Title :
A VLSI architecture for adaptive signal processing
Author_Institution :
Dept. of Electr. & Comput. Eng., Old Dominion Univ., Norfolk, VA, USA
Abstract :
A VLSI architecture for direct parallel implementation of N -tap LMS (least mean square) adaptive filters is presented which uses N multiplier-accumulators and N-1 parallel adders. It will take two multiply delays and log2 N add delays for each sampled datum. A prototype VLSI processor which can implement 16-tap adaptive filters on a single chip was also developed. The processor has very good architectural features, such as modularity, regularity, and cascadability. By cascading the processors, a very-high-order adaptive filter can be implemented in real time with a several-megahertz sampling rate
Keywords :
VLSI; adaptive filters; computerised signal processing; digital filters; digital signal processing chips; parallel architectures; DSP; LMS adaptive filter; VLSI architecture; VLSI processor; adaptive signal processing; cascadability; direct parallel implementation; least mean square; megahertz sampling rate; multiplier-accumulators; parallel adders; real time; Adaptive algorithm; Adaptive filters; Adaptive signal processing; Computer architecture; Data communication; Finite impulse response filter; Least squares approximation; Resonance light scattering; Signal processing algorithms; Very large scale integration;
Conference_Titel :
Southeastcon '91., IEEE Proceedings of
Conference_Location :
Williamsburg, VA
Print_ISBN :
0-7803-0033-5
DOI :
10.1109/SECON.1991.147709