Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Minimizing End-to-End Latency for Joint Source-Channel Coding Systems

Kaiyi Chi, Qianqian Yang, Yuanchao Shu, Zhaohui Yang, Zhiguo Shi

TL;DR

The paper addresses minimizing the end-to-end latency in uplink DL-based joint source-channel coding by modeling encoding, transmission, and decoding times and enforcing SSIM constraints. It proves the resulting optimization is NP-hard and develops a problem-transformation approach that yields a closed-form optimal allocation of compression ratios, truncation thresholds, and edge computing resources, along with a low-complexity heuristic. The proposed algorithms achieve significant latency reductions, with the heuristic closely matching the optimal performance while requiring far less computation. This work provides practical resource-management strategies to deploy DL-based JSCC in latency-sensitive uplink scenarios, bridging theory and real-world constraints. All mathematical expressions are presented with precise notation to support replication and integration into systems analyses.

Abstract

While existing studies have highlighted the advantages of deep learning (DL)-based joint source-channel coding (JSCC) schemes in enhancing transmission efficiency, they often overlook the crucial aspect of resource management during the deployment phase. In this paper, we propose an approach to minimize the transmission latency in an uplink JSCC-based system. We first analyze the correlation between end-to-end latency and task performance, based on which the end-to-end delay model for each device is established. Then, we formulate a non-convex optimization problem aiming at minimizing the maximum end-to-end latency across all devices, which is proved to be NP-hard. We then transform the original problem into a more tractable one, from which we derive the closed form solution on the optimal compression ratio, truncation threshold selection policy, and resource allocation strategy. We further introduce a heuristic algorithm with low complexity, leveraging insights from the structure of the optimal solution. Simulation results demonstrate that both the proposed optimal algorithm and the heuristic algorithm significantly reduce end-to-end latency. Notably, the proposed heuristic algorithm achieves nearly the same performance to the optimal solution but with considerably lower computational complexity.

Minimizing End-to-End Latency for Joint Source-Channel Coding Systems

TL;DR

The paper addresses minimizing the end-to-end latency in uplink DL-based joint source-channel coding by modeling encoding, transmission, and decoding times and enforcing SSIM constraints. It proves the resulting optimization is NP-hard and develops a problem-transformation approach that yields a closed-form optimal allocation of compression ratios, truncation thresholds, and edge computing resources, along with a low-complexity heuristic. The proposed algorithms achieve significant latency reductions, with the heuristic closely matching the optimal performance while requiring far less computation. This work provides practical resource-management strategies to deploy DL-based JSCC in latency-sensitive uplink scenarios, bridging theory and real-world constraints. All mathematical expressions are presented with precise notation to support replication and integration into systems analyses.

Abstract

While existing studies have highlighted the advantages of deep learning (DL)-based joint source-channel coding (JSCC) schemes in enhancing transmission efficiency, they often overlook the crucial aspect of resource management during the deployment phase. In this paper, we propose an approach to minimize the transmission latency in an uplink JSCC-based system. We first analyze the correlation between end-to-end latency and task performance, based on which the end-to-end delay model for each device is established. Then, we formulate a non-convex optimization problem aiming at minimizing the maximum end-to-end latency across all devices, which is proved to be NP-hard. We then transform the original problem into a more tractable one, from which we derive the closed form solution on the optimal compression ratio, truncation threshold selection policy, and resource allocation strategy. We further introduce a heuristic algorithm with low complexity, leveraging insights from the structure of the optimal solution. Simulation results demonstrate that both the proposed optimal algorithm and the heuristic algorithm significantly reduce end-to-end latency. Notably, the proposed heuristic algorithm achieves nearly the same performance to the optimal solution but with considerably lower computational complexity.
Paper Structure (18 sections, 5 theorems, 34 equations, 5 figures, 2 algorithms)

This paper contains 18 sections, 5 theorems, 34 equations, 5 figures, 2 algorithms.

Table of Contents

  1. Introduction
  2. System Model and Problem Formulation
  3. System Model
  4. Encoding at Local Device
  5. Transmission Model
  6. Decoding at the BS
  7. Performance Metrics
  8. Problem Formulation
  9. Proposed Solution
  10. Problem Transformation
  11. Optimal Solution
  12. The Proposed Low-Complexity Algorithm
  13. Simulation Results
  14. Conclusion
  15. Proof of Lemma 1
  16. ...and 3 more sections

Key Result

Lemma 1

Solution to $\mathcal{P}2$ with $T<T^*$ is infeasible while the solution with $T>T^*$ is feasible.

Figures (5)

  • Figure 1: The considered JSCC system model.
  • Figure 2: Average SSIM of the reconstructed images vs. SNR under different compression ratios on ImageNet dataset.
  • Figure 3: Delay vs. number of devices.
  • Figure 4: Delay vs. edge computation resource. (The percentage means the percentage of a CPU core, for example, 300% means 3 CPU cores.)
  • Figure 5: Computation resource allocation vs. local computation resource of device 1. (200% in total, which means two CPU cores.)

Theorems & Definitions (6)

  • Lemma 1
  • Theorem 1
  • Theorem 2
  • Theorem 3
  • Remark 1
  • Lemma 2