Joint Source-channel Coding Using Machine Learning Techniques
Abstract
Most modern communication systems rely on separate source encoding and channel encoding schemes to transmit data. Despite the long-lasting success of separate schemes, joint source channel coding schemes have been proven to outperform separate schemes in applications such as video communications. The task of this research is to develop a joint source-channel coding scheme that mitigates some of the limitations of current separate coding schemes. My research will attempt to leverage recent advances in machine/deep learning techniques to develop resilient schemes that do not depend on explicit codes for compression and error correction but automatically learn end-to-end mapping schemes for source signals. The success of the developed scheme will depend on its ability to correctly approximate an input vector under inconsistent channel conditions.
Citation
Akpabio, Inimfon (2019). Joint Source-channel Coding Using Machine Learning Techniques. Undergraduate Research Scholars Program. Available electronically from https : / /hdl .handle .net /1969 .1 /175391.