Energy-Efficient Resource Allocation for Adaptive Semantic-Bit Communications in Emergency Networks

In this paper, we propose an energy-efficient resource allocation scheme for the multi-user multi-channel adaptive semantic-bit communication in emergency networks with the text transmission emergency task. Specifically, we first analyze the semantic energy efficiency for both semantic and bit communication modes. Then, we formulate a network-level semantic energy efficiency maximization problem by jointly optimizing the communication mode decision, channel selection, semantic coding length, power control, and computation resource allocation. To solve the formulated problem, we propose a fractional programming based hierarchical coalitional game (FP-HCG) algorithm. To be specific, we first use the Dinkelbach method to transform the fractional objective function into a parametric form such that the optimal solution of the original problem can be obtained by updating the parameter through iterations. In each iteration, we solve the transformed problem by formulating a two-layer coalitional game, where the coalitional game is adopted to determine the channel selection in the upper layer, and the genetic algorithm is adopted to determine the communication mode decision and resource allocation of each coalition in the lower layer. Simulation results demonstrate that the proposed scheme can efficiently improve the network energy efficiency for emergency communications.