This paper investigates the multihop task offloading problem in a vehicular network. The problem is formulated as an optimization problem with an objective to select a set of optimal service vehicles for the tasks of a task vehicle such that the maximum task service delay among all the tasks is minimized while the delay constraints of the tasks and the resource constraints of the service vehicles are guaranteed. To solve the problem, a service vehicle searching mechanism is proposed to search for candidate service vehicles for a task vehicle by exploring available multihop paths based on the connection statuses of the paths between the task vehicle and the candidate service vehicles, and the driving statuses of all vehicles in the network. Moreover, an adaptive service vehicle selection (AdSVS) algorithm is further proposed to select a set of optimal service vehicles among all the candidate service vehicles for offloading the tasks of the task vehicle. The proposed algorithm incorporates a bat algorithm and a particle swarm optimization (PSO) algorithm, which can improve the performance of task offloading. Simulation results show that the proposed Ad-SVS algorithm outperforms several benchmark algorithms in terms of the maximum task service delay.

With the development of 5G communication technologies and smart mobile devices, various computation-intensive and delay-sensitive tasks continue to increase. The combination of Mobile Edge Computing (MEC) and Ultra-Dense Networks (UDN) increases the network capacity and improves the computing capability of mobile devices, which effectively meets the transmission and computing demands of tasks. However, the ultra-dense deployment of network infrastructures causes energy shortage and channel interference, making it challenging to reduce the system cost. In this paper, we investigate the task offloading and resources scheduling problem in UDN with MEC. In order to minimize the total system cost including delay and energy consumption in the intensive deployment environment of edge servers and base stations (BSs) simultaneously, we design the strategy of task offloading, BS selection and resources scheduling of mobile devices. Because of the complex coupling of decision variables, the original problem is decomposed into two sub-problems. We propose Newton-IPM based Computing Resource Allocation (NICRA) algorithm and Genetic Algorithm based BS Selection and Resources Scheduling (GABSRS) algorithm to solve these two sub-problems, respectively. Then, we prove the number of iterations can be reduced effectively by the GABSRS algorithm while reaching the optimal solution through mathematical analysis. Through experiments analysis, the effectiveness of the GABSRS algorithm is validated.

With the development of Internet of Things (IoT) and network technologies, traditional networks cannot cope with the growth of network traffic and the changes in service requirements. The 5-th Generation Mobile Communication (5G) technology improves network transmission performance. In the communication network, 5G combines Software Defined Network (SDN) and Network Function Virtualization (NFV), through the deployment of end-to-end network slicing, to meet the challenge of differentiated service requirements in the complex environment. In the mobile network, users need to choose appropriate slices for access. Its performance is related to the quality of service and determines the efficiency of system resources utilization. We research the problem of slice re-access and slice resource scheduling caused by user mobility in 5G network slicing architecture and propose a slice access mechanism based on maximum throughput. A slice access selection algorithm based on genetic algorithm (GA) is proposed. Related simulations and comparative tests are carried out to prove the effectiveness and superiority of the algorithm.

Offloading tasks from computing intensive mobile devices (MDs) to neighboring edge servers in the form of incentive mechanism can effectively reduce latency and increase utility in mobile edge computing (MEC). In this paper, we design an auction mechanism for a MEC system, and the system consists of multiple MDs and one service provider (SP). As the auctioneer, SP receives the bidding information from MDs, making resource allocation strategies by optimizing social welfare. An exact algorithm to solve social welfare maximization problem and a perturbation-based randomized allocation algorithm to achieve (1 - α) optimal social welfare approximation rate are proposed. Furthermore, we prove that the truthful random auction mechanism can achieves the auction properties, including individual rationality, incentive compatibility, weakly budget balance and computational efficiency in theory. Finally, simulation results show the effectiveness of the auction mechanism.