System Identification Using LMS Algorithm and Huber Cost Function Minimization

Modelling a FIR Filter using LMS Algorithm and, Huber's Cost Function for presence of outliers

Modelling a FIR Filter using LMS Algorithm and, Huber's Cost Function Minimization for presence of a certain percentage of outliers.
Here we have to identify and model a 3-tap FIR filter with weights [0.26 0.93 0.26].
This has to be done using:
1) Mean Square error minimization (LMS Algorithm)-
The reference signal is corrupted by additive white gaussian noise (mean=0, standard deviation=0.1)
2) Huber Loss Minimization (with 10 to 20 percent outlier added to the noise)
The reference signal is corrupted by additive white gaussian noise (mean=0, standard deviation=0.05)