Patent US5677951A successfully refreshed from Google Patents.
US5677951A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Adaptive filter and method for implementing echo cancellation
Assignee: Avago Technologies International Sales Pte Ltd
•
Filing: 1995-06-19
•
Granted: 1997-10-14
•
Expires: 2015-06-19
Abstract
Abstract A dynamically regularized fast recursive least squares (DR-FRLS) adaptive filter and method are disclosed which provide O(N) computational complexity and which have dynamically adjusted regularization in real-time without requiring restarting the adaptive filtering. The adaptive filter includes a memory for storing at least one program, including a dynamically regularized fast recursive least squares (DR-FRLS) program; and a processor which responds to the DR-FRLS program for adaptively filtering an input signal and for generating an output signal associated with the input signal.
Inventors
- 👤 Steven Leslie Gay
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-05 16:33
CPC Classifications
H; H04; H04B; H04B3/00; H04B3/02; H04B3/20; H04B3/23; H03; H03H; H03H21/00; H03H21/0012; H03H21/0043
Claims Analysis
Total Claims: 21
Claim 1
Independent Claim
62 words
An adaptive filter comprising: a memory for storing at least one program, including a dynamically regularized fast recursive least squares (DR-FRLS) program; and a processor, responsive to the DR-FRLS program, for dynamically generating regularization signals from the input signal for adaptively filtering an input signal and for generating an output signal associated with the adaptively filtered input signal from the regularization signals.
Claim 2
Dependent Claim
62 words
An adaptive filter comprising: a memory for storing at least one program, including a dynamically regularized fast recursive least squares (DR-FRLS) program; and a processor, responsive to the DR-FRLS program, for dynamically generating regularization signals from the input signal for adaptively filtering an input signal and for generating an output signal associated with the adaptively filtered input signal from the regularization signals.
Claim 3
Dependent Claim
34 words
The adaptive filter of claim 1 wherein the processor is responsive to a dynamically adjustable forgetting factor signal corresponding to a forgetting factor λ for dynamically controlling the adaptive filtering of the input signal.
Claim 4
Dependent Claim
64 words
The adaptive filter of claim 1 wherein the processor further includes: a first parameter generator for generating data related update signals from the input signal; and a second parameter generator for generating regularization related update signals from a refresh signal as the regularization signals; wherein the processor uses the data related update signals and the regularization related update signals to generate the output signal.
Claim 5
Dependent Claim
34 words
The adaptive filter of claim 3 wherein the likelihood variable stabilization unit is responsive to a forgetting factor signal corresponding to a forgetting factor λ for controlling the stabilizing of the likelihood variable estimates.
Claim 6
Dependent Claim
37 words
The adaptive filter of claim 4 wherein the likelihood variable stabilization unit is responsive to the forgetting factor signal being dynamically adjustable for dynamically controlling the stabilizing of the likelihood variable estimates during the adaptive filtering process.
Claim 7
Dependent Claim
45 words
The adaptive filter of claim 1 wherein the processor further includes: a likelihood variable stabilization unit for stabilizing likelihood variable estimates associated with the input signal; and a joint process extension processing unit responsive to the stabilized likelihood variable estimates for generating the output signal.
Claim 8
Dependent Claim
79 words
An echo cancellation system to compensate for an echo in an environment, the echo being associated with an excitation signal, the echo cancellation system comprising: an adaptive filter for dynamically generating regularization signals for adaptively filtering the excitation signal using a dynamically regularized fast recursive least squares (DR-FRLS) technique and a dynamically adjusted forgetting factor signal to generate an echo residual signal associated with the excitation signal, the echo residual signal for compensating for the echo in the environment.
Claim 9
Dependent Claim
79 words
An echo cancellation system to compensate for an echo in an environment, the echo being associated with an excitation signal, the echo cancellation system comprising: an adaptive filter for dynamically generating regularization signals for adaptively filtering the excitation signal using a dynamically regularized fast recursive least squares (DR-FRLS) technique and a dynamically adjusted forgetting factor signal to generate an echo residual signal associated with the excitation signal, the echo residual signal for compensating for the echo in the environment.
Claim 10
Dependent Claim
68 words
The echo cancellation system of claim 7 wherein the adaptive filter further includes: a first parameter generator for generating data related update signals from the excitation signal; and a second parameter generator for generating regularization related update signals from a refresh signal as the regularization signals; wherein the adaptive filter uses the data related update signals and the regularization related update signals to generate the echo residual signal.
Claim 11
Dependent Claim
61 words
The echo cancellation system of claim 7 wherein the adaptive filter further includes: a likelihood variable stabilization unit for dynamically stabilizing likelihood variable estimates using the dynamically adjusted forgetting factor signal, the likelihood variable estimates associated with the excitation signal; and a joint process extension processing unit responsive to the dynamically stabilized likelihood variable estimates for generating the echo residual signal.
Claim 12
Independent Claim
68 words
A method for adaptively filtering an input signal using an adaptive filter comprising the steps of: (a) operating a processor using a dynamically regularized fast recursive least squares (DR-FRLS) program; (b) processing the input signal by adaptive filtering using the processor operated by the DR-FRLS program to dynamically generate regularization signals; and (c) generating an output signal associated with the adaptively filtered input signal from the regularization signals.
Claim 13
Independent Claim
68 words
A method for adaptively filtering an input signal using an adaptive filter comprising the steps of: (a) operating a processor using a dynamically regularized fast recursive least squares (DR-FRLS) program; (b) processing the input signal by adaptive filtering using the processor operated by the DR-FRLS program to dynamically generate regularization signals; and (c) generating an output signal associated with the adaptively filtered input signal from the regularization signals.
Claim 14
Dependent Claim
44 words
The method of claim 10 wherein the step (b) of processing the input signal includes the steps of: (b1) receiving a forgetting factor signal corresponding to a forgetting factor; and (b2) controlling the adaptive filtering of the input signal using the forgetting factor signal.
Claim 15
Dependent Claim
62 words
The method of claim 11 further including the steps of: (a1) dynamically adjusting the forgetting factor signal; the step (b1)of receiving the forgetting factor signal includes the step of receiving the dynamically adjusted forgetting factor signal; and the step (b2) of controlling the adaptive filtering of the input signal includes dynamically controlling the adaptive filtering using the dynamically adjusted forgetting factor signal.
Claim 16
Dependent Claim
44 words
The method of claim 10 wherein the step (b) of processing the input signal includes concurrently performing the steps of: (b1) performing data related updating using the input signal; and (b2) performing regularization related updating using a refresh signal to generate the regularization signals.
Claim 17
Dependent Claim
53 words
The method of claim 10 wherein the step (b) of processing the input signal includes concurrently performing the steps of: (b1) performing likelihood variable stabilization for stabilizing likelihood variable estimates associated with the input signal; and (b2) performing joint process extension processing on the stabilized likelihood variable estimates for generating the output signal.
Claim 18
Independent Claim
82 words
A method for performing echo cancellation to compensate for an echo in an environment, the echo being associated with an excitation signal, the method comprising the steps of: (a) receiving the excitation signal; (b) operating an adaptive filter using a dynamically regularized fast recursive least squares (DR-FRLS) technique; (c) adaptively filtering the excitation signal using the DR-FRLS technique; (d) generating an echo residual signal associated with the adaptively filtered excitation signal; and (e) compensating for the echo using the echo residual signal.
Claim 19
Independent Claim
82 words
A method for performing echo cancellation to compensate for an echo in an environment, the echo being associated with an excitation signal, the method comprising the steps of: (a) receiving the excitation signal; (b) operating an adaptive filter using a dynamically regularized fast recursive least squares (DR-FRLS) technique; (c) adaptively filtering the excitation signal using the DR-FRLS technique; (d) generating an echo residual signal associated with the adaptively filtered excitation signal; and (e) compensating for the echo using the echo residual signal.
Claim 20
Dependent Claim
45 words
The method of claim 15 wherein the step (c) of adaptively filtering the excitation signal includes the steps of: (c1) receiving a forgetting factor signal corresponding to a forgetting factor; and (c2) controlling the adaptive filtering of the excitation signal using the forgetting factor signal.
Claim 21
Dependent Claim
63 words
The method of claim 16 further including the steps of: (b1) dynamically adjusting the forgetting factor signal; the step (c1) of receiving the forgetting factor signal includes the step of receiving the dynamically adjusted forgetting factor signal; and the step (c2) of controlling the adaptive filtering of the excitation signal includes dynamically controlling the adaptive filtering using the dynamically adjusted forgetting factor signal.
References Cited (Backward Citations)
US4817081A
Cited
US5216629A
Cited
US5303264A
Cited
US5153875A
Cited
US5293402A
Cited
US5285480A
Cited
US5351291A
Cited
US5371789A
Cited
US5359656A
Cited
US5408530A
Cited
US5327495A
Cited
US5353348A
Cited
US5428562A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2000-10-13 | 2001-04-14 | 2001-10-14 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2004-10-13 | 2005-04-14 | 2005-10-14 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2008-10-13 | 2009-04-14 | 2009-10-14 | $7700.00 | Window Closed / Lapsed |