Radio Resource Management In Heterogeneous Integrated Networks Oriented To Smart Distribution Grid

A communication system with the characteristics of high-speed, bidirectional and integrated is a stable basis for the implementation of the interaction between dynamic real-time information and electric power in smart distribution grid (SDG). The weak link of communication system in SDG is the access network, and there is no single network that is able to meet the industrial grade requirements of the quality of service (QoS) for the communication services in SDG at present. The integration of heterogeneous network is one of the main features in the development of the new generation of wireless communication system. Heterogeneous integrated network can provide safe and reliable real-time communication services through overlapping a coverage with a variety of wireless access technologies, which is promising for application to the access network in SDG Owning to the differences of control mechanisms and resource forms among the heterogeneous networks, it is not a simple superposition of the various wireless access technologies, but an effective integration and utilization of resources in the coordination and integration of the heterogeneous networks. Therefore, wireless resource management is the crux to realize the orderly organization and efficient use of heterogeneous integrated networks. Under the above background and centring on the QoS requirements of communication services in SDG, a heterogeneous network model oriented to SDG is presented and the key technologies of radio resource management of this model are researched in this paper, which is intended to provide a novel mentality of building an efficient stable communication network and reliable real-time business service mechanism in SDG.On the basis of variable studies of the heterogeneous integrated network interconnection architecture, the requirements of the communication system in SDG and the QoS demand of various types of communication services, a heterogeneous network model oriented to SDG is presented. This model supports to overlap a coverage with a variety of wireless access technologies, breaking through the current situation that there is only one network on the access layer in traditional distribution grid,thus provides more reliable and safer security for the realization of communication services in SDG. The functional architecture of heterogeneous integrated networks that adapts the concept of Ambient Network(AN) is able to provide feasible support of information interaction and collaborative resource management among the heterogeneous networks, afford transparent service for distribution terminals, as well as sustain efficient heterogeneous wireless network resource management optimization decisions.Two access selection algorithms are proposed to deal with the problem of selecting a suitable network for a distribution communication service in heterogeneous integrated networks. First of all, from the objective perspective of network performance, the entropy-weight and grey relational analysis (E-GRA) based access selection algorithm is presented.A judgment matrix is constructed of the received signal strength, the remaining available resources and the current business blocking rate, and we can get the gray relational matrix reflecting the level of performance of each access network through standardized processing and gray relational calculation. Further, the entropy weight method is adopted to calculate the weight of each parameter reasonably. Combining with the correlation matrix we can get the sort of each heterogeneous network performance. The optimal network will be picked up according to the results of the sort when distribution terminal requests network access or switch. Secondly, by synthesizing network performances and service requirements, a rough set(RS) and analytic hierarchy process(AHP) based access network selection algorithm is discussed. On the basis of satisfying the multi-attribute judgment, the RS theory and AHP are respectively used to determine the subjective and objective weights of network selection indexes. Meanwhile, the least square method is adopted to acquire the compromised weight of each performance index justifiably. The comprehensive performance level of networks and user preferences are both considered in order to improve the accuracy and reasonability of the selection.Aiming at working out the problem that existing load balancing(LB) algorithms failed to consider the requirements of different types of services sufficiently in heterogeneous integrated networks, two new load balance scheduling algorithms are proposed. In the first place, a multi-objective optimal based service admission control algorithm is proposed, which can be applied in balancing the load rate of heterogeneous networks. An optimized multi-objective control model adopting improved membership function is designed, in which the minimum total occupied resources of distribution services, the minimum service blocking rate and the maximum network load balancing degree of each candidate network are taken as objective functions. The objective membership function, composed of gaussian and upper limit functions, is used to fuzz each objective function, then the fuzzy satisfaction maximum-minimum technique is adopted to reformulate this multi-objective optimization problem into a single objective nonlinear programming problem, and the optimal service access strategy is obtained by the genetic algorithm. In the second place,from the point of service transformation, the needs of different services combined with the actual performance of each available network in heterogeneous environment are taken into account, and proper utility functions and fuzzy logic based load balancing algorithm are developed. Three utility functions are designed to reflect three implementation-specific aspects of network performance, including the effective bandwidth, real-time (RT) calls blocking rate and the average non-real-time (NRT) calls transmission time. Further, a fuzzy logic system is adopted to fulfill the fuzzy decision values of three utility functions. Finally, the load of each network is adjusted according to the fuzzy decision value to achieve equilibrium.On the basis of the above research, considering the combination of load transfer scheme and call admission control scheme of heterogeneous integrated networks in SDG, two hybrid LB algorithms are discussed. Firstly,based on resource reservation and preemptive priority, the hybrid LB algorithm is proposed. An adequate number of ongoing services can be transferred from overload cells into the overlapping ones with light workload according to the load rate of each cell and terminal mobility.New calls with different priorities are differentially served by means of resource reservation and preemptive priority based admission control strategy. Simulation results show that the algorithm can guarantee the system resource utilization and the QoS of real time and non-real time services, as well as decrease the system call blocking probability and handover rate effectively compared to the reference algorithms.Scondly, we present a0/1programming model and GA based dynamic LB algorithm. Services are dynamically transferred from overload networks into ones with light workload according to the load rate and QoS assurance degree of real time and non-real time services of each candidate networks. The0/1planning based mathematical model is structured to describe the admission control optimization problem of the new power distribution communication services. The genetic algorithm is adopted to obtain the optimal solution to rationally allocating the network resources for the services, so as to effectively control load rates of different access networks.