Implementation Of LTE/LTE-A Link Level Simulation Platform And L2S Interface Study

System level simulation is an important method of evaluating the performance of wireless communication system.A wireless communication system is usually contains many communication links,and high accuracy of system level simulation will lead to huge computational cost,due to the complexity of communication environment.So it's very necessary to split the link-level simulation and system level simulation and provide a general interface between them to improve simulation efficiency further.The accuracy of L2S interface method is crucial to the whole system simulation,the effective SNR mapping method is widely used in MIMO-OFDM system at present.Such interface methods map a serial of SNR of sub-carriers to an effective SNR value by a mapping function,to predict link performance.A link level simulation platform is built and this paper places lots of emphasis on the EESM and MIESM method based on this simulation platform.For EESM method,this paper gives an improved method based on S-curve model under the premise of analyzing the influence factor.The Exponential effective SNR mapping method is based on chernoff bound of symbol error rate and it's implementation is simple,can obtain effective SNR through mapping function and parameter ?,and find the corresponding BLER on the performance curve of AWGN channel.In the link simulation,it's impossible to simulate all the points on the performance curve,this paper simulates in a certain interval of SNR,and provides a mathematic model of regression curve,then fits the discrete points into a smooth reference curve,reduces interpolation error.Then the performance of EESM method is analyzed and validates the effectiveness of improved method by comparing the simulation result of the two sorts of EESM method.Result shows that the improved method has a better performance,and could predict link performance effectively.MIESM method is a physical layer abstraction algorithm based on mutual information to predict link performance,this paper presents improved method for the computing of mean mutual information,and the simulation for the improved method in different conditions with transmission modes and bandwidth proves its validity.All subcarriers are not required to use the same modulation and coding scheme in this method,so it can be applied to adaptive modulation and coding system,and has high prediction accuracy and versatility.MIESM method firstly calculates bit mutual information of signal over all sub-carriers using lookup table or matnmatical formula,then derives the mean bit mutual information by averaging all the mutual information.MIESM method can be devided into MMIB method and RBIR method.This paper introduces the principle and implementation of these two methods and analyzes performance according to the link simulation results.Finally,this paper compares and analyzes the prediction error of the two method,and gives the graphic-design of L2S Interface.It is observed that the accuracy of the improved method is higher,and it can work well for LTE-A system.