基于深度学习的1-比特超大规模MIMO信道估计
2021年电子技术应用第8期
蒲旭敏1,2,吴 超1,2,杨小珑1,2
1.重庆邮电大学 通信与信息工程学院,重庆400065;2.重庆邮电大学 移动通信技术重庆市重点实验室,重庆400065
摘要: 超大规模多输入多输出(Multiple Input Multiple Output,MIMO)技术在未来的移动通信中具有巨大的潜力。然而,采用超大型的天线阵列会出现空间非平稳性信道特征,如果为基站的每根天线都配备一个高精度量化器,系统功耗将大幅度增加,从而阻碍超大规模MIMO系统的广泛应用。因此,假设基站的每根天线都配备有一对1-比特模数转换器(Analog-to-Digital Converters,ADC),利用子阵列与用户之间的映射关系来描述非平稳信道特征,借助深度神经网络(Deep Neural Network,DNN)强大的泛化能力,设计一种新的生成式监督DNN模型,该模型可以使用合理数量的导频进行训练。仿真结果表明,所提出的网络可以使用较少的导频得到较好的估计性能,在性能与计算复杂度之间取得良好的平衡。
中圖分類號: TN92
文獻標識碼: A
DOI:10.16157/j.issn.0258-7998.211341
中文引用格式: 蒲旭敏,吳超,楊小瓏. 基于深度學習的1-比特超大規(guī)模MIMO信道估計[J].電子技術應用,2021,47(8):87-90,96.
英文引用格式: Pu Xumin,Wu Chao,Yang Xiaolong. Channel estimation for 1-bit extremely massive MIMO via deep learning[J]. Application of Electronic Technique,2021,47(8):87-90,96.
文獻標識碼: A
DOI:10.16157/j.issn.0258-7998.211341
中文引用格式: 蒲旭敏,吳超,楊小瓏. 基于深度學習的1-比特超大規(guī)模MIMO信道估計[J].電子技術應用,2021,47(8):87-90,96.
英文引用格式: Pu Xumin,Wu Chao,Yang Xiaolong. Channel estimation for 1-bit extremely massive MIMO via deep learning[J]. Application of Electronic Technique,2021,47(8):87-90,96.
Channel estimation for 1-bit extremely massive MIMO via deep learning
Pu Xumin1,2,Wu Chao1,2,Yang Xiaolong1,2
1.School of Communication and Information Engineering,Chongqing University of Posts and Telecommunications, Chongqing 400065,China; 2.Chongqing Key Laboratory of Mobile Communications Technology,Chongqing University of Posts and Telecommunications, Chongqing 400065,China
Abstract: Extremely massive multiple input multiple output(MIMO) has shown considerable potential in future mobile communications. However, the use of extremely massive aperture arrays will lead to spatial non-stationary channel conditions,and each antenna of the base station is equipped with a high-precision quantizer, the power consumption of the system will be greatly increased, which will hinder the widespread application of ultra-large-scale MIMO systems. Therefore, this article assumes that each antenna of the base station is equipped with a pair of 1-bit analog-to-digital converters(ADC), and uses the mapping relationship between the sub-array and the user to describe the non-stationary channel characteristics. Based on the powerful generalization ability of neural network(DNN), this paper designs a new generative supervised DNN model that can be trained with a reasonable number of pilots. The simulation results show that the proposed network can achieve better estimation performance with less pilots and achieve a good balance between performance and complexity.
Key words : channel estimation;deep learning;spatial non-stationary;1-bit ADC
0 引言
大規(guī)模多輸入多輸出(Multiple Input Multiple Output,MIMO)技術是第五代(Fifth Generation,5G)移動通信的關鍵技術之一。隨著天線陣列尺寸數(shù)量級的增加,形成了超大規(guī)模MIMO系統(tǒng)。通過增加天線陣列的尺寸,可以提升頻譜效率、能量效率和空間分辨率等,還有望獲取超高的數(shù)據(jù)速率和系統(tǒng)吞吐量。超大規(guī)模MIMO技術也因此成為第六代(Sixth Generation,6G)移動通信關鍵技術的候選[1]。然而,大口徑陣列的使用會造成不同的信道條件。當整個陣列的孔徑有限并且服務相同的用戶時,空間穩(wěn)定性是成立的。但是,對于大孔徑陣列,由于天線陣列不同區(qū)域所服務的用戶不同,接收功率的級別也因此不同,這稱為空間非平穩(wěn)性[2]。因此,可以引入子陣列和用戶可見區(qū)域(Visibility Region,VR)來描述信道非平穩(wěn)性[3]。
本文詳細內容請下載:http://ihrv.cn/resource/share/2000003707。
作者信息:
蒲旭敏1,2,吳 超1,2,楊小瓏1,2
(1.重慶郵電大學 通信與信息工程學院,重慶400065;2.重慶郵電大學 移動通信技術重慶市重點實驗室,重慶400065)

此內容為AET網(wǎng)站原創(chuàng),未經(jīng)授權禁止轉載。
