nation function,’’ IEEE J. Sel. Areas Commun., vol. 18, no. 3, pp. 535–547, [27] Q. T. Ngo, D. N. M. Dang, and Q. Le-Trung, ‘‘Extreme power saving
Mar. 2000. directional MAC protocol in IEEE 802.11ah networks,’’ IET Netw., vol. 9,
[6] A. Šljivo, D. Kerkhove, L. Tian, J. Famaey, A. Munteanu, I. Moerman, no. 4, pp. 180–188, Jul. 2020.
J. Hoebeke, and E. D. Poorter, ‘‘Performance evaluation of IEEE 802.11ah [28] J. Lei, J. Tao, J. Huang, and Y. Xia, ‘‘A differentiated reservation MAC
networks with high-throughput bidirectional traffic,’’ Sensors, vol. 18, protocol for achieving fairness and efficiency in multi-rate IEEE 802.11
no. 2, p. 325, 2018. WLANs,’’ IEEE Access, vol. 7, pp. 12133–12145, 2019.

136518 VOLUME 9, 2021

S. P. Badarla, V. P. Harigovindan: RAW-RA Scheme

[29] M. Laddomada, F. Mesiti, M. Mondin, and F. Daneshgaran, ‘‘On the SRI PAVAN BADARLA (Associate Member,
throughput performance of multirate IEEE 802.11 networks with variable- IEEE) received the Bachelor of Technology degree
loaded stations: Analysis, modeling, and a novel proportional fairness in electronics and communication engineering
criterion,’’ IEEE Trans. Wireless Commun., vol. 9, no. 5, pp. 1594–1607, from JNTU Hyderabad, India, and the Master
May 2010. of Technology degree (First class with Distinc-
[30] D.-Y. Yang, T.-J. Lee, K. Jang, J. B. Chang, and S. Choi, ‘‘Performance tion) in digital communication engineering from
enhancement of multirate IEEE 802.11 WLANs with geographically scat- Visvesvaraya Technological University, Belagavi,
tered stations,’’ IEEE Trans. Mobile Comput., vol. 5, no. 7, pp. 906–919,
Karnataka, India. He is currently pursuing the
Jul. 2006.
Ph.D. degree with the Department of Electronics
[31] M. A. Yazici and N. Akar, ‘‘Running multiple instances of the
distributed coordination function for air-time fairness in multi-rate and Communication Engineering, National Insti-
WLANs,’’ IEEE Trans. Commun., vol. 61, no. 12, pp. 5067–5076, tute of Technology Puducherry, Karaikal, Puducherry, India. His research
Dec. 2013. interests include in the areas of wireless networks, wireless communications,
[32] U. Sangeetha and A. V. Babu, ‘‘Fair and efficient resource allocation in the Internet of Things, IEEE 802.11 MAC layer mechanisms, and IEEE
IEEE 802.11ah WLAN with heterogeneous data rates,’’ Comput. Com- 802.11 multi-rate WLANs.
mun., vol. 151, pp. 154–164, Feb. 2020.
[33] M. Mahesh, B. S. Pavan, and V. P. Harigovindan, ‘‘Data rate-
based grouping to resolve performance anomaly of multi-rate IEEE V. P. HARIGOVINDAN (Senior Member, IEEE)
802.11ah IoT networks,’’ IEEE Netw. Lett., vol. 2, no. 4, pp. 166–170, received the Bachelor of Technology degree in
Dec. 2020. electronics and communication engineering from
[34] F. Daneshgaran, M. Laddomada, F. Mesiti, and M. Mondin, ‘‘Unsaturated the University of Calicut, the Master of Technol-
throughput analysis of IEEE 802.11 in presence of non ideal transmission ogy degree in digital electronics and communi-
channel and capture effects,’’ IEEE Trans. Wireless Commun., vol. 7, no. 4, cation systems from Visvesvaraya Technological
pp. 1276–1286, Apr. 2008. University with First Rank, and the Ph.D. degree
[35] E. Kocan, B. Domazetovic, and M. Pejanovic-Djurisic, ‘‘Range extension from the National Institute of Technology Calicut,
in IEEE 802.11ah systems through relaying,’’ Wireless Pers. Commun., in 2013. He is currently an Assistant Professor
vol. 97, no. 2, pp. 1889–1910, May 2017.
with the Department of Electronics and Commu-
[36] W. Kuo, ‘‘Energy efficiency modelling for IEEE 802.11a distribution
nication Engineering and the Associate Dean (Faculty Welfare) with the
coordination function system without finite retry limits,’’ IET Commun.,
vol. 1, no. 2, pp. 165–172, Apr. 2007. National Institute of Technology Puducherry (under Ministry of Education,
[37] M. Mahesh and V. P. Harigovindan, ‘‘Fuzzy based optimal and Government of India). He has more than 40 international publications to his
traffic-aware restricted access window mechanism for dense IoT credit. Currently, he is serving as the Principal Investigator of three funded
networks,’’ J. Intell. Fuzzy Syst., vol. 37, no. 6, pp. 7851–7864, projects each from the Department of Science and Technology, Government
Dec. 2019. of India, the Science and Engineering Research Board, and Microsoft AI for
[38] L. Tian, A. Šljivo, S. Santi, E. De Poorter, J. Hoebeke, and J. Famaey, Earth. His research interests include in the areas of wireless networks and
‘‘Extension of the IEEE 802.11ah Ns-3 simulation module,’’ in Proc. 10th wireless communications.
Workshop, New York, NY, USA, 2018, pp. 53–60.

VOLUME 9, 2021 136519