Research On Connectivity Maximization Of Industrial Internet-of-Things With Multiple Services

In recent years,with the development of Information and Communications Technology(ICT),the Internet of Things(IoT)has become an integral part of people’s lives and various industries.Industrial Internet of Things(IIoT)has reshaped the visage of industrial production and accelerated the development of Industry 4.0 to achieve more efficient and sustainable production.Nevertheless,traditional IIoT technologies,such as fieldbus,are no longer applicable at the presence of the rapidly growing number of multi-type services.The advent of the 5th Generation Mobile Communication(5G)has opened up unprecedented possibilities for deploying complex IIoT networks with high density and multiple services.In which,non-orthogonal multiple access(NOMA)can make full use of time/frequency domain resources to support access to large-scale devices,which improves system throughput and spectrum efficiency and becomes one of the key technologies for 5G.Meanwhile,network slicing,another key technology of 5G,can satisfy the demand of multi-type services’ feature of IIoT by slicing the traditional physical network into multiple mutually independent virtual networks according to the needs of different types of devices.In order to better solve the problem of multi-type services and large-scale devices access faced in the incoming IIoT networks,this thesis combines NOMA and network slicing to more properly schedule physical resources.This thesis includes the following two aspects:1.In this thesis,we propose a system model for maximizing the connection density of devices under orthogonal allocation of service resources.In this model,three different service types of devices are considered to exist,and three different slices are introduced in order to allow different types of services to coexist in the IIoT network.The physical resources between different slices are orthogonal,and NOMA is used to access the devices in the same slice.To maximize the connection density of devices,the optimization problem of joint power and sub-carrier resource allocation is constructed and formulated as a mixed-integer nonlinear programming problem.To solve this problem,first,the original problem is split into three subproblems using resource orthogonality and introducing power allocation weights.Second,a corresponding heuristic optimization algorithm is designed for each subproblem.Finally,the dichotomous-like search algorithm is proposed to find the optimal power allocation weights.2.To further improve the system resource utilization,a system model for maximizing the connection density of devices under non-orthogonal service resources is proposed in this thesis.In this model,physical resources between different slices are considered to be non-orthogonal,and NOMA is used to access devices in the same or different slices.To maximize the connection density of devices,the optimization problem of joint power and sub-carrier resource allocation is constructed and formulated as a mixed-integer nonlinear programming problem.To solve this problem,first,it is transformed into a mixed-integer linear programming problem.Second,it is further simplified to an integer linear programming problem by designing a simple and efficient power allocation scheme.Finally,heuristic optimization algorithms based on best-effort pairing are proposed to further reduce the computational complexity.We evaluate the performance of the proposed schemes by comparing them with various benchmark algorithms and conclude that our proposed algorithms achieve near-optimal performance with significant improvements compared with existing methods.Meanwhile,our approaches achieve a good balance between computational complexity and system performance.