POLAR CODES WITH ENHANCED MACHINE LEARNING

Abstract

Arikan (2009) introduced a revolutionary approach to achieving capacity on binary memoryless channels, known as polar coding. The list decoding, particularly accompanied by use of CRC outer code, gives dramatic improvement in block error and bit error probabilities. We look for managing performance across varying signal-to-noise ratios (SNR), for digital wireless communication system. Machine learning presents a promising solution to develop more robust and intelligent decoding strategies. We study and do the comparison of traditional and Machine Learning Performance with Polar codes.