Implementation of an autocorrelation-based spectrum sensing algorithm in real-world channels with frequency offset

Abstract

This work presents a testbed implementation of a spectrum sensing algorithm for cognitive radio that is based on the autocorrelation function. Much of the work in current literature uses simulation based approaches to characterize functionality. In contrast here, the algorithm is applied in real-world channels and compared with appropriate simulations. It is shown how the algorithm may be improved to overcome the problem of frequency offset, which is a hardware-based impairment that current literature on the algorithm generally does not consider.