Research On Interface Between Link Simulation And System Simulation For LTE-A Downlink System

The wireless communication system has two kinds simulation:system-level simulation and link level simulation. The concern of Link Simulation is performance of end to end communication. Modeling the physical layer from the bit-level data, the performance impact of new technology is detailed analyzed. System Level Simulation focuses on higher-level performance such as the whole system coverage, capacity and other properties. Link simulation is to verify the physical layer technology, and system simulation is for network planning, community collaboration, and to provide convenient location of the base station. However, system simulation has a huge amount of data link, so we use the link to system interfaces to simplify complexity. Firstly, the link simulation platform is established and simulation results based on this platform will be acquired to analyze the interface algorithms. This paper will focus on two interfaces:RBIR and EESM which are widely used.An adaptive search algorithm was proposed in this paper to determine the reasonable range of SNR of the simulation, which can save a lot of time and makes the link simulation platform run automatically. The adaptive algorithm based on the principle of binary search has increased the efficiency of the searching for the range of signal to noise ratio. The accuracy of the simulation platform has a very profound influence on the verification of physical link layer algorithm and data analyzing, however, most of the existing link-level platforms are too simple. Referring to the standard protocol from 3GPP Working Group, the platform is built in a modular fashion, it can reproduce the complete procedure of the data transferring from end to end. Through the actual simulation, the platform established in this paper has high availability which can satisfy the needs of the interface algorithm analysis.For EESM algorithm, the paper gives an improved algorithm based on the error function. Exponential effective SNR mapping algorithm is based on the symbol error rate, but the error is some big due to the approximate boundary. The improved algorithm improves the prediction performance of higher-order modulation, but weakens the performance of lower-order part and increases the computational complexity as side effect. The improved algorithm can play a role in some scenarios.Based on the compression of RBIR, this paper gives the equivalent SNR exponential mapping algorithm on Mutual Information (MIEESM). Mutual information effective SNR mapping algorithm has higher accuracy of prediction than EESM and can be applied to a communication system using adaptive modulation, but the RBIR compression of MIESM is too simple which increases the error of high-end part. Compared with the original RBIR algorithm, the enhanced algorithm plays a similar performance in low-end part but do better in high-end part.Actual simulation has proved the feasibility of the enhanced algorithm.