Differential Evolution-based Service Caching and Offloading in Digital Twin-enabled ITSs

Due to the evolving requirements of onvehicle applications and the demand for higher quality of experience (QoE) for users, digital twin (DT)-enhanced intelligent transport systems (ITSs) are growing rapidly. This work aims to make effective offloading decisions that minimize delay and energy consumption, in DT-enabled ITS (DTITS), leveraging non-orthogonal multiple access (NOMA) and service caching for both popular and emerging demands. A $\mathbf{0 - 1}$ linear optimization problem is constructed, and the discretized differential evolution (DDEDTITS) algorithm is introduced to address popularityaware caching and offloading problems. Simulations reveal that DDE-DTITS is more effective than benchmark methods, significantly reducing delays and energy use.