System and method for adaptive reduced-rank parameter estimation using an adaptive decimation and interpolation scheme

摘要

The present invention describes a system and method for general parameter estimation using adaptive processing that provides a performance that significantly exceeds existing reduced-rank schemes using reduced computational resources with greater flexibility. The adaptive processing is accomplished by calculating a reduced-rank approximation of an observation data vector using an adaptive decimation and interpolation scheme. The new scheme employs a time-varying interpolator finite impulse response (FIR) filter at the front-end followed by a decimation structure that processes the data according to the decimation pattern that minimizes the squared norm of the error signal and by a reduced-rank FIR filter. According to the present invention, the number of elements for estimation is substantially reduced, resulting in considerable computational savings and very fast convergence performance for tracking dynamic signals. The current invention is aimed at communications and signal processing applications such as equalization, interference suppression of CDMA systems, echo cancellation and beamforming with antenna arrays. Amongst other promising areas for the deployment of the present technique, we also envisage biomedical engineering, control systems, radar and sonar, seismology, remote sensing and instrumentation.