PERFORMANCE OF ENERGY-EFFICIENT COOPERATIVE SPECTRUM SENSING FOR DETECTING LICENSED USERS IN ASPECTRUM SHARING SYSTEM USING EIGENVALUE DETECTOR

Abstract

Licensed User (LU) detection is a critical operation in Spectrum Shearing System (SSS) to prevent
Licensed User (LU) from harmful interference. However, channel impairment such as multipath
fading, shadowing, and path loss affect the sensing accuracy of SSS. Non-Cooperative SS (NCSS)
used to solve this problem suffers from receiver uncertainty resulting in a high Probability of Missing
(PM) that causes interference, while Cooperative SS (CSS) used to solve the problem of NCSS also
suffers from large reporting overhead resulting in long sensing time, bandwidth inefficiency and
energy inefficiency. Hence, this work proposes an Energy-Efficient CSS (EECSS) for signal
detection in a CRN using cluster and Eigenvalue Detector (EVD). Secondary Users (SUs) 'L'
numbers (4, 5, 6) are jointly used to carry out SS and each SU performs local sensing using EVD. The
received signals are then used to form a square matrix to determine the ratio of maximum to minimum
eigenvalue and used as test statistics to determine the idleness of the spectrum. SU selection is carried
out based on the Signal Noise Ratio (SNR) of the received signal to determine the SUs that participate
in the local sensing at that time. The 'L' number of SUs is used to form 'N' (3, 4, 5) clusters to reduce the
reporting overhead between SUs. Global sensing result is determined by combining the sensing results from individual clusters using the OR fusion rule. The model is simulated using MATLAB R2018a and evaluated using Probability of Detection (PD) and Sensing Time (ST).

Keywords: Eigenvalue Detector (EVD), Probability of Detection (PD), Sensing Time (ST), Secondary User (SU), Primary User (PU)