Study On Chaotic Communication Optimization Using Neural Network

Due to the broad bandwidth,noise-like,and long-term unpredictable nature of chaotic signals,chaotic secure optical communication has a wide application prospect in the field of all-optical secure communication.External-cavity semiconductor lasers have important applications in chaotic optical communication because of their simple structure.However,lasers with the same parameters at the transmitter and receiver are required to achieve high-quality chaotic synchronization and it is difficult to find multiple lasers with identi-cal parameters.This limits the development of chaotic optical communication.In recent years,thanks to the rapid improvement of computer computing power,neural networks have been developed rapidly and have achieved breakthrough research results in several fields.There are few studies on using neural networks to realize chaotic communication,therefore the application of neural networks in chaotic communication needs further re-search.In this thesis,a chaotic communication scheme using the neural network to achieve chaos synchronization is introduced and numerically demonstrated.At the transmitter,an external-cavity semiconductor laser is used to generate a chaotic carrier.At the receiver,a neural network is trained using the original chaotic carrier and the modulated chaotic sig-nal so that the neural network can recover the original chaotic carrier from the modulated chaotic signal.To compensate dispersion in the optical fiber,a neural network is used in the receiver,which can reduce the cost of the system compared with the optical disper-sion compensation equipment.The synchronization performance and the communication performance of the system are studied by numerical simulation.The results show that chaotic synchronization can be effectively achieved using neural networks,and chaotic communication can be achieved using the synchronized chaotic carrier.The use of neu-ral networks simplifies the structure of the receiver in chaotic optical communication and makes it easier to apply to point-to-multipoint communication.Based on the chaotic synchronization of the neural network,a secure PON using neu-ral network and subcarrier modulation is introduced.In OLT,the chaotic carrier generated by the external-cavity semiconductor laser is used to encrypt the downstream message after subcarrier modulation.In ONU,a trained neural network is used for chaotic syn-chronization and decryption.We numerically demonstrate that the neural network used in the ONU can effectively achieve chaos synchronization.Thus,the system can realize upstream and downstream bidirectional transmission and resist the attack from the eaves-dropper.