Efficient Data Caching and Transmission for Power Line Communication (PLC) Systems

In power line communication (PLC) systems, popular data is frequently requested and delivered by PLC access units (PAUs). To improve the data delivery efficiency and reduce the transmission delay, edge caching is applied into PLC systems, and finite storage capacity is available for each PAU. In this context, this paper takes the transmission delay as an important performance metric in PLC systems. The probabilistic caching design is employed, where a higher caching probability means the data is more likely to be locally cached. In order to fully utilize the finite spectrum resource, the non-orthogonal multiple access (NOMA) transmission is adopted for each PAU to simultaneously serve the users connecting to it. With the proposed caching and transmission design, the partial transmission delays under cache hit and miss scenarios are derived, and the total transmission delay is then obtained. Under the constraints of limited cache size, the delay minimization problem is formulated. It is then solved after some effective transformations, and the optimal caching probability for each data is obtained. Simulation results show how the data size, cache size and backhaul rate affect the transmission delay, and also demonstrate the effectiveness of the proposed caching design in terms of reducing the delay.