Research map / Architectures and efficiency / RNNs and normalization
KO codes: inventing nonlinear encoding and decoding for reliable wireless communication via deep-learning
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use nonlinear real-valued learned encoding and decoding on the RM(9,2) tree | RM(9,2) · AWGN | BER | Helped |
| Use nonlinear real-valued learned encoding and decoding on the RM(8,2) tree | RM(8,2) · AWGN | BLER | Helped |
| Use a KO encoder with successive-cancellation decoding | Polar(64,7) · AWGN | BER: almost 0.7 dB | Helped |
| Use KO decoding instead of successive-cancellation decoding | KO(64,7) · AWGN | BER | Helped |
| Test the AWGN-trained code without retraining on Rayleigh fast fading | KO(8,2) · Rayleigh fast fading | BER | Helped |
| Test the AWGN-trained code without retraining on a bursty channel | KO(8,2) · Bursty channel | BER | Helped |
| Constrain transmitted codewords to be binary while retaining a nonlinear encoder (ablation) | KO(8,2) · AWGN | BER | Hurt |
| Constrain the encoder to be linear over real values (ablation) | KO(8,2) · AWGN | BER | Hurt |
| Compress decoder neural blocks to 1 hidden layer with 4 nodes (ablation) | KO(8,2) · AWGN, SNR -4 dB | BER: 1.18919e-4 ± 7e-9 -> 0.64702e-4 ± 4e-9 | Hurt |
| Add list decoding to the existing KO(8,2) decoder without retraining | KO(8,2) · AWGN | BER | Helped |
Closest papers by meaning
- Communication Algorithms via Deep Learning (ICLR 2018 · similarity 0.90)
- CRISP: Curriculum based Sequential neural decoders for Polar code family (ICML 2023 · similarity 0.88)
- DeepPolar: Inventing Nonlinear Large-Kernel Polar Codes via Deep Learning (ICML 2024 · similarity 0.87)
- Learning Linear Block Error Correction Codes (ICML 2024 · similarity 0.85)
- Friendly Attacks to Improve Channel Coding Reliability (AAAI 2024 · similarity 0.84)
- Structured Masked Diffusion for Joint Multiuser Decoding (NeurIPS 2026 · similarity 0.81)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.