Knowledge Sharing-enabled Semantic Rate Maximization for Multi-cell Task-oriented Hybrid Semantic-Bit Communication Networks

TL;DR

This paper proposes a knowledge sharing-enabled hybrid semantic-bit communication framework for multi-cell networks, optimizing semantic transmission rate by jointly managing knowledge sharing, semantic extraction, and SBS association to improve task performance.

Contribution

It introduces a novel mechanism combining semantic and bit communications with knowledge sharing, along with optimized algorithms for multi-cell scenarios.

Findings

Proposed algorithms outperform traditional methods in semantic transmission rate.

Knowledge sharing reduces mismatched knowledge issues effectively.

Simulation results validate the approach's superiority in diverse network conditions.

Abstract

In task-oriented semantic communications, the transmitters are designed to deliver task-related semantic information rather than every signal bit to receivers, which alleviates the spectrum pressure by reducing network traffic loads. Effective semantic communications depend on the perfect alignment of shared knowledge between transmitters and receivers, however, the alignment of knowledge cannot always be guaranteed in practice. To tackle this challenge, we propose a novel knowledge sharing-enabled task-oriented hybrid semantic and bit communications mechanism, where a mobile device (MD) can proactively share and upload the task-related mismatched knowledge to associated small base station (SBS). The traditional bit communications can be adopted as an aid to transmit the rest data related to unshared mismatched knowledge to guarantee the effective execution of target tasks. Considering the heterogeneous transceivers in multi-cell networks, target task demands, and channel conditions, an optimization problem is formulated to maximize the generalized effective semantic transmission rate of all MDs by jointly optimizing knowledge sharing, semantic extraction ratio, and SBS association, while satisfying the semantic accuracy requirements and delay tolerances of MD target tasks. The formulated mixed integer nonlinear programming problem is decomposed into multiple subproblems equivalently. An optimum algorithm is proposed and another efficient algorithm is further developed using hierarchical class partitioning and monotonic optimization. Simulation results demonstrate the validity and superior performance of proposed solutions.