| With the rapid development of wireless communication technology and the increasing number of users, the user's demand for data is also increasing, which leads to the rapid increase in the load of the traditional cellular network. At present, the main solution is to deploy a number of micro cells in the middle of the cell, but with the increase of the number of cells, the complexity and cost of the implementation of the scheme is improved. In addition, the use of cognitive radio technology has led to the introduction of the lower spectral efficiency and the discovery of the spectrum holes in the licensed frequency bands. Cognitive radio technology is composed of three parts: spectrum sensing, spectrum management and spectrum sharing. On the other hand, the current communication network will be for data transmission, high rate and high quality of data security as the goal, how to use the limited spectrum resources effectively to realize the wireless broadband is particularly important. Through the D2D technology came into being, it requires no infrastructure support to end users without base station and direct information transmission, not only reduces the load rate of the cellular network, and enhance the overall network throughput.In this paper, as a starting point from the spectrum access strategy, to the maximum extent possible to ensure that the communication between different devices are found to share spectrum resources and not being used to improve the utilization of spectrum resources under the premise of mutual interference in time. As a specific spectrum sharing technology,spectrum access strategy can be divided into overlay access, underlay access and hybrid access. In cognitive D2D networks, the spectrum reuse between primary and secondary users is realized by using the cognitive ability of D2D users, and the transmission rate of the whole network reaches the maximum. In order to analyze the accurate and detailed model of the system, the first system network model, channel model,signal to interference plus noise model (SINR) model, and then combined with the control from the two angles of the spectrum allocation and power to describe the formation of a problem of joint optimization of a power spectrum. See the first thought of using the optimization problem is solved by convex optimization theory, prove that this function is non convex function, so we introduce the four-color theorem by using the genetic algorithm heuristic algorithm with chaos theory to transform the problem into a mixed integer optimization problem (MINLP). By comparing the simulation results with the brute force search method, this algorithm can not only obtain a sub optimal solution in theory, but also has a significant improvement in time complexity and feasibility.Then, based on the underlay model of spectrum access strategy, the stochastic geometry is used to model the outage probability. In this mode,the secondary users can communicate with the main users to reuse the frequency band in the case that the interference of the main user is lower than a certain threshold value. Therefore, we define the cognitive D2D network interrupt event as a D2D transceiver for SINR less than a predetermined threshold. In addition, we all know that the channel capacity is closely related to the channel characteristics, different channel fading models and different user locations will bring different outage probability distribution and change curve. In this paper, using the Poisson point process in stochastic geometry on the base station, D2D user location description, on the assumption that the channel fading is Ruili fading and interference under the condition of exponential distribution,the typical D2D network transceiver of the outage probability are pushed into the analysis, and compares it with the traditional grid model and empirical model.Finally, this paper combines the D2D technology with cognitive radio, which is in line with the development trend of the future heterogeneous wireless communication network. |
PDF Full Text Request