Research On Dual-threshold Energy Detection In Cognitive Radio

With the advancements in wireless communication, the shortage of spectrum is getting more serious. The current static spectrum allocation schemes can never meet with the development requirements. As a result, cognitive radio arises to improve the spectrum utilization via taking advantage of dynamic spectrum access technology which allows cognitive user to make use of the spectrum by causing no harmful interference to the primary user. Spectrum sensing is one of the key technologies of cognitive radio.Firstly,several common spectrum sensing technologies are introduced, among which energy detection is discussed in details. Referred to relevant literatures, the detection performance and parameters are analyzed. Secondly, due to these disadvantages of traditional energy detection, two improved algorithms based on dual-threashold energy detection are proposed.Taking advantages of the sparsity of signal, a dual-threshold cooperative spectrum sensing algorithm based on differential energy detection is proposed. The energy difference is small when only noise is in the band and it is large when there`s the primary user`s signal, thus the band state can be determined from the energy difference. The proposed algorithm applies a novel differential algorithm between the two thresholds of traditional dual-threshold energy detection algorithm. Simulation results show that the proposed algorithm performs better than the traditional energy detection.Taking into account of the variability and uncertainty of the wireless channel, a dual-threshold energy detection based on adaptive sampling number is proposed. The two thresholds are given accordingly to noise uncertainty. And between the two thresholds, a novel adaptive sampling number algorithm is used. In order to avoid too much energy consumption, an optimal upper bound of the sampling number is given by trading-off the energy consumption and throughput. Both theoretical and emulational practicability and advantage are proved, which demonstrated that the proposed algorithm can efficiently prove the detection performance although there is a little more energy consumption.