Science and Technology Production
Congress
Authorship
Date
2019
Publishing House and Editing Place
UNS
Summary
Information provided by the agent in
SIGEVA
In this paper, a Neural Network based decoder ispresented, which is based on the Successive Cancellation decoderfor Polar Codes. The purpose is to test a classic decoding scheme,while replacing the functions which characterize the structure ofthe Polar Code Decoder by Neural Networks. The topology ofthe resulting system was studied by simulation, and the best parameters,selected. The Frame Error Rate was computed as well,for transmission over three different channels (Additive WhiteGaussian Noi...
In this paper, a Neural Network based decoder ispresented, which is based on the Successive Cancellation decoderfor Polar Codes. The purpose is to test a classic decoding scheme,while replacing the functions which characterize the structure ofthe Polar Code Decoder by Neural Networks. The topology ofthe resulting system was studied by simulation, and the best parameters,selected. The Frame Error Rate was computed as well,for transmission over three different channels (Additive WhiteGaussian Noise, Rayleigh fading and Impulsive Noise channels).Notably, the proposed decoder does not require channel sideinformation, and shows a good performance independently ofthe characteristics of the channel.
Show more
Show less
Key Words
NEURAL NETWORKSDECODINGPOLAR CODES