Cognitive detector for complex multi-source environment: Preliminary results based on background perception for HFSWR

Abstract

A new target detection architecture, designated as Cognitive Detector, is proposed which aims at solving the problems arising from detecting targets in various non-stationary, time-variant and multi-source clutter environment. Being aware of the change of the detection environment can help select an optimized detection strategy. Thus image character of detection background and the scene analysis result, which is the information that hasn't been make full use of in existing detection systems, is utilized to adjust the detection algorithm adapting for the various background and add more information about the target/clutter to enhance the system's performance. After describing the definition of cognitive detection and its operation processes, the paper focuses on the specific application of high frequency surface wave radar (HFSWR), for which the cognitive detector is rather well suitable. Compared with the classical and the Knowledge-Based (KB) detector, the proposed methods based on the background perception can extract the information of detection environment on-line by a single type of sensor and be expected to achieve a better performance.