Multifunction cognitive radar task scheduling using Monte Carlo tree search and policy networks

作     者：Shaghaghi, Mahdi Adve, Raviraj S. Ding, Zhen 

作者机构：Univ Toronto Dept Elect & Comp Engn Toronto ON Canada DRDC Ottawa ON Canada 

出 版 物：《IET RADAR SONAR AND NAVIGATION》 (IET雷达、声纳与导航)

年 卷 期：2018年第12卷第12期

页      面：1437-1447页

核心收录：

学科分类：0810[工学-信息与通信工程] 0808[工学-电气工程] 08[工学] 

基　　金：Natural Sciences and Engineering Research Council (NSERC) of Canada Defence Research and Development Canada (DRDC) 

主　　题：approximation theory computational complexity decision making learning (artificial intelligence) Monte Carlo methods radar computing telecommunication scheduling tree searching 

摘      要：A modern radar may be designed to perform multiple functions, such as surveillance, tracking, and fire control. Each function requires the radar to execute a number of transmit-receive tasks. A radar resource management (RRM) module makes decisions on parameter selection, prioritisation, and scheduling of such tasks. RRM becomes especially challenging in overload situations, where some tasks may need to be delayed or even dropped. In general, task scheduling is an NP-hard problem. In this work, the author develops the branch-and-bound (B&B) method which obtains the optimal solution but at exponential computational complexity. On the other hand, heuristic methods have low complexity but provide relatively poor performance. They resort to machine learning-based techniques to address this issue;specifically, they propose an approximate algorithm based on the Monte Carlo tree search method. Along with using bound and dominance rules to eliminate nodes from the search tree, they use a policy network to help to reduce the width of the search. Such a network can be trained using solutions obtained by running the B&B method offline on problems with feasible complexity. They show that the proposed method provides near-optimal performance, but with computational complexity orders of magnitude smaller than the B&B algorithm.