System Modeling For Satellite Communication And Event-driven Optimization Control Of Transmission Power Rate

Satellite communication is the radio communication on the earth carried out by many earth stations using aerial artificial communication satellites as relay stations.The satellite communication system consists of the satellite and earth stations.Driven by social demand and technological development,satellite communication in the 21 st century is climbing to a new level,and lots of new technologies or developments have aroused people's concern.Using communication satellite as a relay station to transmit high rate communication service between ground stations is a combination of broadband service demand and modern satellite communication technology,and this is also one of the main development directions of the current satellite communication.The key to the success of broadband satellite communication service is to guarantee the Quality of Service(QoS)required by users.In this paper,we study the optimization control of transmission power rate(TPR)for a communication system.In this system,a communication satellite with random energy source provides data transmission service for users(Base stations).The users randomly send out data transmission service requests to the communications satellite,and then the satellite immediately provides the data transmission service to the user in a random fading channel after immediately.In this paper,the stochastic model of the communication system of a communications satellite in the fading channel is given under the random service request and random energy harvesting,which includes communication satellite service request process,communication channel fading process and energy harvesting process.There three process constitute the background process of the communication satellite system discussed in this paper,and we use a discrete-state stochastic process to describe this background process.At the same time,we use a continuous fluid process to model the evolution of the energy left in the buffer.Then the problem of event-driven TPR control based on maximum throughput is presented.The event-driven TPR control mechanism allows the system to readjust TPR to adapt to the current changed system state only when the system state changes.Consequently,an event-driven communications satellite control system with continuous-time and hybrid-state is obtained.In order to solve the control problem,the communication system control model is transformed into a hybrid-state and discrete-timeMarkov decision processes(MDPs)with infinite-horizon and finite action sets,by embedding a discrete-time decision epoch sequence,and the existence of the optimal stationary policy is illustrated.According to the above model,the algorithm design is given in this paper by using the standard state-space discretization method.Firstly,we partition the state-space.Secondly,the sequence of Markov decision process in finite discrete state is constructed.Then,the optimal control problem corresponding to this sequence is solved and the optimal reward function and the stationary optimal policies are obtained.Finally,the reward function and the optimal policies obtained above are extended.After the algorithm is given,an numerical example of event-driven TPR is given to illustrate the theoretical result,and the influences of system parameters on performance are also investigated,such like initial energy,buffer capacity and the maximum TPR.This study has a positive significance to the communication design and management,resource allocation optimization and sustainable development of communication satellites.