Esearch On Improving Sensing Time Of OFDM Blind Sensing Algorithm

Spectrum sensing is one of the key technologies of cognitive radio. As primary users’ information is mostly confidential, it is difficult in a certain degree to obtain priori information. Meanwhile, cognitive radio should be able to run on different platforms and communication standards. Thus efficient spectrum sensing method should be blind and require no priori information of any communication environment. This paper conducts optimization study of the perception time of blind OFDM spectrum sensing algorithm in response to the problem of too long perception time needed at present.Firstly, the paper reviews the research status and significance of spectrum sensing, the key technology of cognitive radio system. Theoretical analysis and simulation research of three classic single-user spectrum sensing algorithms are included in this paper. It analyze the effect generated by perception algorithms, collaborative approach, perception time, perception cycle and other various parameters on perceived performance from three aspects: local sensing technology, cooperative sensing technology and optimization of perception mechanism.Secondly, in response to shortcomings of long sensing time and not conducive to implementation issue of the blind sensing algorithm mentioned by Guzzon E, the paper raises the double-blind adaptive spectrum threshold sensing algorithm, which costs shorter sensing time. This algorithm combines easy-used energy detection and blind sensing algorithm with using a self-adaptive blind double threshold, and improves step formula of unreliable region based on self-adaptive double threshold algorithm. Then it compares the simulation of the double-blind adaptive spectrum threshold sensing algorithm and the blind sensing algorithms of computing speed and sensing performance. Simulation results show that the performance of the two algorithms are of equal rank, and the double-blind adaptive spectrum threshold sensing algorithm raised in this paper need shorter time. Meanwhile we simulate the effects of the two key parameters on the algorithm performance.Finally, to resolve the problem, raised by Den Lilin, of long perception time needed finding the optimal threshold vector by using genetic algorithm, this paper will make use of particle swarm optimization, which is easily implemented, highly precise and with fast convergence, into multi-band joint sensing to find the optimal threshold vector and also give the flowcharts and algorithm steps of multi-band sensing algorithm based on PSO. The simulation results show that multi-band joint sensing algorithm based on PSO presented in this paper achieves a higher sensing probability within the same time than genetic multi-band joint sensing algorithm. It also conducts a simulation analysis of the influence of various parameters of PSO on perceived performance of multi-band joint sensing algorithm.