TY - GEN

T1 - Almost sure convergence of the normed LMS algorithm with error feedback delay

AU - Voltz, Peter J.

PY - 1993

Y1 - 1993

N2 - In some applications of LMS type adaptive algorithms, it is necessary to implement a variant of the algorithm with feedback delay in the weight update calculation. In this paper we consider the normed version of such an algorithm and show that the algorithm converges exponentially if the update gain parameter, μm, is sufficiently small. The result is first proved for inputs which satisfy a standard deterministic mixing condition, and then the development is extended to the case when the input may not be strictly mixing, but is instead a stationary ergodic vector sequence with positive definite autocorrelation.

AB - In some applications of LMS type adaptive algorithms, it is necessary to implement a variant of the algorithm with feedback delay in the weight update calculation. In this paper we consider the normed version of such an algorithm and show that the algorithm converges exponentially if the update gain parameter, μm, is sufficiently small. The result is first proved for inputs which satisfy a standard deterministic mixing condition, and then the development is extended to the case when the input may not be strictly mixing, but is instead a stationary ergodic vector sequence with positive definite autocorrelation.

UR - http://www.scopus.com/inward/record.url?scp=0027814135&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0027814135&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0027814135

SN - 0818641207

T3 - Conference Record of the Asilomar Conference on Signals, Systems & Computers

SP - 179

EP - 183

BT - Conference Record of the Asilomar Conference on Signals, Systems & Computers

A2 - Singh, Avtar

PB - Publ by IEEE

T2 - Proceedings of the 27th Asilomar Conference on Signals, Systems & Computers

Y2 - 1 November 1993 through 3 November 1993

ER -