Low power DFT filter bank based two-stage spectrum sensing

Abstract

The paper presents a low power discrete Fourier transform (DFT) filter bank based two-stage spectrum sensing with energy detector (ED) for the first stage coarse sensing, complemented by a cyclostationary feature detector (CFD) for fine spectrum sensing. The dynamic power of the sensing operation depends on the sampling rate at which computations are performed. By employing polyphase DFT filter bank prior to spectrum sensing, the sampling rate can be reduced by M (the number of subbands). This in effect reduces the computational complexity of the design which consequentially leads to lowering the power consumed. As the spectrum is sensing can be done in parallel, the speed will be M times that of serial spectrum sensing algorithm.