Acoustic Signal Processing For Acoustic Source Localisation In An Elastic Solid

Many research projects have been carried out to achieve close range positioning in the context of indoor localisation using propagation's arrival time difference or multi-path structure with defined features such as Received Signal Strength (RSS). However, most researches used electromagnetic waves as signal carriers, for example, modern Wi-Fi routers, can provide a sufficient coverage for short range positioning and indoor localisation applications. Electromagnetic waves, however, are sensitive to the physical environment and its high traveling speed is prone to resulting into a low accuracy for short range localisation because of the limited bandwidth. In this paper, a probabilistic algorithm using acoustic signal pattern-matching templates is proposed in order to overcome the disadvantages associated with electromagnetic wave localisation approaches, while maintaining the precision. The resolution of the proposed passive acoustic locating method is verified to 3 cm. The structure of the system and relevant experimental parameters in an application of Human Computer Interface (HCI) are described and discussed.