近年來有許多深度學習應用在通訊解碼的研究，但其中有研究認為應該要可以適應不同的編碼。因此提出了基於CNN的解碼架構，但即使這樣還是有模型數量過多的問題。我們認為還有可以改進的空間。在本文中，我們將研究如何只使用一個模型即可同時應用於相同碼率的多個卷積碼上，而不需要使用多個模型來回切換來處理不同的卷積碼。為了解決這一問題，我們提出了將卷積碼的編碼類別作為條件輸入的神經網路解碼器架構。通過這種方法，我們可以顯著減少所需的模型數量，同時保持有效的解碼性能。此外，我們比較了解碼器和分類器的獨立及聯合訓練方式，結果表明我們的聯合訓練方法在多種情況下都有優越的性能。

In recent years, there have been extensive research on applying deep learning to communication decoding. However, some studies suggest that the models should be adaptable to different coding schemes. Consequently, a CNN-based decoding architecture has been proposed. However, this approach still faces the issue of having too many models. We believe there is room for improvement. In this thesis, we aim to develop a method that allows a single model to be used for multiple convolutional codes with the same code rate, without the need for switching between different models for different codes. To solve this problem, we propose a neural network decoder architecture that takes the encoding category of a convolutional code as a conditional input. With this approach, we can significantly reduce the number of models required while maintaining effective decoding performance. Furthermore, we compared the independent and joint training methods of decoder and classifier, and the results show that our joint training method has superior performance in various situations.

[1] A. Viterbi, "Error bounds for convolutional codes and an asymptotically optimum decoding algorithm," IEEE Transactions on Information Theory, vol. 13, no. 2, pp. 260-269, 1967.
[2] R. Vijay, S. Mrinalee, G. Mathur. "Comparison between Viterbi algorithm soft and hard decision decoding," Journal of Information Systems and Communication, vol. 3, issue 1, pp. 193-198, 2012.
[3] Y. Sun, G. Dou and H. Yin, "A Bit Flipping Viterbi Decoding Algorithm Based on CRC Check for Convolutional Code," IEEE 21st International Conference on Communication Technology, pp. 1292-1295, 2021.
[4] H. Yoshikawa, "Error Performance Analysis of the K-best Viterbi Decoding Algorithm," International Symposium on Information Theory and Its Applications, pp. 257-260, 2018
[5] J. King, W. Ryan, C. Hulse, R. D. Wesel, "Efficient Maximum-Likelihood Decoding for TBCC and CRC-TBCC Codes via Parallel List Viterbi," 2023 12th International Symposium on Topics in Coding, pp. 1-5, 2023
[6] V. V. Zolotarev, N. N. Grinchenko, G. V. Ovechkin, P. V. Ovechkin, "Modified Viterbi algorithm for decoding of block codes," 2017 6th Mediterranean Conference on Embedded Computing, pp. 1-4, 2017
[7] T. Agrawal, A. Kumar and S. K. Saraswat, "Comparative analysis of convolutional codes based on ML decoding," International Conference on Communication Control and Intelligent Systems, pp. 41-45, 2016.
[8] A. Middya, A. S. Dhar, "Real-time area efficient and high speed architecture design of Viterbi decoder," International Conference on Advances in Electrical, Electronics, Information, Communication and Bio-Informatics, pp. 246-250, 2016.
[9] R. Surya, K. Balasubramanian, B. Yamuna, "Design of a low power and high-speed Viterbi decoder using T-algorithm with normalization," International Conference on Advances in Computing and Communications, pp. 1-6, 2021.
[10] G. D. Korde, S. L. Haridas. "Design of Asynchronous Viterbi decoder using pipeline architecture," International Journal for Research in Applied Science & Engineering Technology, Vol. 6, Issue I, pp. 464-469, 2018.
[11] W. Lyu, Z. Zhang, C. Jiao, K. Qin and H. Zhang, “Performance Evaluation of Channel Decoding with Deep Neural Networks,” IEEE International Conference on Communications, pp. 1-6, 2018.
[12] Z. Zhang, D. Yao, L. Xiong, B. Ai and S. Guo, "A convolutional neural network decoder for convolutional codes," International Conference on Communications and Networking in China, pp. 113-125, 2019.
[13] L. Li-fu, L. Hai-wen, L. Hong-liang, G. Yong-jun, "Research and implementation of Viterbi decoding in TD-LTE system," 2017 IEEE 2nd Advanced Information Technology, Electronic and Automation Control Conference, pp. 890-894, 2017
[14] H. Yang, E. Liang, R. D. Wesel, "Joint Design of Convolutional Code and CRC under Serial List Viterbi Decoding", 2018 IEEE Global Communications Conference, 2018.
[15] T. Gruber, S. Cammerer, J. Hoydis S. t. Brink, "On deep learning-based channel decoding," 51st Annual Conference on Information Sciences and Systems, pp. 1-6, 2017
[16] E. Nachmani, Y. Be'ery D. Burshtein, "Learning to decode linear codes using deep learning," 54th Annual Allerton Conference on Communication, Control, and Computing, pp. 341-346, 2016
[17] E. Nachmani, E. Marciano, L. Lugosch, W. J. Gross, D. Burshtein, Y. Be’ery, "Deep Learning Methods for Improved Decoding of Linear Codes," IEEE Journal of Selected Topics in Signal Processing, vol. 12, no. 1, pp. 119-131
[18] M. Rowshan, M. Qiu, Y. Xie, X. Gu, J. Yuan, "Channel Coding Toward 6G: Technical Overview and Outlook," IEEE Open Journal of the Communications Society, vol. 5, pp. 2585-2685, 2024
[19] L. Fanari, E. Iradier, I. Bilbao, R. Cabrera, J. Montalban, P. Angueira, O. Seijo, I. Val, "A Survey on FEC Techniques for Industrial Wireless Communications," IEEE Open Journal of the Industrial Electronics Society, vol. 3, pp. 674-699, 2022
[20] J. Wang, C. R. Tang, H. Huang, H. Wang and J. Q. Li, “Blind Identification of Convolutional Codes Based on Deep Learning,” Digital Signal Processing, vol. 115, 2021.
[21] Y. Yang, F. Gao, X. Ma, S. Zhang, "Deep Learning-Based Channel Estimation for Doubly Selective Fading Channels," IEEE Access, vol. 7, pp. 36579-36589, 2019
[22] N. Farsad, A. Goldsmith, "Neural Network Detection of Data Sequences in Communication Systems," IEEE Transactions on Signal Processing, vol. 66, no. 21, pp. 5663-5678, 2018
[23] L. Huang, W. Chen, E. Chen, H. Chen, "Blind recognition of k/n rate convolutional encoders from noisy observation," Journal of Systems Engineering and Electronics, vol. 28, no. 2, pp. 235-243, 2017
[24] P. Yu, H. Peng, J. Li, "On Blind Recognition of Channel Codes Within a Candidate Set," IEEE Communications Letters, vol. 20, no. 4, pp. 736-739, 2016
[25] Y. Wang, H. You, X. Wang Z. Huang, "Blind Recognition of Convolutional Codes: A Matrix Transformation-Aided Deep Learning Approach," IEEE International Conference on Signal Processing, Communications and Computing, pp. 1-5, 2022
[26] A. Goldsmith, “Wireless Communications,” Cambridge University Press, 2005.
[27] E. Jang, S. Gu, B. Poole, “Categorical reparameterization with Gumbel-softmax,” International Conference on Learning Representations, 2017.
[28] Chris J. Maddison, Andriy Mnih, Yee Whye The, “The Concrete Distribution: A Continuous Relaxation of Discrete Random Variables,” arXiv preprint arXiv:1611.00712, 2017.
[29] A. Paszke, S. Gross, F. Massa, A. Lerer, J. Bradbury, G. Chanan, T. Killeen, Z. Lin, N. Gimelshein, L. Antiga, A. Desmaison, A. Köpf, E. Yang, Z. DeVito, M. Raison, A. Tejani, S. Chilamkurthy, B. Steiner, L. Fang, J. Bai, S. Chintala, “Pytorch: An imperative style, high-performance deep learning library”, Advances in neural information processing systems, vol. 32, 2019.
[30] D. P. Kingma, J. Ba, “Adam: A method for stochastic optimization,” arXiv preprint arXiv:1412.6980, 2017