Ultrasonic bearing estimation using a MEMS microphone array and spatiotemporal filters

Simulation results are presented for a method to process ultrasonic signals from an array of MEMS microphones to determine the angle of incidence of the original signal. Specifically, noise immunity is a primary concern, as is low-power operation and compact implementation. To achieve these objectives, a bank of spatiotemporal filters is the method chosen for computation. This approach focuses on sensing apparent velocity of the wavefront across the array, as opposed to standard methods concentrating on relative temporal or delay characteristics of the sensor responses. It utilizes mixed-mode band-pass VLSI filter banks to enable bearing processing. Simulation demonstrated one-degree accuracy with SNR down to 14 dB.