Beamforming and Computing Capacity Allocation for ISAC-Assisted Secure Mobile Edge Computing

This work deals with the design of an integrated sensing and communication (ISAC)-assisted secure mobile edge computing (MEC) system, where latency-sensitive users offload partial tasks to edge servers near a full-duplex (FD) base station (BS) for computation. Unlike the conventional secure MEC systems that employ artificial noise to ensure security, the FD BS in our scheme transmits radar signals to jam and sense the eavesdropper (Eve), which can simultaneously achieves secure MEC and target sensing without consuming extra power. Considering that the BS can only acquire imperfect channel state information (CSI) in practical scenarios, we aim to minimize the maximum energy consumption among users under the constraints on latency, security as well as sensing performance. As the formulated optimization problem is non-convex due to the coupled variables, we decompose it into four sub-problems and alternately optimize these sub-problems to obtain the locally optimal solution, leading to a block coordinate descent (BCD)-based algorithm. Numerical results demonstrate that our proposed scheme outperforms benchmarks in terms of security performance and energy consumption.