Joint Computation Offloading and Resource Allocation Optimization in Heterogeneous Networks With Mobile Edge Computing

Zhang, Jing; Xia, Weiwei; Yan, Feng; Shen, Lianfeng

doi:10.1109/access.2018.2819690

Cited by 217 publications

(114 citation statements)

References 31 publications

Supporting

Mentioning

114

Contrasting

Order By: Relevance

“…The baseline algorithm QA, the MUMTO algorithm, the proposed Neural-ICFMO algorithm, and the Neural-ICO algorithm are simulated when the number of IoT devices N is varied from 2 to 45. From Fig.4, it can be observed that the proposed Neural-ICFMO algorithm constantly outperforms the other three algorithms in terms of the weighted sum of delay and power consumption, which is the optimization objective as given in (9) for both the Neural-ICFMO and Neural-ICO algorithms. The performance of the QA and the MUMTO algorithms are quite similar, and the performance improvements of the proposed Neural-ICFMO algorithm as well as the Neural-ICO algorithm over these two algorithms increase with the increasing IoT device number.…”

Section: A Performance Vs Varying Iot Device Numbersmentioning

confidence: 97%

See 1 more Smart Citation

Multiuser Resource Control With Deep Reinforcement Learning in IoT Edge Computing

Lei

Xiong

et al. 2019

IEEE Internet Things J.

View full text Add to dashboard Cite

By leveraging the concept of mobile edge computing (MEC), massive amount of data generated by a large number of Internet of Things (IoT) devices could be offloaded to MEC server at the edge of wireless network for further computational intensive processing. However, due to the resource constraint of IoT devices and wireless network, both the communications and computation resources need to be allocated and scheduled efficiently for better system performance. In this paper, we propose a joint computation offloading and multi-user scheduling algorithm for IoT edge computing system to minimize the long-term average weighted sum of delay and power consumption under stochastic traffic arrival. We formulate the dynamic optimization problem as an infinite-horizon average-reward continuous-time Markov decision process (CTMDP) model. One critical challenge in solving this MDP problem for the multi-user resource control is the curse-of-dimensionality problem, where the state space of the MDP model and the computation complexity increase exponentially with the growing number of users or IoT devices. In order to overcome this challenge, we use the deep reinforcement learning (RL) techniques and propose a neural network architecture to approximate the value functions for the post-decision system states. The designed algorithm to solve the CTMDP problem supports semi-distributed auction-based implementation, where the IoT devices submit bids to the BS to make the resource control decisions centrally. Simulation results show that the proposed algorithm provides significant performance improvement over the baseline algorithms, and also outperforms the RL algorithms based on other neural network architectures.

show abstract

Section: A Performance Vs Varying Iot Device Numbersmentioning

confidence: 97%

“…From (9) and using Little's Law to derive the delay, we can derive the expression of c(s k , π(s k )) as below:…”

Section: E Reward Functionmentioning

confidence: 99%

Multiuser Resource Control With Deep Reinforcement Learning in IoT Edge Computing

Lei

Xiong

et al. 2019

IEEE Internet Things J.

View full text Add to dashboard Cite

show abstract

“…El Haber et al [24] jointed optimization of computational cost and devices energy for task offloading in multi-tier edge-clouds while respecting the devices' latency requirement. In [26], the authors proposed a jointly optimization policy in heterogeneous networks in order to minimize the cost of system respect to the energy consumption, computation and transmission cost. The authors of [37] investigated the task allocation problem in MEC environment with mmWave technology.…”

Section: Joint Computation Offloading and Resource Allocationmentioning

confidence: 99%

“…As wireless channel conditions of UEs have significant differences, the communication resource allocation and computational resource allocation can improve communication and computing efficiency, respectively. Some works addressed the joint allocation of radio and server resource allocation algorithms in various MEC systems [21][22][23][24][25][26]. The joint computation and communication resources can balance the delay cost and computation cost according to the UEs' servicing delay and energy cost.…”

mentioning

confidence: 99%

Collaborative Computation Offloading and Resource Allocation in Cache-Aided Hierarchical Edge-Cloud Systems

Lan

Wang

et al. 2019

Electronics

View full text Add to dashboard Cite

The hierarchical edge-cloud enabled paradigm has recently been proposed to provide abundant resources for 5G wireless networks. However, the computation and communication capabilities are heterogeneous which makes the potential advantages difficult to be fully explored. Besides, previous works on mobile edge computing (MEC) focused on server caching and offloading, ignoring the computational and caching gains brought by the proximity of user equipments (UEs). In this paper, we investigate the computation offloading in a three-tier cache-assisted hierarchical edge-cloud system. In this system, UEs cache tasks and can offload their workloads to edge servers or adjoining UEs by device-to-device (D2D) for collaborative processing. A cost minimization problem is proposed by the tradeoff between service delay and energy consumption. In this problem, the offloading decision, the computational resources and the offloading ratio are jointly optimized in each offloading mode. Then, we formulate this problem as a mixed-integer nonlinear optimization problem (MINLP) which is non-convex. To solve it, we propose a joint computation offloading and resource allocation optimization (JORA) scheme. Primarily, in this scheme, we decompose the original problem into three independent subproblems and analyze their convexity. After that, we transform them into solvable forms (e.g., convex optimization problem or linear optimization problem). Then, an iteration-based algorithm with the Lagrange multiplier method and a distributed joint optimization algorithm with the adoption of game theory are proposed to solve these problems. Finally, the simulation results show the performance of our proposed scheme compared with other existing benchmark schemes. and transmission techniques applied in the conventional cellular networks may not be efficient to meet UEs' requirements of high throughput and adequate computational power. To improve the delay performance and operational costs of services in fifth-generation (5G) wireless networks, future communication networks not only need to support seamless wireless access but also to offer the provisioning of computational offloading for UEs [3,4].To meet with such a challenge of limited computational capability of UEs, a typical paradigm of mobile edge computing (MEC) is proposed which combines wireless network service and cloud computing at the edge of the small cell networks (SCNs) [5]. In an MEC system, a huge number of heterogeneous ubiquitous and decentralized devices are enabled to communicate, potentially cooperate and perform computation offloading by uploading their computational tasks to the MEC server via access ratio networks [6,7]. UEs no longer need to offload all of their tasks (e.g., high-quality video streaming, mobile gaming, etc.) to the central and remote cloud. Thus their requirement can be satisfied at any time and anywhere. However, the limited computational capabilities of edge servers may not be sufficient when there exists competition for resources by a large number of device...

show abstract

“…With the advent of MEC technology, MEC caching is used in many applications with strict delay requirements. A joint optimization scheme was proposed in wireless cellular networks with mobile edge computing, taking into consideration computation offloading decision, physical spectrum resource allocation, MEC computation resource allocation, and content caching strategy [8]. Literature [9] proposed a storage resource allocation scheme of the MEC server taking each BS traffic load into consideration.…”

Section: Introductionmentioning

confidence: 99%

Hit Ratio Driven Mobile Edge Caching Scheme for Video on Demand Services

Chen

et al. 2019

2019 IEEE International Conference on Multimedia and Expo (ICME)

View full text Add to dashboard Cite

More and more scholars focus on mobile edge computing (MEC) technology, because the strong storage and computing capabilities of MEC servers can reduce the long transmission delay, bandwidth waste, energy consumption, and privacy leaks in the data transmission process. In this paper, we study the cache placement problem to determine how to cache videos and which videos to be cached in a mobile edge computing system. First, we derive the video request probability by taking into account video popularity, user preference and the characteristic of video representations. Second, based on the acquired request probability, we formulate a cache placement problem with the objective to maximize the cache hit ratio subject to the storage capacity constraints. Finally, in order to solve the formulated problem, we transform it into a grouping knapsack problem and develop a dynamic programming algorithm to obtain the optimal caching strategy. Simulation results show that the proposed algorithm can greatly improve the cache hit ratio.Index Terms-MEC, video-on-demand service, video cache, hit ratio, the grouping knapsack problem

show abstract

Joint Computation Offloading and Resource Allocation Optimization in Heterogeneous Networks With Mobile Edge Computing

Cited by 217 publications

References 31 publications

Multiuser Resource Control With Deep Reinforcement Learning in IoT Edge Computing

Multiuser Resource Control With Deep Reinforcement Learning in IoT Edge Computing

Collaborative Computation Offloading and Resource Allocation in Cache-Aided Hierarchical Edge-Cloud Systems

Hit Ratio Driven Mobile Edge Caching Scheme for Video on Demand Services

Contact Info

Product

Resources

About