Intelligent Degree Distribution of Rateless Codes in Industrial Scenarios

The main performance challenge of industrial automation communication networks is the end to end latency-bounded reliability from an industry 4.0 perspective. However, the existing degree distributions for the design of rateless codes have not been optimized for industrial automation scenarios, especially lack of adaptability, flexibility, modifiability, scalability and robustness for different industrial tasks. To solve this problem, this paper proposes an intelligent degree distribution (IDD) based on deep reinforcement learning. The simulation results show that the IDD can greatly reduce the latency and improve the reliability for matching the different industrial scenarios compared with the other methods.