Research On Cooperative Spectrum Sensing And Sharing Technology For LEO Satellites

With the rapid development of air-space-ground-sea integrated network,electromagnetic environment is facing increasingly prominent challenges.On the one hand,the sharp increase of various frequency demands in the system leads to the shortage of spectrum resources,resulting in serious co-frequency interference problems.On the other hand,due to the over low utilization rate of the allocated licensed spectrum,there is a problem of "pseudo depletion" of resources.The traditional static frequency allocation method has been difficult to meet the growing frequency demand,and the contradiction between "the increasingly serious spectrum resource deficit" and "the low spectrum utilization" is very prominent.Therefore,the research on the global electromagnetic spectrum sensing theory and key technologies with accurate sensing capability in complex wireless environment is of great significance for the safe control of electromagnetic spectrum space and the efficient utilization of spectrum resources.LEO(low earth orbit)constellation can achieve global(including bipolar)seamless coverage,with advantages of large coverage,short distance between satellite and ground,low transmission loss,etc.,and has great attraction for realizing global spectrum sensing.However,due to the requirements of light and miniaturization,the perception and processing capabilities of a single LEO satellite are limited.However,satellite communication has the characteristics of large coverage area,wide monitoring frequency band and complex electromagnetic environment,which lead to the challenges of LEO satellite global spectrum sensing in terms of monitoring bandwidth and monitoring accuracy,monitoring demand and monitoring capability,sensing performance and spectrum efficiency.In view of the above challenges,we have explored theories and methods to better realize the accurate perception and efficient sharing of global electromagnetic spectrum.The main research contents of this dissertation are summarized as follows:(1)Aiming at the contradiction between high real-time sensing performance requirements of non-cooperative signals and poor monitoring resolution of satellite wide beam antennas,a multisatellite cooperative sensing algorithm based on coalitional game theory is proposed.We have established a multi-satellite cooperative sensing framework for broadband signals with coarse and fine granularity,and divided the broadband spectrum to be monitored into multiple narrowband spectrum for sensing.First,a coarse-grained cooperative spectrum sensing algorithm is proposed to sense multiple sub-bands simultaneously.Considering the influence of inter-satellite links,the algorithm models individual benefits as a bargaining game with transferable utility,and allocates the coalition utility among alliance members according to individual contributions.According to the individual benefits,the coalitional game operation is carried out to realize the adaptive grouping of cooperative satellites to sense different subchannels.The whole algorithm can give consideration to both individual benefits and coalition benefits,stimulate the initiative of individuals to participate in cooperation,and improve the perception performance of the entire broadband.For interested single channel sensing,a fine-grained cooperative spectrum sensing algorithm is proposed.The algorithm is constructed as a coalitional game model with non transferable utility,and the coalition utility is equal to the individual benefits.The game takes individual interests as the priority,drives the cooperative satellite to form the strongest coalition,and completes the fine perception of a single channel.The simulation results show that the proposed algorithm can effectively improve the sensing performance while meeting the sensing requirements of different spectral resolutions.(2)Aiming at the contradiction between the high demand for weak signal sensing and the weak sensing ability of a single LEO satellite in the complex electromagnetic environment,a task driven multi-satellite array sensing enhancement method is proposed.Specifically,for the sensing task with known target location information but unknown signal characteristics,a multi-satellite array sensing modeling and theoretical analysis algorithm under perturbation is proposed.In this algorithm,the average power pattern function of distributed satellite cluster(DSC)is derived by using the random antenna array theory.On this basis,the optimal beamforming weight vector is obtained by constructing a constrained optimization problem to maximize the signal-to-noise ratio.Then the closed form expression of correct detection probability based on signal level fusion is derived,and the sensing performance analysis of DSC under perturbation is completed.For the sensing task with known target signal characteristics but unknown location information,a multi-satellite array sensing algorithm based on feedback control is proposed.In this algorithm,a two bit feedback random phase optimization algorithm is proposed,which can adjust the disturbance step size according to the changes of the surrounding environment,achieving faster convergence speed and better stability at the same time.Finally,the closed form expression of the correct detection probability is derived by using the optimized phase value,and the sensing performance analysis of the multi-satellite array based on feedback control is completed.The simulation results show that the proposed method can achieve spatial filtering of the target emitter by using beamforming technology,and effectively improve the perception ability of weak signals when strong and weak signals coexist.(3)Aiming at the contradiction between the shortage of available spectrum resources of current satellite systems and the low efficiency of existing spectrum sharing algorithms,a joint optimization method of inter satellite co frequency sharing based on spectrum sensing is proposed.A generic spectrum sharing framework for LEO and Geostationary Earth Orbit(GEO)satellite systems is designed.On this basis,the interference effect of LEO satellite movement on GEO system is analyzed.The spectrum activity state of users is modeled as a hidden Markov model,and the average spectrum occupation time under different initial states is analyzed.In order to improve the spectrum utilization of the sharing system,under the constraint of certain detection probability,transmission power and interference threshold,a multivariate constraint optimization model is established to maximize the throughput of the LEO system.It is a balance optimization problem between the sensing duration,transmission power,interference intensity and average throughput.The linear search algorithm and Lagrangian multiplier method are used to solve the problem.The simulation results show that the proposed method can not only achieve higher average throughput,but also achieve better time utilization compared with the traditional scheme.