Neural network channel equalizer for wireless signals to prove that LMS is equivalent to a Neural Network in very special conditions. Desired feature list after main goal is accomplished: -Reimplement ANN as a convolutional network in order to pipeline data processing -Actually implement a full TF implementation of an equalizer with RELU activation functions -->Look into RNN as a means to keep track of channel "history" -Experiment with a neural net turbo decoder?
Added testing framework including: -RLS and LMS as baselines -Naive Tensorflow implementation of an equalizer