Research Of Theory And Application On Multi-Channel Dynamic Access And Networking In Unmanned System

Complex and variable spectrum environments and decreasing available spectrum resources make reliable and efficient network communication in unmanned systems particularly difficult,but Cognitive Radio(CR)technology for dynamic spectrum access and communication using the “spectrum hole”(ie,the idle band temporarily used by unauthorized users)in the licensed band provides a solution to the above difficulties.CR-based unmanned nodes can 'move' affected or interfered communications to idle frequency bands obtained by spectrum sensing,thus mitigating the effects of dynamic changes in the spectrum environment.However,how to coordinate the access of each unmanned node to the idle frequency band is the key problem for unmanned systems to use CR technology for link building.Secondly,since the imperfect spectrum sensing in physical layer and the access mechanism of Media Access Control(MAC)both affect the delay of link building of unmanned nodes,it is necessary to jointly optimize spectrum sensing and access for time-sensitive unmanned systems from a cross-layer perspective.Furthermore,for largescale unmanned systems,there is an urgent need for reliable communication network to achieve efficient collaboration in unmanned systems.On the one hand,the use of CR technology enables the unmanned nodes to transmit information between multiple frequency bands,which undoubtedly increases the flexibility of network communications;on the other hand,the information exchanging between unmanned nodes allocated on different frequency bands needs frequent channel hopping,which seriously affects the connectivity of network communications.Therefore,there is a need for networking design based on tradeoff between network connectivity and flexibility in large-scale unmanned systems using CR technology.Except for the theoretical research on the above-mentioned unmanned system communication and networking,it is also necessary to study the practical application of unmanned system communication and networking.Taking the UAV-based data collection of ground nodes in the typical air-to-ground unmanned system as an example,considering the urgent need to minimize mission time in emergency or military scenarios,it is necessary to design multi-dimensional joint optimization of UAV-based data collection.In general,this thesis focuses on the theory and application of spectrum access and networking for unmanned systems.Relaying on the CR technology based spectrum access networking and application with minimizing mission time,considering specific network scenarios such as practical spectrum access conditions(non-perfect spectrum sensing and transmission collisions of spectrum access)and typical spectrum access protocols(multi-channel access and contention access).The main work and significance of the thesis include:To solve the distributed access and communication link building in unmanned system under dynamic spectrum environment,the characteristics of unmanned system under dynamic spectrum environment are considered,a channel hopping based blind rendezvous mechanism is proposed to achieve spectrum access and communication links building in unmanned system.The mechanism,which is inspired by the operation of circular clock,enables any pairwise nodes achieve rendezvous by using the channel hopping frequency difference of the transmitting and receiving nodes.At the same time,the practical situations that the nodes are difficult to perform distributed time synchronization and that the available channel inconsistency are considered in the algorithm's parameter design.Furthermore,through theoretical analysis and simulation,the optimal parameter configuration(the receiving node's channel stay duration)that minimizes the average time to rendezvous is obtained.Simulation results verify the performance of the proposed mechanism in terms of average time to rendezvous,maximum average time to rendezvous,rendezvous fairness,rendezvous robustness,etc.,and the impacts of imperfect spectrum detection on rendezvous are also analyzed.To solve the tradeoff problem between imperfect spectrum sensing and spectrum access in channel hopping based communication link building,the energy detection model based spectrum sensing in physical layer and multi-user contention access and non-contention access in media access control(MAC)layer are considered comprehensively.An analysis model and an optimization method to minimize the link-building delay are proposed.Firstly,through the analysis of channel hopping process,the mathematical model of linkbuilding delay is obtained.Secondly,a method for analyzing rendezvous performance of channel hopping algorithm by using periodicity and regularity is proposed.On this basis,the impacts of contention access and non-contention access on the link-building delay are considered respectively.And two heuristic search algorithms for two access schemes to obtain optimal sensing parameters are proposed.Finally,the simulation results verify the correctness of the theoretical model,and the performance comparison and analysis verify the effectiveness of the heuristic search algorithms and the advancement of the link-building delay minimization optimization design.To solve the problem of network connectivity and flexibility in the multi-channel networking of unmanned systems,a multi-channel distributed clustering mechanism for networking is proposed.The clustering mechanism,which is based on the practical assumptions in distributed networking(i.e.,hard to achieve time synchronization,and possible transmission collisions),solves the isolated node problem in existing algorithms.Firstly,the optimal clustering size based on channel quality is obtained through theoretical analysis.Then,based on the improved reaction-diffusion model in biological systems,a bio-inspired distributed clustering algorithm is proposed,which enables nodes to tend to optimal cluster size in distributed clustering process.Furthermore,based on theoretical analysis,the optimal parameters of the algorithm are derived to improve the performance and applicability of the algorithm.Finally,the simulation comparison and analysis verify the effectiveness of the algorithm and prove the advantages of the algorithm in computational complexity and communication complexity.To solve the optimization design problem of air-to-ground UAV data collection,joint optimization design of UAV trajectory,UAV height,UAV speed and data link scheduling is conducted to minimize the mission time.Through the analysis of the relationship between the original data collection time and the UAV trajectory,the time-related original problem is equivalently transformed into the trajectory minimization problem.By analyzing the relationship between the trajectory and altitude,the altitude optimization problem is modeled as the problem of maximizing the transmission radius on 2D plane of the ground nodes.On this basis,two methods of trajectory optimization are proposed combining the traveling salesman problem algorithm and the convex optimization technique,namely STOA and GTOA.STOA avoids repeated visiting ground nodes,while GTOA greatly reduces the amount of computation of optimization at the expense of reducing a small amount of search space.Secondly,the UAV speed and data link scheduling optimization model is obtained by discretizing the trajectory,and the problem is solved by using the block coordinate descent method.Finally,the simulation results verify the advanced nature of the algorithm,and the factors affecting the trajectory and mission time are analyzed through simulations.