Chinese Journal of Electronics
Vol.23, No.2, Apr. 2014
Time Agility Optimization for Dual-stage
Collaborative Spectrum Sensing
∗
CUI Cuimei, WANG Yiming
(School of Electronics and Information Engineering, Soochow University, Suzhou 215021, China )
Abstract — Cooperative spectrum sensing has been
shown to be an effective method to mitigate the impact
of hidden terminal and shadow fading in cognitive radio
networks. Currently most works focused on either coop-
erative sensing or sensing scheduling as a viable means to
improve the detection performance without studying their
interactions. This paper proposed a dynamic and variable
time-division multiple-access scheduling mechanism that
was incorporated into a Dual-stage collaborative spectrum
sensing (DCSS) model in a less ideal radio environment.
Moreover, this paper derived closed-form expressions of
sensing time for DCSS, and addressed the critical range of
timeslot length. An optimized algorithm to minimize sens-
ing time was deduced and verified. The simulation results
indicate that average sensing time with DCSS is shortened
by 11.5% when compared with that of the Single collabora-
tive spectrum sensing (SCSS) while guaranteeing detection
error rate requirement of less than 1%.
Key words — Cognitive radio, Dual collaboration spec-
trum sensing, Time-division multiple-access, Time agility,
Optimization.
I. Introduction
In Cognitive radio networks (CRNs), spectrum sensing is
known to be a very challenging task due to hidden Primary
users (PUs) and the channel fading and shadowing. The co-
operative spectrum sensing method is one viable solution for
tackling these problems by taking advantage of the spatial di-
versity from multiple Cognitive radio users (CRs).
A comprehensive survey on Cooperative spectrum sensing
(CSS) was captured in Ref.[1]. In Refs.[2-3], the authors ex-
ploited the spatial diversity inherent in a multiuser scenario
to analyze the relay-based CSS schemes with the Amplify-
and-forward (AF) protocol. We note, however, that each
approach has its own set of advantages and drawbacks and
hence, in practice one may need to utilize a combination of
various CSS techniques. We also note that performance of
spectrum sensing not only relies on detection methods, but
also depends on sensing actions’ scheduling. CSS schedul-
ing has been studied in recent literature
[4]
. However, this
scheduling scheme is unsuitable for CRs at lower received
power.
In this paper, a relay-based DCSS model is proposed to
improve the detection accuracy and reliability to better pro-
tect PUs and capture the spectral opportunities in deep fading
environments. This scheme is conducted by following two suc-
cessive collaboration processes, i.e., AF cooperation and data
fusion. A natural question to be asked is: how much time is
spent for this DCSS? While intuitively, we hold that it will
take more detection time for this DCSS scheme to achieve
better sensing accuracy when compared with Single cooper-
ative spectrum sensing (SCSS), in practice we don’t know if
this belief is true. In order to find a solution while consid-
ering the dynamic nature of CRNs, a Dynamic and variable
time-division multiple-access (DV-TDMA) sensing scheduling
mechanism is proposed in substitution of the conventional
fixed TDMA
[5]
scheduling scheme. Using the mathematical
expectation, we came up with closed-form solution of the av-
erage sensing time for the proposed DCSS scheme over fad-
ing channels. Finally, in order to meet the requirements
in IEEE 802.22 that PUs for all signal types should be de-
tected within 2 seconds with the probabilities of missed de-
tection and false alarm detection no greater than 0.1
[6]
,and
inspired by the method in Ref.[7], we propose an algorithm
in term of the optimal number of collaborating CRs which
helps to determine how many CRs are needed to collaborate
in the data fusion process. In this scheme, the average sens-
ing time is minimized while specified detection error rate is
ensured.
II. The Dcss Scheme
1. System model
The system setup is a CRN consisting of N cooperating
pairs and a receiver (CR Rx), and cooperating pair i ∈ [1,N]
is composed of a CR user U
i
and a CR relay user R
i
.All
CRs opportunistically operate on a set of channels allocated
to PUs located in the same geographical region and experience
independent and identically distributed (i.i.d.) Rayleigh fading
∗
Manuscript Received Sept. 2012; Accepted July. 2013. The work is supported by National Natural Science Foundation of China
under Grant (No. 60872003 and No. 61172056), Doctoral Fund of Ministry of Education of China under Grant (No. 20093201110005).