多制式无线通信L2S仿真方法

发布时间：2018-07-04 23:58

本文选题：L2S + 平均值　；参考：《东南大学》2015年硕士论文

【摘要】：无线通信系统发展至今,已形成多制式系统共存、多网络覆盖的新特性。如何准确仿真网络性能至关重要。无线系统仿真中采用分离链路级和系统级仿真的方法来降低计算的负担,提高仿真的速度。L2S方法是指Link to System方法,即链路级到系统级仿真映射方法。L2S方法能够通过物理层信道状态得到无线传输差错概率,为系统级仿真提供数据交互。本文主要研究了平均值、真实值以及有效信噪比映射L2S方法,并针对LTE-A 8天线双码本给出了一种基于波束运算和分组平均的预编码选择算法。本文首先建立了基于平均值和真实值的GSM系统L2S接口。对于平均值接口分别仿真分析了直接映射和两级映射的性能。仿真结果表明直接映射方式误差很大,采用两级映射的平均值接口准确性较高。与平均值接口相比,引入RBER标准差的真实值接口在低RBER时提高了接口准确性。本文给出了CDMA2000 1x EV-DO系统中解扩后符号等效信噪比的计算方式,以此信噪比为输入,研究了基于有效信噪比映射(ESM)的1x EV-DO系统L2S方法。通过仿真表明,与采用平均值接口的1x EV-DO L2S接口相比,有效信噪比映射算法的准确性更高。然后研究比较了四种采用不同信息压缩函数的ESM算法的性能。通过仿真分析表明,MIESM和EESM具有较高的准确性,而LESM和CESM准确性较低,并探讨了有效信噪比映射中信息压缩函数需要满足的一般性规律,阐述了对数函数和容量函数适合做为信息压缩函数的原因。LTE-A下行支持多达8个发射天线进行数据传输,并采用了双码本的预编码方案，双码本设计提高了码本选择的复杂度。为降低码本选择的计算复杂度,本文在基于互信息量最大化准则的基础上,给出了一种基于波束运算和分组平均的预编码选择算法。该算法采用基于波束运算的互信息量计算方式,解决互信息量计算中存在的大量重复计算问题；并且该算法采用分组平均的思想,大幅减少所需计算的信道数。该算法被成功的应用在LTE-A链路级仿真中,仿真结果表明该算法在不损失预编码性能的前提下,大幅降低了算法复杂度。本文重点研究了基于EESM和MIESM算法的LTE-A下行8天线空间复用传输模式的L2S方法,探讨了不同层数(14层)映射,不同带宽配置(1.4MHz-20MHz),不同调制编码方式,不同信道模型(EPA、EVA、ETU)下L2S方法的准确性。仿真结果表明EESM和MIESM算法能够很好的用作LTE-A 8天线传输的L2S方法,在大部分场景下都有很高的准确性,误差小于±0.1dB。准确性在大带宽、高延时扩展信道下有所下降,特别是与高阶调制和高码率相结合时,并且调制阶数对准确性的影响大于码率。
[Abstract]:With the development of wireless communication system, a new characteristic of multi-system coexistence and multi-network coverage has been formed. How to simulate network performance accurately is very important. In wireless system simulation, the method of separating link-level and system-level simulation is used to reduce the burden of calculation, and the method of improving the speed of simulation is the Link to system method. That is link level to system level simulation mapping method. L2S method can get wireless transmission error probability through physical layer channel state and provide data interaction for system level simulation. This paper mainly studies the L2S method of mean value, real value and effective signal-to-noise ratio mapping, and presents a precoding selection algorithm based on beam operation and block average for LTE-A8 antenna dual codebook. In this paper, the L 2S interface of GSM system based on average value and real value is established. The performance of direct mapping and two-stage mapping are simulated and analyzed respectively for the average value interface. The simulation results show that the error of direct mapping is very large, and the accuracy of the average interface using two-level mapping is higher. Compared with the average interface, the real value interface with RBER standard deviation improves the interface accuracy at low RBER. In this paper, the equivalent signal-to-noise ratio (SNR) of despreading symbols in CDMA2000 1x EV-DO system is presented. Based on the signal-to-noise ratio (SNR) input, the L2S method of 1x EV-DO system based on effective signal-to-noise ratio (ESM) is studied. The simulation results show that the effective SNR mapping algorithm is more accurate than the 1x EV-DO L2S interface with average value interface. Then the performance of four ESM algorithms using different information compression functions is studied and compared. The simulation results show that MIESM and EESM have higher accuracy, but LESM and CESM have lower accuracy. The general rule of information compression function in effective signal-to-noise ratio mapping is discussed. The reason why logarithmic function and capacity function are suitable for information compression function. LTE-A downlink supports as many as 8 transmitting antennas for data transmission. The dual codebook precoding scheme is adopted, and the design of double codebook increases the complexity of codebook selection. In order to reduce the computational complexity of codebook selection, a precoding selection algorithm based on beam operation and block average is proposed based on the maximization criterion of mutual information. The algorithm uses the mutual information calculation method based on beam operation to solve the problem of repeated computation in the mutual information calculation, and the algorithm adopts the idea of grouping average to reduce the number of channels that need to be calculated. The algorithm is successfully applied to LTE-A link level simulation. The simulation results show that the algorithm reduces the complexity of the algorithm without losing the precoding performance. In this paper, the L2S method of LTE-A downlink 8 antenna spatial multiplexing transmission mode based on EESM and MIESM algorithm is studied, and the mapping of different layers (14 layers), different bandwidth configuration (1.4MHz-20MHz), different modulation and coding methods are discussed. The accuracy of L2S method under different channel models (EPA-EVAV ETU). The simulation results show that the EESM and MIESM algorithms can be used as L2S method for LTE-A8 antenna transmission, and the accuracy is very high in most scenarios, and the error is less than 卤0.1 dB. The accuracy is decreased in large bandwidth and high delay spread channels, especially when combined with high order modulation and high bit rate, and the modulation order has more influence on the accuracy than the bit rate.
【学位授予单位】：东南大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TN92

【相似文献】