PPDO: a privacy-preservation-aware delay optimization task-offloading algorithm for collaborative edge computing

Abstract

Although collaborative edge computing (CEC) systems are beneficial in enhancing the performance of mobile edge computing (MEC), the issue of user privacy leakage becomes prominent during task offloading. To address this issue, we design a privacy-preservation-aware delay optimization task-offloading algorithm (PPDO) in a CEC system. By considering location and usage pattern privacy protection, we establish a privacy task model to interfere with the edge server and ensure user privacy. To address the extra delay arising from privacy protection, we subsequently leverage a Markov decision processing (MDP) policy-iteration-based algorithm to minimize delays without compromising privacy. To simultaneously accelerate the MDP operation, we develop an extension that improves the PPDO by optimizing the action set. Finally, a comprehensive simulation was conducted using the edge user allocation (EUA) dataset. The results demonstrated that PPDO achieves an optimal trade-off between privacy protection and delay with a minimum delay compared with existing algorithms. Moreover, we examined the advantages and disadvantages of improving PPDO.

1. Introduction

With the rapid growth of the scale of mobile networks, intelligence-based mobile devices have become widely used in numerous fields. Owing to the many requirements of resource-intensive applications (Ouyang et al., 2019), there is a significant increase in data volume and demand for less delay (Wang F et al., 2018), especially for applications that require a high quality of service (QoS) and are vulnerable to delays (Taleb et al., 2017; Chen et al., 2020), such as mobile gaming (Yang et al., 2020), virtual reality (VR) (Lin et al., 2021), image video processing (Mach and Becvar, 2017), and speech recognition (Yousaf et al., 2018). However, the resources of mobile cloud computing (MCC) are usually located far from the mobile devices, with high latency and energy consumption for data transmission owing to network congestion and speed limitations (Gao et al., 2023a). Fortunately, these issues can be addressed by the emerging technology of mobile edge computing (MEC) (Abbas et al., 2018). By situating servers at the edge of the network, edge computing can minimize latency (Dong et al., 2023).

Collaborative edge computing (CEC) systems have attracted significant attention owing to the constrained edge server (ES) resources of MEC as compared to MCC (Lee et al., 2021). CEC, characterized by the distribution of computational tasks among distributed edge nodes (Sahni et al., 2019), ensures appropriate collaboration on ESs when faced with a risk of privacy leakage owing to a reduced communication overhead in selecting a closer ES. Moreover, when the network conditions are poor or the ES processing queues are congested, certain tasks can be processed by mobile devices instead of being offloaded to the ES. By appropriately offloading tasks, CEC can greatly improve the performance of MEC. Several studies have addressed the costs associated with task offloading in edge computing, generally including execution delay and energy consumption. Mao YY et al. (2017) proposed an algorithm based on Lyapunov optimization to solve the cost optimization problem in edge computing systems with limited computational resources. Similarly, Cao et al. (2021) adopted the Lyapunov optimization framework to minimize user QoS losses stemming from network delay and low frame rates under cost constraints. Chu et al. (2023) proposed a distributed online version of the mechanism to maximize user QoS in MEC by jointly optimizing service caching, resource allocation, and task offloading decisions. Qin et al. (2021) devised a threshold-based distributed task offloading algorithm that allows MEC users to update their thresholds based on their cost functions. Ren et al. (2019) considered collaboration between cloud and edge computing, with a joint communication resource and computation resource allocation problem formulated to minimize the latency of mobile devices. Wang JD et al. (2021) designed a resource allocation scheme that rationally assigns computational and network resources under changeable edge computing conditions. However, most of these studies focused on cost optimization, ignoring the importance of privacy protection during offloading.

When ESs are located closer to mobile devices, the QoS of users can benefit from the advantages of edge computing. However, edge computing raises the severe issue of privacy leakage.