Analog adaptive filtering based on a modified hopfield network

Research output: Contribution to journalArticleResearchpeer-review

1 Citation (Scopus)

Abstract

In the last few years analog adaptive filters have been a subject of active research because they have the ability to handle in real time much higher frequencies, with a smaller size and lower power consumption than their digital counterparts. During this time several analog adaptive filter algorithms have been reported in the literature, almost all of them use the continuous time version of the least mean square (LMS) algorithm. However the continuous time LMS algorithm presents the same limitations than its digital counterpart, when operates in noisy environments, although their convergence rate may be faster than the digital versions. This fact suggests the necessity to develop analog versions of recursive least square (RLS) algorithm, which is known to have a very low sensitivity to additive noise. However a direct implementation of the RLS in analog way would require a considerable effort. To overcome this problem, we propose an analog RLS algorithm in which the adaptive filter coefficients vector is estimated by using a fully connected network that resembles a Hopfield network. Theoretical and simulations results are given which show that the proposed and conventional RLS algorithms have quite similar convergence properties when they operate with the same sampling rate and signal-to-noise ratio. However, because of its analog realization form, the proposed algorithm can operate in real time with smaller sampling period than its digital counterpart, and then it can achieve a smaller misadjustment with a much better tracking ability than the conventional RLS algorithm.
Original languageAmerican English
Pages (from-to)2245-2252
Number of pages2019
JournalIEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences
StatePublished - 1 Jan 1997

Fingerprint

Hopfield Network
Adaptive Filtering
Adaptive filtering
Least Square Algorithm
Recursive Algorithm
Analogue
Adaptive Filter
Least Mean Square
Adaptive filters
Continuous Time
filter
Additive Noise
Sampling
Convergence Properties
Power Consumption
Least Squares
Convergence Rate
Additive noise
sampling
signal-to-noise ratio

Cite this

@article{b8fe77b070dd4b5195ffb099bc948dd4,
title = "Analog adaptive filtering based on a modified hopfield network",
abstract = "In the last few years analog adaptive filters have been a subject of active research because they have the ability to handle in real time much higher frequencies, with a smaller size and lower power consumption than their digital counterparts. During this time several analog adaptive filter algorithms have been reported in the literature, almost all of them use the continuous time version of the least mean square (LMS) algorithm. However the continuous time LMS algorithm presents the same limitations than its digital counterpart, when operates in noisy environments, although their convergence rate may be faster than the digital versions. This fact suggests the necessity to develop analog versions of recursive least square (RLS) algorithm, which is known to have a very low sensitivity to additive noise. However a direct implementation of the RLS in analog way would require a considerable effort. To overcome this problem, we propose an analog RLS algorithm in which the adaptive filter coefficients vector is estimated by using a fully connected network that resembles a Hopfield network. Theoretical and simulations results are given which show that the proposed and conventional RLS algorithms have quite similar convergence properties when they operate with the same sampling rate and signal-to-noise ratio. However, because of its analog realization form, the proposed algorithm can operate in real time with smaller sampling period than its digital counterpart, and then it can achieve a smaller misadjustment with a much better tracking ability than the conventional RLS algorithm.",
author = "Mariko Nakano-Miyatake and Hector Perez-Meana",
year = "1997",
month = "1",
day = "1",
language = "American English",
pages = "2245--2252",
journal = "IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences",
issn = "0916-8508",
publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",

}

TY - JOUR

T1 - Analog adaptive filtering based on a modified hopfield network

AU - Nakano-Miyatake, Mariko

AU - Perez-Meana, Hector

PY - 1997/1/1

Y1 - 1997/1/1

N2 - In the last few years analog adaptive filters have been a subject of active research because they have the ability to handle in real time much higher frequencies, with a smaller size and lower power consumption than their digital counterparts. During this time several analog adaptive filter algorithms have been reported in the literature, almost all of them use the continuous time version of the least mean square (LMS) algorithm. However the continuous time LMS algorithm presents the same limitations than its digital counterpart, when operates in noisy environments, although their convergence rate may be faster than the digital versions. This fact suggests the necessity to develop analog versions of recursive least square (RLS) algorithm, which is known to have a very low sensitivity to additive noise. However a direct implementation of the RLS in analog way would require a considerable effort. To overcome this problem, we propose an analog RLS algorithm in which the adaptive filter coefficients vector is estimated by using a fully connected network that resembles a Hopfield network. Theoretical and simulations results are given which show that the proposed and conventional RLS algorithms have quite similar convergence properties when they operate with the same sampling rate and signal-to-noise ratio. However, because of its analog realization form, the proposed algorithm can operate in real time with smaller sampling period than its digital counterpart, and then it can achieve a smaller misadjustment with a much better tracking ability than the conventional RLS algorithm.

AB - In the last few years analog adaptive filters have been a subject of active research because they have the ability to handle in real time much higher frequencies, with a smaller size and lower power consumption than their digital counterparts. During this time several analog adaptive filter algorithms have been reported in the literature, almost all of them use the continuous time version of the least mean square (LMS) algorithm. However the continuous time LMS algorithm presents the same limitations than its digital counterpart, when operates in noisy environments, although their convergence rate may be faster than the digital versions. This fact suggests the necessity to develop analog versions of recursive least square (RLS) algorithm, which is known to have a very low sensitivity to additive noise. However a direct implementation of the RLS in analog way would require a considerable effort. To overcome this problem, we propose an analog RLS algorithm in which the adaptive filter coefficients vector is estimated by using a fully connected network that resembles a Hopfield network. Theoretical and simulations results are given which show that the proposed and conventional RLS algorithms have quite similar convergence properties when they operate with the same sampling rate and signal-to-noise ratio. However, because of its analog realization form, the proposed algorithm can operate in real time with smaller sampling period than its digital counterpart, and then it can achieve a smaller misadjustment with a much better tracking ability than the conventional RLS algorithm.

UR - https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=0031270908&origin=inward

UR - https://www.scopus.com/inward/citedby.uri?partnerID=HzOxMe3b&scp=0031270908&origin=inward

M3 - Article

SP - 2245

EP - 2252

JO - IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences

JF - IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences

SN - 0916-8508

ER -