Analysis of combined data-decision fusion scheme for cognitive radio networks

Abstract

Cognitive radio (CR) is a promising solution to solve the problem of radio spectrum scarcity faced by wireless devices. Cooperative Spectrum sensing improves the detection performance of the Primary User (PU). For ideal case, reporting channels between different Secondary Users (SUs) and Fusion Centre (FC) are considered as non-fading. However, in practical scenario fading introduces some error data transmitted from the SUs to FC. In this paper we study a combined hard decision and soft data fusion scheme which provides a mutual concession between hard and soft fusion scheme by offering advantages of both schemes. We first calculate the Minimum Mean Squared Error (MMSE) for all Rayleigh faded reporting channels using the training symbols. SUs having least MMSE take hard decisions, while other remaining SUs takes soft decision. The FC finally decides presence of PU based on the analysis of the received sensing information. We used diversity techniques such as maximal ratio combining and equal gain combining to abbreviate the effect of fading. We have also discussed the effect of number of SUs on the missed detection probability and total error probability.