在即將到來的5G移動通信中，延遲問題是提高整體通信效率的主要挑戰之一，根據國際電信聯盟提出的IMT-2020中，不僅要擴大通信帶寬以包含更多的傳輸數據，整體延遲也應該要相對應的減低。5G的應用場景超可靠度和低延遲通訊(Ultra-reliable and Low Latency Communications, URLLC)中提出了一個技術，無允諾上行(Uplink Grant-free)傳輸，藉由預先配置給用戶特殊編碼或是特殊訊號，以此省去上傳時需要等待接收端的上傳許可所造成的延遲。
此篇論文的重點在於提高前傳上行網路的傳輸效率並降低整體延遲，提出了基於無允諾上傳DRoF系統，把傳統用於CDMA的展頻碼，金氏碼，和類神經網路與現有的DRoF系統組合，金氏碼依靠其高自相關與不論任何時間延遲的低互相關可以在各種光纖傳輸差異下始終檢索金氏碼多路並行DRoF信號，而類神經網路可以同時達成降低展頻碼的碼長並且維持相近的錯誤率，同時我將不同波長的雷射應用於兩個傳輸端，結果表明金氏碼多路DRoF與DWDM完全兼容，並且可以在沒有任何光學濾波機制的情況下成功解訊號，本篇論文將以兩路使用者分別將訊號載在不同雷射波長與傳輸不同的光纖長度後，創造出在接收端將會在不同時間接收訊號的情境，而使用金氏碼編碼後並以類神經網路解調的DROF訊號依然可以順利在FEC threshold下還原兩路不同使用者的訊號。

In the coming 5G mobile communication, the latency issue is one of the major challenges to improve overall communication efficiency. As indicated by ITU in IMT-2020, not only the communication bandwidth should be enlarged to enclose more transmission data, but also the overall latency should be revised down. 5G application scenarios in Ultra-reliable and Low Latency Communications (URLLC) propose a technique, grant-free uplink transmission, which is pre-configured to give users particular coding or particular signal, in order to reduce the delay caused by waiting uploading license of the receiver.
The focus of this paper is to improve the transmission efficiency of the forward transmission network and reduce the overall delay. A grant-free uplink DRoF system is proposed, which combines the traditional CDMA spreading code, the Gold code, and neural network with the DRoF system. In the DRoF system, the Gold code depends on its high autocorrelation and low cross-correlation regardless of any time delay to always retrieve the Gold code multi-channel parallel DRoF signal under various fiber transmission differences, and neural network can simultaneously reduce the code length of the spreading code and maintain a similar error rate. At the same time, we apply different wavelengths of laser to both transmitter. The results show that the Gold code multi-channel DRoF is compatible with DWDM and can be demodulate without any additional optical filtering mechanism. In this paper, we will separately modulate the signals at different laser wavelengths and transmit different fiber lengths to demonstrate a situation where receiver will receive signals at different times. The DROF signal with Gold code demodulated by the neural network can still successfully recover the signals of two different users under the FEC threshold.

[1] "Future mobile data usage and traffic growth," 2019. [Online]. Available: https://www.ericsson.com/en/mobility-report/future-mobile-data-usage-and-traffic-growth.
[2] G. Brown, “White Paper: Exploring 5G New Radio: Use Cases, Capabilities & Timeline,” Heavy Reading on behalf of Qualcomm (2016)
[3] N. Shibata, T. Tashiro, S. Kuwano, N. Yuki, J. Terada, and A. Otaka, “Mobile front-haul employing Ethernet-based TDM-PON system for small cells,” 2015 Optical Fiber Communication Conference (OFC), paper M2J.1.
[4] P. Chanclou, L. Anet Neto, K. Grzybowski, Z. Tayq, F. Saliou, and N. Genay, “Mobile Fronthaul Architecture and Technologies: A RAN Equipment Assessment [Invited],” J. Opt. Commun. Netw. 10, A1-A7 (2018)
[5] "History of spread spectrum and CDMA," 2008. [Online]. Available: http://wcdma3g.blogspot.com/2008/07/history-of-wcdma.html.
[6] D. Iida, S. Kuwano, J. Kani, J. Terada, “Dynamic TWDM-PON for Mobile Radio Access Networks,” Opt. Express 21, 26209-26218 (2013)
[7] A. de la Oliva, J. A. Hernandez, D. Larrabeiti and A. Azcorra, “An overview of the CPRI specification and its application to C-RAN-based LTE scenarios,” IEEE Commun. Mag. 54, 152-159 (2016)
[8] Y.-x. L. Jing Zhang, Şahin Kaya Özdemir, Re-Bing Wu, Feifei Gao, Xiang-Bin Wang, Lan Yang & Franco Nori "Quantum Internet using code division multiple access," 2013. [Online]. Available: https://www.nature.com/articles/srep02211
[9] D. L. S. R.L. PICKHOLTZ, L.B. MILSTEIN "Theory of Spread-Spectrum Communications-A Tutorial," 1982. [Online]. Available:
https://pdos.csail.mit.edu/archive/decouto/papers/pickholtz82.pdf
[10] 粘添壽, "無線網路之傳輸技術," 2002. [Online]. Available: http://www.tsnien.idv.tw/Network_WebBook/chap15/15-2%20%E7%84%A1%E7%B7%9A%E7%B6%B2%E8%B7%AF%E4%B9%8B%E5%82%B3%E8%BC%B8%E6%8A%80%E8%A1%93.html
[11] R. Gold, “Maximal recursive sequences with 3-valued cross-correlation functions,” IEEE Trans. Inf. Theory 14, 154-156 (1968)
[12] Wikipedia, "Correlation function." [Online]. Available: https://en.wikipedia.org/wiki/Correlation_function.
[13] Hecht-Nielsen, "Theory of the backpropagation neural network," International 1989 Joint Conference on Neural Networks, Washington, DC, USA, 1989, pp. 593-605 vol.1.
[14] "Backpropagation demo," 2018. [Online]. Available: https://google-developers.appspot.com/machine-learning/crash-course/backprop-scroll/
[15] "Gradient Descent," 2018. [Online]. Available: https://www.wandouip.com/t5i10130/
[16] I. c. s. I. am. (2018). Gradient Descent. Available: https://ithelp.ithome.com.tw/articles/10198147