Consolidate Viability and Information Theories for Task-Oriented Communications: A Homeostasis Solution

Abstract

The next generation of cellular networks, 6G, is expected to offer a range of exciting applications and services, including holographic communication, machine-to-machine communication, and data sensing from millions of devices. There is an incremental exhaustion of the spectral resources. It is crucial to efficiently manage these resources through value-driven approaches that eliminate waste and continually enhance the communication process. These management principles align with the Task-Oriented Communications (TOC) philosophy. The aim is to allocate the minimum necessary communication resource according to the receiver's objective and continuously improve the communication process. However, it is currently unclear how to build knowledge of the receiver's goal and operate accordingly for efficient-resource utilization. Our management approach combines viability theory and transfer entropy to ensure that the actor remains within a viable space as per their goal and to gradually reduce the information exchange through knowledge accumulation. We discuss these theories in the context of TOC and examine their application in the plant process control case. Finally, we provide insights into future research directions from computational, performance, and protocol perspectives.