Mobile edge computing has been an important supplement for traditional cloud architecture to offer low-delay services mobile users. However, it is in general impossible service providers overdeploy so much resources satisfy the rapidly increasing while diverse user demands. In this paper, we study a system consisting of platform, cloudlets joining system, and study, focus on profit-driven perspective such that platform purchases computation resource from resource-rich makes profit by...

Mobile edge computing (MEC), as a key enabling technology in 5G, provides distributed services for mobile devices to offload their computing-intensive tasks proximately deployed servers. The incentive mechanism is crucial issue with limited resources stimulate resource sharing among different participants improve utilization and also quality of user experience. Design efficient mechanisms has attracted lot attention, much work been carried out. In this article, we present comprehensive...

Mobile edge computing has been an important paradigm for providing delay-sensitive and computation-intensive services to mobile users. In this paper, we study the problem of joint optimization task assignment energy management in a mobile-server-assisted network, where servers can provide assisted offloading on behalf fixed at network edge. The design objective is minimize system delay. As far as know, our paper presents first work that improves quality service whole from long-term aspect by...

Multi-access edge computing (MEC) has been an important and promising paradigm for offering services to mobile users with computation-intensive latency-critical tasks. In this paper, we study a D2D collaboration based MEC system, where the service platform purchases resources from resource-rich collaborative devices when task arrival rate exceeds platform's capability providing satisfactory QoS. The design objective is maximize profit while maximally satisfying delay requirements of We...

This paper presents a novel routing planning method based on multi-objective optimization to tackle the problem in computing power networks. The proposed aims improve performance and efficiency of by considering multiple objectives. In this study, we first model network formulate as problem. To address problem, introduce non-dominated sorting genetic algorithm incorporating ratio parameter adjustment strategy reinforcement learning. Extensive simulations are conducted evaluate algorithm....

An appealing technology that merges computing and network resources to offer convergent services is the power network, which has attracted a lot of interest. In this paper, we study task routing problem in heterogeneous minimize loss value system. We first formulate as mixed integer nonlinear programming (MINP). To solve problem, propose Value aware Ant Colony algorithm(VACO) deduced complexity algorithm. The proposed algorithm VACO searches for optimal traversal directed graph by multiple...

Mobile crowdsensing is a novel paradigm by exploiting mobility, sensing, computation, and communication capability of smart devices. In this paper, we study data offloading problem for mobile in opportunistic social networks. scenario, users can upload sensing directly via cellular networks using various plans. A user also resort to another forwarding that short-range communications (when they encounter). To minimize total uploading cost while meeting given deadlines, plan assignment...

Mobile crowdsensing is a sensing paradigm such that mobile users need to move task locations perform tasks. In this paper, we focus on studying the assignment problem of eco-friendly which aims minimize carbon emissions while meeting various resource limits including deadlines and transportation constraints. We first describe system model formulate problem. Then divide into two subproblems selection best type user-task matching. matching as unbalanced minimum-cost bipartite matching,...

To achieve the dual goal of high economic revenue and low carbon emissions, optimal operation micro grid in electricity market trading markets should be investigated. A collaborative planning model electricity-hydrogen integrated energy system with PV hydrogen storage equipment is established this paper. In model, a typical daily output curve derived based on Monte Carlo scenario generation method minimum operating cost day maximum annual net profit are calculated. Additionally, proposed...

Abstract Considering the uncertainty of probability distribution electricity price in multi-energy markets and taking maximum expected profits as optimization object, this paper establishes optimal power purchasing portfolio model based on distributional robust CVaR. An problem proportion real-time market, day-ahead mid-long term contract market is transformed into a semi-definite programming problem. The analysis case study shows efficiency proposed model, which paves new way for companies...

Under the trend of multiple big data technologies, development multi-node collaborative parallel computing has made great advances. Limited by security and permissions, federated learning become current mainstream method to solve difficult situation. However, for huge number nodes based on learning, there are many redundancies in client nodes. Therefore, resource utilization model training efficiency can be negatively affected. In this paper, order improve scenarios such as we study a node...

Mobile edge computing (MEC) has been a promising architecture for providing delay-sensitive services to mobile users. In the provisioning of MEC services, one big issue is potential mismatching between dynamic task arrivals and resources allocated different types applications at servers, which can cause degraded system performance. this paper, we study joint optimization application deployment assignment time scales in constituent multiple cloudlet service providers. The design objective...

The modular multilevel converter (MMC) has been widely used in the high-voltage direct current transmission system. To ensure safe and reliable operation of system, it is necessary to conduct reliability evaluation MMC. In previous studies, dependent relationship among submodules (SMs) tends be neglected, which may result error modeling for fill this gap, combined with conditional probability method, model MMC constructed considering dependency SMs. First, effect levels flowing through SMs...

Computing power network (CPN), which integrates heterogeneous computing resources and communication network, can tackle the challenges brought by pervasiveness of mobile Internet Things applications. In this paper, we study optimization task scheduling in a CPN considering unbalancing between distribution resource cost. The design objective is to maximize system profit while satisfying tasks' delay requirements. We formulate problem as an integer programming problem. To address NP-hard...

Scientific (e.g. genetic) data and relative processing tools are usually located at different Web sites; scientists have been strongly expecting a universal solution to compare, integrate, analyze these heterogeneous large size with from distant sites for years. This paper presents pipelined services workflow framework: SPL (service pipeline logic) that can reduce frequent interaction between mitigate network traffic incurred by huge scientific transmission during the course of...

Abstract Mobile edge computing is an important paradigm for delay-sensitive and computation-intensive tasks, which are growing explosively. In this paper, we study the joint optimization of task assignment energy management in mobile servers assisted network order to minimize system delay. Considering constraints capacity heterogeneous servers, paper introduces protection threshold dynamically determine energy-critical prevents them from participating processing, so as effectively extend...