A Deep Learning-Assisted Multi-Relay DCSK Communication System

Area of Science:

• Wireless Communication
• Deep Learning
• Signal Processing

Background:

• Chaos-based communication systems offer potential but are sensitive to channel quality.
• Existing methods for channel quality assessment require explicit Channel State Information (CSI).
• Cooperative communication systems can enhance reliability but require efficient relay coordination.

Purpose of the Study:

• To propose a novel Multi-Relay Deep Learning-assisted Differential Chaos Shift Keying (MR-DL-DCSK) system.
• To develop a method for joint channel quality assessment and symbol demodulation without explicit CSI.
• To implement a channel quality-aware relay coordination strategy for improved transmission reliability.

Main Methods:

• A Deep Neural Network (DNN) classifier is utilized at the receiver for joint channel quality assessment and symbol demodulation.
• A channel quality-aware relay coordination strategy is proposed, where relays align decoded bits based on channel quality.
• The destination selects the signal with the highest channel quality probability for final demodulation.

Main Results:

• The proposed MR-DL-DCSK system achieves superior bit error rate (BER) performance compared to existing systems.
• The system demonstrates reliable demodulation without requiring explicit CSI.
• The channel quality-aware relay coordination ensures prioritization of signals from the most reliable links.

Conclusions:

• The novel MR-DL-DCSK system effectively enhances chaos-based cooperative communication.
• The DNN-based approach provides reliable performance and excellent generalization capabilities in various wireless environments, including vehicle-to-vehicle (V2V) channels.
• This work validates the practical applicability of advanced deep learning techniques in improving wireless communication reliability.