Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Joint Communication and Computation Design for Probabilistic Semantic Communications

Zhouxiang Zhao, Zhaohui Yang, Mingzhe Chen, Zhaoyang Zhang, H. Vincent Poor

TL;DR

The paper tackles the problem of jointly optimizing transmission and computation resources in a multi-user probabilistic semantic communication (PSC) network, where data are semantically compressed using local and shared probability graphs. It introduces a three-stage algorithm combining MMSE-based receive beamforming, a rough AO-driven search over semantic compression ratios with a piecewise linear computation-load model, and a gradient-based refinement to achieve higher sum equivalents rates, while respecting per-user power budgets. The approach accounts for the non-convex and non-smooth nature of the problem by transforming the computation load into a solvable form and using alternating optimization plus gradient ascent, showing improved performance over non-semantic baselines and ZF/MSE variants. The results highlight a meaningful trade-off between computation and communication, demonstrating that semantic compression can substantially boost uplink efficiency with proper resource allocation, and setting a path for PSC-enabled deployments in future wireless networks.

Abstract

In this paper, the problem of joint transmission and computation resource allocation for a multi-user probabilistic semantic communication (PSC) network is investigated. In the considered model, users employ semantic information extraction techniques to compress their large-sized data before transmitting them to a multi-antenna base station (BS). Our model represents large-sized data through substantial knowledge graphs, utilizing shared probability graphs between the users and the BS for efficient semantic compression. The resource allocation problem is formulated as an optimization problem with the objective of maximizing the sum of equivalent rate of all users, considering total power budget and semantic resource limit constraints. The computation load considered in the PSC network is formulated as a non-smooth piecewise function with respect to the semantic compression ratio. To tackle this non-convex non-smooth optimization challenge, a three-stage algorithm is proposed where the solutions for the receive beamforming matrix of the BS, transmit power of each user, and semantic compression ratio of each user are obtained stage by stage. Numerical results validate the effectiveness of our proposed scheme.

A Joint Communication and Computation Design for Probabilistic Semantic Communications

TL;DR

The paper tackles the problem of jointly optimizing transmission and computation resources in a multi-user probabilistic semantic communication (PSC) network, where data are semantically compressed using local and shared probability graphs. It introduces a three-stage algorithm combining MMSE-based receive beamforming, a rough AO-driven search over semantic compression ratios with a piecewise linear computation-load model, and a gradient-based refinement to achieve higher sum equivalents rates, while respecting per-user power budgets. The approach accounts for the non-convex and non-smooth nature of the problem by transforming the computation load into a solvable form and using alternating optimization plus gradient ascent, showing improved performance over non-semantic baselines and ZF/MSE variants. The results highlight a meaningful trade-off between computation and communication, demonstrating that semantic compression can substantially boost uplink efficiency with proper resource allocation, and setting a path for PSC-enabled deployments in future wireless networks.

Abstract

In this paper, the problem of joint transmission and computation resource allocation for a multi-user probabilistic semantic communication (PSC) network is investigated. In the considered model, users employ semantic information extraction techniques to compress their large-sized data before transmitting them to a multi-antenna base station (BS). Our model represents large-sized data through substantial knowledge graphs, utilizing shared probability graphs between the users and the BS for efficient semantic compression. The resource allocation problem is formulated as an optimization problem with the objective of maximizing the sum of equivalent rate of all users, considering total power budget and semantic resource limit constraints. The computation load considered in the PSC network is formulated as a non-smooth piecewise function with respect to the semantic compression ratio. To tackle this non-convex non-smooth optimization challenge, a three-stage algorithm is proposed where the solutions for the receive beamforming matrix of the BS, transmit power of each user, and semantic compression ratio of each user are obtained stage by stage. Numerical results validate the effectiveness of our proposed scheme.
Paper Structure (15 sections, 2 theorems, 48 equations, 11 figures, 1 table, 3 algorithms)

This paper contains 15 sections, 2 theorems, 48 equations, 11 figures, 1 table, 3 algorithms.

Table of Contents

  1. Introduction
  2. System Model and Problem Formulation
  3. Semantic Communication Model
  4. Transmission Model
  5. Computation Model
  6. Problem Formulation
  7. Algorithm Design
  8. Stage 1: MMSE for Receive Beamforming Matrix
  9. Stage 2: Rough Search for Semantic Compression Ratio
  10. Stage 3: Refined Search for Semantic Compression Ratio
  11. Algorithm Analysis
  12. Simulation Results
  13. Conclusion
  14. Proof of Lemma \ref{['lemma1']}
  15. Proof of Theorem \ref{['theorem1']}

Key Result

Lemma 1

For any given transmit power of each user, i.e., $\mathbf{p}$, the optimal linear receive beamforming matrix $\mathbf{W}$ of the BS under MMSE strategy can be written as where $\mathbf{P}=\mathrm{diag}\{\mathbf{p}\}$ represents a diagonal matrix with $[\mathbf{P}]_{i,i}=[\mathbf{p}]_{i}$, and $\mathbf{I}_M$ is an identical matrix of size $M\times M$.

Figures (11)

  • Figure 1: Illustration of a knowledge graph.
  • Figure 2: An illustration of the considered PSC network.
  • Figure 3: Illustration of the probability graph considered in the PSC system.
  • Figure 4: The framework of considered PSC network.
  • Figure 5: Illustration of computation load versus semantic compression ratio $\rho$.
  • ...and 6 more figures

Theorems & Definitions (4)

  • Lemma 1
  • Proof
  • Theorem 1
  • Proof