Method of performing convergence in a, least mean square, adaptive filter, echo canceller

• US 5,295,136 A
• Filed: 04/13/1992
• Issued: 03/15/1994
• Est. Priority Date: 04/13/1992
• Status: Expired due to Term

• Current Assignee: Motorola Mobility, Inc. (Lenovo Group Ltd.)
• Original Assignee: Motorola, Inc. (Motorola Solutions, Inc.)
• Inventors: Ashley, James P., Nguyen, Quoc
• Primary Examiner(s): Olms, Douglas W.
• Assistant Examiner(s): Kizou, Hassan
• Application Number:
  US07/867,555
• Time in Patent Office: 701 Days
• Field of Search: 370/32.1, 370/32, 379/410, 379/411, 375/103, 375/14, 364/724.16, 364/724.19
• US Class Current: 370/290
• CPC Class Codes: H04B 7/015 Reducing echo effects

Abstract

A method is described of decreasing convergence time in an adaptive echo canceller. The method includes the step of locating a primary echo within a filter vector based upon relative tap values within the filter vector. The filter vector is then narrowed, based upon the located taps. An estimated error is determined based, in part, upon the narrowed filter vector. An updated filter vector is produced based, in part, upon the estimated error.

Claims

1. A method of converging an adaptive echo canceller, the method comprising the steps of:

locating a representation of a primary echo within a filter vector of a received signal based upon a comparison of relative magnitues of a plurality of taps of the filter vector;

dividing the filter vector into a narrowed filter vector proximate the primary echo and a secondary filter vector;

estimating a primary error by vector multiplying the narrowed filter vector by a correlative origin signal vector of the narrowed filter vector and subtracting the product from a return signal;

estimating a secondary error by vector multiplying the secondary filter vector by a correlative origin signal vector of the secondary filter vector and subtracting the product from the primary error;

generating a primary adaptation vector by multiplying the estimated primary error by the correlative origin signal vector of the narrowed filter vector and a narrow vector adaptation factor;

generating a secondary adaptation vector by multiplying the estimated secondary error by the correlative origin signal vector of the secondary filter vector and secondary vector adaptation factor; and

updating the filter vector by summing with the filter vector within an area defined by the narrowed filter vector the primary adaptation vector and summing with the filter vector within an area defined by the secondary vector the secondary adaptation vector.

2. The method as in claim 1 wherein the step of generating the primary and secondary adaptation vector further comprises the step of dividing the primary and secondary errors by a scaling factor.

3. An apparatus for converging an adaptive echo canceller, the apparatus comprising:

means for locating a representation of a primary echo within a filter vector of a received signal based upon a comparison of relative magnitudes of a plurality of taps of the filter vector;

means for dividing the filter vector into a narrowed filter vector proximate the primary echo and a secondary filter vector;

means for estimating a primary error by vector multiplying the narrowed filter vector by a respective correlative origin signal vector and summing the vector product with a return signal, and for estimating a secondary error by vector multiplying the secondary filter vector by a respective correlative origin signal vector and summing the product vector with the primary error; and

means for updating the filter vector within an area defined by the narrowed filter vector using the primary error and a narrow vector adaptation factor and within an area defined by the secondary vector by using the estimated secondary error and a secondary adaptation factor.

4. In an adaptive echo canceller having a filter vector with taps, each with a tap value, a method of decreasing convergence time, the method including the steps of:

locating a representation of a primary echo within the filter vector based upon a determination of a location of the largest tap value;

dividing the filter vector into a narrowed filter vector and a secondary filter vector based upon the located primary echo;

estimating a primary and secondary error based, in part, upon the narrowed filter vector and secondary filter vector; and

, producing an updated filter vector based, in part, upon the estimated errors, a narrow vector adaptation factor and a secondary adaptation factor.