Fair Computation Efficiency Scheduling in NOMA-Aided Mobile Edge Computing

Abstract

Splitting the inference model between device, edge server, and cloud can improve the performance of EI greatly. Additionally, the non-orthogonal multiple access (NOMA), which is the key supporting technologies of B5G/6G, can achieve massive connections and high spectrum efficiency. Motivated by the benefits of NOMA, integrating NOMA with model split in MEC to reduce the inference latency further becomes attractive. However, the NOMA based communication during split inference has not been properly considered in previous works. Therefore, in this paper, we integrate the NOMA into split inference in MEC, and propose the effective communication and computing resource allocation algorithm to accelerate the model inference at edge. Specifically, when the mobile user has a large model inference task needed to be calculated in the NOMA-based MEC, it will take the energy consumption of both device and edge server and the inference latency into account to find the optimal model split strategy, subchannel allocation strategy (uplink and downlink), and transmission power allocation strategy (uplink and downlink). Since the minimum inference delay and energy consumption cannot be satisfied simultaneously, and the variables of subchannel allocation and model split are discrete, the gradient descent (GD) algorithm is adopted to find the optimal tradeoff between them. Moreover, the loop iteration GD approach (Li-GD) is proposed to reduce the complexity of GD algorithm that caused by the parameter discrete. Additionally, the properties of the proposed algorithm are also investigated, which demonstrate the effectiveness of the proposed algorithms.

A fairness‐aware task offloading method in edge‐enabled IIoT with multi‐constraints using AGE‐MOEA and weighted MMF

“…The number of multi-constrained task sets [2][3][4][5][6][7][8][9][10][11][12][13][14] The number of edge servers [14][15][16] The virtual machine capacity of edge servers [15][16][17][18][19][20][21][22][23][24][25][26][27][28][29][30] The CPU frequency of edge servers [2000-2500] MHz…”

“…For improving resource allocation efficiency and promoting multiple resource sharing, Lin et al 27 used dominant resource fairness and MMF to allocate available and remaining resources, respectively. To cope with the unfairness among mobile devices due to the near‐far effect, Huang et al 28 adopted a new objective function for fair computational resource scheduling in MEC scenarios and verified the superiority of the proposed scheme in large‐scale and time‐sensitive tasks. For fairness‐aware resource management in the MEC network with multi‐user and multiserver, Guo et al 29 adopted the MMF resource management algorithm while considering the supply and demand of resources in task allocation, thereby greatly reducing the total time consumption.…”

FairHealth: Long-Term Proportional Fairness-Driven 5G Edge Healthcare in Internet of Medical Things

“…The computation of the MEC server and local IoMT devices is 30 and 5 GHz, respectively. The healthcare data size of five patients is [5,10,15,20,25] Mbit, respectively. The required CPU cycles of each data packet are 500 cycles/bit.…”

“…Despite this, most existing MEC task offloading algorithms [5], [6] consider altruistic nodes just struggle to achieve a global optimum, which could not be applied in scenarios where multiple parties participate. There are also some works [8]- [10] focusing on general fair resource allocation in MEC but do not consider priority-aware and deadline-sensitive service characteristics in healthcare sectors. Consequently, we propose a long-term proportional fairness-driven edge healthcare scheme, referred to as FairHealth.…”

A Max-Min Task Offloading Algorithm for Mobile Edge Computing Using Non-Orthogonal Multiple Access

“…Therefore, different from the existing literature, we consider the problem of maximizing the minimum number of bits (of the high-complexity tasks) offloaded by users to an MEC server using NOMA by constraining the total energy, transmission power and the offloading delay within given limits. In this context, Huang et al studied a max-min computation efficiency problem in NOMA-MEC system [14], and a similar problem in a millimeter wave settings was explored in [15]. In order to improve the physicallayer security of offloaded data in a NOMA-MEC system, a max-min antieavesdropping poroblem was presented in [16].…”