Research On Resource Scheduling Based On Discrete Spectrum

The standardization program of Long Term Evolution-Advanced (LTE-Advanced) system has been launched by the3GPP, following the advanced version of International Mobile Telecommunications (IMT-Advanced) system’s work plan. Compared with LTE, Carrier aggregation(CA), Relay, Coordinated Multiple Point transmission,(CoMP), Multiple-Input Multiple-Output(MIMO) have been introduced in LTE-Advanced. As the wireless spectrum is limited and scarce, the traditional way of continuous spectrum allocation will produce a large number of spectrum fragments causing the lower spectrum utilization. Cognitive radio technology is put forward to solve this problem. Besides, in the next wireless communication system, the discrete spectrum resource allocation faces a large challenge to meet the demand of higher bandwidth, higher data rate and higher QoS performance. That is to say, the current way of static spectrum allocation cannot satisfy the100MHz bandwidth needs in LTE-Advanced system. Research topic of this paper is based on the spectrum resource allocation in the cognitive LTE-Advanced network.At first, this paper generalizes the research background, expounds the research significance, and illustrates the paper structure. On the basis of analyzing the research status at home and abroad, resource scheduling scheme with carrier aggregation and relay network is emphatically discussed. And based on this, we respectively investigate a multi-user multi-service resource scheduling scheme with CA and relay network.In cognitive LTE-Advanced scenarios, we propose a carrier scheduling scheme with CA, named as QSRUS (QoS-based Separated Random User Scheduling), to acquire the compromise of performance and complexity for the two classical carrier scheduling schemes. With establishing the preferred spectrum lists corresponding to the four types of service model based on service priority and spectrum priority, balancing the load for CC-level scheduler, and formatting the unified utility function to differentiate QoS level for RB-level scheduler, the scheduling metric is calculated for all potential users. From the simulation results, the proposed scheme can achieve significant improvements in scheduling fairness, QoS performance, system throughput and switching frequency, which is an excellent carrier scheduling solution with CA.In the meantime, aiming at the characteristic of the resource competition between two hops in relay network, we consider advancing the traditional scheduling schemes in QoS performance, and present a resource scheduling scheme with relay in the cognitive LTE-Advanced network, named as QTHPF (QoS-based Two Hop Proportional Fair). Through establishing the four types of preferred spectrum lists and formatting the unified utility function, the QoS requirements and scheduling fairness among multiple services are well satisfied simultaneously. Besides, the system throughput, switching frequency and overall system performance are effectively optimized.