Research On Spectrum Sensing And Resource Allocation Algorithm Based On Cognitive Database

In recent years, according with the diversity of wireless communication application and wireless business growth, spectrum resource is becoming scarce. In order to alleviate the spectrum shortage problem in the TD-LTE system, cognitive radio technology for the coexistence between TD-LTE system and Ultra High Frequency band radio television system can be used to achieve spectrum sharing and improve spectrum efficiency.At present, cognitive database can be utilized to obtain the deployment of BS nearby a secondary device to overcome the hidden terminal problem and avoid harmful interference to the primary users. But it is not guaranteed to access the white space resources when the secondary device send requirement only using the cognitive database technology. So we can combine the cognitive database technology and spectrum sensing technology to ensure the access of the secondary device when requesting the white space resources. However, the secondary device still need to sense the channels provided by the cognitive database one by one before finding the available channel under the premise of without causing harmful interference to the primary users, thus the cost of sensing and information feedback overhead are still large,while the access rate of secondary device will also be affected.In this thesis, in order to reduce the cost of sensing, the information exchange between the secondary device and cognitive database is improved based on the present algorithm by using cognitive database and spectrum sensing, and a channel allocation algorithm based on geo-location and positioning accuracy is proposed. The algorithm makes full use of positioning accuracy which is an element within the decisions of cognitive database. The improved algorithm first allocates the available channels to White Space Device which locates outside the keep-out region of TV stations, thus sensing stage can be avoided, so sensing costs and information feedback overhead can be reduced. Then it allocates the channels to the WSD within the scope of protection of TV station, and in this case the WSD has to sense the allocated channels to determine whether they are idle using the energy detection algorithm. Thus the times of collusions can be reduced for better protecting the incumbent users. Finally, the simulation results are given to prove the feasibility of the improved algorithm.Second, for the application of TD-LTE system on vacant TV frequency spectrum, we research how to allocate the white space obtained from the cognitive database in the D2D (Device to Device) communication. The power control between D2D link and cellular link is improved based on the present algorithm by using sensing-aware resources allocation, and an efficient resources allocation scheme for balancing the signal to noise ratio between the cellular and D2D terminal is proposed. Considering all possible interference in the uplink and downlink, we establish a convex optimization problem to maximize total system throughput on the basis of optimizing the received interference power and balancing SNR of all users. Then, the convex optimization problem is solved by using Lagrange multiplication method, and thus the closed-form solution is obtained for this convex optimization problem. Finally, simulation results are presented to prove the feasibility of the improved algorithm.