Research On Energy Detection And Resource Schadule In MAC Layer In Cognitive Radio Systems

Nowadays, with the rapid development of the wireless technology application and the increasing demand of the radio spectrum resources, the radio spectrum resources increasingly tense. However, the actual spectrum using measure report given by FCC shows that the utilization of the radio spectrum resources is quite uneven in deferent regions or deferent time, and the utility is deferent for deferent spectrums. Dynamic spectrum access is a suitable scheme for its flexibility in alleviating the radio spectrum resource scarcity, and the cognitive radio technology is the leading one for realizing the dynamic spectrum sharing. Therefore, the research on cognitive radio technology is becoming more and more hotter and deepening. Therefore, cognitive radio technology is my main research direction in this thesis. The major works are outlined as follows:1,The accurate detection of the licensed spectrum is the key step for dynamic spectrum sharing in cognitive radio systems. In this thesis, we propose a double-threshold detection algorithm in energy detection when the cognitive user detects independently. It can attain better effect on protecting the primary users than other ways. At the meanwhile, we put forward a dynamic detection threshold algorithm in energy detecton. Theoritical analysis and simulation results show that in lower SNR the dynamic detection threshold algorithm can get better detecton probability than others, at the same time, the double threshold detection algorithm can do better for protecting the primary user when SNR is not too low.2,In spectrum detection algorithm, cooperative detection scheme is regarded as a more effective way in order to overcome the fading and interference in physical lay detection in cognitive radio systems. We propose a double-threshold cooperative energy detection algorithm in cognitive radio systems. In the scheme we adopt the centrrlized cognitive radio network,. and the cognitive node sends its detection result or detection value to the centre, and the final decision is confirmed in the centre, and then it is sent back to the related cognitive nodes. Meanwhile, we also turn out the cooperation chance selection and the partner selection. From theory analysis and simulation results, it can be know that the double-threshold cooperative energy detection algoritm not only can resolve the detection results falling between the two detection thresholds but also can improve the cooperative detecton probability, and then, for the tradeoff between the detection performance and the realization complexity we'd better select the suitable chance and the advisable partner to cooperate.3,Along with the fast development of the cognitive radio technology, how to manage the radio resource, such as spectrum allocating dynamically, spectrum mobility management, Quality of service (Qos), etc. is concerning in Media Access Control (MAC) in cognitive radio systems. We propose an access strategy of primary protection under energy constraint in cognitive radio systems, and discuss the throughput of the cognitive network applying Partially Observable Markov Decision Process (POMDP). Theory analysis and simulation results indicate that the transmittability of the cognitive network relies strongly on the channel's occupation of the primary network and the strength of the primary's protection, at the same; it is restricted by the initialized energy of the cognitive users.4,On the other way, we give an access strategy with collision avoiding in cognitive radio systems. We also apply the POMDP model to resolve the throughput of the cognitive network when there are more than one cognitive user trying to occupy the same channel. We can draw a conclusion that the throughput of the cognitive network strongly relies on the channel occupation of the primary network when collision among the cognitive users happens. If collision appears, when there are enough idle channels in primary networks this strategy can improve the cognitive network throughput obviously, on the other hand, it has little improvement for cognitive network throughput.