Research On The Joint User Grouping And Resource Allocation In VMIMO-SC-FDMA System

The desire for high speed data transmission forces researchers to continually put forward some technical which can greatly increase the efficiency of spectrum resources.MIMO technology is an effective way to increase the data transfer rate.However,due to the limit of small volume and low battery capacity of user terminal in the Long Term Evolution Advanced uplink system,MIMO technology is hardly to exert its advantages in LTE-A system.In order to solve this problem,the concept of virtual MIMO was put forward.Virtual MIMO see a number of users with equipped single antenna as a virtual entity equipped with multiple antennas by user grouping and form a multi antenna array with the multiple receiving antennas configured at the base station,so as to achieve the effect of traditional MIMO.In order to solve this problem,the concept of virtual MIMO technology has been proposed.It forms a virtual MIMO array between the base station side and the users within a user group,so as to improve the system throughput effect.Firstly,in virtual MIMO user grouping strategy is the key technology.But the existing user grouping strategy can only have some fixed umber of user grouping,and rarely consider the system reliability performance and the quality of service in MAC layer.Therefore,this paper proposed a dynamic user grouping criterion which takes BER performance of system into account,and gives the calculation formula of each user group index in the user group set after dynamic grouping.Then,since a typical application scenario of the user grouping strategy is combined with a resource allocation strategy in the SC-FDM uplink which has a frequency selective fading channel.Therefore,in order to obtain the multi-user diversity gain and frequency selectivity,this paper proposed a 0-1 optimized mathematical model for the joint user grouping and resource allocation algorithms considering the delay constraint of service and BER performance.The model combines the delay constraint of service,dynamic user grouping and resource allocation in SC-FDMA uplink together,which will extend the delay constraint of service in the MAC layer to the physical layer by using the effective bandwidth expansion.So the data service rate in the physical layer can meet the delay constraint of service,ensuring the quality of service.In addition,this algorithm also can satisfy the required BER performance of system and maximize the system service rate.Finally,this paper presents a new iterative Hungarian algorithm which with high complexity and optimal solution to solve the joint user grouping and resource allocation problem.The most commonly used existing algorithm is the branch and bound method and Hungary method,where the branch and bound method is complex and the Hungarian algorithm can only assigne individual resource block to one user group,so it is difficult to meet the requirements of SC-FDMA system.The IHA algorithm obtain the optimal system throughput by iteration and traversal method based on the complete user grouping set,the complete resource grouping set and the Hungarian algorithm.Although the algorithm can guarantee the optimality of the solution,its complexity is overhigh and its practicability and maneuverability is poor.In order to further simplify the user grouping problem,a quick and efficient solution to obtain the optimal matching of user group and the resource block groups,an iterative Hungarian algorithm based on user and resource partition is proposed innovatively in this paper based on iterative Hungarian algorithm.The algorithm decompose the combinational optimization problem of user grouping and resource allocation into several sub-problems with low complexity by using integer partition theory,which greatly reduces the search scope and the complexity of joint user grouping and resource allocation problem.In addition,the proposed algorithm is simulated in the SC-FDMA uplink system with the software of MATLAB.We compare the running time and its solution of the BNB algorithm and the iterative Hungarian algorithm based on the user and resource partitions algorithm.The simulation results show that the proposed suboptimal algorithm can obtain the solution which is approximate to the optimal solution but has a lower complexity than BNB method.Then,we compare the proposed joint user grouping and resource allocation algorithm with the fixed number of user grouping and resource allocation algorithms.The simulation results show that the proposed algorithm can guarantee the system's BER performance and maximize the system actual service rate with the condition of guaranteeing the delay constraint of service.