Resource Allocation And Performance Analysis For New-Type Hybrid Satellite-Terrestrial Networks

In recent decades,satellite communication,with its advantages of wide coverage,flexible networking,and being less susceptible to disasters,has provided coverage for remote areas and emergency communication scenarios that cannot be served by terrestrial fiber or cellular networks.However,the disadvantages of satellite communication,such as highly dependence on line of sight transmission and limited system capacity,restrict the promotion and development of satellite communications to a great extent.Therefore,making use of terrestrial networks with rich infrastructure and high transmission rates to complement satellite networks,i.e.,forming a hybrid satellite-terrestrial network,can achieve seamless coverage of high-speed mobile broadband networks and is an inevitable trend for the development of communication networks.However,traditional hybrid satellite-terrestrial network has the following two disadvantages.Firstly,satellite and terrestrial components in the network adopt their own network management centers and different air interface technologies,which lead to poor system compatibility and cannot achieve global optimization of resources.Moreover,satellite and terrestrial components employ exclusive spectrum management,easily resulting in waste of spectrum resources in time and space.In view of above problems,this paper conducts research on two new-type hybrid satellite-terrestrial network architectures,including integrated satellite-terrestrial network and cognitive satellite-terrestrial network.In the integrated satellite-terrestrial network,comprehensive resource management can be achieved by adopting a unified control center and compatible air interfaces.Moreover,satellite and terrestrial networks are allowed to use the same frequency band.In the cognitive satellite-terrestrial network,the introduction of cognitive radio technology enables flexible spectrum sharing between satellite and terrestrial networks,which effectively mitigates the contradiction between limited spectrum and the growing demand for broadband services.Compared with traditional hybrid satellite-terrestrial networks,these two new-type hybrid satellite-terrestrial network architectures can improve network compatibility,management flexibility,and spectrum resource utilization.However,their characteristics,such as diverse scenarios and complex interference environments,also bring new challenges to network resources allocations.Moreover,since satellite and terrestrial communications in the hybrid network have much difference in channel fading,how to accurately evaluate and analyze network performance is another difficult problem for hybrid satellite-terrestrial networks.This article focuses on investigating the resource allocation and transmission performance for the integrated satellite-terrestrial network and cognitive satellite-terrestrial network.Main achievements are listed as follows:1.An energy-efficient adaptive transmission scheme is proposed for the integrated satelliteterrestrial network.In existing satellite networks,satellite can communicate with terrestrial destination nodes either directly or with the aid of terrestrial relay nodes.In the direct transmission mode,the satellite link suffers from high symbol error rate(SER)or low transmission rate due to serious shadowing effects.Although cooperative transmission can improve transmission reliability through diversity gain,the extra power consumption of the relay node can easily bring about low energy efficiency.Therefore,taking the requirements of both efficiency and reliability in satellite communications into account,we propose an energy-efficient adaptive transmission scheme with SER constraints.Firstly,the SER performance of the direct transmission and cooperative transmission are analyzed theoretically,and closed-form expressions of average SER and asymptotic SER are obtained.Taking the derived SERs as constraints,we formulate the proposed adaptive transmission scheme as an optimization problem with energy efficiency as the optimization goal,and study the tradeoff among energy efficiency,spectrum efficiency,and SER.In order to reflect the impact of system parameters on the transmission mode more intuitively,we analyze and discuss the economic efficiency as a complementary performance measure.2.For two typical communication scenarios in cognitive satellite-terrestrial networks,the performance of secondary communications in cognitive satellite-terrestrial networks is theoretically analyzed while guaranteeing the communication quality of primary communications.In the first scenario,the terrestrial communication is regarded as the cognitive user to access the spectrum where the primary satellite communication is operating on.In this case,the cellular base station needs to guarantee the interference power constraints imposed by satellite receiver through power control,and the cell edge users will suffer from the inter-component co-channel interference(CCI)from the satellite and intra-component CCI from adjacent cells.By deriving the statistical characteristics of the hybrid CCI,we analyze the outage performance of the cell edge user and obtain a closed-form expression of the outage probability.In the second scenario,the satellite communication is regarded as the cognitive user to access the spectrum where the primary terrestrial communication is operating on.Considering that the interference power constraints imposed by terrestrial communications and the shadow fading of satellite channels may cause poor performance of satellite communications,we introduce cooperative communications into the cognitive satellite-terrestrial network.Since the signal-to-noise ratio formula is complex in cooperative communications and the distribution of satellite link fading involves in special functions,the traditional theoretical analysis method based on the probability density function is not applicable.Therefore,we propose a derivation method based on the moment generating function theory.According to the definition of moment generating function,a threevariable joint distribution problem is decomposed into a univariate and bivariate distribution problems,and then closed-form expressions of outage probability and SER are obtained.Simulation results confirm the correctness of the theoretical analysis,and show the effects of system parameters such as fading conditions and interference power constraints on the network performance.3.We investigate the cognitive satellite-terrestrial network oriented to quality-of-service(QoS)guarantees and interference power constraints.Existing researches on cognitive satelliteterrestrial networks mainly focus on how to maximize system spectrum efficiency or energy efficiency.However,they do not consider the diverse QoS guarantees for different services,which cannot meet the service requirements of future broadband satellite networks.To solve this problem,we firstly analyze the throughput that secondary satellite communications can support when guaranteeing certain delay QoS requirements from the perspective of effective capacity.Based on this,taking the requirements of both low energy consumption and diverse delay QoS guarantees in satellite communications into account,we propose a power optimization scheme that maximizes effective energy efficiency.Through Charnes-Cooper theory,we firstly convert the concave-convex fractional programming problem into a convex optimization problem,and utilize the Karush-Kuhn-Tucker(KKT)condition to obtain the expression of the optimal transmit power.In order to further theoretically evaluate the impact of parameters such as delay exponent and interference constraint on system performance,based on the obtained optimal solution of the transmit power,we analyze the outage performance and finally obtain a closed-form expression of the outage probability.Simulation results not only verify the correctness of the theoretical analysis,but also show that the proposed scheme can support the real-time transmission of multimedia services over wireless fading channels,while guaranteeing the interference constraints imposed by primary terrestrial communications and different statistical QoS limitations of satellite communications.