Patent US4208441A successfully refreshed from Google Patents.
Patent US5282225A successfully refreshed from Google Patents.
US5282225A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Adaptive blind channel equalizer system
Assignee: Northeastern University
•
Filing: 1992-02-04
•
Granted: 1994-01-25
•
Expires: 2012-02-04
Abstract
Abstract A channel equalizer system is provided that employs a non-linear technique and a data symbol memory for rapidly adjusting a plurality of equalizer coefficients without using a training sequence. Past data symbols stored in the data symbol memory are used to achieve significantly fast convergence of the equalizer coefficients to their optimum values. Statistical properties of the transmitted data symbols are also exploited. Global convergence is guaranteed by traversing local minima in the convergence process. An expression of performance, referred to as a cost function, includes a plurality of weighted terms that results in more rapid convergence than a conventional cost function with a single term. Still further rapid convergence results when the weights are adaptively adjusted in real-time. The improved performance is achieved with a negligible increase in computational complexity.
Inventors
- 👤 Chrysostomos L. Nikias
- 👤 John G. Proakis
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-10-28 22:56
CPC Classifications
H; H03; H03H; H03H21/00; H03H21/0012; H04; H04L; H04L25/00; H04L25/02; H04L25/03; H04L25/03006; H04L25/03012; H04L25/03114
Claims Analysis
Total Claims: 28
Claim 1
Independent Claim
201 words
An adaptive blind channel equalizer system comprising: a channel equalizer with a plurality of equalizer coefficients, responsive to a transmitted data signal distorted by a channel and to a coefficient update signal, for providing an equalized signal; a coefficient update module, connected to the channel equalizer and responsive to the transmitted data signal distorted by the channel, for providing a plurality of updated equalizer coefficients to the channel equalizer for replacing the plurality of equalizer coefficients; a non-linear function evaluation module, connected to the output of the channel equalizer, for evaluating a non-linear function in response to said channel equalizer output to provide an error signal to said coefficient update module, the evaluation module including a memory for storing a plurality of past data symbols; an adaptive weight module, connected to the non-linear function evaluation module and to the channel equalizer, for providing a set of adaptively determined weights to be incorporated in the non-linear function; a scaling module, connected to the channel equalizer and responsive to the equalized signal, for providing a scaled signal; and a decision module, connected to the scaling module and responsive to the scaled signal, for providing a signal point in accordance with the scaled signal.
Claim 2
Dependent Claim
201 words
An adaptive blind channel equalizer system comprising: a channel equalizer with a plurality of equalizer coefficients, responsive to a transmitted data signal distorted by a channel and to a coefficient update signal, for providing an equalized signal; a coefficient update module, connected to the channel equalizer and responsive to the transmitted data signal distorted by the channel, for providing a plurality of updated equalizer coefficients to the channel equalizer for replacing the plurality of equalizer coefficients; a non-linear function evaluation module, connected to the output of the channel equalizer, for evaluating a non-linear function in response to said channel equalizer output to provide an error signal to said coefficient update module, the evaluation module including a memory for storing a plurality of past data symbols; an adaptive weight module, connected to the non-linear function evaluation module and to the channel equalizer, for providing a set of adaptively determined weights to be incorporated in the non-linear function; a scaling module, connected to the channel equalizer and responsive to the equalized signal, for providing a scaled signal; and a decision module, connected to the scaling module and responsive to the scaled signal, for providing a signal point in accordance with the scaled signal.
Claim 3
Dependent Claim
24 words
The adaptive blind channel equalizer system of claim 1 wherein the coefficient update module includes a memory for storing a plurality of equalizer coefficients.
Claim 4
Dependent Claim
49 words
The adaptive blind channel equalizer system of claim 1 wherein the coefficient update module generates said plurality of updated equalizer coefficients by subtracting a multiplicative product of an incremental step-size, the transmitted data signal distorted by the channel, and the error signal, from a plurality of previous equalizer coefficients.
Claim 5
Dependent Claim
32 words
The adaptive blind channel equalizer system of claim 1 wherein the non-linear function evaluation module evaluates said non-linear function and subtracts the result from the equalized signal to provide the error signal.
Claim 6
Dependent Claim
19 words
The adaptive blind channel equalizer system of claim 4 wherein the non-linear function includes a plurality of weighted terms.
Claim 7
Dependent Claim
28 words
The adaptive blind channel equalizer system of claim 4 wherein the non-linear function is a function of a current equalized signal and a plurality of previous equalized signals.
Claim 8
Dependent Claim
43 words
The adaptive blind channel equalizer system of claim 4 wherein the non-linear function is a product of a current equalized signal z n and an expected value of the square of the norm of the current equalized signal E(|z n | 2 ).
Claim 9
Dependent Claim
32 words
The adaptive blind channel equalizer system of claim 4 wherein the non-linear function is: g[Z.sub.n ]=z.sub.n -z.sub.n [4w.sub.0 |z.sub.n |.sup.2 +2w.sub.1 |z.sub.n-1 |.sup.2 +. . . +2w.sub.M |z.sub.n-M |.sup.2 -4w.sub.0 R.sub.2 ]
Claim 10
Dependent Claim
168 words
The adaptive blind channel equalizer system of claim 1 wherein the adaptive weight module includes: a moment module, responsive to the equalized signal, for determining a current moment J 0 and a plurality of subsequent moments J 1 ; a deviation module, connected to the moment module, for providing a deviation including the norm of the difference between a subsequent moment J i and a moment at perfect equalization J i .sup.(0) ; and a weight determination module, connected to the deviation module, for providing a first weight that is proportional to a deviation of the current moment from the moment at perfect equalization, only if the first weight is less than a stability threshold, otherwise providing the stability threshold as the first weight, and for providing at least one subsequent weight that is proportional to a deviation of the current moment from the moment at perfect equalization, only if the subsequent weight is less than the stability threshold, otherwise providing the stability threshold as the subsequent weight.
Claim 11
Dependent Claim
31 words
The adaptive blind channel equalizer system of claim 9 wherein the plurality of subsequent moments J i is equal to |E(z n z* n-i ) for a statistically independent, uncorrelated sequence.
Claim 12
Dependent Claim
42 words
The adaptive blind channel equalizer system of claim 9 wherein the plurality of subsequent moments J i is equal to |E(z n z* n-i )-R(i)| 2 for a statistically dependent sequence of known autocorrelation R(i), where i=1, . . . , M.
Claim 13
Dependent Claim
31 words
The adaptive blind channel equalizer system of claim 1 wherein the scaled signal is characterized by a variance that is substantially the same as the variance of the transmitted data signal.
Claim 14
Dependent Claim
146 words
Apparatus for adjusting a plurality of coefficients of a channel equalizer that is responsive to a distorted transmitted data signal and that provides an equalized signal, the apparatus comprising: a coefficient update module, connected to the channel equalizer and responsive to the distorted transmitted data signal, for providing a plurality of updated equalizer coefficients to the channel equalizer for replacing the plurality of equalizer coefficients; a non-linear function evaluation module, connected to the output of the channel equalizer, for evaluating a non-linear function in response to said channel equalizer output to provide an error signal to said coefficient update module, the evaluation module including a memory for storing a plurality of past data symbols; and an adaptive weight module, connected to the non-linear function evaluation module and to the channel equalizer, for providing a set of adaptively determined weights to be incorporated in the non-linear function.
Claim 15
Dependent Claim
146 words
Apparatus for adjusting a plurality of coefficients of a channel equalizer that is responsive to a distorted transmitted data signal and that provides an equalized signal, the apparatus comprising: a coefficient update module, connected to the channel equalizer and responsive to the distorted transmitted data signal, for providing a plurality of updated equalizer coefficients to the channel equalizer for replacing the plurality of equalizer coefficients; a non-linear function evaluation module, connected to the output of the channel equalizer, for evaluating a non-linear function in response to said channel equalizer output to provide an error signal to said coefficient update module, the evaluation module including a memory for storing a plurality of past data symbols; and an adaptive weight module, connected to the non-linear function evaluation module and to the channel equalizer, for providing a set of adaptively determined weights to be incorporated in the non-linear function.
Claim 16
Dependent Claim
36 words
The apparatus for adjusting a plurality of coefficients of a channel equalizer of claim 13, further comprising: a scaling module, connected to the channel equalizer and responsive to the equalized signal, for providing a scaled signal.
Claim 17
Dependent Claim
55 words
The apparatus for adjusting a plurality of coefficients of a channel equalizer of claim 13, wherein the coefficient update module generates said plurality of updated equalizer coefficients by subtracting a multiplicative product of an incremental step-size, the transmitted data signal distorted by the channel, and the error signal, from a plurality of previous equalizer coefficients.
Claim 18
Dependent Claim
38 words
The apparatus for adjusting a plurality of coefficients of a channel equalizer of claim 13 wherein the non-linear function evaluation module evaluates a non-linear function and subtracts the result from the equalized signal to provide the error signal.
Claim 19
Independent Claim
142 words
A method for adaptive blind equalization comprising the steps of: providing an equalized signal from a channel equalizer with a plurality of equalizer coefficients, the channel equalizer being responsive to a transmitted data signal distorted by a channel; providing a set of adaptively determined weights in accordance with the equalized signal; storing a plurality of past data symbols of the equalized signal in a memory; evaluating a non-linear function using the set of adaptively determined weights and the plurality of past data symbols to provide an error signal; providing a plurality of updated equalizer coefficients to the channel equalizer for replacing the plurality of equalizer coefficients in accordance with the error signal and the transmitted data signal distorted by a channel; providing a scaled signal in accordance with the equalized signal; and providing a signal point in accordance with the scaled signal.
Claim 20
Independent Claim
142 words
A method for adaptive blind equalization comprising the steps of: providing an equalized signal from a channel equalizer with a plurality of equalizer coefficients, the channel equalizer being responsive to a transmitted data signal distorted by a channel; providing a set of adaptively determined weights in accordance with the equalized signal; storing a plurality of past data symbols of the equalized signal in a memory; evaluating a non-linear function using the set of adaptively determined weights and the plurality of past data symbols to provide an error signal; providing a plurality of updated equalizer coefficients to the channel equalizer for replacing the plurality of equalizer coefficients in accordance with the error signal and the transmitted data signal distorted by a channel; providing a scaled signal in accordance with the equalized signal; and providing a signal point in accordance with the scaled signal.
Claim 21
Dependent Claim
51 words
The method of adaptive blind equalization of claim 17 wherein the step of providing a plurality of updated equalizer coefficients includes the steps of: subtracting a multiplicative product of an incremental step-size, the transmitted data signal distorted by the channel, and the error signal, from a plurality of previous equalizer coefficients.
Claim 22
Dependent Claim
38 words
The method for adaptive blind equalization of claim 17 wherein the step of evaluating a non-linear function includes the steps of: evaluating a non-linear subexpression and subtracting the result from the equalized signal to provide the error signal.
Claim 23
Dependent Claim
42 words
The method for adaptive blind equalization of claim 17 wherein the step of providing a scaled signal is executed such that the equalizer output is scaled dynamically, and the equalizer output attains the same variance as that of the transmitted data signal.
Claim 24
Dependent Claim
148 words
The method for adaptive blind equalization of claim 17 wherein the step of providing a set of adaptively determined weights includes the steps of: determining a current moment J 0 and a plurality of subsequent moments J i ; providing a deviation including the norm of the difference between a subsequent moment J i and a moment at perfect equalization J i .sup.(0) ; and providing a first weight that is proportional to a deviation of the current moment from the moment at perfect equalization, only if the first weight is less than a stability threshold, otherwise providing the stability threshold as the first weight, and providing at least one subsequent weight that is proportional to a deviation of the current moment from the moment at perfect equalization, only if the subsequent weight is less than the stability threshold, otherwise providing the stability threshold as the subsequent weight.
Claim 25
Dependent Claim
31 words
The method for adaptive blind equalization of claim 21 wherein the plurality of subsequent moments J i is equal to |E(z n z* n-i ) for a statistically independent, uncorrelated sequence.
Claim 26
Dependent Claim
42 words
The method for adaptive blind equalization of claim 21 wherein the plurality of subsequent moments J i is equal to |E(z n z* n-i )-R(i)| 2 for a statistically dependent sequence of known autocorrelation R(i), where i=1, . . . , M.
Claim 27
Independent Claim
129 words
A method of updating a plurality of equalizer coefficients of a channel equalizer comprising the steps of: providing an equalized signal from the channel equalizer with a plurality of equalizer coefficients, the channel equalizer being responsive to a transmitted data signal distorted by a channel; providing a set of adaptively determined weights in accordance with the equalized signal; storing a plurality of past data symbols of the equalized signal in a memory; evaluating a non-linear function using the set of adaptively determined weights and the plurality of past data symbols to provide an error signal; and providing a plurality of updated equalizer coefficients to the channel equalizer for replacing the plurality of equalizer coefficients in accordance with the error signal and the transmitted data signal distorted by a channel.
Claim 28
Independent Claim
129 words
A method of updating a plurality of equalizer coefficients of a channel equalizer comprising the steps of: providing an equalized signal from the channel equalizer with a plurality of equalizer coefficients, the channel equalizer being responsive to a transmitted data signal distorted by a channel; providing a set of adaptively determined weights in accordance with the equalized signal; storing a plurality of past data symbols of the equalized signal in a memory; evaluating a non-linear function using the set of adaptively determined weights and the plurality of past data symbols to provide an error signal; and providing a plurality of updated equalizer coefficients to the channel equalizer for replacing the plurality of equalizer coefficients in accordance with the error signal and the transmitted data signal distorted by a channel.
References Cited (Backward Citations)
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1997-01-24 | 1997-07-26 | 1998-01-25 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2001-01-24 | 2001-07-26 | 2002-01-25 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2005-01-24 | 2005-07-26 | 2006-01-25 | $7700.00 | Window Closed / Lapsed |