The Least mean squares filter solution converges to the Wiener filter solution, assuming that the unknown system is LTI and the noise is stationary.
Both filters can be used to identify the impulse response of an unknown system, knowing only the original input signal and the output of the unknown system.
By relaxing the error criterion to reduce current sample error instead of minimizing the total error over all of n, the LMS algorithm can be derived from the Wiener filter.
Given a known input signal
, the output of an unknown LTI system
is an unknown filter tap coefficients and
The model system
, using a Wiener filter solution with an order N, can be expressed as:
are the filter tap coefficients to be determined.
The error between the model and the unknown system can be expressed as:
The total squared error
Use the Minimum mean-square error criterion over all of
Substitute the definition of
Distribute the partial derivative:
Using the definition of discrete cross-correlation:
{\displaystyle R_{xy}(i)=\sum _{n=-\infty }^{\infty }x[n]y[n-i]}
{\displaystyle {\frac {\partial E}{\partial {\hat {h}}_{i}}}=-2R_{xs}[i]+2\sum _{k=0}^{N-1}{\hat {h}}_{k}R_{ss}[i-k]=0}
Rearrange the terms:
{\displaystyle R_{xs}[i]=\sum _{k=0}^{N-1}{\hat {h}}_{k}R_{ss}[i-k]}
This system of N equations with N unknowns can be determined.
The resulting coefficients of the Wiener filter can be determined by:
{\displaystyle W=R_{xx}^{-1}P_{xs}}
is the cross-correlation vector between
By relaxing the infinite sum of the Wiener filter to just the error at time
, the LMS algorithm can be derived.
The squared error can be expressed as:
Using the Minimum mean-square error criterion, take the gradient:
Apply chain rule and substitute definition of y[n]
Using gradient descent and a step size
This is the LMS update equation.