AI-ENABLED SEMANTIC COMMUNICATION ARCHITECTURE FOR 6G NETWORKS: AN INTELLIGENT, CONTEXT-AWARE AND RESOURCE-EFFICIENT FRAMEWORK WITH HYBRID OPTIMIZATION

Communication architectures that can enable intelligent services, vast connectivity, ultra-low latency, and extremely heterogeneous applications are necessary due to the sixth-generation (6G) wireless networks' rapid expansion. Large amounts of redundant data are communicated with significant bandwidth, energy, and computational overhead since conventional communication systems are primarily concerned with correct bit-level transmission. By conveying task-relevant meaning rather than full raw data, semantic communication overcomes this constraint. This research presents an AI-Enabled Semantic Communication Architecture for 6G Networks that combines context-aware decision-making, intelligent semantic extraction, and hybrid optimization. Gray Wolf Optimization (GWO) optimizes transmission parameters and semantic reconstruction quality, whereas Particle Swarm Optimization (PSO) selects semantic features and allocates communication resources in the suggested framework. The optimization method is dynamically modified by a context-aware semantic controller based on network resources, channel conditions, and application needs. The framework is intended to maximize semantic similarity and communication efficiency while minimizing semantic distortion, latency, energy usage, and bandwidth utilization. Text-based semantic communication is used in an experimental assessment framework with different network loads and signal-to-noise ratios. For upcoming 6G applications, the suggested architecture offers a basis for resource-efficient, flexible, and intelligent semantic communication.