Comparison of spectrum sensing in MDFT modulated filter banks and OFDM based cognitive radio systems

When using for cognitive radios (CR), the traditional Orthogonal Frequency Division Multiplexing (OFDM) technology with its drawback of the spectral leakage will cause interference between neighboring zone and also reduce the primary user signal detection performance. This paper proposed a multi-band spectral detection scheme based on Modified Discrete Fourier Transform (MDFT) modulated filter bank. The detection probability expression of the cognitive radio system under MDFT modulated filter bank structure was derived. Theoretical analysis and simulation results show that the CR system based on MDFT filter bank has higher stop-band attenuation than the system based on OFDM with using rectangular window function as prototype filter. And with the ideal channel or the pedestrian channel of 3GPP TS 25.104 protocol standards, especially at lower SNR, the spectrum sensing performance of the CR system based on MDFT filter bank structure is significantly improved compare to the system based on OFDM.