Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Performance Limits of a Deep Learning-Enabled Text Semantic Communication under Interference

Tilahun M. Getu, Walid Saad, Georges Kaddoum, Mehdi Bennis

TL;DR

A principled probabilistic framework for SemCom is introduced, which shows that DeepSC produces semantically irrelevant sentences as the power of (multi-interferer) RFI gets very large, and derives DeepSC’s practical limits and a lower bound on its outage probability under multi-interferer RFI and proposes a (generic) lifelong DL-based IR2 SemCom system.

Abstract

Although deep learning (DL)-enabled semantic communication (SemCom) has emerged as a 6G enabler by minimizing irrelevant information transmission -- minimizing power usage, bandwidth consumption, and transmission delay, its benefits can be limited by radio frequency interference (RFI) that causes substantial semantic noise. Such semantic noise's impact can be alleviated using an interference-resistant and robust (IR$^2$) SemCom design, though no such design exists yet. To stimulate fundamental research on IR2 SemCom, the performance limits of a popular text SemCom system named DeepSC are studied in the presence of (multi-interferer) RFI. By introducing a principled probabilistic framework for SemCom, we show that DeepSC produces semantically irrelevant sentences as the power of (multi-interferer) RFI gets very large. We also derive DeepSC's practical limits and a lower bound on its outage probability under multi-interferer RFI, and propose a (generic) lifelong DL-based IR$^2$ SemCom system. We corroborate the derived limits with simulations and computer experiments, which also affirm the vulnerability of DeepSC to a wireless attack using RFI.

Performance Limits of a Deep Learning-Enabled Text Semantic Communication under Interference

TL;DR

A principled probabilistic framework for SemCom is introduced, which shows that DeepSC produces semantically irrelevant sentences as the power of (multi-interferer) RFI gets very large, and derives DeepSC’s practical limits and a lower bound on its outage probability under multi-interferer RFI and proposes a (generic) lifelong DL-based IR2 SemCom system.

Abstract

Although deep learning (DL)-enabled semantic communication (SemCom) has emerged as a 6G enabler by minimizing irrelevant information transmission -- minimizing power usage, bandwidth consumption, and transmission delay, its benefits can be limited by radio frequency interference (RFI) that causes substantial semantic noise. Such semantic noise's impact can be alleviated using an interference-resistant and robust (IR) SemCom design, though no such design exists yet. To stimulate fundamental research on IR2 SemCom, the performance limits of a popular text SemCom system named DeepSC are studied in the presence of (multi-interferer) RFI. By introducing a principled probabilistic framework for SemCom, we show that DeepSC produces semantically irrelevant sentences as the power of (multi-interferer) RFI gets very large. We also derive DeepSC's practical limits and a lower bound on its outage probability under multi-interferer RFI, and propose a (generic) lifelong DL-based IR SemCom system. We corroborate the derived limits with simulations and computer experiments, which also affirm the vulnerability of DeepSC to a wireless attack using RFI.
Paper Structure (26 sections, 7 theorems, 77 equations, 13 figures)

This paper contains 26 sections, 7 theorems, 77 equations, 13 figures.

Table of Contents

  1. Introduction
  2. System Description and Problem Formulation
  3. System Model
  4. Problem Formulation
  5. Problems on the Asymptotic Performance Analysis of DeepSC under RFI
  6. Problems on the Asymptotic Performance Analysis of DeepSC under MI RFI
  7. Problems on the Non-Asymptotic Performance Analysis of DeepSC under MI RFI
  8. Performance Limits of DeepSC
  9. Performance Limits of DeepSC Subjected to RFI
  10. Performance Limits of DeepSC Subjected to MI RFI
  11. The Practical Limits and Outage Probability of DeepSC
  12. The Practical Limits of DeepSC under MI RFI
  13. The Outage Probability of DeepSC under MI RFI
  14. Optimization Based On the Outage Probability of DeepSC
  15. Toward IR$^2$ 6G Wireless Systems
  16. ...and 11 more sections

Key Result

Theorem 1

Per the approximation in (simantic_similarity_function_2) and Sec. subsec: sysetm_model's settings, DeepSC manifests the following performance limits -- for a given $K$ -- under infinitesimally small RFI: $i) \lim_{\sigma^2 \to \infty} p(0) = 0$; $ii) \lim_{P_{\textnormal{max}}^s\to 0 } p(0) = 0$,

Figures (13)

  • Figure 1: A trained DeepSC under RFI from one or more single-antenna RFI emitters.
  • Figure 2: A (generic) lifelong DL-based IR$^2$ SemCom system.
  • Figure 3: $p(0)$ versus $\beta$ under infinitesimally small RFI, fixed $P_{\textnormal{max}}^s=5$ W, and varying $\sigma^2$: $N=10^7$.
  • Figure 4: $p(0)$ versus $\beta$ under infinitesimally small RFI, fixed $\sigma^2=100$ W, and varying $P_{\textnormal{max}}^s$: $N=10^7$.
  • Figure 5: $p(0)$ versus $\beta$ under an RFI, fixed $P_{\textnormal{max}}^s=10$ W, and varying $P_{\textnormal{min}}^i$: $N=10^7$.
  • ...and 8 more figures

Theorems & Definitions (17)

  • Definition 1
  • Theorem 1
  • Remark 1
  • Remark 2
  • Theorem 2
  • Remark 3
  • Theorem 3
  • Remark 4
  • Remark 5
  • Remark 6
  • ...and 7 more