混合比特量化接收机在Massive MIMO下的应用
发布时间：2018-12-05 20:28
【摘要】：随着现代无线通信技术的发展,为了进一步增强无线通信系统的传输性能,Massive MIMO系统作为下一代无线通信系统的核心技术已经被越来越广泛的认可。而massive MIMO系统需要在基站端配置成百上千的巨大规模的天线数量,因为天线数量的急剧增加,对整个系统中的量化ADC来说,占用资源消耗、复杂度和功耗成为了亟待解决的问题。所以在未来的massive MIMO 5G无线通信中,如何配置一种高速率高性能的ADC变得至关重要。目前对于在massive MIMO下的高速率和高性能ADC,主流的方法是采用了低量化比特的结构。区别于12比特量化的ADC,低量化比特的ADC具有功耗低,速率快的特点。在低量化比特的ADC中为了追求和高量化比特ADC同等级的性能,人们提出了各种算法的检测器,其中主流的有基于最大似然估计的EM算法和GAMP算法。这三种算法都是通过循环迭代来代替具有复杂计算量的最大似然估计。然而低比特量化的结构在量化估计后与高比特量化相比性能毕竟有所差异。为了精准估计出一些CSI信息,高比特量化的结构是必要的。所以混和量化结构的接收机是一种在massive MIMO中非常实用的结构。本文介绍了GAMP算法的DQ理想检测器。并在此基础上提出了GAMP算法的PDQ检测器和线性检测器。这两种检测器在结构上混合了高比特量化和低比特量化,兼顾了高性能和高速率ADC的要求以及精确估计CSI信息的要求。同时这两种检测器相比于DQ理想检测器计算量更简单更容易硬件实现。接着,本文对GAMP算法的DQ理想检测器、PDQ检测器和线性检测器做了仿真。本文建立了混合比特量化接收机的系统模型,并且在该模型下验证了三种检测器的误码率和MSE性能。同时本文还针对量化步长,量化方差和混合结构的混合度做了大量仿真。分析了量化步长,量化方差和混合结构对不同检测器的影响。最后本文的仿真验证了PDQ检测器比DQ检测器的性能下降不大,同时计算复杂度得到了大幅度减少。从而证明了GAMP算法的PDQ检测器是一种对混合比特量化接收机的优良结构,这种改良的混合比特量化接收机是一种非常适用的结构。
[Abstract]:With the development of modern wireless communication technology, in order to further enhance the transmission performance of wireless communication system, Massive MIMO system as the core technology of the next generation wireless communication system has been more and more widely recognized. The massive MIMO system needs to deploy hundreds of large antennas in the base station, because of the rapid increase of the number of antennas, the consumption of resources, complexity and power consumption have become a problem to be solved urgently for the quantized ADC in the whole system. Therefore, in the future massive MIMO 5 G wireless communication, how to configure a high speed and high performance ADC becomes very important. At present, the main method of high speed and high performance ADC, under massive MIMO is to adopt low quantization bit structure. ADC, which is different from 12 bit quantization ADC, with low quantization bit, has the characteristics of low power consumption and high rate. In order to achieve the same performance as high quantization bit ADC in low-quantized bit ADC, a variety of detectors are proposed, among which the main algorithms are EM algorithm based on maximum likelihood estimation and GAMP algorithm. These three algorithms replace the maximum likelihood estimation with complex computational complexity by cyclic iteration. However, the performance of low bit quantization structure is different from that of high bit quantization after quantization estimation. In order to estimate some CSI information accurately, high bit quantization structure is necessary. So the receiver with mixed quantization structure is a very practical structure in massive MIMO. In this paper, the DQ ideal detector of GAMP algorithm is introduced. On this basis, the PDQ detector and linear detector of GAMP algorithm are proposed. The two detectors are structured as a mixture of high bit quantization and low bit quantization, taking into account the requirements of high performance and high rate ADC as well as the requirement of accurate estimation of CSI information. At the same time, these two detectors are simpler and easier to implement in hardware than DQ ideal detectors. Then, the DQ ideal detector, PDQ detector and linear detector of GAMP algorithm are simulated. In this paper, the system model of the hybrid bit quantization receiver is established, and the error rate and MSE performance of the three detectors are verified under the model. At the same time, the quantization step size, quantization variance and mixing degree of hybrid structure are simulated. The effects of quantization step size, quantization variance and mixed structure on different detectors are analyzed. Finally, the simulation results show that the performance of the PDQ detector is not much worse than that of the DQ detector, and the computational complexity is greatly reduced. It is proved that the PDQ detector of the GAMP algorithm is an excellent structure for the hybrid bit quantization receiver, and the improved hybrid bit quantization receiver is a very suitable structure.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TN851;TN919.3
本文编号：2365427
[Abstract]:With the development of modern wireless communication technology, in order to further enhance the transmission performance of wireless communication system, Massive MIMO system as the core technology of the next generation wireless communication system has been more and more widely recognized. The massive MIMO system needs to deploy hundreds of large antennas in the base station, because of the rapid increase of the number of antennas, the consumption of resources, complexity and power consumption have become a problem to be solved urgently for the quantized ADC in the whole system. Therefore, in the future massive MIMO 5 G wireless communication, how to configure a high speed and high performance ADC becomes very important. At present, the main method of high speed and high performance ADC, under massive MIMO is to adopt low quantization bit structure. ADC, which is different from 12 bit quantization ADC, with low quantization bit, has the characteristics of low power consumption and high rate. In order to achieve the same performance as high quantization bit ADC in low-quantized bit ADC, a variety of detectors are proposed, among which the main algorithms are EM algorithm based on maximum likelihood estimation and GAMP algorithm. These three algorithms replace the maximum likelihood estimation with complex computational complexity by cyclic iteration. However, the performance of low bit quantization structure is different from that of high bit quantization after quantization estimation. In order to estimate some CSI information accurately, high bit quantization structure is necessary. So the receiver with mixed quantization structure is a very practical structure in massive MIMO. In this paper, the DQ ideal detector of GAMP algorithm is introduced. On this basis, the PDQ detector and linear detector of GAMP algorithm are proposed. The two detectors are structured as a mixture of high bit quantization and low bit quantization, taking into account the requirements of high performance and high rate ADC as well as the requirement of accurate estimation of CSI information. At the same time, these two detectors are simpler and easier to implement in hardware than DQ ideal detectors. Then, the DQ ideal detector, PDQ detector and linear detector of GAMP algorithm are simulated. In this paper, the system model of the hybrid bit quantization receiver is established, and the error rate and MSE performance of the three detectors are verified under the model. At the same time, the quantization step size, quantization variance and mixing degree of hybrid structure are simulated. The effects of quantization step size, quantization variance and mixed structure on different detectors are analyzed. Finally, the simulation results show that the performance of the PDQ detector is not much worse than that of the DQ detector, and the computational complexity is greatly reduced. It is proved that the PDQ detector of the GAMP algorithm is an excellent structure for the hybrid bit quantization receiver, and the improved hybrid bit quantization receiver is a very suitable structure.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TN851;TN919.3
【相似文献】
相关期刊论文 前1条
1 周云龙;彭颖;;基于图像法的稀相气/固两相流混合比及空隙率分析[J];化工自动化及仪表;2013年10期
相关会议论文 前2条
1 李继军;叶海华;秦国泰;;空间大气环境探测中N_2/O_2混合比的地面校测[A];中国空间科学学会空间探测专业委员会第十八次学术会议论文集（下册）[C];2005年
2 许华;蒋伟阳;吴卫东;黄勇;常富华;陈志梅;唐晓红;;金、钆混合平面膜靶及腔靶制备工艺[A];中国工程物理研究院科技年报（1998）[C];1998年
相关硕士学位论文 前2条
1 周淼;基于丙酮平面激光诱导荧光气流混合比测量研究[D];哈尔滨工业大学;2015年
2 刘果;混合比特量化接收机在Massive MIMO下的应用[D];电子科技大学;2016年
，本文编号：2365427
本文链接：https://www.wllwen.com/kejilunwen/xinxigongchenglunwen/2365427.html
