Resource Allocation And Handover Optimization Based On Extensible Multi-layer LEO-MSS

Low earth orbit mobile satellite system(LEO-MSS)is the major system to provide communication support for mobile terminals beyond the coverage of terrestrial communication systems.However,two inherent weeknesses limit the communication service capability of LEO-MSS.One is the coverage radius of most LEO satellite beams is more than 100 km,which is much larger than that of terrestrial base stations and limit the capacity of the whole system.The other is the quick movement of LEO satellites,causing frequent passive handovers and great difficulties to guarantee the quality of service(QoS).Therefore,we focus on improving the system capacity and handover in the regions beyond the coverage of terrestrial networks.Based on LEO-MSS,we introduce new network nodes into the system and form a multi-layer network.Then we propose resource management methods and handover management methods based on this extensible multi-layer network.Specifically,we carry out the research work from the following four aspects.(1).Traditional LEO-MSS single layer network can't solve the undercapacity and frequent passive handover perblems fundamentally,and the multi-hop dynamic routings between LEO satellites are complicated.Thus,we solve the above problems via introducing a novel multi-layer network based on the traditional LEO-MSS.In the novel extensible multi-layer network,Geostationary earth orbit(GEO)satellites are introduced as the relay nodes to replace the complicated multi-hop dynamic routings between LEO satellites.High altitude platform(HAP)and terrestrial relay(TR)are introduced to cover the hot-spot regions to increase the capacity and reduce the handover rate,while the original LEO satellites provide coverage in the remote regions.The formulations of the average capacity and handover rate are derived when HAPs share the same bandwidth with LEO satellites.The theoretical analysis and numerical simulations verify the performance improvement in the capacity and handover rate,and ascertain the influence factors.(2).In the extensible multi-layer network,LEO satellites and HAPs share the same spectrum to maximize the total capacity,increasing the chanllege of the system resource management.The original resource management methods are not suitable for the extensible multi-layer network any more.Besides,the large propagation delay of LEO-ground links,the quick movement and multi-beam structure increase the difficulties of resource management compared with terrestrial networks.Thus,according to the multi-layer and multi-beam structure,we design a hierarchical resource management framework to achieve the efficient resource management.Considering the rate demands of users and delay demands of services,we make use of the regular and predictable movement of LEO satellites to reduce the impact of the non-real-time channel state information(CSI)feedback,and propose a forecast based multi-layer multi-beam joint dynamic radio resource optimization method to solve the complicated multi-layer multi-beam downlink resource allocation problem.Via establishing hierarchical optimization,we give the mathematical solutions and low-complexity algorithms to achieve the optimal downlink throughput.The numerical simulations verify the significant downlink throughput improvement with acceptable complexity.(3).There are many differences e.g.interference model and the power constraint between the uplink and downlink in the extensible multi-layer network based on LEO-MSS.Thus,the same optimization method can't achieve the optimal throughput of both downlink and uplink.For the uplink,we make full use of the four colour frequency reuse scheme,the power constraint of mobile terminals(MT),the uplink channel and interference model,and propose a novel resource optimization method suitable for the uplink for the first time.The novel method is the joint of the uplink power-domain non-orthogonal multiple access(PD-NOMA)and the cooperative transmission of multiple beams.We derive the formulation of the optimal power allocation results with three users and verify the significant uplink throughput improvement.We also formulate the problem of user pairing to maximize the uplink throughput and compare the throughput improvement with different solution methods.The simulation results verify the significant uplink capacity and throughput improvement.(4).The handover rate of the extensible multi-layer network based on LEO-MSS is much lower than that of the traditional LEO-MSS.But the handover problems still exist,including the lack of efficient handover protocol procedures and multi-layer passive handover strategy.Traditional handover procedures in the terrestrial network cause large handover delay and signaling cost in satellite systems.Thus,in tenns of the large propagation delay,to reduce handover delay and signalling cost,we design simplified handover procedures suitable for the satellite systems via using the positioning and timing techniques to replace the random access process to achieve the uplink synchronization.Also,considering the multi-layer and multi-beam structure and the predictability of the passive handovers,we propose an efficient centralized handover management framework and a forecast based handover method to optimize the handover time,dropping probability and throughput.The simulation results verify the improvement on the handover delay,signaling cost,dropping probability and throughput.