Design and Performance Evaluation of an Indoor Ranging System Using Audio Chirp and MUSIC Algorithms

In most existing indoor acoustic ranging systems, a range is estimated based on measuring the TOA of an ultrasonic wave. Due to high directivity of ultrasonic transducers and high attenuation of ultrasonic waves with distance in air, these systems cannot provide a range in large indoor environments. To achieve longer ranges, a TOA-based acoustic ranging system that uses a low frequency audible signal is proposed. The audible sound is a spread spectrum LFM signal, which is generated with a standard speaker and received with a standard microphone. The indoor acoustic environment is modeled with a multipath propagation channel whose characteristics are different from those of outdoor environments. It is shown that the classical ranging techniques used in traditional systems, such as the GPS, Sonar, Radar, etc., may not provide the optimum performance in indoor environments. The complexity and site-specific nature of indoor acoustic propagation, when using traditional correlation techniques, often lead to the creation of undetectable direct-line-of-sight situations. It is shown in this paper that by using MUSIC super-resolution techniques, namely the root-MUSIC, which is applied to the estimated frequency response of the multipath indoor propagation channels, the substantial TOA errors can be reduced significantly and the range accuracy improved by a factor of at least 3. Key-Words: indoor localization, chirp, root-MUSIC, LFM.