Analysis of Multiband Sensing-Time Joint Detection Framework for Cognitive Radio Systems

An optimal multiband spectrum sensing framework is explored which jointly detects a wideband orthogonal frequency-division multiple access primary signal over multiple non-contiguous narrowband channels. The spectrum sensing problem optimizes the aggregate opportunistic throughput of a cognitive radio network restraining the interference to the primary user keeping the individual energy detector thresholds and sensing slot duration as optimization variables. The problem is solved as convex optimization under certain practical constraints for any input signal and channel noise conditions. Simulated results for BPSK modulated signal and real-valued Gaussian noise show the increase in secondary throughput when both subchannel thresholds and sensing time is adaptively chosen. The effect of sub-channel parameter variation on the overall problem is also shown.