CSL professor works to improve digital communication

Reliable digital communication, both wireline and wireless, is the primary workhorse of the modern information age. A critical aspect of reliable communication involves the design of codes that allow transmissions to be robustly (and computationally efficiently) decoded under noisy conditions. According to CSL professor

Pramod Viswanath, great strides have been made in the field of coding. He posits that the continuing development of these codes will rely on deep learning methods which will play a critical role in solving various goals of coding theory.  

“Basic progress in coding and communication has generally been driven by individual human ingenuity and, befittingly, is sporadic” says Viswanath, and ECE Gilmore Family Endowed Professor. “Deep learning is fast emerging as capable of learning sophisticated algorithms from observed data alone and has been remarkably successful in a large variety of human endeavors.”

Motivated by these successes, Professor Viswanath envisions that deep learning methods could play a crucial role in making the next breakthroughs of coding and communication theory and practice.

“Our research aims to bring the tools of deep learning to design new family of encoding and decoding methods for canonical communication models; the codes so generated are naturally built for finite block lengths” says Viswanath.

Two major activities are planned for the neural network architectures that will be the outcomes of this research. First, an online repository will make available all algorithms, maintained with source code and documentation, developed in this project. Second, the practical applications of these new codes will be tested and explored within the context of commercial wireless deployments such as cell phones.