Abstract

The evolution of the cognitive radio is for the purpose of meeting the optimization goal of the limited spectrum availability and this purpose is achieved by the spectrum sensing process. The spectrum sensing problem is formulated as a detection problem with the two basic hypotheses namely Hypo-0 (H0) and Hypo-1(H1). The formulation of the Hypo-0 takes various parameters including: a time instance of signal sample collection, number of samples, Gaussian white noise at particular variance and the formulation of Hypo-1, it takes a parameter propagation channel that evaluates the power arrival as interference. There are various methods evolved for this purpose and out of that the typical energy detection is one of the popular methods which operates without any prior or heuristic of the signal information but it does not perform well in the low noise conditions. This paper proposes an evaluation framework for the problem formulation of the spectrum sensing with two fundamental hypotheses: H0 and H1, where both optimal energy detection methods and the methods for low SRN conditions are benchmarked by statistical analysis. This is one of the unique frameworks which is not found in the literature for the purpose of the benchmarking the spectrum sensing techniques. The performance evaluation of ten cooperative spectrum sensing methods including: a) Maximum Eigen value detection (MED), b) Generalized likelihood ratio (GLR), c) maximum-minimum eigenvalue detection (MMED), d)energy detection(ED), e) AGM (arithmetic to geometric mean(AGM) f) Hadamard Ratio(HR), g) Volume based(VD), h) Gershgorin radii centres ratio(GRCR), i) Gini index detection(GID) and j) Proposed Rician , rice factor based detection(RFD). The benchmarking results is performed for the probability of detection verses varying signal to noise from low to high SNR and varying number of Secondary users or cognitive users. The proposed: Racian rice factor based detection(RFD) demonstrates higher probability of detection in both low and high noise conditions and also in the increasing number of secondary or cognitive radio.