Research On Resource Management Technology For Phased Array Radar And Its Network In Tracking System

Sensor resource management can be described as a system or process that provides automatic or semi-automatic control of a suite of sensors or measurement devices, which improve the performance of sensing and give an effective usage of sensor resources. The sensor resource management problems can be classified as the management problems of single sensor and multiple sensors. We focus our attention on the management problem of phased array radar (PAR) resources and multi-radar tracking fusion system.The first part of this dissertation studies the PAR management problem, which include the modeling for a phased array radar and its resource management algorithms. The second part disscusses the sensor management problem in multi-radar tracking fusion system, where the centralized and distributed architecture of tracking fusion system are the key topics.In the first part of this dissertation, a set of algorithms are presented for the PAR management, which use the interacting multiple model (IMM) estimator in combination with the modified probabilistic data association (MPDA) technique for tracking highly maneuvering targets in the presence of FA and ECM. Techniques for neutralizing the effects of two types of ECM, namely, an SOJ and RGPO, are presented. Algorithms for estimating the RCS of the target and the unknown jammer power level and schemes for the adaptive selection of the radar waveform, detection threshold, and sampling interval are also given. The performance of these algorithms for management is analyzed under different conditions.In the next part, a general optimization model of sensor management based on covariance control strategy is presented. In centralized multi-radar tracking fusion system, an off-line steady-state and an on-line adaptive sensor allocation algorithm are presented for different time in submission of allocation. Two algorithms are presented to resolve the problem of variable sampling interval, which are combined with the adaptive sensor allocation algorithm result in two different joint adaptive sensor management algorithms. This dissertation analyzes the performance of these algorithms under different conditions, some simulation results are also given.Finally, the sensor management problem in distributed multi-radar tracking fusion system is studied. For two distributed fusion architectures, which are sensor-level tracking and fully distributed network, this dissertation proposes two covariance control-based distributed joint management algorithms. Since there is no fixed superior/subordinate relationship in fully distributed network, this dissertation proposes an ordered node algorithm which use a token for each node to making sensor management decision in turn. Experiment results validates the above two algorithms.