The Research Of Multi-user OFDMA System Resource Allocation Algorithm

Future wireless communication services require system has capability of high-quality and high-speed, however radio spectrum resource is extremely scarce. To provide reliable quality of service, it is important to effectively improve the utilization of spectrum resource. Orthogonal frequency division multiple access(OFDMA) has high spectral efficiency and the characteristic of resisting to frequency selective fading, which makes it become the key technology of the fourth generation of mobile communication. OFDMA adopts orthogonal frequency division multiple as core technology. In the system, each subcarrier is orthogonal to each other. And this makes OFDMA resource allocation system more flexible. Different from the past static resource allocation in OFDMA systems, through dynamic resource allocation algorithm, it is reasonable to allocate system resource so that the system performance greatly improves.Resource allocation optimization of multi-user OFDMA system is to maximize the system spectral efficiency. Furthermore, resource optimization has other different constraints. Such as the total power constraint and user fairness constraint. Based on user channel information and user business requirements, Resource allocation algorithm allocates subcarriers to corresponding users and assigns certain power resource to the subcarriers. So that algorithm improves system performance, at the same time algorithm also guarantees system constraints. This paper optimizes the performance of spectral efficiency and user fairness in OFDMA resource allocation. Through improving the inadequate part of the original resource allocation algorithms, and adding novel resource allocation schemes, resource allocation algorithm based on subcarrier tradeoff and fairness threshold algorithm are proposed. The core idea of the two algorithms and system performance are expressed as follows.In resource allocation algorithm of considering the fairness, proportional fairness algorithm makes user fairness be guaranteed by introducing user proportional fairness constraint. Algorithm finds the worst current fairness user, and gives high allocation priority to this user. So that the ratio of each user’s rate tends to system constraint. However, while allocating subcarrier to the worst fairness user, the subcarrier cannot get the best subcarrier allocation. To solve this problem, this paper proposed a resource allocation algorithm based on subcarrier tradeoff. While allocating subcarrier to the worst fairness user, system executed the subcarrier tradeoff. The tradeoff user got the subcarrier which should be allocated to the worst fairness user. Then the worst fairness user got a subcarrier from the tradeoff user. System executed the subcarrier allocation based on the channel gain variation after subcarrier tradeoff. The simulation result shows that the algorithm ensures the user fairness, while reducing about 36 percent resource allocated loss in spectral efficiency.In the iterative water-filling algorithm, during each subcarrier allocation, algorithm calculates each subcarrier’s power based on water water-filling algorithm. And this makes performance of user fairness better compared to proportional fairness algorithm. However, the iterative water-filling algorithm for resource allocation is under the situation of low signal to noise ratio, and in this scenario the spectrum efficiency of the system becomes more important. Therefore, the balance between system performance in spectral efficiency and user fairness becomes particularly critical. This paper proposed fairness threshold algorithm, system calculated the user fairness during subcarrier allocation. If the current user fairness had not reached the fairness threshold, system would allocate subcarriers to optimize the user fairness. If the current user fairness had already reached the fairness threshold, system would allocate subcarriers to optimize the spectral efficiency. Simulation results show that system effectively adjusts resource allocation performance of spectrum efficiency and user fairness by setting fairness threshold.