A review of the noise uncertainty impact on energy detection with different OFDM system designs

Abstract Cognitive radio networks (CRN) based on spectrum sensing represent intelligent wireless communication technology dedicated to a more efficient exploitation of the available frequency spectrum. Although the energy detection (ED) method was found to be a promising candidate for spectrum sensing in the CRN, its detection performance is challenged by the noise fluctuations. These fluctuations, known as noise uncertainty (NU), may vary beyond what is estimated due to changes in temperature, interference and filtering. In this work, the influence of NU on the performance of ED for signals transmitted using an orthogonal frequency division multiplexing (OFDM) technique is reviewed. Besides that, thorough analyses are performed by means of extensive simulations of the ED process for three different OFDM system designs based on rate adaptation, margin adaptation and mutual rate and margin adaptation. The analyses presented in this review paper give a systematic insight into how various OFDM modulations, NU levels, probabilities of a false alarm, number of samples used in the ED process and levels of signal-to-noise ratio impact the probability of signal detection and the overall ED performance of different OFDM system designs. The results obtained through simulations show that the trade-off among the parameters analyzed can bring improvements in the ED process of different OFDM system designs. The research challenges for improvement of the main ED weaknesses have been further discussed, with a performance comparison of the ED method with other prominent local spectrum sensing methods. The survey results presented constitute a reference for improvements of the broadly-accepted ED approach.