| The continuous progress of information technology and the booming development of mobile Internet of Things have brought about an explosive growth of mobile data traffic.Among them,the traffic of maritime network applications has become an important part of it,which is also the trend of the development of space-air network,intelligent ocean.Explore the deployment of high-speed,high-reliability,low-cost new maritime communication network is extremely important for the development of future maritime related industries and marine safety and security.This puts forward higher requirements for the information service capacity of maritime communication network in terms of all-round coverage,data transmission quality,transmission timeliness,etc.The existing mobile communication network and satellite network with multi-dimensional coverage,to provide a good basis for the maritime wireless network in distributed space-air coverage,which includes the existing cellular network,shore-based,interstellar links and other access to the heterogeneous network.Space-air coverage of the maritime network,the era of Internet of everything is inseparable from the maritime edge network,in order to achieve high speed,low latency and other diversified and differentiated user needs,the computation is delivered to the maritime edge network,reducing the dependence on the core network and significantly improving the efficiency of the service.On this basis,the study of edge computing-based network service optimization is carried out for the effective combination of communication,computing and caching for multi-layer heterogeneous networks in the sea area,providing technical and theoretical support for improving the performance and future development of maritime communication systems.The main contributions of this thesis are as follows.(1)Ship video data scheduling algorithm in maritime heterogeneous networkingCentering on the task of marine/maritime information interaction,such as video data collected by shipboard voyage data recorders,video data can be efficiently transmitted through the maritime information physical system at a lower cost compared to expensive low-band satellite communications.Aiming at the maritime CPSs,a framework for near-shore wireless transmission network is constructed to jointly consider the characteristics of intermittent network connections,and an online and offline scheduling scheme for ship video packet transmission is proposed with the goal of maximizing the network throughput as much as possible.First,the scheduling problem is transformed into a mathematics-based job-machine problem,and then,the initial intermittent connectivity problem of video packet communication transmission is converted into a single-machine scheduling problem using a time-capacity mapping approach,and three algorithms,namely,a two-stage algorithm for offline timecapacity mapping,an online admission algorithm,and an exponential capacity algorithm,are proposed to maximize the total weight of packet transmission.The effectiveness as well as the superiority of the proposed scheme is verified by simulation based on the data extracted from the dedicated ship navigation software.(2)Multi-vessel multitasking offloading algorithm for edge computing in maritime heterogeneous networkingWith the increase of maritime Io T devices,it brings an increase of ship users’ demands(such as maritime services and entertainment),which greatly exceeds the supply of the network.In order to meet the requirements of high bandwidth and low latency for the high-speed development of mobile networks and to reduce the network load,computational offloading techniques are introduced into the maritime mobile edge network,combining communication and computation to study the multi-vessel multi-computational task offloading problem.First,determine whether computation-intensive tasks should be offloaded to edge cloud servers.Second,select the appropriate edge server to run the computational tasks.To reduce time delay and device energy consumption,a multi-vessel computation offloading algorithm based on an improved Hungarian algorithm for maritime mobile edge computing networks is proposed based on a channel selection scheme,and the proposed scheme is validated to efficiently and orderly utilize edge cloud resources and guarantee the performance of maritime network services.(3)Cloud-edge-end offloading interaction of ship computing tasks in maritime heterogeneous networkingWith the large number of maritime Io T devices accessing and exploring maritime networks and the growing maritime activities,it creates a huge challenge for the processing cost and energy efficiency of maritime data.To solve the problem,combining communication and computation,considering the demand of Quality of Service(Qo S)of different maritime applications in complex maritime environments,and offloading optimization of mobile devices by means of computational offloading,a two-stage joint optimal offloading algorithm is proposed to optimize the allocation of computational and communication resources under energy-constrained and delay-sensitive conditions.In the first stage,offshore users decide whether to offload computational tasks considering their needs and the ocean environment.Then,considering the dynamic tradeoffs of latency and energy consumption,the channels and power are rationally allocated to optimize the offloading strategy in the second phase in coordination with the central cloud server.Finally,the simulation results show the effectiveness of the proposed algorithm to achieve an efficient allocation of computing resources to meet the diverse business needs of the users of the maritime communication network.(4)Caching strategy optimization for files of ship users in maritime heterogeneous networkingFor the port navigation and near-shore navigation scenarios of ship users,the ship users are considered as device users,and the efficiency is further improved by combining communication and caching through reasonable deployment of caching strategy on mobile devices in combination with the traditional base station transmission method.The mobile-aware caching strategy is divided into two problems to solve.The first problem is to transform the cache placement problem that solves the user’s latency minimization into a decision problem by proposing a low-complexity algorithm that approximates the optimal solution,and proving the method using the properties of submodular functions.The second problem involves external restriction parameters such as cache file type,cache upper limit,and deadline.Simulations reveal that there is a bottleneck in the performance improvement of the whole system as the external parameters change.A suitable formulation of these parameters can put the system in the most efficient range of inputs and outputs,based on the principle of marginal effects,and using Bayesian optimization to address the selection of these parameters and further enable the performance of the optimization method.Finally,it is validated by simulation with real data,which in turn improves cache resource utilization and network quality of service.To sum up,this thesis is based on the two research points of developing a more efficient integrated maritime communication network and studying the corresponding efficient maritime communication network resource optimization allocation,fully considering the complexity of cross-network and giant system bearing multiple users and multiple services,exploring the key technology research of heterogeneous networking and resource optimization scheduling in sea area,which can provide networked information services for maritime hydro-meteorology,resource survey,disaster mitigation,ship data transmission,intelligent transportation systems and other applications,especially the research of flexible resource scheduling support for sudden application mode at sea,so as to optimize the system service access control and networking service performance. |
PDF Full Text Request