Journal
IEEE COMMUNICATIONS LETTERS
Volume 27, Issue 8, Pages 1984-1988Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/LCOMM.2023.3282307
Keywords
Channel capacity; capacity-achieving distribution; deep learning; capacity learning
Categories
Ask authors/readers for more resources
In this letter, the problem of determining the capacity of a communication channel is approached as a cooperative game between a generator and a discriminator. Deep learning techniques are used to solve this problem. The generator's task is to produce channel input samples that can be ideally distinguished by the discriminator. The approach, known as cooperative channel capacity learning (CORTICAL), provides both the optimal input signal distribution and the estimated capacity of the channel. Numerical results show that the proposed framework can learn the capacity-achieving input distribution even in non-Shannon settings.
In this letter, the problem of determining the capacity of a communication channel is formulated as a cooperative game, between a generator and a discriminator, that is solved via deep learning techniques. The task of the generator is to produce channel input samples for which the discriminator ideally distinguishes conditional from unconditional channel output samples. The learning approach, referred to as cooperative channel capacity learning (CORTICAL), provides both the optimal input signal distribution and the channel capacity estimate. Numerical results demonstrate that the proposed framework learns the capacity-achieving input distribution under challenging non-Shannon settings.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available