World Action Model Empowered Integrated Sensing and Communication: Opportunities and Challenges

Abstract

As a key enabling technology for 6G networks, Integrated sensing and communication (ISAC) is recognized as an AI-native core enabler for sixth-generation (6G) wireless networks. Conventional ISAC frameworks, however, rely on isolated channel modeling and static resource allocation. Moreover, they lack environment prediction and reasoning capabilities, severely limiting their effectiveness in complex time-varying wireless environments. World action models, characterized by multimodal unified representation, physical spatial-temporal forecasting, and general-purpose reasoning, present a new paradigm for intelligent ISAC systems. This paper provides a systematic overview of World action model-enabled ISAC. The survey first examines the fundamental limitations of traditional ISAC and the enabling value of World action models. A three-layer ISAC architecture, namely, comprising sensing input, model cognition, and collaborative execution is then proposed, along with two dominant implementation paradigms. Four core technical pillars are further identified, which consist of wireless World action model training, physical-layer joint optimization, network resource cooperative scheduling, and edge-lightweight inference. Six critical research challenges are identified: data scarcity, physical consistency, distributed synchronization, real-time inference, lack of rigorous theoretical performance bounds, and security/privacy concerns. Finally, future research directions are outlined from the perspectives of theory, model architecture, engineering deployment, and multi-technology convergence.