为应对大规模多输入多输出(Multiple⁃input multiple⁃output,MIMO)系统中信道状态信息(Channel state information,CSI)反馈开销的日益增长,基于深度学习的CSI反馈网络(如Transformer网络)受到了广泛的关注,是一种非常有应用前景的智能...为应对大规模多输入多输出(Multiple⁃input multiple⁃output,MIMO)系统中信道状态信息(Channel state information,CSI)反馈开销的日益增长,基于深度学习的CSI反馈网络(如Transformer网络)受到了广泛的关注,是一种非常有应用前景的智能传输技术。为此,本文提出了一种基于数据聚类的CSI反馈Transformer网络的简化方法,采用基于聚类的近似矩阵乘法(Approximate matrix multiplication,AMM)技术,以降低反馈过程中Transformer网络的计算复杂度。本文主要对Transformer网络的全连接层计算(等效为矩阵乘法),应用乘积量化(Product quantization,PQ)和MADDNESS等简化方法,分析了它们对计算复杂度和系统性能的影响,并针对神经网络数据的特点进行了算法优化。仿真结果表明,在适当的参数调整下,基于MADDNESS方法的CSI反馈网络性能接近精确矩阵乘法方法,同时可大幅降低计算复杂度。展开更多
针对卷积神经网络(convolutional neural network,CNN)计算复杂度高和内存占用大的问题,本文提出了一种基于轻量级CNN的信道状态信息(channel state information,CSI)图像指纹被动定位(LCNNLoc)方法.离线训练阶段,将幅值差矩阵和相位矩...针对卷积神经网络(convolutional neural network,CNN)计算复杂度高和内存占用大的问题,本文提出了一种基于轻量级CNN的信道状态信息(channel state information,CSI)图像指纹被动定位(LCNNLoc)方法.离线训练阶段,将幅值差矩阵和相位矩阵构造成类似于“RGB”的三通道特征图像;同时设计了一个轻量级CNN架构,利用特征图像作为该框架的输入进行训练,在训练结束时将CNN模型保存为指纹数据库.在线定位阶段,采用概率加权质心方法实现了实时的位置估计.实验结果表明,相较于传统方法,LCNNLoc不仅提升了定位精度,还降低了算法运行耗时.展开更多
Massive MIMO is one of tile enabling technologies tbr beyond 4G and 5G systems due to its ability to provide beamforming gain and reduce interference Dual-polarized antenna is widely adopted to accommodate a large num...Massive MIMO is one of tile enabling technologies tbr beyond 4G and 5G systems due to its ability to provide beamforming gain and reduce interference Dual-polarized antenna is widely adopted to accommodate a large number of antenna elements in limited space. However, current CSI（channel state information） feedback schemes developed in LTE for conventional MIMO systems are not efficient enough for massive MIMO systems since the overhead increases almost linearly with the number of antenna. Moreover, the codebook for massive MIMO will be huge and difficult to design with the LTE methodology. This paper proposes a novel CSI feedback scheme named layered Multi-paths Information based CSI Feedback （LMPIF）, which can achieve higher spectrum efficiency for dual-polarized antenna system with low feedback overhead. The MIMO channel is decomposed into long term components （multipath directions and amplitudes） and short term components （multipath phases）. The relationship between the two components and the optimal precoder is derived in closed form. To reduce the overhead, different granularities in feedback time have been applied for the long term components and short term components Link and system level simulation results prove that LMPIF can improve performance considerably with low CSI feedback overhead.展开更多
在设备到设备通信的车联网场景(Vehicle to Everything-Device to Device,V2X-D2D)下,信道的快速时变会导致基站(Base Station,BS)端通常无法获取完美信道状态信息(Channel State Information,CSI).为解决现有频谱分配方案不适用于V2X-...在设备到设备通信的车联网场景(Vehicle to Everything-Device to Device,V2X-D2D)下,信道的快速时变会导致基站(Base Station,BS)端通常无法获取完美信道状态信息(Channel State Information,CSI).为解决现有频谱分配方案不适用于V2X-D2D场景的问题,考虑车对车(Vehicle-to-Vehicle,V2V)链路可靠性、最大发射功率、频谱复用的约束,建立V2X的场景模型与通信模型.明确了在满足V2V链路可靠性的前提下,最大化车与基础设施(Vehicle to Infrastructure,V2I)链路的遍历容量的优化目标;在考虑信道快速时变影响的情况下,推导V2V链路的中断概率、V2I链路遍历容量的闭式表达式;针对一对一模式和一对多模式下的频谱分配问题,分别提出基于改进匈牙利算法的快速频谱分配方案和基于图着色-偏好列表的频谱分配方案.仿真结果表明:与现有算法相比,基于改进匈牙利算法的快速频谱分配方案接入率更高、复杂度更低,基于图着色-偏好列表的频谱分配方案也具有接入率、频谱利用率高的优势.展开更多
针对基于RSSI和CSI的指纹定位技术易受环境干扰、定位精度较低的问题,提出了一种基于RSSI指纹和相位修正信道状态信息(phase correct based channel state information,PC-CSI)指纹的加权融合指纹定位技术。基于PC-CSI的指纹定位在传统...针对基于RSSI和CSI的指纹定位技术易受环境干扰、定位精度较低的问题,提出了一种基于RSSI指纹和相位修正信道状态信息(phase correct based channel state information,PC-CSI)指纹的加权融合指纹定位技术。基于PC-CSI的指纹定位在传统基于CSI幅值的指纹定位基础上增加相位信息对定位结果进行修正,之后对RSSI指纹和PC-CSI指纹的定位结果加权重定位。实验结果表明,提出的加权融合指纹定位算法与基于CSI的主动定位算法相比,平均定位误差(mean position error,MPE)降低了36.2%,能满足室内定位需求。展开更多
信道状态信息(Channel State Information,CSI)反馈是大规模多输入多输出(Multiple-Input Multiple-Output,MIMO)系统的一个关键问题。大规模MIMO系统中基站天线数量巨大,CSI反馈出现了反馈开销大、反馈精度低等问题。为了降低反馈开销...信道状态信息(Channel State Information,CSI)反馈是大规模多输入多输出(Multiple-Input Multiple-Output,MIMO)系统的一个关键问题。大规模MIMO系统中基站天线数量巨大,CSI反馈出现了反馈开销大、反馈精度低等问题。为了降低反馈开销,提高反馈精度,采用深度学习方法,提出了一种基于特征融合的CSI反馈网络(Feature Fusion Net,FFNet)。利用基于注意力机制的特征融合在编码器中融合不同尺度的CSI特征,并在解码器中使用多通道多分辨率卷积网络以及通道重排,从而高精度地重建压缩后的CSI。仿真结果表明,与几种经典的深度学习CSI反馈方法相比,在室内和室外信道条件下,均具有更高的反馈精度。展开更多
This paper considers an intelligent reflecting surface(IRS)-assisted multiple-input multiple-output(MIMO)system.To maximize the average achievable rate(AAR)under outdated channel state information(CSI),we propose a tw...This paper considers an intelligent reflecting surface(IRS)-assisted multiple-input multiple-output(MIMO)system.To maximize the average achievable rate(AAR)under outdated channel state information(CSI),we propose a twin-timescale passive beamforming(PBF)and power allocation protocol which can reduce the IRS configuration and training overhead.Specifi-cally,the short-timescale power allocation is designed with the outdated precoder and fixed PBF.A new particle swarm opti-mization(PSO)-based long-timescale PBF optimization is pro-posed,where mini-batch channel samples are utilized to update the fitness function.Finally,simulation results demonstrate the effectiveness of the proposed method.展开更多
随着人们对人数统计需求的不断增长,基于信道状态信息(channel state information,CSI)的人流量监测技术因其易于部署、保护隐私和适用性强等优势而备受关注.然而,在现有的人流量监测工作中,人数识别的准确率容易受到人群密集程度的影响...随着人们对人数统计需求的不断增长,基于信道状态信息(channel state information,CSI)的人流量监测技术因其易于部署、保护隐私和适用性强等优势而备受关注.然而,在现有的人流量监测工作中,人数识别的准确率容易受到人群密集程度的影响.为了保证监测精度,通常只能在人群稀疏的情况下进行监测,这导致了基于CSI的人流量监测技术缺乏实用性.为了解决这一问题,提出了一种能够识别连续性人流的监测方法.该方法首先利用解卷绕和线性相位校正算法,对原始数据进行相位补偿并消除随机相位偏移;然后通过标准差和方差提取连续性人流数据中的有效数据包;最后将时域上的相位差信息作为特征信号输入到深度学习的CLDNN(convolutional,long short-term memory,deep neural network)中进行人数识别.经过实验测试,该方法在前后排行人距离不小于1 m的情况下,分别实现了室外96.7%和室内94.1%的准确率,优于现有的人流量监测方法.展开更多
该文针对闭环多用户MIMO-OFDM系统提出一种基于线性预测的低速率CSI(Channel State Information)反馈方法。根据相关带宽将OFDM子载波划分成多个子带,移动台对每个子带的CSI作线性预测,并对预测误差进行量化编码后反馈给基站;基站使用...该文针对闭环多用户MIMO-OFDM系统提出一种基于线性预测的低速率CSI(Channel State Information)反馈方法。根据相关带宽将OFDM子载波划分成多个子带,移动台对每个子带的CSI作线性预测,并对预测误差进行量化编码后反馈给基站;基站使用相同的线性预测滤波器将反馈来的预测误差恢复成CSI,然后在每个子带上通过迫零-波束赋形实现多用户空间复用。同时,该文还在采用注水定理分配发射功率的条件下,从理论上分析了下行链路信道容量。数值仿真结果显示,每个反馈数据的实部或虚部仅用1bit量化时,本方法仍能够以较高的精度恢复CSI。与目前3GPP LTE标准所采用的基于码书的反馈方案相比,该方法能够在反馈开销相同情况下,有效地抑制同信道干扰,大幅提高系统容量。展开更多
文摘针对卷积神经网络(convolutional neural network,CNN)计算复杂度高和内存占用大的问题,本文提出了一种基于轻量级CNN的信道状态信息(channel state information,CSI)图像指纹被动定位(LCNNLoc)方法.离线训练阶段,将幅值差矩阵和相位矩阵构造成类似于“RGB”的三通道特征图像;同时设计了一个轻量级CNN架构,利用特征图像作为该框架的输入进行训练,在训练结束时将CNN模型保存为指纹数据库.在线定位阶段,采用概率加权质心方法实现了实时的位置估计.实验结果表明,相较于传统方法,LCNNLoc不仅提升了定位精度,还降低了算法运行耗时.
基金supported by the National High-Tech R&D Program(863 Program 2015AA01A705)
文摘Massive MIMO is one of tile enabling technologies tbr beyond 4G and 5G systems due to its ability to provide beamforming gain and reduce interference Dual-polarized antenna is widely adopted to accommodate a large number of antenna elements in limited space. However, current CSI（channel state information） feedback schemes developed in LTE for conventional MIMO systems are not efficient enough for massive MIMO systems since the overhead increases almost linearly with the number of antenna. Moreover, the codebook for massive MIMO will be huge and difficult to design with the LTE methodology. This paper proposes a novel CSI feedback scheme named layered Multi-paths Information based CSI Feedback （LMPIF）, which can achieve higher spectrum efficiency for dual-polarized antenna system with low feedback overhead. The MIMO channel is decomposed into long term components （multipath directions and amplitudes） and short term components （multipath phases）. The relationship between the two components and the optimal precoder is derived in closed form. To reduce the overhead, different granularities in feedback time have been applied for the long term components and short term components Link and system level simulation results prove that LMPIF can improve performance considerably with low CSI feedback overhead.
文摘在设备到设备通信的车联网场景(Vehicle to Everything-Device to Device,V2X-D2D)下,信道的快速时变会导致基站(Base Station,BS)端通常无法获取完美信道状态信息(Channel State Information,CSI).为解决现有频谱分配方案不适用于V2X-D2D场景的问题,考虑车对车(Vehicle-to-Vehicle,V2V)链路可靠性、最大发射功率、频谱复用的约束,建立V2X的场景模型与通信模型.明确了在满足V2V链路可靠性的前提下,最大化车与基础设施(Vehicle to Infrastructure,V2I)链路的遍历容量的优化目标;在考虑信道快速时变影响的情况下,推导V2V链路的中断概率、V2I链路遍历容量的闭式表达式;针对一对一模式和一对多模式下的频谱分配问题,分别提出基于改进匈牙利算法的快速频谱分配方案和基于图着色-偏好列表的频谱分配方案.仿真结果表明:与现有算法相比,基于改进匈牙利算法的快速频谱分配方案接入率更高、复杂度更低,基于图着色-偏好列表的频谱分配方案也具有接入率、频谱利用率高的优势.
文摘针对基于RSSI和CSI的指纹定位技术易受环境干扰、定位精度较低的问题,提出了一种基于RSSI指纹和相位修正信道状态信息(phase correct based channel state information,PC-CSI)指纹的加权融合指纹定位技术。基于PC-CSI的指纹定位在传统基于CSI幅值的指纹定位基础上增加相位信息对定位结果进行修正,之后对RSSI指纹和PC-CSI指纹的定位结果加权重定位。实验结果表明,提出的加权融合指纹定位算法与基于CSI的主动定位算法相比,平均定位误差(mean position error,MPE)降低了36.2%,能满足室内定位需求。
基金supported by the National Natural Science Foundation of China(62271068)the Beijing Natural Science Foundation(L222046).
文摘This paper considers an intelligent reflecting surface(IRS)-assisted multiple-input multiple-output(MIMO)system.To maximize the average achievable rate(AAR)under outdated channel state information(CSI),we propose a twin-timescale passive beamforming(PBF)and power allocation protocol which can reduce the IRS configuration and training overhead.Specifi-cally,the short-timescale power allocation is designed with the outdated precoder and fixed PBF.A new particle swarm opti-mization(PSO)-based long-timescale PBF optimization is pro-posed,where mini-batch channel samples are utilized to update the fitness function.Finally,simulation results demonstrate the effectiveness of the proposed method.
文摘该文针对闭环多用户MIMO-OFDM系统提出一种基于线性预测的低速率CSI(Channel State Information)反馈方法。根据相关带宽将OFDM子载波划分成多个子带,移动台对每个子带的CSI作线性预测,并对预测误差进行量化编码后反馈给基站;基站使用相同的线性预测滤波器将反馈来的预测误差恢复成CSI,然后在每个子带上通过迫零-波束赋形实现多用户空间复用。同时,该文还在采用注水定理分配发射功率的条件下,从理论上分析了下行链路信道容量。数值仿真结果显示,每个反馈数据的实部或虚部仅用1bit量化时,本方法仍能够以较高的精度恢复CSI。与目前3GPP LTE标准所采用的基于码书的反馈方案相比,该方法能够在反馈开销相同情况下,有效地抑制同信道干扰,大幅提高系统容量。
