Antenna array signal processing based methods for target detection and source estimation

• Main Author: Huang, Ling
• Other Authors: Lu Yilong
• Format: Theses and Dissertations
• Language: English
• Published: 2014
• Subjects: DRNTU::Engineering::Electrical and electronic engineering::Antennas, wave guides, microwaves, radar, radio; DRNTU::Engineering::Electrical and electronic engineering::Electronic systems::Signal processing
• Online Access: https://hdl.handle.net/10356/58904
• Institution: Nanyang Technological University

Antenna arrays provide significant contributions in areas such as wireless communications, radar and sonar. Antenna array signal processing has been gaining great interests in two major types of applications, target detection and source estimation. Target detection is important in areas such as fire rescuing, anti-terrorist and surveillance, in which through-the-wall (TTW) target detection and line-of-sight (LOS) target detection are equally demanded. Source estimation is important in areas such as vehicle tracking, patient safety, medical instrumentation monitoring and direction finding. This thesis studies antenna array signal processing methods for radar target detection and source direction estimation. For radar target detection, both TTW and LOS target detections using radar imaging methods are studied. Since there are not many research works reported on tomographic imaging in TTW target detection, a schematic study on static target detection behind walls using single-view and multiple-view tomographic imaging methods are carried out. The findings show that multiple-view method gives relatively good performances both with and without walls. This study is only theoretical but it provides a potential direction of using the multiple-view method in TTW applications with loose time and space constraints. Besides, a random sparse array (RSA) and compressed sensing (CS) based target detection method in LOS condition is proposed since the approach of using a RSA and CS on target detection has not been investigated by other researchers. A RSA reduces hardware cost and simplifies system implementations and CS reduces the amount of data required. The RSA is characterized into two modes, the SAR mode and the array mode, for corresponding static and moving target detections. Both simulation and experiment results show that the proposed method can detect static and moving targets robustly over conventional methods with much less data and hardware cost. This proposed method can find its potential applications in imaging systems requiring large data storage and processing time. For source direction estimation, direction of arrival (DOA) and polarization estimations are studied. Conventional uniform circular array (UCA) with vertically polarized antenna elements is well known for DOA estimation of vertically polarized source signals from low grazing angles. However, recently there are increasing demands for DOA and polarization estimations of arbitrarily polarized signals from high grazing angles in military areas such as airborne applications. A UCA cannot meet the requirements, therefore a vector circular array (VCA) with identical elements polarized in radial directions on a circular grid is proposed along with a simple and intuitive estimation method. Extensive simulations and error analyses show that the VCA can estimate DOAs and polarizations of sources from high grazing angles with very small estimation errors. DOA ambiguity problems due to large inter-element spacing of antenna arrays have been noticed since the first study of DOA estimation. Although some research works have been tried to solve the problem, no optimum solution has come out. In view of this and the property of a VCA, an ambiguity resolving approach based on a VCA, namely the composite-multiple-resolution (CMR) approach, is proposed. The basic idea is to make use of different portions of the array, namely the 2-element ULA and the vector arc array (VAA). Simulations and case studies are provided to investigate the performance of the approach. The results show that DOA ambiguity problem can be solved by using the approach and more interestingly, the operating bandwidth of the VCA can be greatly increased to more than 9:1 with the proposed approach. The VCA based approach may find its potential applications in high speed airborne applications.