README_bib.md

framework

Interference Cancellation

Jakubisin, Daniel J., and R. Michael Buehrer. "Approximate joint MAP detection of co-channel signals in non-Gaussian noise." IEEE Transactions on Communications 64.10 (2016): 4224-4237.

• [jakubisin2016approximates]

Jakubisin, Daniel J., and R. Michael Buehrer. "Approximate joint MAP detection of co-channel signals." Military Communications Conference, MILCOM 2015-2015 IEEE. IEEE, 2015.

• [jakubisin2015approximate]

BCJR

• nib no
• NCTU TW
• Caltech

CMAP

Concurrent MAP Detector Jiang, Wei, and Daoben Li. "Iterative single-antenna interference cancellation: algorithms and results." IEEE Transactions on vehicular technology 58.5 (2009): 2214-2224

• c16
• [jiang2009iterative]

Rake Gaussian

Ping, Li, Lihai Liu, and W. K. Leung. "A simple approach to near-optimal multiuser detection: interleave-division multiple-access." Wireless Communications and Networking, 2003. WCNC 2003. 2003 IEEE. Vol. 1. IEEE, 2003.

• c13
• [ping2003simple]

Neural Network Alternative

general

• Qin, Zhijin, et al. "Deep learning in physical layer communications." arXiv preprint arXiv:1807.11713 (2018).
  • [qin2018deep]
  • why ML
    • Mathematical model versus practical imperfection
    • Block structures versus global optimality
    • Robust signal processing algorithms versus low costs
  • GAN to accelerate classfication
• O’Shea, Timothy, and Jakob Hoydis. "An introduction to deep learning for the physical layer." IEEE Transactions on Cognitive Communications and Networking 3.4 (2017): 563-575.
  • [o2017introduction]
  • 
  • DeepSig & Virginia Tech
  • BPSK, multi-user
  • without CSI

detection

• Ye, Hao, Geoffrey Ye Li, and Biing-Hwang Juang. "Power of deep learning for channel estimation and signal detection in OFDM systems." IEEE Wireless Communications Letters 7.1 (2018): 114-117.
  • [ye2018power]
  • end-to-end without CSI
  • OFDM
  • Baseline: LS, MMSE
• Farsad, Nariman, and Andrea Goldsmith. "Detection algorithms for communication systems using deep learning." arXiv preprint arXiv:1705.08044 (2017).
  • [farsad2017detection]
  • moleceluar when channel model N/A

decode

• Nachmani, Eliya, Yair Be'ery, and David Burshtein. "Learning to decode linear codes using deep learning." Communication, Control, and Computing (Allerton), 2016 54th Annual Allerton Conference on. IEEE, 2016.
  • [nachmani2016learning]
  • independence of the performance on the txed codeword

Extensive

• Ye, Hao, and Geoffrey Ye Li. "Initial results on deep learning for joint channel equalization and decoding." Vehicular Technology Conference (VTC-Fall), 2017 IEEE 86th. IEEE, 2017.
  • [ye2017initial]

  • the back propagation is very similar to belief propagation, the traditional decoding algorithm

  • more than white gaussian noise, OFDM
  • Baseline: Gaussian process for classification (GPC)+successive cancellation(SC), MMSE+SC
• Xu, Weihong, et al. "Polar decoding on sparse graphs with deep learning." arXiv preprint arXiv:1811.09801 (2018).
  • [xu2018polar]
  • LDPC decode
  • Baseline: sum-product algorithm(SPA),Min-Sum
• Gruber, Tobias, et al. "On deep learning-based channel decoding." Information Sciences and Systems (CISS), 2017 51st Annual Conference on. IEEE, 2017.
  • [gruber2017deep]
  • 
  • Baseline: MAP
  • Conclusion:
    • structured codes are easier to learn
    • the neural network is able to generalize to codewords that it has never seen during training for structured, but not for random codes.
• Kim, Minhoe, et al. "Deep Learning-Aided SCMA." IEEE Communications Letters 22.4 (2018): 720-723.
  • SCMA encode and decode