Toward Goal-Oriented Semantic Communications: New Metrics, Framework, and Open Challenges

Abstract

Since Shannon's pioneering masterpiece which established the foundation of modern information theory, the design target of communications has long been promising bit-level message reconstruction and achieving Shannon capacity. However, this focus neglects the semantics and effectiveness aspects of information. Recently, the development of wireless technologies and the spurt of deep learning (DL) techniques allow us to reclaim the meaning/usefulness aspect in the design of 6G communication paradigms, where goal-orient-ed communication is becoming a trend. Age of information (AoI), a well-known metric that captures the significance of information by recording the time elapsed from the generation time slot, has been extended to various variants, such as value of information (Vol), urgency of information (Uol), age of incorrect information (Aoll), and so on. While each of them proposes novel ways to measure the semantics/effectiveness aspect of information, there is not yet a unified framework encompassing all of them. To this end, we propose a novel tensor-based approach, the goal-ori-ented tensor (GoT), to unify them, which allows more flexible and fine-grained goal characterizations. Following the proposed GoT, we architect a holistic goal-oriented semantic communication framework, in which semantics perception, dissemination, and control-plane decision making collaboratively work toward realizing specific goals. Finally, we outline several open challenges to fulfill the vision of the GoT framework.

Topics & Concepts