Maximizing Semantic Spectral Efficiency Using an Improved Frilled Lizard Optimization Algorithm

With the rapid development of 5G wireless communication technology and the continuous evolution of the digital society, traditional wireless communication networks are gradually approaching maturity, with their system capacity based on Shannon’s theorem nearing theoretical limits. Simultaneously, the burgeoning growth of artificial intelligence and the demand for intelligent communication in 6G networks have driven the resurgence of semantic communication technologies, making them one of the forefront research topics in contemporary communication technologies. This paper proposes a resource allocation strategy based on an improved Frilled Lizard Optimization Algorithm (FLO). By integrating immune network-based population initialization, memory mechanisms, reverse elite strategies, and fitness difference adjustment strategies, the FLO is significantly improved. The proposed strategy aims to maximize semantic spectrum efficiency in a text transmission semantic communication system. By jointly optimizing channel allocation and the number of transmitted semantic symbols, this strategy demonstrates superior performance in resource allocation, providing theoretical and practical support for the further development of semantic communication systems.