基于关联成像和深度学习的光通信图像传输机理研究

Currently, the problem of channel noise interference, spectrum resource shortage and high bit error rate in optical communication limits the quality and efficiency of image transmission and reduces the reliability of optical communication channel..In this project, a theoretical model of optical communication image transmission based on ghost imaging and Deep Learning is established to reveal the image transmission mechanism in optical communication. We study the mechanism of image transmission based on temporal ghost imaging, the new imaging transmission model of joint source-channel coding by temporal ghost imaging and the new algorithm of image reconstruction based on deep learning and temporal ghost imaging. .The innovation method of line to point transmission in optical communication based on temporal ghost imaging by chaotic laser is obtained. The new image transmission model based on source compression encoding by deep learning, channel encoding and temporal ghost imaging is established. The new theory of high resolution image reconstruction based on the Double layer convolutional neural network and temporal ghost imaging is obtained..The research of this project can improve the resolution of the image and improve the reliability and security of the channel while reducing the amount of image transmission information. The research of the project is not only used in the field of wireless optical communication, but also can be extended to the field of cable optical communication.

目前光通信中的信道噪声干扰、频谱资源紧张和误码率较高等问题，限制了光通信图像传输质量和效率，降低了光通信信道的可靠性。.本项目建立基于关联成像和深度学习的光通信图像传输理论模型，揭示光通信图像编码传输机理。拟研究①基于时间关联成像的光通信图像传输机理；②联合信源信道编码的时间关联成像传输模型；③基于深度学习时间关联成像的图像重构方法。创新点①基于混沌光时间关联成像线到点的光通信传输新方法；②基于深度学习信源压缩编码、信道级联编码以及时间关联成像的图像传输新模型；③基于双层卷积神经网络时间关联成像的高分辨率图像重构新理论。.本项目的研究能够在降低图像传输信息量的同时提高传输图像的分辨率，进而提高信道的可靠性和安全性。项目的研究不仅为无线光通信领域所用，还可推广到光纤通信领域中。

目前光通信中的信道噪声干扰、频谱资源紧张和误码率较高等问题，限制了光通信图像传输质量和效率，降低了光通信信道的可靠性。本项目建立基于关联成像和深度学习的光通信图像传输理论模型，揭示光通信图像编码传输机理。研究基于时间关联成像的光通信图像传输机理；联合信源信道编码的时间关联成像传输模型；基于深度学习时间关联成像的图像重构方法。形成了基于混沌光时间关联成像线到点的光通信传输新方法；基于深度学习信源压缩编码、信道级联编码以及时间关联成像的图像传输模型；基于双重卷积神经网络时间关联成像的高分辨率图像重构新理论。本项目的研究能够在降低图像传输信息量的同时提高传输图像的分辨率，进而提高信道的可靠性和安全性。项目的研究不仅为无线光通信领域所用，还可推广到光纤通信领域中。