Deep learning based autoencoder for m-user wireless interference channel physical layer design
- Submitting institution
-
University of Sussex
- Unit of assessment
- 12 - Engineering
- Output identifier
- 410738_93968
- Type
- D - Journal article
- DOI
-
10.1109/ACCESS.2020.3025597
- Title of journal
- IEEE Access
- Article number
- -
- First page
- 174679
- Volume
- 8
- Issue
- -
- ISSN
- 2169-3536
- Open access status
- Compliant
- Month of publication
- September
- Year of publication
- 2020
- URL
-
https://doi.org/10.1109/ACCESS.2020.3025597
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
2
- Research group(s)
-
-
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- We introduce and develop a Deep Learning (DL) communications systems that aims to solve the problem of m-user wireless interference inherent to 5G and beyond-5G systems operating in the same frequency. We developed a fundamental new way for design of the physical layer as an end-to-end DL reconstruction optimization task. We develop a novel DL based method using Auto encoders for interference-adaptive constellation as a solution to wireless interference problem. Our method demonstrates a significant achievable improvement from the conventional 5G system. Our method is being adopted by key players including Samsung and Nokia for 5G and beyond-5G/6G communication standards.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
