学位论文简介
随着第五代移动通信技术(5-th generation, 5G)的商用,智能家居、智慧城市、万物互联、增强现实娱乐等高带宽、低时延、多连接的应用带来了用户对数据流量的需求增长,也对无线接入网络提出了新的挑战。然而,受限于系统的器件以及信道特性带来的衰落效应,影响系统的接收性能和可实现速率。本论文对光OFDM系统中基于旋转QAM和机器学习方法进行了深入的研究,通过信道与系统模型的仿真分析以及实验平台搭建与结果讨论,对提出的方法进行了验证和分析。主要取得了以下创新性成果:
第一,针对器件带宽受限对光OFDM高频子载波性能的影响,提出了基于旋转QAM的增强自适应方案,该方案通过将旋转QAM自适应调制后的高频子载波和低频子载波符号进行交织,减轻了子载波之间的信噪比(Signal-to-Noise Rate, SNR)的不平衡,能有效地提升高频子载波的BER性能。
第二,针对基于旋转QAM的传统解调中对信道信息的依赖,从而在信道估计不准以及信道条件变化时解调性能受损的问题,提出了改进的误差纠正的最大似然判决(Deviation Correction-based Maximum Likelihood Decision,DC-MLD)和机器学习(Machine Learning ,ML)解调方法。改进的DC-MLD可以提高接收灵敏度,机器学习解调方法可以达到与MLD近似的性能而无需CSI。
第三,针对基于旋转QAM的OFDM系统中带宽受限的问题,设计了基于旋转QAM的PAS OFDM系统,在衰落信道中进行了系统的仿真分析并搭建了实验平台进行验证,表明基于旋转QAM的PAS可以在抵抗信道衰落的同时提升系统的可实现速率。在此基础上,为了避免在旋转QAM的解调过程中对CSI的估计,提出了ANN的解调方法,在1m自由空间以及50km光纤和1m自由空间传输后,与PAS方案相比能提高接收灵敏度。
第四,为了进一步降低降低训练和测试开销的同时保证性能,提出了改进的分组ANN(Grouped-ANN,G-ANN)的均衡方法,该方法对训练集的数据进行了分组和量化处理。搭建了实验平台对G-ANN网络在不同参数设置下的性能进行了分析,选择了最优网络性能的参数。经20km光纤和1m自由空间传输后,与传统均衡和ANN均衡方法相比,能提高接收灵敏度的同时可以提升系统的可实现速率。
主要学术成果
He J, Xu L, He J, et.al. Performance comparison of different rotated QAM based DMT schemes in underwater optical wireless communication. Journal of Lightwave Technology, 2023, 41(2): 610-617. (SCI-1)
He J, He J, Shi J. An enhanced adaptive scheme with pairwise coding for OFDM-VLC system. IEEE Photonics Technology Letters, 2018, 30(13): 1254-1257. (SCI-3)
He J, Zhou Z, He J. A rotated QAM-based probabilistically shaped OFDM with ANN scheme for W-band RoF system. 2022 Optical Fiber Communications Conference and Exhibition (OFC). IEEE, 2022: 1-3. (顶级会议,EI)
He J, He J. A rotated QAM based DMT scheme for underwater optical wireless communication system. 2021 International Conference on Optical Instruments and Technology: Optical Communication and Optical Signal Processing. SPIE, 2022, 12278: 56-62. (EI)
He J, He J. Grouped-ANN based equalizer for the rotated and probabilistically shaped 64QAM OFDM in W-band ROF system. Photonics Technology Letter. (SCI,Under review)
He J, Li Z, He J, et.al. Visible laser light communication based on LDPC-coded multi-band CAP and adaptive modulation. Journal of Lightwave Technology, 2019, 37(4): 1207-1213. (SCI)
He J, Tang K, He J, et.al. Effective vehicle-to-vehicle positioning method using monocular camera based on VLC. Optics Express, 2020, 28(4): 4433-4443. (SCI)
Shi J, He J, He J, et.al. Enabling user mobility for optical camera communication using mobile phone. Optics express, 2018, 26(17): 21762-21767. (SCI)
Shi J, He J, He J, et al. Multilevel modulation scheme using the overlapping of two light sources for visible light communication with mobile phone camera. Optics express, 2017, 25(14): 15905-15912. (SCI)
He J, Yin L, He J, et.al. NOMA-LFM Waveform and ANN Scheme for W-Band Integrated Sensing and Communication System. IEEE Photonics Technology Letters, 2024, 36(8): 559-562. (SCI)