Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Learning-based Big Data Sharing Incentive in Mobile AIGC Networks

Jinbo Wen, Yang Zhang, Yulin Chen, Weifeng Zhong, Xumin Huang, Lei Liu, Dusit Niyato

TL;DR

This work tackles incentive design for big data sharing in mobile AIGC networks under information asymmetry. It introduces an AoI-based Quality of Data (QoD) metric that jointly accounts for data freshness and service latency, and formulates a contract-theoretic incentive mechanism to motivate mobile devices to contribute high-quality sensing data. To cope with the complex, dynamic environment, the authors implement a Proximal Policy Optimization (PPO) approach to learn the optimal contract items that maximize the edge server's utility while satisfying Individual Rationality and Incentive Compatibility constraints. Numerical results demonstrate that the PPO-based contract design achieves higher utility and adapts effectively to network states, highlighting its potential to enable reliable, low-latency mobile AIGC services in edge-enabled ecosystems.

Abstract

Rapid advancements in wireless communication have led to a dramatic upsurge in data volumes within mobile edge networks. These substantial data volumes offer opportunities for training Artificial Intelligence-Generated Content (AIGC) models to possess strong prediction and decision-making capabilities. AIGC represents an innovative approach that utilizes sophisticated generative AI algorithms to automatically generate diverse content based on user inputs. Leveraging mobile edge networks, mobile AIGC networks enable customized and real-time AIGC services for users by deploying AIGC models on edge devices. Nonetheless, several challenges hinder the provision of high-quality AIGC services, including issues related to the quality of sensing data for AIGC model training and the establishment of incentives for big data sharing from mobile devices to edge devices amidst information asymmetry. In this paper, we initially define a Quality of Data (QoD) metric based on the age of information to quantify the quality of sensing data. Subsequently, we propose a contract theoretic model aimed at motivating mobile devices for big data sharing. Furthermore, we employ a Proximal Policy Optimization (PPO) algorithm to determine the optimal contract. Numerical results demonstrate the efficacy and reliability of the proposed PPO-based contract model.

Learning-based Big Data Sharing Incentive in Mobile AIGC Networks

TL;DR

This work tackles incentive design for big data sharing in mobile AIGC networks under information asymmetry. It introduces an AoI-based Quality of Data (QoD) metric that jointly accounts for data freshness and service latency, and formulates a contract-theoretic incentive mechanism to motivate mobile devices to contribute high-quality sensing data. To cope with the complex, dynamic environment, the authors implement a Proximal Policy Optimization (PPO) approach to learn the optimal contract items that maximize the edge server's utility while satisfying Individual Rationality and Incentive Compatibility constraints. Numerical results demonstrate that the PPO-based contract design achieves higher utility and adapts effectively to network states, highlighting its potential to enable reliable, low-latency mobile AIGC services in edge-enabled ecosystems.

Abstract

Rapid advancements in wireless communication have led to a dramatic upsurge in data volumes within mobile edge networks. These substantial data volumes offer opportunities for training Artificial Intelligence-Generated Content (AIGC) models to possess strong prediction and decision-making capabilities. AIGC represents an innovative approach that utilizes sophisticated generative AI algorithms to automatically generate diverse content based on user inputs. Leveraging mobile edge networks, mobile AIGC networks enable customized and real-time AIGC services for users by deploying AIGC models on edge devices. Nonetheless, several challenges hinder the provision of high-quality AIGC services, including issues related to the quality of sensing data for AIGC model training and the establishment of incentives for big data sharing from mobile devices to edge devices amidst information asymmetry. In this paper, we initially define a Quality of Data (QoD) metric based on the age of information to quantify the quality of sensing data. Subsequently, we propose a contract theoretic model aimed at motivating mobile devices for big data sharing. Furthermore, we employ a Proximal Policy Optimization (PPO) algorithm to determine the optimal contract. Numerical results demonstrate the efficacy and reliability of the proposed PPO-based contract model.
Paper Structure (16 sections, 20 equations, 3 figures, 1 algorithm)

This paper contains 16 sections, 20 equations, 3 figures, 1 algorithm.

Table of Contents

  1. Introduction
  2. System Model
  3. Contract Modelling
  4. Quality of Data Metric
  5. Problem Formulation
  6. Utility of mobile devices
  7. Utility of the BS
  8. Contract formulation
  9. PPO-based Optimal Contract Design
  10. MDP Formulation
  11. State space
  12. Action space
  13. Immediate reward
  14. PPO Algorithm for Optimal Contract Design
  15. Numerical Results
  16. ...and 1 more sections

Figures (3)

  • Figure 1: An illustration of big data sharing in mobile AIGC networks.
  • Figure 2: Performance analysis of the proposed PPO-based algorithm for optimal contract design, where $\alpha = 0.75$ and the seed is set to $312$.
  • Figure 3: Utilities of the BS and mobile devices under different $\alpha$.

Theorems & Definitions (2)

  • Definition 1
  • Definition 2