Skip to content

Wireless Institute Seminar – ProductAE: Towards Training Larger Channel Codes based on Neural Product Codes

May 27, 2022 at 11:00 am
258 Fitzpatrick Hall

Dr. Homayoon Hatami

Abstract:

There have been significant research activities in recent years to automate the design of channel encoders and decoders via deep learning. Due to the dimensionality challenge in channel coding, it is prohibitively complex to design and train relatively large neural channel codes via deep learning techniques. Consequently, most of the results in the literature are limited to relatively short codes having less than 100 information bits. In this work, we present ProductAEs, a computationally efficient family of deep-learning driven (encoder, decoder) pairs, that aim at enabling the training of relatively large channel codes (both encoders and decoders) with a manageable training complexity. We build upon the ideas from classical product codes and propose constructing large neural codes using smaller code components. More specifically, instead of directly training the encoder and decoder for a large neural code of dimension k and block-length n, we provide a framework that requires training neural encoders and decoders for the code parameters (k1,n1) and (k2,n2) such that k1k2=k and n1n2=n. Our training results show significant gains, over all ranges of signal-to-noise ratio (SNR), for a code of parameters (100,225) and a moderate-length code of parameters (196,441), over polar codes under successive cancellation (SC) decoder. Moreover, our results demonstrate meaningful gains over Turbo Autoencoder (TurboAE) and state-of-the-art classical codes. This is the first work to design product autoencoders and pioneering work on training large channel codes.

Bio:

Homayoon Hatami was born in Kermanshah, Iran, in 1987. He received his B.Sc. and M.Sc. degrees in electrical engineering – communication systems from the University of Tehran, Tehran, Iran, in 2010 and 2013, respectively. He received his Ph.D. degree in electrical engineering – communication systems from the University of Notre Dame, Notre Dame, IN, USA, in 2019. Since 2019, he has been with Samsung Semiconductor Inc., San Diego, CA, USA, where he is currently a Staff Engineer. Homayoon and his colleagues were recently named co-recipient of the 2022 ICC Best Paper Award in the category “Selected Areas in Communications.” IEEE ICC and arXiv:2110.04466.

Add To Calendar

Wireless Institute Events

There are no events scheduled at this time.