Loop-shaping techniques applied to the least-mean-squares algorithm

@article{Moir2011LoopshapingTA,
  title={Loop-shaping techniques applied to the least-mean-squares algorithm},
  author={Tom J. Moir},
  journal={Signal, Image and Video Processing},
  year={2011},
  volume={5},
  pages={231-243}
}
  • T. Moir
  • Published 1 June 2011
  • Mathematics
  • Signal, Image and Video Processing
The least-mean-squares (LMS) algorithm is analysed as a feedback control system. It is shown that despite the fact that LMS is a time-variant system that in fact it behaves much as a linear time-invariant (LTI) closed-loop control system. Therefore, it is possible to treat the LMS algorithm as a control system in the classical sense and define properties such as bandwidth to determine how fast a response (and hence convergence) is maximally possible. Similarly, the steady-state error response… 

FIR System Identification Using Feedback

TLDR
A new approach to finite-impulse (FIR) system identification that uses deconvolution as a means of separating the impulse-response from the system input data and the stability limit is far simpler to calculate.

Stability and convergence of a class of nonlinear algorithms in digital signal processing

  • T. Moir
  • Computer Science
    2017 International Conference on Signals and Systems (ICSigSys)
  • 2017
TLDR
This paper examines twosquare-root algorithms and shows that the convergence is guaranteed provided the closed-loop discrete-time system is stable, but that the upper limit on stability is determined by the values of the square-root coefficients themselves.

References

SHOWING 1-10 OF 25 REFERENCES

Least mean squares as a control system

Abstract The least mean squares (LMS) algorithm is analysed as a control system rather than the conventional statistical approach. This approach greatly simplifies the understanding of the problem

Control-theoretic design of the LMS and the sign algorithms in nonstationary environments

  • C. Kwong
  • Engineering
    IEEE Trans. Acoust. Speech Signal Process.
  • 1990
TLDR
A frequency- domain approach to the latter problems that has the advantages of transparency, ease of computation, and generality compared with the time-domain approach previously used is presented.

Further results on the control-theoretic analysis of the LMS algorithm

  • C. Kwong
  • Engineering
    IEEE Trans. Signal Process.
  • 1993
TLDR
The results of the control-theoretic analysis of the least-mean-square algorithm proposed by the author (1990) to include the case of nonwhite input data are extended and redefined and it is shown that the contribution of the cross power spectra to the total error is negligible.

Least mean squares algorithm for fractionally spaced blind channel estimation

TLDR
The authors consider the problem of blind estimation and equalisation of digital communication finite impulse response (FIR) channels using fractionally spaced samples and shows that the latter can be derived from the proposed algorithm.

Convergence analysis of LMS filters with uncorrelated Gaussian data

TLDR
It is found that the adaptive coefficient μ, which controls the rate of convergence of the algorithm, must be restricted to an interval significantly smaller than the domain commonly stated in the literature.

Performance advantage of complex LMS for controlling narrow-band adaptive arrays

In narrow-band adaptive-array applications, the mean-square convergence of the discrete-time real least mean-square (LMS) algorithm is slowed by image-frequency noises generated in the LMS loops. The

Research on Least Means Squares adaptive control for automotive active suspension

According to the evaluation method of riding comfort and handling safety of automotive, the acceleration of the sprung mass, dynamic tyre load between wheels and road and dynamic deflection between

Adaptive Filter Theory

Background and Overview. 1. Stochastic Processes and Models. 2. Wiener Filters. 3. Linear Prediction. 4. Method of Steepest Descent. 5. Least-Mean-Square Adaptive Filters. 6. Normalized

A variable leaky LMS adaptive algorithm

  • M. KamenetskyB. Widrow
  • Computer Science
    Conference Record of the Thirty-Eighth Asilomar Conference on Signals, Systems and Computers, 2004.
  • 2004
TLDR
A variable leaky LMS algorithm, designed to overcome the slow convergence of standard LMS in cases of high input eigenvalue spread, which uses a greedy punish/reward heuristic together with a quantized leak adjustment function to vary the leak.

Least-mean-square adaptive filters

TLDR
This paper presents a meta-modelling framework for estimating the energy conservation and the learning ability of LMS Adaptive Filters using a number of simple and scalable algorithms.