AN EFFECTIVE ADAPTIVE WAVEFORM SELECTION ALGORITHM FOR TARGET TRACKING IN COGNITIVE RADAR SYSTEM

摘要：

During these years, a new idea called cognitive radar has been proposed, in which the waveforrns can be adaptively selected to be used in future epochs according to the current estimate of the environments. In this paper, we describe an effective adaptive waveform selection algorithm for target tracking in cognitive radar system. An infinite horizon partially observable Markov decision process (POMDP) framework is presented to solve the waveform selection problem, which do not relying on analytic expression for belief state. Furthermore, we develop a Monte Carlo solution method that combines particle filtering for non-Gaussian, nonlinear belief-state estimation and a sampling-based Q-value approximation for solving the FOMDP via lookahead. The proposed method minimizes target tracking errors and is useful for very large state space.

关键词： Adaptive waveform selection POMDP Particle filter Q-value approzimation

作者: JINKUAN WANG JING LI FULAI LIU BIN WANG

作者单位: Engineering Optimization and Smart Antenna Institute, Northeastern University at Qinhuangdao, China

会议类型: 国际会议

会议名称: The Second International Conference on Information & Systems Sciences(ICISS2008)(第二届信息与系统科学国际会议)

会议地点: 大连

会议语种:英文

页码: 683-688

在线出版日期: 2008-12-18（万方平台首次上网日期，不代表论文的发表时间）