Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Latency-Aware Resource Allocation for Mobile Edge Generation and Computing via Deep Reinforcement Learning

Yinyu Wu, Xuhui Zhang, Jinke Ren, Huijun Xing, Yanyan Shen, Shuguang Cui

TL;DR

A new deep reinforcement learning-based algorithm is proposed to solve the joint communication, computation, and the AIGC resource allocation problem in an MEGC system and results demonstrate that the proposed algorithm can achieve lower latency than several baseline algorithms.

Abstract

Recently, the integration of mobile edge computing (MEC) and generative artificial intelligence (GAI) technology has given rise to a new area called mobile edge generation and computing (MEGC), which offers mobile users heterogeneous services such as task computing and content generation. In this letter, we investigate the joint communication, computation, and the AIGC resource allocation problem in an MEGC system. A latency minimization problem is first formulated to enhance the quality of service for mobile users. Due to the strong coupling of the optimization variables, we propose a new deep reinforcement learning-based algorithm to solve it efficiently. Numerical results demonstrate that the proposed algorithm can achieve lower latency than two baseline algorithms.

Latency-Aware Resource Allocation for Mobile Edge Generation and Computing via Deep Reinforcement Learning

TL;DR

A new deep reinforcement learning-based algorithm is proposed to solve the joint communication, computation, and the AIGC resource allocation problem in an MEGC system and results demonstrate that the proposed algorithm can achieve lower latency than several baseline algorithms.

Abstract

Recently, the integration of mobile edge computing (MEC) and generative artificial intelligence (GAI) technology has given rise to a new area called mobile edge generation and computing (MEGC), which offers mobile users heterogeneous services such as task computing and content generation. In this letter, we investigate the joint communication, computation, and the AIGC resource allocation problem in an MEGC system. A latency minimization problem is first formulated to enhance the quality of service for mobile users. Due to the strong coupling of the optimization variables, we propose a new deep reinforcement learning-based algorithm to solve it efficiently. Numerical results demonstrate that the proposed algorithm can achieve lower latency than two baseline algorithms.
Paper Structure (11 sections, 22 equations, 6 figures, 1 algorithm)

This paper contains 11 sections, 22 equations, 6 figures, 1 algorithm.

Table of Contents

  1. Introduction
  2. System Model and Problem Formulation
  3. Communication Model
  4. Latency Model
  5. Latency of MU $U_{\mathsf{comp}}$
  6. Latency of MU $U_{\mathsf{AIGC}}$
  7. Latency of MU $U_{\mathsf{VE}}$
  8. Problem Formulation
  9. DRL-based Resource Allocation Algorithm
  10. Numerical Results
  11. Conclusion

Figures (6)

  • Figure 1: MEGC system model.
  • Figure 2: Two stages of data offloading and result backhaul transmission in MEGC.
  • Figure 3: The reward value versus the training episodes under different learning rates.
  • Figure 4: Latency versus the training episodes: (1) the MU $U_{\mathsf{comp}}$, (2) the MU $U_{\mathsf{AIGC}}$, (3) the MU $U_{\mathsf{VE}}$.
  • Figure 5: The average latency vs. different algorithms.
  • ...and 1 more figures