Self-encoding algorithm based on convolutional neural network and MIMO visible light communication system thereof

Abstract

The invention discloses a self-encoding algorithm based on a convolutional neural network and a multiple-input-multiple-output MIMO visible light communication system thereof. At a transmitting end, X(X is a positive integer) bit data is encoded by using an encoding module of a trained auto-encoder, a data stream composed of an n * n (n is a positive integer) array is output, and the n * n arraydata stream drives an n * n LED array to send out an optical signal; and at a receiving end, after the receiving array receives the optical signals, decoding is carried out through a decoding module of the auto-encoder, and original data is restored. Wherein the auto-encoder based on the convolutional neural network is trained by adopting an information set, and X-bit data in the information set is sequentially input into the optical communication system to train the auto-encoder in the optical communication system. The MIMO visible light communication system is simple and feasible, is low inbit error rate, and has a wide application prospect.

Description

technical field [0001] The present invention relates to the technical field of visible light communication, in particular to a self-encoding algorithm based on a convolutional neural network and a MIMO visible light communication system thereof. Background technique [0002] Visible light communication technology uses light-emitting devices such as fluorescent lamps or light-emitting diodes to transmit information with high-frequency flickering signals that cannot be recognized by the naked eye. It only needs to be modified on the basis of existing lighting devices to realize visible light communication. Compared with traditional wireless communication technology, it has the advantages of fast information transmission speed, low construction cost and no electromagnetic interference, and has high security. Safe communication can be realized by blocking light, which can effectively prevent information leakage. Therefore, visible light communication technology is one of the nex...

AI Technical Summary

Problems solved by technology

[0003] However, there will be many problems in the specific implementation of visible light communication. Two of the more critical problems are path loss and mutual interference between multiple channels. These two problems have not been fully studied and solved yet.

The influence of path loss in optical communication can be considered. This loss increases with the increase of the distance between the transmitting end and the receiving end. This loss is also related to related factors such as channel quality. The complexity of the influencing factors leads to This kind of loss is difficult to estimate accurately, which makes it difficult to make up for this loss with precise mathematical transformation

Moreover, when multiple light-emitting devices communicate at the same time, due to factors such as the randomness of the emission direction of light particles, there will be serious interference problems between different optical signals, which will lead to a high bit error rate, which will eventually affect Normal progress of visible light communication

Images