Reduced complexity transform-domain adaptive filter using selective partial updates

Abstract

A transform-domain adaptive filter uses selective partial updating of adaptive filter parameters. This updating may be based on a constrained minimization problem. The adaptive filter parameters are separated into subsets, and a subset is selected to be updated at each iteration. A normalization process applied to the frequency bins prior to multiplication by the adaptive filter parameters is used to prevent adaptive filter lock-up that may be experienced in the event of high energy levels of signals in particular frequency bins. Convergence of the transform domain filter is ensured at a rate generally faster than a corresponding time-domain adaptive filter. The transform-domain adaptive filter may be used for various applications, including system identification, channel equalization, or echo cancellation.

1 . A method to perform adaptive filtering, comprising: 
 adaptively filtering frequency domain signals at a present filtering iteration in at least two frequency bins using respective filter coefficients;    updating at least one but fewer than all of the respective filter coefficients at the present filtering iteration based on a comparison of the respective filter coefficients with each other; and    adaptively filtering the frequency domain signals at a next filtering iteration using the updated respective filter coefficients.    
   
   
       2 . The method according to  claim 1  further including normalizing the frequency domain signals in the at least two frequency bins.  
   
   
       3 . The method according to  claim 2  wherein normalizing the frequency domain signals includes dividing instantaneous energy levels of the frequency domain signals in each of the at least two frequency bins by respective average energy levels of the frequency domain signals in each of the at least two frequency bins.  
   
   
       4 . The method according to  claim 1  wherein updating the respective filter coefficients includes solving a constrained minimization equation.  
   
   
       5 . The method according to  claim 4  wherein solving the constrained minimization equation includes calculating a squared-Euclidean-norm.  
   
   
       6 . The method according to  claim 1  wherein updating the respective filter coefficients includes determining which filter coefficient to update according to a least mean square (LMS) process.  
   
   
       7 . The method according to  claim 1  wherein updating the respective filter coefficients occurs at every iteration of the adaptive filtering.  
   
   
       8 . The method according to  claim 1  further including predefining the respective filter coefficients.  
   
   
       9 . The method according to  claim 2  wherein normalizing the frequency domain signals is done as a function of a previous average energy level and a present instantaneous energy level of the respective normalized frequency domain signals.  
   
   
       10 . The method according to  claim 9  wherein normalizing the frequency domain signals includes applying a scale factor to the previous average and present instantaneous energy levels of the respective, normalized frequency domain signals to minimize a potential for adaptive filtering lock-up.  
   
   
       11 . The method according to  claim 1  further including transforming a time domain signal to provide the frequency domain signals in the at least two frequency bins.  
   
   
       12 . The method according to  claim 11  wherein transforming the time domain signal includes orthogonally transforming the time domain signal.  
   
   
       13 . The method according to  claim 1  further including minimizing an echo signal in a time domain signal related to the frequency domain signals.  
   
   
       14 . The method according to  claim 13  wherein updating the respective filter coefficients includes targeting a burst echo signal in the time domain signal.  
   
   
       15 . The method according to  claim 1  used in at least one of the following applications: 
 system identification, channel equalization, or echo cancellation.    
   
   
       16 . An adaptive filter, comprising: 
 at least one mathematical processing unit to filter, in an adaptive manner, frequency domain signals at a present filtering iteration in at least two frequency bins using respective adaptive filter coefficients; and    an update unit to update at least one but fewer than all of the respective adaptive filter coefficients based on a comparison of the respective adaptive filter coefficients with each other at the present filtering iteration, the at least one mathematical processing unit configured to filter, in an adaptive manner, the frequency domain signals at a next filtering iteration using the updated respective adaptive filter coefficients.    
   
   
       17 . The adaptive filter according to  claim 16  further including normalizers to normalize the frequency domain signals in the at least two frequency bins.  
   
   
       18 . The adaptive filter according to  claim 17  wherein the normalizers include at least one mathematical processing unit to divide instantaneous energy levels of the frequency domain signals in each of the at least two frequency bins by respective average energy levels.  
   
   
       19 . The adaptive filter according to  claim 16  wherein the update unit determines whether to update the respective adaptive filter coefficients by solving a constrained minimization equation.  
   
   
       20 . The adaptive filter according to  claim 19  wherein the update unit solves the constrained minimization equation by calculating a squared-Euclidean-norm.  
   
   
       21 . The adaptive filter according to  claim 16  wherein the update unit includes a least mean square (LMS) processing unit to determine which adaptive filter coefficient to update.  
   
   
       22 . The adaptive filter according to  claim 16  wherein the update unit updates the respective adaptive filter coefficients at every filtering iteration.  
   
   
       23 . The adaptive filter according to  claim 16  wherein the respective adaptive filter coefficients are predefined.  
   
   
       24 . The adaptive filter according to  claim 16  wherein the adaptive filter updates the respective adaptive filter coefficients as a function of a previous average energy level and a present instantaneous energy level of the respective frequency domain signals.  
   
   
       25 . The adaptive filter according to  claim 24  wherein the adaptive filter updates the respective adaptive filter coefficients by applying a scale factor to the previous average and present instantaneous energy levels of the respective frequency domain signals to minimize a potential for adaptive filter lock-up.  
   
   
       26 . The adaptive filter according to  claim 16  further including a transform unit to transform a time domain signal to provide the frequency domain signals in the at least two frequency bins.  
   
   
       27 . The adaptive filter according to  claim 26  wherein the transform unit is an orthogonal transform unit.  
   
   
       28 . The adaptive filter according to  claim 16  wherein the at least one mathematical processing unit outputs an echo cancellation signal, further including a summing unit coupled to the adaptive filter and a far end of a communications system to calculate an error signal by summing (i) a returning far end signal having an echo signal and (ii) the echo cancellation signal.  
   
   
       29 . The adaptive filter according to  claim 28  wherein a feedback path couples the update unit to the summing unit to return the error signal to the update unit to allow the update unit to update the respective adaptive filter coefficients to minimize the echo signal in the returning far end signal.  
   
   
       30 . The adaptive filter according to  claim 16  wherein the update unit updates the respective adaptive filter coefficients in a manner that targets a burst echo signal in the far end signal.  
   
   
       31 . The adaptive filter according to  claim 16  used in at least one of the following applications: 
 system identification, channel equalization, or echo cancellation.    
   
   
       32 . A computer-readable medium having stored thereon sequences of instructions, the sequences of instructions, when executed by a digital processor, causing the processor to: 
 adaptively filter frequency domain signals at a present filtering iteration in at least two frequency bins using respective filter coefficients;    update at least one but fewer than all of the respective filter coefficients at the present filtering iteration based on a comparison of the respective filter coefficients with each other; and    adaptively filter the frequency domain signals at a next filtering iteration using the updated respective filter coefficients.