Micropower gradient flow acoustic localizer

Abstract

Abstract-A micropower mixed-signal system-on-chip for three-dimensional localization of a broadband acoustic source is presented. Direction cosines of the source are obtained by relating spatial and temporal differentials in the acoustic traveling wave field acquired across four coplanar microphones at sub-wavelength spacing. Correlated double sampling and leastsquares adaptive cancellation of common-mode leakthrough in the switched-capacitor analog differentials boost localization accuracy at very low aperture. A second stage of mixed-signal least-squares adaptation directly produces digital estimates of the direction cosines. The 3 mm × 3 mm chip in 0.5 µm CMOS technology quantizes signal delays with 250 ns resolution at 16 kHz sampling rate, and dissipates 54 µW power from a 3 V supply. Field tests of the processor with acoustic enclosure demonstrated its utility and endurance in tracking ground and airborne vehicles. Applications include acoustic surveillance, interactive multimedia, and intelligent hearing aids.

Statistical Estimation and Adaptation for Visual Compensation in Object Tracking

“…1. A three-dimensional acoustic localizer obtains the direction of arrival (DOA) [10] at an acoustic sensor, and detects two angle components and from the arrival time difference between embedded adjacent microphones. Then, the object estimation errors are corrected by the association with two visual sensors.…”

Mixed-signal real-time adaptive blind source separation

“…Initial experimental results from a prototype micropower chip demonstrated 30 dB real-time separation of mixed speech signals presented to the chip. Further experimentation is directed towards integrating the ICA VLSI architecture with previously developed circuits implementing gradient flow [4], for real-time adaptive acoustic source separation and localization, using miniature acoustic arrays. The ICA architecture is directly amenable to integration with sensor arrays for small, compact, battery-operated "smart" sensor applications in hearing aids, personal digital assistants, and surveillance networks.…”

“…The update is performed locally by once or repeatedly incrementing, decrementing or holding the current value of counter based on the learning rule served by the micro-controller. The 8 most significant bits of the 14-bit counter holding and updating the coefficients are presented to a multiplying D/A capacitor array [4] to linearly unmix the separated signal. The remaining 6 bits in the coefficient registers provide flexibility in programming the update rate to tailor convergence.…”

“…The differential analog input channels directly interface with gradient output signals from a previously developed gradient flow processor for acoustic localization [4], to extend its functionality to joint separation and localization of up to three acoustic sources. The mixed-signal architecture is implemented using fully differential switched-capacitor sampled-data circuits.…”

“…Mixed-signal VLSI implementation of the method [4] has demonstrated improved performance in terms of power dissipation and bearing resolution over conventional bearing estimation localizers. Besides its use in bearing estimation, gradient flow provides an efficient signal representation as a front-end for blind source separation.…”