ScholarLens Back to search
1991•Unpublished venueRequires access

Dl.1 Adaptive Pilot Filtering for LMS Algorithm

W. Y. Chen

Open publisher page 0 citations

Abstract

The convergence speed of the Least Mean Square (LMS) algorithm is slow for applications where the received signal is not white. An adaptive pilot filtering procedure is proposed in this paper to increase the convergence speed of the LMS algorithm. The procedure proposed uses an adaptive filter with only a few filter coefficients to filter both the received signal and the original signal for the purpose of whitening the received signal. The convergence speed of the adaptive pilot filtering procedure combined with the LMS algorithm is comparable to the Kalman algorithm for some applications. The adaptive pilot filtering procedure combined with the main LMS algorithm only requires less than 2 times that of a conventional LMS algorithm. The name of the adaptive pilot filter comes from the analogy to the pilot parachute.

About this research paper

What this paper is about

The convergence speed of the Least Mean Square (LMS) algorithm is slow for applications where the received signal is not white. An adaptive pilot filtering procedure is proposed in this paper to increase the convergence speed of the LMS algorithm. The procedure proposed uses an adaptive filter with only a few filter coefficients to filter both the received signal and the original signal for the purpose of whitening the received signal. The convergence speed of the adaptive pilot filtering procedure combined with the LMS algorithm is comparable to the Kalman algorithm for some applications. The adaptive pilot filtering procedure combined with the main LMS algorithm only requires less than 2 times that of a conventional LMS algorithm. The name of the adaptive pilot filter comes from the analogy to the pilot parachute.

Why it matters

A significance statement is not available in the OpenAlex record.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

The convergence speed of the Least Mean Square (LMS) algorithm is slow for applications where the received signal is not white. An adaptive pilot filtering procedure is proposed in this paper to increase the convergence speed of the LMS algorithm. The procedure proposed uses an adaptive filter with only a few filter coefficients to filter both the received signal and the original signal for the purpose of whitening the received signal. The convergence speed of the adaptive pilot filtering procedure combined with the LMS algorithm is comparable to the Kalman algorithm for some applications. The adaptive pilot filtering procedure combined with the main LMS algorithm only requires less than 2 times that of a conventional LMS algorithm. The name of the adaptive pilot filter comes from the analogy to the pilot parachute.

Key concepts: Adaptive filter, Least mean squares filter, Kernel adaptive filter, Algorithm, Convergence (economics), Filter (signal processing), Adaptive algorithm, SIGNAL (programming language)

Related papers

Related paper 1Related paper 2Related paper 3
Back to paper searchBrowse research topicsOriginal source
Dl.1 Adaptive Pilot Filtering for LMS Algorithm — Research Paper | ScholarLens