International Journal of Pure and Applied Mathematics

Volume 119 No. 14 2018, 247-257

ISSN: 1314-3395 (on-line version)
url: http://www.ijpam.eu
Special Issue
ijpam.eu

   

CWA-AI MODEL TO CONTROL NETWORK DELAYS FOR A-GPS MOBILE BASED AD HOC
NETWORKS
1
Sunil Chandolu and 2Dr. P. Sanyasi Naidu
1
Research Scholar, GITAM University, Andhra Pradesh, India.
2
Professor, GITAM University, Andhra Pradesh, India.

Abstract: Because of increasing the popularity of 1. INTRODUCTION

Global Positioning Systems (GPS) in outside sky
environments, in some areas like indoors and urban Present days, there are more number of applications
canyons maintenance the performance of GPS is an offers location based service representations are enabled
emerging and challenging task in present scenarios. An to process extended positioning systems like United
assisted GPS is used to operate communications in these States (U.S) global positioning systems with their
locations; it is an open source framework for many satellite systems. Some of the extended versions of GPS
wireless networks to support testing scenarios and systems like assisted GPS by U.S to support
implementation with different maintenance simulated communications with increase of location accuracy
parameters. To maintain network delays in MANET type services in wireless ad hoc networks. While traditional
of wireless networks, Delay measurement and Modeling GPS recipients perform well outside in open-sky
approach is used to processed static data channel situations, the A-GPS idea improves regular GPS
maintenance in wireless networks. This approach is a beneficiaries to work in certain powerless flag
standard with secure location maintenance and exploits conditions, for example, urban and indoor conditions,
channel representation in wireless network where the signs are unequivocally lessened proliferating
communications. Delay modeling with different network through structures and contorted due to multipath
simulated parameters over dynamic location in wireless wonders [4]. Mobile Station Based (MS-Based) A-GPS
networks is an aggressive concept because of congestion innovation investigates extra earthbound remote connect
and contention in dynamic channel maintenance in to help give a recipient GPS route and other information
network communication with respect to Bandwidth from elective sources [2]. Receivers are then diminished
Delay Product (BDP) and total time consistency (i.e. from the errand of demodulating most route information
round trip time (RTT)) in location based data in testing conditions with high gathering blunder rates.
transmission using Transfer Control Protocol in wireless For instance, specialist organizations may circulate route
networks. To address this issue, in this paper we propose information containing orbital parameters that are
an innovative and novel approach i.e. Congestion gathered by their unhindered GPS beneficiaries. The
Window Adaptation through Argument Identification procedure of the data sharing and processing with
(CWA-AI) to provide wider measurements in dynamic different formations in global positioning systems shown
data transmission for wireless networks. This is an in figure 1.
accurate method to provide argument status from
overshooting in to decrease of congestion in real time
wireless ad hoc networks. Our simulation results show
effective performance over network data transmission
with respect to network delay, throughput and time
efficiency levels with congestion control in wireless
networks.

Keywords: Wireless Ad hoc Networks, Global

Positioning System, Congestion Window Adaption,
Round Trip Time, Bandwidth Delay Product and
Transfer Control Protocol.

Figure 1. Global positioning systems operations in

wireless ad hoc networks.

  

247
International Journal of Pure and Applied Mathematics Special Issue

   

Giving help information is just a single part of the A- overshooting problem in statistical standard in wireless
GPS operation that test systems bolster. What test ad hoc networks.
systems don't regularly do is the displaying of system
delays for help conveyance. A few evaluations were b) We introduce variable, variance of contention
already given [2], yet no methodical measurably RTT per each hop (VCRH) to define link connections
advocated A-GPS arrange to defer models have been with respect random evaluations with status of
proposed to the best of our insight. Upgrading GPS- bottleneck network communications.
based instrumentation is a typical pattern [8] – [12], and
this paper concentrates on tending to more satisfactory c) We make modified TCP congestion window
demonstrating of the MS-Based A-GPS operation. adaption approach to maintain acknowledgement with
respect to VCRH. This approach timely accurate and
The primary factual model for the MS-Based A-GPS effective method to manage overshooting appearance in
information conveyance delays for frameworks utilizing wireless networks.
the Secure User Plane Location (SUPL) standard. These
information exchanges include various system The remaining of this paper organized as follows:
correspondences of the SUPL-indicated information Section 2 introduces related work with respect to GPS
records and can't utilize regular organize defer models. system with TCP in wireless networks. Section 3
Abusing different model segments detailed for general describes Mobile station based delay measurements and
system defer thinks about and their blends, a committed data delaying models with data assistance in wireless as
blend A-GPS delay channel model (AGDCM) is hoc networks. Section 4 defines procedure of CWA-AI
acquired, which gives better execution analyzed factual in dynamic network communication. Section 5 simulates
dispersions utilized for traditional system postpone the experimental results with comparison of traditional
models. The obtained model for the MS-Based A-GPS GPS system in wireless ad hoc networks. Sections 6
information deferrals can be used in a large portion of concludes overall conclusion with wireless
the test systems. At last, the MS-Based A-GPS testbed communication.
configuration is depicted that can be repeated by
different specialists to encourage additionally examines. 2. REVIEW OF RELATED WORK
A TCP blockage control technique is intended to build
the quantity of sections up to organize limit, which is Normally, organize defer displaying approaches include
spoken to as a data transfer Bandwidth Delay Product two stages: 1) utilizing a postpone estimation procedure
(BDP), and Chen et al. [5] investigated the BDP in both and gathering postpone estimation information in chose
wired and remote systems. In customary systems, systems and 2) choosing factual models and afterward
fragments can go through connections back to back, and evaluating model parameters. For general system
afterward, these sections can chock up the entire system postpone demonstrating undertakings, a few
pipe. In any case, a recipient should begin to forward a methodologies have been accounted for, for example,
portion after totally accepting it from the sender in multi- inward and outside displaying strategies [3], [4].
hop impromptu systems. Furthermore, considering the The interior demonstrating system considers arranging
channel dispute and channel impedance issue, the BDP delay as a superposition of back to back deferrals
of a transfer control protocol (TCP) association is happening between organizing components along the
considerably littler than in wired systems. information engendering ways. While all the more
satisfactorily demonstrating physical engendering, the
To understand the previously mentioned TCP clog deferral estimation crusades are exceptionally difficult
window overshooting issue, we propose a novel system because of access limitations to organize components.
called Congestion Window Adaptation through Outside demonstrating structures gather end-to-end defer
Argument Identification (CWA-AI). The main estimations without investigating inner system
contributions of our proposed approach are as follows: complexities. The system postpone thinks about allude
a) For effective time consistence, we split RTT into to factual system topology test systems, including
congestion RTT and Argument or Contention RTT, p2psim [9], BRITE [3], GT-ITM [3], ns-2 [2], and even
Argument RTT do not perform any action with BDP and modern estimation foundations are intended to examine
only congestion RTT performs best accuracy to cause information spread between various has, for example,
IMPA [3], CAIDA [4], NIMI [5], and Surveyor [6].
  

248
International Journal of Pure and Applied Mathematics Special Issue

   

Estimations are commonly taken utilizing devoted aftereffects of reproduction, Fu et al. demonstrated that
system estimation devices, for example, netdyn [7], the estimation of clog window may not be kept up at the
TCPDUMP [8], Wildpackets [9], and Wireshark [4], or range around W∗ amid the commonsense work of TCP;
demonstrative devices, for example, ping and traceroute. however it changes until the point that it achieves an
General system postpone demonstrating has been the esteem considerably bigger than W∗. In view of these
subject of broad research. Commonly notable two critical conclusions, they proposed a versatile pacing
appropriations are considered for factual system plan. In the plan, a hub that transmits a parcel should sit
postpone models, including gamma, lognormal, tight for a specific time after the back-off time. We can
circulation in [5], gamma dispersion in [6], exponential watch that versatile pacing can moderate the intra-stream
and its blend disseminations in [7], and Weibull and conflict issue, yet the holding up time isn't exact. Also,
blend of Weibull models in [8]. Dissemination Luo et al. [8] proposed a paced TCP component in which
parameters are evaluated utilizing different procedures. the pace time between bundles is balanced by fluctuating
Strategies, including the probability work, multi-scale RTT esteems.
most extreme punished probability assess nonparametric
thickness estimation, the Expatiation and Maximization Our proposed component [16] embraced a comparable
(EM) calculation, and the quick Fourier change based plan to the TCP dispute control system [15] with two
EM calculation were looked into in [1], though [2] parameters for evaluating throughput and dispute, and it
alludes to the consecutive Monte Carlo technique. While is more precise. In any case, there is as yet an issue in
organize postpone displaying approaches have been this component, since these two parameters are
broadly examined, they are not straightforwardly characterized by the status at the bottleneck connect
material for A-GPS help correspondence because of a without considering the instance of the entire way. The
few information trade cycles as per the SUPL bottleneck is, in reality, a vital factor on the off chance
convention. As a few message trade cycles are required, that of the clog. Then again, conflicts cause very much a
the aggregate postpone estimation is made out of distinctive circumstance. We will clarify this disputed
numerous roundtrip time estimations of SUPL-indicated issue in the following segment.
information records. After estimation gathering, the
MLE [4] and TRR calculation [4] is utilized to outline 3. MS-BASED GPS TESTBED & DELAYING
half and half appropriation models for help delays. As flag twists are normally remarkable and unrepeatable
The TCP clog window overshooting issue has been for certifiable testing, engineers utilize the GPS test
considered, and the overshooting issue happens because systems to create agent and quantifiable situations [11]–
of the accompanying two viewpoints: 1) the serious [15]. As A-GPS operation modes wind up noticeably
dispute in the MAC layer and 2) the off-base adjustment normal, these test systems ought to be improved to give
of blockage window in the TCP layer. In view of these the MS-Based help information, for example, orbital
reasons, past examinations could be grouped into four parameters, and coarse appraisals of position and time
classes: 1) the lessening of intra-stream conflict; 2) the [2], [5], [6].
decrease of interflow dispute; 3) the estimation of the The testbed utilized for arranging defer considers in this
provincial limit; and 4) versatile blockage control paper utilizes the GPS Simulation Toolkit from NI [14]
components. Intraflow dispute alludes to the conflict that concentrates orbital chronicle information in SEM
between the bundles that have a place with a similar design [7] and the ephemeris information in RINEX 2
TCP association. It incorporates two noteworthy parts: organize [8]. In view of gave client area and time
1) the dispute between information bundles and 2) the information, the test system decides noticeable satellites
conflict amongst information and ACK parcels. Fu et al. and produces their signs. In the testbed, SUPL help
[7] examined the effect of the multi-jump remote conveyance engineering is created by us and
channel on TCP throughput. They displayed two vital incorporated into the test system. For the case examined
conclusions. To begin with, for a given system topology in this paper, a handset (SET) started helps information
and TCP association mode, there is a unique TCP clog ask for mode is utilized for postponing demonstrating.
window W∗. Second, in a static remote specially The SUPL makes utilize of the client information
appointed system, the significant explanations behind channels, for example, Internet convention (IP) joins,
parcel misfortunes are struggle and impedance caused by generally called client plane, to empower the MS-Based
interface layer conflicts. At the same time, in light of the A-GPS help correspondence. The SUPL characterizes a
  

249
International Journal of Pure and Applied Mathematics Special Issue

   

help correspondence convention between the SUPL- acquired. The support interaction includes both TCP/IP
empowered terminal called SET and an A-GPS server and SUPL particular information bundle signals.
called a SUPL location platform (SLP). Contrasted and Overall, the procedure starts with the transmitting of a
early control plane conventions, the SUPL-construct TCP/IP SYN banner bundle to link the SET to an
engineering limits dependence with respect to organizing original TCP slot of the SLP server. After all of the
foundations. This permits its less demanding mix with transactions are finished, the procedure ends with a
the GPS test systems. TCP/IP FIN banner bundle transmitting. SYN, FIN, and
other TCP/IP packages are resolved in Fig. 2. Please be
aware that all five SUPL concept packages are
interchanged over the Online between SYN and FIN
flag’ transmitting. The wait statistic is acquired as the
main distinction of appearance periods of FIN and SYN
packages at the SLP and will be known as MS-Based A-
GPS system wait statistic hereafter

Tdelay (i)  TFIN (i)  TSYN (i )

where TSYN and TFIN are the appearance periods of the

SYN and the FIN packages, respectively, and i is the
statistic cycle number, i = 1, . . . , N. To make simpler
the statistic procedure, the first SYN timestamp is
calculated at SLP, and transmission time from the device
to SLP is ignored as it is just a fraction of complete wait.
Figure 2. MS-based GPS framework procedure with Now, the SLP server may instantly detect both SYN and
delay measurements. FIN packages when devices demand assistance data.
The entire cycle of the utilized testbed is outlined in Fig.
1. One can see included ephemeris and chronological
registry orbital parameters alongside coarse time and
area references as essential components of the MS-Based
help information. As the test system produces route
information, one can utilize it for the MS-Based help age
in that capacity information are accessible at A-GPS
disjoins gave from unhindered GPS recipients. The
coordinated A-GPS module extricates this information
and proselytes it to Abstract Syntax Notation One
(ASN.1) documentation [5]. The SUPL-sort operations
include a few information trade cycles between the A-
GPS server and the SET, and the greater part of the trade
documents are encoded ASN.1 records as per the
determined conventions, for example, the radio resource
location protocol (RRLP) [5], which is actualized in the
utilized testbed.

3.1. Delay Measurements

Figure 3. Packet flow procedure over data transmission
The MS-Based A-GPS system wait dimensions over in wireless networks.
TCP/IP relationships have been acquired using
timestamps measured by a system analyzer device Fig. 2 demonstrates the bundle return procedure, i.e.,
created in [4]. The visitors are supervised at the SLP PSH packets of 9, 8, 22, 114, and 798 KB are
server. The analyzer timestamps message delivering and communicated between the SLP (port variety 9999) and
getting circumstances from which the overall weight is the SET over the Online and the weight of 3.14 s is
  

250
International Journal of Pure and Applied Mathematics Special Issue

   

approximated according to (1). For mathematical There are three reasons that we embrace the V CR to
research, a statistic strategy is conducted by gathering assess the level of connection conflicts. To start with, it
3000 wait dimensions for the selected systems, such as can speak to the level of connection dispute. On the off
LAN, WLAN, 3G, HSDPA, and 4G LTE. Information is chance that the likelihood of channel conflicts is bigger,
also gathered for various SET-to-SLP distance machines the MAC dispute window should be bigger, and the
described as a town, nationwide, and international. To back-off time is chosen in a substantially more extensive
provide Dynamic model to data transmission with territory; in this manner, higher dispute likelihood brings
different levels of data process discussed in next section about a bigger variety of the conflict delay. We assess
with suitable maintenance of different network this unsteady through a difference, and it could
demonstrations. specifically mirror the dispute plausibility. Second, the
change is an irregular variable that mirrors the dispute
4. CWA-AI circumstance seen amid the current perception window.
Third, the difference can likewise mirror the status of a
We recommend a new procedure known as CWA–CD. bottleneck. Assuming that there is a bottleneck in one of
In this area, we present a new parameter known as the V the two ways, the way with the bottleneck has a bigger V
CRH and then explain the key purpose why that this CR. In the event that there is a bottleneck interface
parameter can be used to estimate the level of MAC because of extreme conflicts, there is a high likelihood
contentions. Depending on this parameter, we improve of connection breakage.
the TCP blockage screen variation procedure to solve the
overshooting problem for dynamic data transmission to In this system, three subtle elements ought to be
wireless network demonstration. clarified. In the first place, we wipe out the impact of
a) V CRH way length. In the wake of acquiring the dispute RTT,
we can ascertain the difference and afterward separate it
We characterize a parameter called the V CRH. In the by the quantity of bounces. At that point, we can get the
first place, the estimation of the conflict RTT is deviation for each bounce. Second, we consider a round-
accumulated through parcel sending. In the MAC trip way rather than a forward way utilized as a part of
specialist, the conflict delay is measured as the interim the traditional strategy in light of the fact that ACKs in
between a bundle landing in the head of the line of the the regressive course are additionally a critical factor for
line and the conveyance time to the physical layer. We identifying join dispute. Third, we have to set a suitable
set a period stamp in the piece (outline) with the goal estimation of n, i.e., the required number of tests for
that it can monitor the dispute deferral of the part in the evaluating the V CR. On the off chance that n is
present hub. The dispute delay is aggregated jump by the extensive; it builds the memory size and estimated the
bounce and recorded in the section, lastly, the TCP goal cost of hubs. Nonetheless, it gradually reacts to a change
duplicates the entirety from the portion into the ACK. because of a long haul normal impact. Then again, if n is
We should take note of the accompanying two focuses little, it is extremely delicate to a little change and may
here: 1) the postponed ACK work [20] ought to be cause a huge estimation of V CRH on various RTTs.
incapacitated and 2) we handicap the discontinuity work
in recreation to rearrange the part testing of the dispute To implement this scenario to update VCR with different
delay. Subsequently, the MAC specialist could duplicate sources with their ACKs:
the entirety of the conflict delay specifically from the
Ek 1  k  1)  CRnew
part to the bundles. In genuine usage, the MAC operator Ek  for (k  n) ------------- (1)
could pick the conflict test in the last part amid the k
discontinuity stage and duplicate it from the parcel. The
example estimation of the conflict delay is likewise Vk 1  Ek21 )  (k  1)  CRnew
2
Vk   Ek2 for (k  n) ... (2)
recorded in the ACK, and the TCP source gets the k
dispute RTT from the recently got ACK. Second, the
deviation of the conflict RTT of the most recent n CRnew  CRold
fragments can be made sense of at the source hub, and it
Ek  Ek 1  for (k  n) ,----------- (3)
n
is known as the change of dispute RTT (V CR). At long
last, the V CRH is assessed as the V CR partitioned by
the quantity of bounces in a given course.
  

251
International Journal of Pure and Applied Mathematics Special Issue

   

Vk 1  Ek21 )  n  CRnew
2
 CRold
2
Vk   Ek2 for (k  n) , --- (4)
k

where k is the testing series of recently obtained ACKs,

n is the most of the examples, and Ek and Vk are the
expectation and the difference of CR (contention RTT)
from the latest k (when k < n) or n (when k >= n) ACKs.
CRnew is the recently acquired CR value, and CRold is
the old CR value. If k < n, Vk is extracted from (1) and
(2), because the number of example records k is different
from n. These two equations are used during the brief
temporary time right after the TCP connection
initialization or the timeout. In this situation, CRold is
unnecessary. If k >= n, Vk is extracted from (3) and (4). Figure 4. TCP/IP Grid Network Demonstration.
In this situation, a new CR testing value included in a
new ACK pushes out an old access n periods ago, and This new component respects a vast variety of dispute
we can upgrade the V CR according to the new CR value RTT as a proof of conflict status conversely with [15].
itself. When Vk is newly obtained, we can get the V This instrument is more exact, in light of the fact that it
CRH by splitting Vk by the number of trips. assesses the dispute from a progression of conflict RTT
esteems. This system is timelier, on the grounds that it
b) TCP Congestion Adaption Approach dispenses with the old-fashioned esteems well before,
yet in the traditional TCP conflict control, the old-
We set an edge parameter V CRH_th, and if the V CRH fashioned dispute data influences the figuring of V CRH.
surpasses this limit, the level of conflict in the system is Then again, our blockage window adjustment instrument
respected to be serious. At whatever point a TCP source is less difficult and more steady. To begin, an endless
gets another ACK, it refreshes the estimation of the V supply of another ACK, the V CRH result shows how
CRH before checking whether this esteem is bigger than we can alter the estimation of cwnd, and regardless of
or equivalent to the V CRH_th esteem. On the off the possibility that the estimation on the dispute is
chance that it is bigger, cwnd is diminished by one MSS unique in relation to reality, the subsequent cwnd is quite
rather than an expansion in the clog window. This recently somewhat adjusted. In any case, in the
approach could adequately limit the clog window from traditional TCP dispute control, once the estimation on
overshooting. On the off chance that the RTO throughput and conflict is more awful, the cwnd is reset.
terminates, which infers that the system is in an awful Second, if a parcel misfortune happens, we decide if we
clog or conflict status, we have to check whether the V can limit the cwnd, contingent upon the dispute
CRH is bigger than or equivalent to V CRH_th before circumstance. On the off chance that the conflict status
splitting the blockage window threshold(ssthresh) [6]. isn't extreme, it is valuable for easing the substantial
On the off chance that the V CRH is bigger, the system variety of cwnd and expanding the throughput by simply
status is respected to be terrible, and we reset the cwnd dividing the clog window. Our new component is more
to 2 ∗ MSS and empower the moderate begin step. powerful and more exact.
Something else, the disputed status of the system isn't so
terrible, and we simply divide the cwnd. We should see 5. SIMULATED PERFORMANCE EVALUATION
that the V CRH testing ought to be reset after an RTO,
and in this manner, the resulting moderate begin of cwnd In this section, we present to discuss about performance
isn't influenced by the low estimation of V CRH got evaluation of proposed approach with traditional
before the RTO terminates. approaches like delay measurement approach [1] and
TCP contention control procedure [15] and our proposed
approach CWA-AI in ns-2/Java network simulator with
comparison of four different techniques. To this
implementation properly, use 4 GB RAM and i5
processor system configurations for efficient execution

  

252
International Journal of Pure and Applied Mathematics Special Issue

   

in real time wireless communication. Using following

network parameters, we develop our network
demonstration effectively. Table 1 shows simulation
parameter to access information in various network
services.

Table 1. Network Simulation Parameters.

Topology Pattern Statistical Representation

Adjacent Node 250m
Distance
Radius for 350m
transmission
Radius Interface 600m
Data segment size 1024bytes
Network Window 48 Figure 5. TCP with CWA-AI operations with time in
Size different segments.
Simulation Time 35-70secs
Number of nodes 10,20,30,40,50 Fig. 5 reveals the TCP cwnd-AI value of a relationship
Number of 4,8,16,24,31,39 during a period. In this determine, the y-axis indicates
Connections the cwnd (window size) value with the device of
Sample VCRH values 8segments sections, each of which has 512 B. First if we evaluate
our suggested TCP CWA–CD procedure with the A-
Protocol TCP/IP, AODV,DSR
GPS with Wait Dimensions, the suggested procedure
successfully reduces the times of cwnd totally reset. In
As shown in table 1, we use network parameters with other terms, the regular totally reset problem proven in
different events in different node connections and the original cwnd is reduced, and this outcome is clearly
network formation based on above parameters. To proven in the time interval including 170 s to 210 s in
concentrate on the consequences of contentions and Fig. 5. In this variety, the proposed procedure gets to the
disturbance and to reduce the effect of path problems as ability of the system pipe after a period of cwnd
much as possible, we concentrate on a grid topology as a improves along with A-GPS with delay measurements.
fixed situation. We evaluate the system efficiency and
define the blockage screen among the four systems. We
also present simulator results in a powerful unique
topology.

Based on above considerations, we investigate the

characteristic parameters for four mechanisms: 1) TCP,
2) TCP Contention Control 3) TCP with Delay
Measurements (A-GPS) and 4) TCP with CWA-AI in
time efficiency, throughput and end-to-end delay with
different parameters.

Figure 6. TCP throughput for four approaches with

different nodes.

In the test, the throughput is assessed regarding the

aggregate number of TCP sections got the goals in a
given time length of 500 s. Fig. 6 demonstrates the
  

253
International Journal of Pure and Applied Mathematics Special Issue

   

throughput execution of the four components in the our suggested CWA–CI procedure, we have divided the
matrix topology. The proposed CWA– CD component real RTT into two parts—the blockage RTT and the
picks up roughly 15.9%, 25.9%, and 55.2% contention RTT—and exposed that the argument RTT
enhancements contrasted and the TCP the component has nothing to do with the BDP and that the BDP is
with most extreme window, the A-GPS Delay driven by only the congestion RTT. We acquired the
Measurements, and the first TCP instrument, separately. argument RTT information from ahead details and in
From the perspective of the first TCP component, our reverse ACK at the MAC part and estimated the
proposed system successfully restricts the TCP clog argument position through the difference of the
window from overshooting. From the perspective of the contention RTT. Then, we created an alteration of the
TCP system with the greatest window, the window adaptation procedure, in which the cwnd value
characterization of the dispute RTT and clog RTT is was changed according to several aspects such as the V
more exact, and the proposed cwnd adjustment CRH and timeout. The difference of the argument RTT
instrument is more adaptable. The aftereffects of the trial reveals the immediate indication of the argument
show that, in the proposed CWA– CD component, the scenario, and this procedure is timelier and more
conflict discovery is more exact, and the clog window is accurate. Furthermore, the appropriate blockage window
appropriately adjusted. adaptation procedure is easy and stable, and it does not
blindly totally reset the cwnd but modify the cwnd value
centered on the argument position. The research reveals
that the proposed mechanism is more accurate and much
constant, and it significantly improves the throughput in
fixed scenarios. Our suggested procedure concentrates
on link fairs contentions and does not consider the
flexibility aspect. Link problems due to mobility are an
aspect of the primary options for link unreliability in
mobile ad hoc systems, which we have previously
investigated. In future work, we will provide a
comprehensive consideration of these aspects to enhance
the efficiency of the network.

7. REFERENCES

[1] Grant Huang, David Akopian, C. L. Philip,”

Figure 7. TCP end-to-end delay measurements with Measurement and Modeling of Network Delays
different nodes. for MS-Based A-GPS Assistance Delivery”, IEEE
TRANSACTIONS ON INSTRUMENTATION
Fig. 7 reveals the TCP end-to-end wait of the four
AND MEASUREMENT, VOL. 63, NO. 8,
mechanisms in the lines topology, and this outcome is
AUGUST 2014.
another important performance aspect of TCP. The TCP
CWA–CD procedure results in roughly 44.6% decrease [2] Federal Communication Commission (FCC).
in contrast to the A-GPS Delay Measurements and it Washington, DC, USA. (2013, Jun. 15). Wireless
reveals that the suggested procedure cuts down on end- 911 Services [Online
to-end wait to a large level and increases the system [3] Spirent Communications, Paignton, U.K. (2013,
efficiency. Based on the results of the experiment, we Jun. 15) [Online
can observe that the proposed mechanism attempts to fix
the adverse effect of link contentions in multi-hop ad hoc [4] Aeroflex Inc., PLAINVIEW, NY, USA. (2013,
networks. Jun. 15) [Online

6. CONCLUSION [5] J. Narisetty, A. Soghoyan, M. Sundaramurthy, and

D. Akopian, “SUPL support for mobile devices,”
This paper proposes TCP with CWA-AI to assistance Proc. SPIE, vol. 8304, pp. 830409-1–830409-12,
model delivers delay measurements for MS based node Jan. 2012.
utilization with global positioning system operations. In

  

254
International Journal of Pure and Applied Mathematics Special Issue

   

[6] D. A. Grejner-Brzezinska, C. K. Toth, H. Sun, X. method in mobile ad hoc networks,” in Proc. IEEE
Wang, and C. Rizos, “A robust solution to high- WICOM, 2008, pp. 1–4.s
accuracy geolocation: Quadruple integration of
[17] J. Li, C. Blake, D. D. Couto, H. Lee, and R.
GPS, IMU, pseudolite, and terrestrial laser
Morris, “Capacity of ad hoc wireless networks,”
scanning,” IEEE Trans. Instrum. Meas., vol. 60,
in Proc. ACM MobiCom, Jul. 2001, pp. 61–69.
no. 11, pp. 3694–3708, Nov. 2011
[18] K. Chen, Y. Xue, S. H. Shah, and K. Nahrstedt,
[7] Technical Specification Group GSM/EDGE Radio
“Understanding bandwidth-delay product in
Access Network; Location Services (LCS);
mobile ad hoc networks,” Comput. Commun., vol.
Mobile Station (MS)—Serving Mobile Location
27, no. 10, pp. 923–934, Jun. 2004.
Centre (SMLC) Radio Resource LCS Protocol
(RRLP), 3GPP, France, 2010. [19] V. Jacobson, “Congestion avoidance and control,”
Comput. Commun. Rev., vol. 18, no. 4, pp. 314–
[8] Navigation Center, Ann Arbor, MI, USA. (2011).
329, Aug. 1988.
Global Positioning System Directorate Systems
Engineering and Integration Interface [20] Z. Fu, H. Luo, P. Zerfos, S. Lu, L. Zhang, and M.
Specification [Online Gerla, “The impact of multihop wireless channel
on TCP throughput and loss,” in Proc. IEEE
[9] NASA Goddard Space Flight Center, Greenbelt,
INFOCOM, 2003, pp. 1744–1753.
MD, USA. (2013, Jun. 15). Ephemeris
Information [Online [21] C. Y. Luo, N. Komuro, K. Takahashi, and T.
Tsuboi, “PACED TCP: A dynamic bandwidth
[10] D. Akopian, A. Soghoyan, S. N. Chayapathi, and
probe TCP with pacing in ad hoc networks,” in
G. V. S. Raju, “Feasibility study of a
Proc. IEEE Pers., Indoor, Mobile Radio
labviewbased GNSS receiver,” in Proc. ION-ITM
Commun., 2007, pp. 1–5.
Conf., San Diego, CA, USA, Jan. 2011, pp. 1270–
1280. [22] S. M. Elrakabawy, A. Klemm, and C. Lindemann,
“TCP with adaptive pacing for multihop wireless
[11] J. Sijbers, A. J. D. Dekker, P. Scheunders, and D.
networks,” in Proc. 6th ACM Int. Symp. Mobile
V. Dyck, “Maximum likelihood estimation of
Ad Hoc Netw. Comput., 2005, pp. 288–299.
Rician distribution parameters,” IEEE Trans. Med.
Imag., vol. 17, no. 3, pp. 357–361, Jun. 1998. [23] M. Shen and D. M. Zhao, “Opportunistic link
scheduling for multihop wireless networks,” IEEE
[12] N. G. Duffield and F. L. Presti, “Network
Trans. Wireless Commun., vol. 8, no. 1, pp. 234–
tomography from measured end-to-end delay
244, Jan. 2009.
covariance,” IEEE Trans. Netw., vol. 12, no. 6,
pp. 978–992, Dec. 2004. [24] Z. Ye, S. V. Krishnamurthy, and S. K. Tripathi,
“Use of congestionaware routing to spatially
[13] Xin Ming Zhang, Wen Bo Zhu, Na Na Li, and
separate TCP connections in wireless ad hoc
Dan Keun Sung,” TCP Congestion Window
networks,” in Proc. 1st IEEE Int. Conf.Mobile Ad
Adaptation Through Contention Detection in Ad
Hoc Sensor Syst., 2004, pp. 389–397.
Hoc Networks” IEEE TRANSACTIONS ON
VEHICULAR TECHNOLOGY, VOL. 59, NO. 9, [25] H. Zhai, X. Chen, and Y. Fang, “Alleviating
NOVEMBER 2010. intraflow and interflow contentions for reliable
service in mobile ad hoc networks,” in Proc. IEEE
[14] G. Holland and N. Vaidya, “Analysis of TCP
Mil. Commun. Conf., 2004, pp. 1640–1646.
performance over mobile ad hoc networks,” ACM
Wireless Netw., vol. 8, no. 2, pp. 275–288, Mar. [26] S. Rangwala, A. Jindal, K. Y. Jang, K. Psounis,
2002. and R. Govindan, “Understanding congestion
control in multihop wireless mesh networks,” in
[15] Z. Fu, X. Meng, and S. Lu, “How bad TCP can
Proc. 14th ACM Int. Conf. Mobile Comput.
perform in mobile ad hoc networks,” in Proc.
Netw., 2008, pp. 291–302.
IEEE ISCC, 2002, pp. 298–303.
[27] D. Triantafyllidou, K. Al Agha, and V. A. Siris,
[16] W. B. Zhu, X. M. Zhang, and N. N. Li, “Improve
“Adaptive setting of TCP’s maximum window in
TCP performance with link-aware warning

  

255
International Journal of Pure and Applied Mathematics Special Issue

   

ad hoc multihop networks with a single flow,” in

Proc. IEEE WCNC, 2009, pp. 1–6.
[28] E. Hamadani and V. Rakocevic, “TCP contention
control: A cross-layer approach to improve TCP
performance in multihop ad hoc networks,” in
Proc. 5th Int. Conf. Wired/Wireless Internet
Commun., 2007, pp. 1–16.
[29] X. M. Zhang, N. N. Li, W. B. Zhu, and D. K.
Sung, “TCP transmission rate control mechanism
based on channel utilization and contention ratio
in ad hoc networks,” IEEE Commun. Lett., vol.
13, no. 4, pp. 280–282, Apr. 2009.
[30] K. Nahm, A. Helmy, and C. J. Kuo, “Cross-layer
interaction of TCP and ad hoc routing protocols in
multihop IEEE 802.11 networks,” IEEE Trans.
Mobile Comput., vol. 7, no. 4, pp. 458–469, Apr.
2008.

  

256
257
258