LMS

Overview

Normalized LMS FIR filter

Discussion

The LMS module implements a single channel N-order FIR adaptive filter. The module takes 2 single channel inputs: desired filter output and filter input (the reference signal). For every data sample the filter output is compared to the desired output and this information is used to update the filter taps according to the normalized LMS algorithm. The module always has 2 single channel ouputs: adaptive filter output and error signal, (desired filter output - actual filter output. The module also optionally outputs the filter coefficients on a third output pin. The array coeffs holds the filter coefficients and is of size numTaps x 1. The coefficients are stored in normal order. The state variable array is of size numTaps x 1. The adaptation constant mu is visible as a parameter.

Type Definition

CODE

typedef struct _ModuleLMS
{
    ModuleInstanceDescriptor instance;            // Common Audio Weaver module instance structure
    INT32 maxTaps;                                // Maximum length of the filter
    INT32 numTaps;                                // Current length of the filter
    FLOAT32 mu;                                   // Adaptation constant
    INT32 stateIndex;                             // Index of the oldest state variable in the array of state variables
    FLOAT32* coeffs;                              // Coefficient array
    FLOAT32* state;                               // State variable array
} ModuleLMSClass;

Variables

Properties

Name	Type	Usage	isHidden	Default value	Range	Units
maxTaps	int	const	0	1024	1:1:5000	samples
numTaps	int	parameter	0	32	1:1:1024	samples
mu	float	parameter	0	0	0:1	linear
stateIndex	int	state	1	0	Unrestricted
coeffs	float*	parameter	0	[1024 x 1]	Unrestricted
state	float*	state	1	[1027 x 1]	Unrestricted

Pins

Input Pins

Name: dataIn

Description: Filter input

Data type: float

Channel range: 1

Block size range: Unrestricted

Sample rate range: Unrestricted

Complex support: Real

Name: reference

Description: Reference input: desired filter output

Data type: float

Channel range: 1

Block size range: Unrestricted

Sample rate range: Unrestricted

Complex support: Real

Output Pins

Name: dataOut

Description: Filter output

Data type: float

Name: errorSignal

Description: Error signal = difference between filter output and reference input

Data type: float

Name: coeffs

Description: Filter coefficients

Data type: float

MATLAB Usage

File Name: lms_module.m

CODE

 M=lms_module(NAME, MAXTAPS, OUTPUTCOEFFS)
 Creates a normalized LMS FIR filter for use in the Audio Weaver environment.
 The module has 2 single channel input pins and up to 3 single channel 
 output pins: 
    Input Pin 1:  dataIn  - Adaptive filter desired output
    Input Pin 2:  reference - Adaptive filter input
    Output Pin 1:  dataOut - output of the adaptive filter
    Output Pin 2:  errorSignal - difference between dataIn and dataOut
    Output Pin 3:  Coefficients (optional)
    
 Arguments:
    NAME - name of the module.
    MAXTAPS - maximum length of the filter (number of taps).
    OUTPUTCOEFFS - Boolean which specifies whether a third output
       pin holding the coefficients should be added. By default,
       OUTPUTCOEFFS=0.