인공 신경망 기반 레이더 표적 우선순위 할당에 대한 연구

Abstract

This paper introduces the radar resource management (RRM) technique for the multi-function radar (MFR). The MFR embarked on an aircraft or a warship conducts various tasks such as surveillance, tracking, target recognition, and missile guidance. In this case, the RRM technique is necessary to operate all these tasks because several resources (i.e., time, energy) used in the MFR are limited. If the RRM technique is applied to the MFR in case of tracking multiple targets, the resource can be efficiently used by calculating priorities of each individual target (target prioritization). It means that the MFR can cope with the hazardous situation by keeping track of the high-priority target first. In this dissertation, a neural network learning algorithm based on the steepest descent method which is more suitable for target prioritization is proposed as compared to the conventional gradient descent method. Several simulation results show that the proposed scheme is much more superior to the traditional neural network model in terms of accuracy and relevance of scoring the priority.