Learning-Based Pricing for Privacy-Preserving Job Offloading in Mobile Edge Computing

Abstract

Mobile edge computing (MEC) is an emerging paradigm where users offload computationally intensive jobs to the Access Point (AP). Given that the AP's resources are shared by selfish mobile users, pricing is a useful tool for incentivising users to internalize the negative externality of delay they cause to other users. Different users have different negative valuations towards delay as some are more delay sensitive. To serve heterogeneous users, we propose a priority pricing scheme where users can get served first for a higher price. Our goal is to find the prices such that in decision making, users will choose the class and the offloading frequency that jointly maximize social welfare. With the assumption that the AP knows users' profit functions, we derive in semi-closed form the optimal prices. However in practice, the reporting of users's profit information incurs a large signalling overhead. Besides, in reality users might falsely report their private profit information. To overcome this, we further propose learning-based pricing mechanisms where no knowledge of individual user profit functions is required. At equilibrium, the optimal prices and average edge delays are learnt, and users have chosen the correct priority class and offload at the socially optimal frequency.

A review on the computation offloading approaches in mobile edge computing: A game‐theoreticperspective

“…In the following, we will review a typical example of calculating the total cost of offloading. According to the proposed method of, 4 the expected delay time spent per job (ie, the sum of execution time and local buffering time) in local computation is as follows:…”

“…Using RL (the exact model is Q-Learning), as a famous method in distributed decision making, helps players improving the coordination among their actions which leads to having better social and individual welfare. Li et al 80 proposed a kind of learning based pricing method to regulate the user's behavior of job computation offloading to the AP in a MEC limited computation power. In addition, they worked on a learning method for optimal pricing learned by AP.…”

“…Due to inherent resource limitations of these novel, popular, and ubiquitous mobile devices, to execute such resource‐intensive applications demanding high data rates and low computation delay, resource capabilities of the remote cloud has been changed with newer technologies such as fog computing (FC) and edge computing (EC) with lower limitations for such real‐time applications. These tree complementary abstraction layers offer distinctive benefits …”

“…In the following, we will review a typical example of calculating the total cost of offloading. According to the proposed method of, the expected delay time spent per job (ie, the sum of execution time and local buffering time) in local computation is as follows: where x k , μ m , λ a , , f m , μ a , L a , and B a and are user k 's offloading frequency, computing service rate of user k (jobs per second), and the Poisson process rate of job's arrival time, local computing probability, CPU‐cycle frequency of the mobile device, average CPU cycles needed by the computation job, the input data size (in Bytes), and processing density (in cycles/bit), respectively.…”

“…These tree complementary abstraction layers offer distinctive benefits. [1][2][3][4][5][6] mobile edge computing (MEC) is an emerging internet-based computing paradigm that is rapidly developing and adopting by different businesses aiming to gain the highest rank among related under progress technologies. 7 MEC is developed to locate near the mobile devices, usually in one hop, with small-scale servers located at the edge of the network to meet computation and delay requirements.…”