Key Parameters Setting And Location Algorithm For A Distributed Sound Source Positioning System

Distributed source localization technology is an important research topic since the First World War owe to its wide range of application. As the improvement of its theoretical research and technical evolution, it can be applied to many new areas such as acoustics signal processing, signal detection, digital signal processing, electronics and software design. In this thesis, we mainly focused on key parameters setting and location algorithm for a distributed sound source positioning system.At the beginning of this thesis, introducing the background information of sound localization technology which include the basic principle and a literature review of this technology. Also investigated three different location algorithms for positioning system which are based on signal arrival (AOA), signal arrival time (TOA) and signal arrival time difference (TDOA). Secondly, we compared the performance of three TDOA based targeting algorithms, namely Chan algorithm, SX algorithm and Taloyer algorithm. On the basis of comparing results, we proposed the overall system scheme and algorithm flow. Thirdly, after the mathematical analysis of positioning error, we found that the value of delay estimation in target system, time synchronization and the form of the distribution stations are three key factors which can affect the performance of positioning system. Fourthly, we combined these three factors with the positioning accuracy (GDOP) together and simulated them in different environments in order to achieve their accuracy. Based on the target system's performance parameters and the status of technology, we also set the range of these three key parameters for positioning system and then conducted target validation. Finally, TDOA location algorithm based on Taylor algorithm for distributed targets was simulated by Matlab.In order to verify the feasibility of the location algorithm, we built up a four-channel indoor recording verification system. Three microphones were used as audio source. Experiment results illustrated that the accuracy of location algorithm is high in the verification environment and it can definitely meet the system requirement. In addition, we presented basic concepts of VisualC++ based positioning system interactive interface design and the serial categories. We also described how to use VisualC++ to set up sound source positioning system interactive interface and the use of man-machine interface via MSComm control Serial Port. Final simulation results were displayed based on the sensor coordinate collection which are implemented by mixed programming of VC++ and Matlab. In this case, Matlab program was implemented in a cross-platform environment which can improved the portability of program.