基于随机集的多主用户多维信息感知算法研究

Random sets theory based multi-dimensional spectrum sensing with multiple primary users

移动认知网络多主用户(Primary User,PU)信号检测问题是当前认知无线电领域面临的主要问题之一.文中提出将随机集(Random Sets,RS)引入移动认知网络的频谱感知过程中,构建多主用户运动模型及观测模型,通过粒子概率假设密度滤波(Particle Probability Hypothesis Density Filter,P-PHDF)算法实现对主用户数量及主用户状态(位置、速度、使用频率、信号接收角度)的实时检测.较传统感知方法,基于随机集的频谱感知方法不仅能准确跟踪检测活跃主用户的数目,同时还能跟踪检测出主用户的具体位置、使用频率及信号接收角度等信息.仿真结果表明,在移动环境下文中提出的多主用户实时感知方法的检测性能良好,并且能有效地抵抗杂波等的干扰,实现了多维移动认知网络中对检测区域内主用户数量及状态的实时检测跟踪.

Multiple primary user signal detection in cognitive mobile network is one of the main cognitive radio problems. This paper introduces random sets theory into the process of mobile network cognitive spectrum sensing, and builds the motion model and the sensor model of multiple primary users, and uses Particle Probability Hypothesis Density Filter to realize real-time detection of the primary users, including the number and status (position, velocity, frequency, signal reception angle) of each primary user. Compared with traditional spectrum sensing methods, the proposed method can track the number of primary users, the position, frequency, as well as arrival of angle. Simulation results show that the random set theory for multidimensional cognitive mobile network can be realized on real-time detection and update of the state of each primary user. It can reliably and effectively detect the number and status of primary users with high capacity of resisting disturbance. Copyright © 2015 by Editorial Department of Chinese Journal of Radio Science