Content uploaded by K.V. Rop. PhD.
Author content
All content in this area was uploaded by K.V. Rop. PhD. on Dec 11, 2019
Content may be subject to copyright.
International Journal of Recent Technology and Engineering (IJRTE)
ISSN: 2277-3878, Volume-7, Issue-6, March 2019
1495
Published By:
Blue Eyes Intelligence Engineering
& Sciences Publication
Retrieval Number: F2649037619/19©BEIESP
Abstract: Realization of better transport experience has become
a global concern with the growth in the number of vehicles on
roads. Various technologies using Intelligent Transportation
System (ITS) have been fronted as the solution. Vehicular Ad Hoc
Networks (VANET) is an ITS technology that can be used to
effectively support many vehicular applications for effective
traffic control as well as information sharing between vehicles on
the same route. With the advancement in wireless technologies,
many applications related to vehicular communication are bound
to be advanced. These applications will ease the exchange of
information from one vehicle to another with the help of modern
wireless technologies. The use of cognitive radio system provides
additional radio resources in the already crowded licensed
spectrum for vehicular communication. Spectrum sensing
capability and effectiveness of the nodes is paramount in the
vehicular environment. Establishing a reliable threshold level for
energy detection has been shown to be essential for efficient
spectrum sensing with double energy detection threshold being
fronted in the recent past. Small scale primary users too like WIFI
span over short range meaning they are not reliable. In this
paper, a triple threshold energy detection method is proposed. This
method improves the spectrum sensing efficiency as well as
addressing the small scale primary users which are unreliable for
use in cognitive radio systems.
Index Terms: Energy Detection, Cognitive Radios, Spectrum
Sensing, VANET.
I. INTRODUCTION
As the number of vehicle on major roads, especially in urban
centers increases, related challenges such as traffic
management, accidents, insecurity, and traffic snarl-ups are
bound to increase too. There is need for effective traffic
control using technology as well as information sharing
between vehicles on the same route [1]. In an attempt to
address these challenges, researches as well as car
manufactures have adopted the use of communication
technology with various applications and services for
vehicular environments are being researched and developed.
This is achieved by use of Intelligent Transportation System
(ITS) which employs the use of wireless communication
systems [1]–[3].
Revised Manuscript Received on March 25, 2019.
K.V. Rop, P.K. Department of Electrical Engineering, Pan African
University, Institute for Basic Science Technology and Innovation, Kenya.
Langat, Department of Telecommunication and Information Engineering,
Jomo Kenyatta University of Agriculture and Technology (JKUAT, Kenya
H.A. Ouma, Department of Electrical and Information Engineering,
University of Nairobi, Kenya
Wireless access in vehicular environments (WAVE) is
therefore an essential part of ITS. Vehicular Ad Hoc Network
(VANET) is a type of WAVE that can effectively support
many vehicular applications. Moving on the road in a
spontaneous and unconstrained manner, vehicles forms an ad
hoc network. These vehicles in VANET act as mobile nodes
in a mobile ad hoc network (MANET) thus creating a high
mobility network with free vehicle movements and the
capability to organize and group themselves arbitrary, whilst
exchanging information between themselves [2], [4], [5].
Various VANET applications span from the highly sensitive
road safety to various optimization of the vehiclar traffic like
routes and link optimization, and congestion/traffic jam
control to applications like free space parking reporting,
internet access, and file sharing and that are generally
commercial [5]–[7]. A VANET can comprise of Vehicular to
Vehicular (V2V), V2P (Vehicular to Person), or Vehicular to
Infrastructure (V2I) Communication [2], [5], [8], [9]. V2V is
cheaper in terms of setting up and managing the
infrastructure thus, it has the potential as the future to
effective vehicle management and communication. There are
various wireless technologies that can be used by vehicles to
communicate with each other or with other communication
devices. The most dominant technology is Dedicated
Short-Range Communication (DSRC). DSRC with 75 MHz
band between 5.850 and 5.925GHz based on IEEE 802.11p
standard is a communication technology set aside to support
various applications that are based on vehicular
communications [2], [10]. Supporting vehicular
communications up to vehicular speeds of 200 km/h, IEEE
802.11p standard uses a reserved frequency band of 5.9 GHz
[8]. With the increase in wireless enabled vehicles, Cognitive
Radio (CR) technology, over the last few years, has been
fronted as the ultimate solution to looming spectral scarcity
disaster [1][11], [12]. CR is a wireless communication
system that is intelligent with capability of adaptively
modifying in real-time its fundamental operating parameters
for efficient and optimized radio spectrum utilization. It has
environmental awareness and learning capability that are
crucial in provision of reliable communication. CRs have the
capability to detect the unused spectrum bands (spectrum
holes also called white spaces), and can subsequently access
these holes when vacant opportunistically [11], [13]. Primary
users (PU) who are the license holders and secondary users
(SU) who are the unlicensed users seeking to
opportunistically use the spectrum are essential members of a
CR system as seen in Figure 1. SU can only temporarily
Cluster Based Triple Threshold Energy
Detection for Spectrum Sensing in Vehicular
Ad-Hoc Networks
K.V. Rop, P.K. Langat, H.A. Ouma
Cluster Based Triple Threshold Energy Detection for Spectrum Sensing in Vehicular Ad-Hoc Networks
1496
Published By:
Blue Eyes Intelligence Engineering
& Sciences Publication
Retrieval Number: F2649037619/19©BEIESP
occupy unused licensed spectrums when the PUs are not
using and vacate as soon as the PUs start to use them. This
must be done effectively without causing interference to the
PUs [1], [2]. In this paper, PU has been categorized as
reliable and unreliable PUs. Reliable PUs are those users
whose network span over a large geographical area.
Examples are licensed television broadcast frequencies.
Unreliable PUs are the short range frequency users like
Wi-Fi, Bluetooth, etc., and they pose a new challenge of
detection. Many papers have been published on various
methods performing spectrum sensing for effective
cooperation and occupation of spectrum. Most researchers
have concentrated on methods of improving the spectrum
sensing as a whole without factoring in the effect of
unreliable small scale users that span over short distances. In
this paper, a triple energy detection scheme is proposed that
provides effective spectrum sensing as well as eliminating
the unreliable PUs.
Figure 1: Illustration of a VANET with PU and SUs [2]
II. SPECTRUM SENSING
Observe, Analyze, Reason, and Act are the four main
phases of the CR cognition cycle [13]. Here, the spectrum
should be detected effectively and occupied by adjusting the
node’s operational parameters. Also, this identified spectrum
can be shared with other users by coordinating with them.
The goal is to utilize these spectrum but vacate the band
immediately the PU appears. Therefore, there are four
cognitive radio networks functionalities that are important.
These are the spectrum sensing, sharing, management, and
spectrum mobility or spectrum handoff. In this work, the
process of spectrum sensing is investigated. Spectrum
sensing is the detection of the PUs process undertaken by
sensing the radio frequency (RF) environment. The
objectives of spectrum sensing are [1], [2], [13].
i. No harmful interference should be caused to Pus.
The SUs can reduce interference caused to PUs to an
acceptable level or vacate the band to switch to
another available band.
ii. Efficient and effective identification and
exploitation of the spectrum holes for the required
quality-of service (QoS) and throughput.
In CR, each SU must determine the frequency bands to use
by sensing the spectral environment of its surroundings and
learning about the presence of interferers or incumbents [14].
SUs with limited sensing capabilities in CR ad hoc networks
strive to acquire information from other SUs about the
available spectrum bands and share with them without
impairing the PU transmission. The design objectives for
sensing strategy includes reliable system performance under
high node mobility, high throughput, non-SU competition,
and distributed implementation. All these are further
complicated by the high mobility nature of nodes in VANETs
[15].
There are three fundamental requirements for spectrum
sensing [1], [2], [16].
i. Continuous spectrum sensing to monitor the
absence or presence of the PU.
ii. Precautions to avoid interference to potential PUs.
iii. Independent detection of the presence of PUs
without their help.
Such spectrum sensing can therefore be conducted
non-cooperatively (individually), in which each SU conducts
radio detection and makes decisions by itself. Spectrum
sensing can be done by using either energy detection,
cyclostationary based detection, matched filter detection,
and Eigen value based detection methods [13], [17]. For
faster and easy spectrum sensing which is ideal for VANETs
without prior PU information, energy detection method has
been chosen.
As shown in [18]–[20], local signal sensing using primary
signal detection can be expressed as;
= , 0
+ (), 1 (1)
where, x(n) is the signal received at the cognitive radio
terminal, w(n) is the Additive White-Gaussian Noise
(AWGN), s(n) - The primary user signal, Ho represents
absence of licensed PU, and H1 represents the presence of
licensed PU
The Signal to Noise Ratio (SNR), can be given as;
= 2
2 (2)
where, is the SNR, 2 is the variance of the signal, and
2
is the variance of the noise.
The following performance metrics are used for the
hypothesis:
Probability of detection (Pd): This is when the
vacant frequency channel is declared vacant. This
can lead to utilization of the spectrum band.
Probability of false alarm (Pf): This is when the
vacant frequency channel is declared occupied. In
this case, the SUs fails to utilize the free band.
Probability of miss detection (Pm): This is where
the occupied channel is declared vacant. This can
cause interference to the PU.
Higher Pd with a low Pf is the goal of the sensing schemes.
But there is always a trade-off between these two
International Journal of Recent Technology and Engineering (IJRTE)
ISSN: 2277-3878, Volume-7, Issue-6, March 2019
1497
Published By:
Blue Eyes Intelligence Engineering
& Sciences Publication
Retrieval Number: F2649037619/19©BEIESP
probabilities. A high Pm (low Pd) results in missing the
presence of PU, meaning there is high probability of causing
interferes to the PU. With a high Pf, the SU observes PU
while it does not exists, which turns out to be less spectrum
utilization.
A band-pass filter in CR system is applied to the received
signal for power measurement in a particular frequency
region in the time domain. The power of received signal
samples is then measured. The received power can be
estimated as;
= 2
=1 (3)
where, E is the received signal, xi is the ith sample of the
received signal, and N = 2TW is the time bandwidth.
Detection probability, missing probability, and false alarm
probability can be given as [13], [18]–[20];
=>|1= 2, (4)
=|1= 1 (5)
=>|0= ,2
(6)
where, λ is the threshold value, Q(a, b) is generalized
Marcum function, Γ(a) is complete gamma function, and Γ(a,
b) represents incomplete gamma function.
III. TRIPLE THRESHOLD ENERGY DETECTION
In [13], [18]–[20], a cooperative spectrum sensing with a
double threshold detection method was proposed to reduce
the PU interference since single threshold detection has high
interference problems. The double threshold therefore,
avoids the unwanted interference by introducing a fuzzy
region (uncertainty region). [13], [21]–[24] proposes the
elimination of unreliable small scale users as it spans over a
short distance which can lead to constant reallocation of
spectrum space to SU. In this work, a triple threshold energy
detection is proposed that seeks to provide reliable and
effective spectrum sensing.
The Figure 1 below shows one threshold conventional
detection, double threshold, and triple threshold methods.
Figure 2: Energy Detection Methods. (a) Conventional single threshold,
(b) Double threshold, and (c) Triple threshold
In Figure 2 (a), Decision H0 and H1 is made when there is a
greater or lesser Ei than the threshold value λ, respectively as
shown in Eq. 1. In Figure 2 (b), the user reports H1 if the
energy value exceeds λ2. If Ei is less than λ1, the decision H0
will be made. Otherwise, if Ei is between λ1 and λ2, then the
SU reports its observational energy value Ei for further
decision making at the fusion center.
Local sensing using primary signal detection for double
threshold method can be expressed as;
=0, >1
,12
1, >2 (7)
By adding two parameters Δ0,i and Δ1,I to represent the
probability of 12 for the ith secondary user under
hypothesis H 0 and H 1 respectively, we have;
1,=12|1 (8)
0,=12|0 (9)
So it can be derived that:
1=>2|1= 2,2 (10)
=1|1= 1 1,1 (11)
=>2|0= ,22
(12)
In this proposed work, the three threshold levels are used
as shown in Figure 2 (c). The first threshold λ1 undresses the
unreliable small scale users by creating a minimum threshold.
While thresholds λ2 and λ3 applies the same concept as double
threshold detection mentioned above. In this model, two
kinds of information: observational values of the SU i.e. local
energy values and local decisions are received at the fusion
center.
=,1
0, 1<<2
,23
1, >3 (13)
By adding two parameters Δ0,i and Δ1,I to represent the
probability of 23 for the ith secondary user under
hypothesis H 0 and H 1 respectively, we have;
1,=23|1
=2,2 2,3
(14)
0,=23|0 =,22
,32
(15)
1=>3|1= 2,3 (16)
0=1<<2= ,12
,22
(17)
Cluster Based Triple Threshold Energy Detection for Spectrum Sensing in Vehicular Ad-Hoc Networks
1498
Published By:
Blue Eyes Intelligence Engineering
& Sciences Publication
Retrieval Number: F2649037619/19©BEIESP
0=1= 1 ,12
(18)
=<2|1= 2,1 2,2
(19)
=>3|0= ,32
(20)
IV. RESULTS AND DISCUSSION
In the simulations, the main emphasis was to show that
triple threshold energy detection method at the local node
stage produces better results in comparison with the single
and double threshold methods. The time bandwidth factor
was chosen as 1000 with the sample points given as N=
=2TW where TW is the time bandwidth factor. A range of
-15dB to 5dB is taken as signal to noise ratio in this work
while the probability of false alarm ranges from 0.01 to 1.
QPSK modulation was also used here with the modulation
index of 4. The minimum threshold λ1 was set at 3W to
eliminate any unreliable small scale PUs since most of these
unreliable PUs have powers of less than 1W. The simulation
results can be seen as follows;
Figure 3: A Plot of SNR vs Pd
From Figure 3, it can be seen that the conventional single
threshold method has low probability of detection as low
SNR. This can be attributed to the fuzzy region which is not
clearly defined.
Figure 4: A Plot of SNR vs Pm
Double and triple energy detection methods performed
relatively better than single threshold method. The double
threshold performed better than triple threshold for very low
SNR but the triple threshold performed better from around
-13dB SNR and reached maximum Pd ahead of the double
threshold method.
Similarly to Figure 3, Figure 4 shows the triple threshold
method had lower probability of missed alarm as compared to
the other two methods.
V. CONCLUSION
The use of cognitive radios has been show in various
researches as the solution for the scarcity in spectrum.
Spectrum sensing is an essential component in CR of which
its inefficiencies will lead to interference to the licensed
primary user. In this paper, a proposed triple threshold has
been shown to be able to provide better spectrum sensing
efficiency than the double threshold and single threshold
energy detection methods. Further research to this work is to
show that cooperation of SUs by fusing the sensing results
can provide results for effective allocation and occupation of
spectrum homes when PU is absent and vacating the same
when the PUs resumes using the spectrum.
REFERENCES
1. K. D. Singh, P. Rawat, and J. Bonnin, “Cognitive Radio for Vehicular
Ad Hoc Networks ( CR-VANETs ): Approaches and Challenges,”
EURASIP J. Wirel. Commun. andNetworking, vol. 1, no. 49, pp. 1–22,
2014.
2. X. He, W. Shi, and T. Luo, “Survey of cognitive radio VANET,” KSII
Trans. Internet Inf. Syst., vol. 8, no. 11, pp. 3837–3859, 2014.
3. S. K. Pawar, A. Awasthi, and M. Gurjar, “Dynamic and Cooperative
Spectrum Sensing in,” pp. 9535–9542, 2018.
4. G. Martuscelli, A. Boukerche, L. Foschini, and P. Bellavista, “V2V
protocols for traffic congestion discovery along routes of interest in
VANETs: a quantitative study,” Wirel. Commun. Mob. Comput., vol. 16,
no. June 2009, pp. 2907–2923, 2016.
5. B. Liu, B. Khorashadi, D. Ghosal, C. Chuah, and H. M. Zhang,
“Analysis of the Information Storage Capability of VANET for
Highway and City Traffic,” Transp. Res. Part C, vol. 23, pp. 68–84,
2012.
6. V. H. La and A. Cavalli, “S ECURITY ATTACKS AND SOLUTIONS
IN V EHICULAR A D HOC N ETWORKS : A S URVEY,” vol. 4, no.
2, pp. 1–20, 2014.
International Journal of Recent Technology and Engineering (IJRTE)
ISSN: 2277-3878, Volume-7, Issue-6, March 2019
1499
Published By:
Blue Eyes Intelligence Engineering
& Sciences Publication
Retrieval Number: F2649037619/19©BEIESP
7. A. Soua, “Vehicular ad hoc Networks : Dissemination , Data Collection
and Routing : Models and Algorithms,” Universite Pierre et Marie
Curie, 2013.
8. V. Malagar, S. Mata, and V. Devi, “Performance Evaluation of
VANETs for Evaluating Node Stability in Dynamic Scenarios,” Int. J.
Comput. Appl. Technol. Res., vol. 7, no. 9, pp. 376–385, 2018.
9. C. Cooper, D. Franklin, M. Ros, F. Safaei, and M. Abolhasan, “A
Comparative Survey of VANET Clustering Techniques,” IEEE
Commun. Surv. Tutorials, vol. 19, no. 1, pp. 657–681, 2017.
10. J. M. Lim, Y. C. Chang, M. Y. Alias, and J. Loo, “Cognitive radio
network in vehicular ad hoc network ( VANET ): A survey,” Cogent
Eng., vol. 29, no. 1, pp. 1–19, 2016.
11. F. Riaz et al., “Multi-Homing Enabled Cognitive Radio Based
Inter-Vehicle Communication System,” Int. J. Comput. Sci. Inf.
Technol., vol. 4, no. 6, pp. 77–95, 2012.
12. C. R. Ad, V. S. Rao, R. V. Prasad, C. Yadati, and I. G. M. M.
Niemegeers, “Distributed Heuristics for Allocating Spectrum in CR Ad
hoc Networks,” in IEEE Globecom 2010, 2010, pp. 1–7.
13. I. F. Akyildiz, B. F. Lo, and R. Balakrishnan, “Cooperative spectrum
sensing in cognitive radio networks : A survey,” Phys. Commun., vol. 4,
no. 1, pp. 40–62, 2011.
14. K. Baraka, L. Safatly, H. Artail, A. Ghandour, and A. El-hajj, “An
Infrastructure-Aided Cooperative Spectrum Sensing Scheme for
Vehicular Ad Hoc Networks,” Ad Hoc Networks, vol. 25, pp. 197–212,
2015.
15. R. Adriano and A. De Souza, “Simulation Platform for Performance
Analysis of Cooperative Eigenvalue Spectrum Sensing with a Realistic
Receiver Model Under Impulsive Noise,” Veh. Technol. - Deploy. Appl.
INTECH Open Sci., pp. 171–198, 2013.
16. J. M. Lim, Y. C. Chang, M. Y. Alias, and J. Loo, “Cognitive radio
network in vehicular ad hoc network ( VANET ): A survey,” Cogent
Eng., vol. 29, no. 1, 2016.
17. E. Souza, I. Nikolaidis, and P. Gburzynski, “A new Aggregate Local
Mobility (ALM) clustering algorithm for VANETs,” in IEEE
International Conference on Communications, 2010, pp. 1–5.
18. D. P.Patil and V. M.Wadhai, “Performance Evaluation of Different
Spectrum Sensing Techniques for Realistic Implementation Oriented
Model in Cognitive Radio Networks,” Int. J. Wirel. Mob. Networks, vol.
6, no. 6, pp. 57–69, 2014.
19. S. Miah, H. Yu, T. K. Godder, and M. Rahman, “A Cluster-Based
Cooperative Spectrum Sensing in Cognitive Radio Network Using
Eigenvalue Detection Technique with Superposition Approach,” Int. J.
Distrib. Sens. Networks, vol. Volume 201, pp. 1–11, 2015.
20. Z. Li, F. R. Yu, and M. Huang, “A Cooperative Spectrum Sensing
Consensus Scheme in Cognitive Radios,” in IEEE Globecom 2009,
2009, pp. 2546–2550.
21. F. A, O. J.B, and M. A. Abubakar, “Threshold Optimization of Energy
Detector Using Fuzzy Logic for Cognitive Radio,” Jorind, vol. 14, no. 1,
pp. 1–12, 2016.
22. A. Bagwari and G. Tomar, “New Cooperative Spectrum
DetectionTechnique in Cognitive Radio Networks,” nternational J.
Wirel. Mob. Commun. Ind. Syst., vol. 3, no. 1, pp. 43–58, 2016.
23. B. Ahuja and K. Sharma, “Adaptive Double Threshold based Spectrum
Sensing for Cognitive Radio Networks,” Int. J. Energy, Inf. Commun.,
vol. 5, no. 6, pp. 1–16, 2014.
24. W. Ejaz, N. Hasan, M. A. Azam, and H. S. Kim, “Improved local
spectrum sensing for cognitive radio networks,” EURASIP J. Adv.
Signal Process., vol. 242, no. 1, pp. 1–12, 2012.
