A fast spectrum sensing for CP-OFDM cognitive radio based on adaptive thresholding

• A spectrum sensing (SS) method for detecting unsynchronized OFDM signals in additive white Gaussian noise (AWGN) is proposed.
• An adaptive CFAR spectrum sensing algorithm is considered based on the generalized likelihood ratio test (GLRT).
• The SS detector has a capability of adaption to noise fluctuations by maintaining a desired probability of false alarm constant.
• The proposed detector can be considered as a good candidate for practical implementation of SS for OFDM signals.

Recently, cyclostationarity (CS) based detection methods exploiting the autocorrelation periodicity property of the orthogonal frequency division multiplexing (OFDM) signals attracted a lot of attention. These detection methods are more complex than energy detection but they have better detection performance in low-SNR regimes. The drawback, however, is their extensive computational complexity. In this paper, we propose a computationally efficient spectrum sensing method for detecting unsynchronized OFDM signals in additive white Gaussian noise (AWGN). The proposed method exploits the second-order CS property of OFDM signals to set an adaptive threshold, which achieves the desired constant false alarm rate (CFAR) property. The complexity of the proposed method is significantly reduced compared with the scheme proposed in [22]. Monte-Carlo simulations illustrate that the performance of the proposed detector outperforms the classical energy detector (ED), the sliding window detector (SW) and Axell's detector in addition to guaranteeing a low sensing time.