Research On Joint Spectrum Allocation Algorithms In Cognitive Radio System

With the rapid development of wireless communication technology and the unceasing increase of wireless users, wireless users need more spectrums, the spectrum resource strain is a rigorous problem in wireless management. Static spectrum allocation of cognitive radio technology lacks the flexibility at present, the rate of the spectrum average utilization is low, and some spectrums are wasted. On the premise of keeping normal communication of authorized users, analysis of the wireless environment, the idle spectrum is secondly occupied, limited spectrum resources are used more effectively, dynamic spectrum allocation technology can solve the contradiction between wireless communications demand and spectrum resources on cognitive radio system. In the future wireless communication field, dynamic spectrum allocation technology has very broad prospects and has become a hot research topic on cognitive radio system. Based on graph theory and game theory and correlative theories, this paper makes research on the cognitive radio spectrum allocation problem by the optimization methods, the main content is as follows:This paper analyzes and studies detailedly spectrum allocation algorithms based on graph theory and game theory and correlative theories, consid ers fully some main factors influenced jointed spectrum allocation, such as system’s fairness, utility of bottleneck cognitive users and jamming decision between cognitive users. This thesis mainly carried out around jointed spectrum allocation algorithm steps, allocation performance and simulation. Additionally, in order to reduce the interference to authorized users and the competition between cognitive users, make sure the cognitive radio system runs normally, the reliability theory could be used to describe the probability that cognitive users use the channels, and the cognitive users continuously use them for a period of time, the article proposes three-dimensional jointly allocation algorithm from space- time- frequency.This thesis proposes four spectrum allocation algorithms on the basis of graph theory and game theory, and analyzes the effectiveness of algorithms. Four innovations of this paper are includes:At present parallel algorithm allocates spectrums by Collaborative-Max-Min-Ba-nd width criterion, system fairness is poor, utility of bottleneck cognitive users is low. On the basis of graph theory, this thesis proposes vertical parallel jointed spectrum allocation algorithm. This paper presents a spectrum allocat ion algorithm based on adopting Collaborative-Max-Min-Bandwidth and Collaborative-Max-Propo-rtional-Fair standards to calculate subgraph,analyses the effect of different scaling factors on system’s fairness. According to optimal scaling factors, when some channels are allocated by Collaborative-Max-Min-Bandwidth rule, the rest of the channel are allocated by Collaborative-Max-Pro-portional-Fair standards rule, calculates each spectrum label what cognitive users used. The simulation shows that the system fairness has improved obviously, at the same time, channel capacities of bottleneck cognitive users have been improved greatly. Vertical parallel jointed spectrum allocation algorithm takes into account both system’s fairness and utility of bottleneck cognitive users, and effectively control communication time at the same time.On the basis of game theory, this thesis proposes a novel non-cooperative game spectrum allocation algorithm, overcomes the effects of disturbances between cognitive users. In the process of multi-user sharing spectrum, cognitive users don’t collaborate together on the practical cognitive radio system. Novel non-cooperative game algorithm is used to study the relationship that multiple participants how to make decision, and maximize their own utility function. The simulation results show that cogntive usres may share with authorized users’ channels in novel non-cooperative game spectrum allocation algorithm, under meeting the limits of jamming power, cognitive users game to achieve reasonable power allocation. Stable rate of joint algorithm is greater than the original non-cooperative game algorithm and the underlay algorithm, simulation indicates that novel non-cooperative game spectrum algorithm is effective.In view of the practical application, this thesis proposes spectrum allocation algorithm based on graph theory and game theory, to overcome the shortcomings that graph theory does not meet practical spectrum utility requirement, and to overcome the shortcomings that game theory algorithm obtains difficultly the spectrum from the spectrum sharing pool. This paper applies firstly graph theory algorithm to allocate rapidly spectrums(channels), at the same time provides fairness during the second allocation, and then this paper uses the game theory algorithm to optimize the first allocation spectrums(channels), and completes spectrum reallocation between cognitive users, spectrum allocation algorithm based on graph theory and game theory makes a virtue out of graph and game theory. Simulation results show that the spectrum resources can be fully utilized, under transmission power constraints, each cognitive user optimizes its own utility(rate) and distribute reasonably transmission power. For spectrum allocation algorithm based on graph theory and game theory, the allocation transmission power is bigger than the single game algorithm on channels what cognitive user used, effectiveness is improved at the same time, total utility of joint algorithm is greater than the single game algorithm, the simulation verifies that spectrum allocation algorithm based on graph theory and game theory is effective.The article proposes three-dimensional joint allocation algorithm from space-time- frequency. The reliability theory could be used to describe the probability of cognitive users using the channel, and the cognitive users continuously using it for a period of time, in order to reduce the competition between cognitive users and the interference to authorized users, and to make sure the cognitive radio system run normally. Simulation results show that when the main users are fixed, with the increase of cognitive users, for the value of system throughput, three-dimensional algorithm from space- time- frequency is better than the greedy algorithm and two dimensional algorithm, when the cognitive users are fixed, with the increase of authorized users, as the idle authorized users amount advanced, so do the volume of system transmission data and the value of system throughput, three-dimensional algorithm from space- time- frequency is better than the greedy algorithm and two dimensional.Three-dimensional algorithm help to reduce the interference to authorized users from cognitive users and lower competition between cognitive users, simulation indicates that this algorithm is effective.