Least squares based channel estimation approach and Bit Error Rate analysis of Cognitive Radio

Cognitive Radio (CR) is very essential in day-to-day life as there is a growing demand for usage of frequency spectrum for different wireless applications. Being an opportunistic communication scenario, cognitive radio performance analysis attains prime significance. Performance analysis of a cognitive radio scenario in a wireless channel is done by a well-known performance metric Bit Error Rate (BER). However, the wireless channel coefficients or channel state information (CSI) are always unknown in any practical cognitive scenario. Hence, it needs to be estimated by estimation approaches like least squares (LS), minimum mean square error (MMSE) and maximum likelihood (ML). Least squares approach of estimation of wireless channel coefficients satisfies linearity property and is simple in computation, however it produces a lesser Mean Square Error (MSE) in comparison to MMSE and ML based approaches. But MMSE and ML approaches are based on the probability density function (PDF) which generally leads to increased computational complexity to determine the estimate of the wireless channel. However, a moderate MSE performance is sufficient for performance analysis of a CR scenario due to spectrum sensing. Least squares based approach provides an estimate of the wireless channel with very less computational complexity. Hence, this paper uses LS to estimate the wireless channel coefficients. In addition bit error rate performance of a cognitive radio scenario is analyzed using the obtained LS estimate of the wireless channel. Simulation results are obtained to analyze the LS performance of CR scenario using mean square error and BER performance metric. The obtained simulation results can be used as a benchmark for analysis of cognitive radio environments.