Spectrum sensing using frequency domain entropy estimation and its FPGA implementation for cognitive radio

Abstract

Spectrum sensing is an important module in cognitive Radio (CR) technology. This paper presents a spectrum sensing technique based on entropy estimation of the received signal in frequency domain. This approach is extended for cooperative sensing using different fusion rules. Performance of the algorithm is analyzed using Monte-Carlo methods. The performance of proposed entropy based detection is compared with energy based detection technique. Simulation results reveal that, entropy based detection algorithm can detect signals of signal-to-noise ratio upto -24 dB using five nodes for detection probability Pd ≥ 0.9 and false alarm probability Pf ≤ 0.1. Sensing algorithm of single node is successfully implemented in Virtex4 FPGA.