Researches On The Surveillance Technology Of Vessel Movement

This dissertation focuses on the surveillance technology of vessel movements. It contains three parts:In the first part, an improvement of radar azimuth resolution is studied. A Tank-Hopfield neural network is proposed and proved for removing the convolution effect which is resulted from azimuth sweeping of Radar antenna pattern, so as to break the boundary of azimuth resolution limited by antenna beam width.In the second part, the Automated Identification System based on Self-Organized TDMA(STDMA) is studied. A thorough evaluation of STDMA protocol is made, thereby the bottleneck of the surveillance performance of STDMA-AIS is highlighted. Then a solution for improving STDMA-AIS is proposed.In the third part, an extended analysis of the AIS's throughput on an interfered channel is conducted specially. A Marcov-random-processes model is employed for the precise assessment and deduction of the throughput of STDMA-AIS.For comparison, a thorough evaluation of Digital Selective Calling protocol for AIS is made as well.