WO2024021440A1 - 一种迭代聚焦式毫米波一体化通信与感知方法 - Google Patents

一种迭代聚焦式毫米波一体化通信与感知方法 Download PDF

Info

Publication number
WO2024021440A1
WO2024021440A1 PCT/CN2022/138958 CN2022138958W WO2024021440A1 WO 2024021440 A1 WO2024021440 A1 WO 2024021440A1 CN 2022138958 W CN2022138958 W CN 2022138958W WO 2024021440 A1 WO2024021440 A1 WO 2024021440A1
Authority
WO
WIPO (PCT)
Prior art keywords
received signal
sensing
variance
channel
environmental information
Prior art date
Application number
PCT/CN2022/138958
Other languages
English (en)
French (fr)
Inventor
张朝阳
童欣
章一晗
Original Assignee
浙江大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 浙江大学 filed Critical 浙江大学
Priority to US18/358,975 priority Critical patent/US12019181B2/en
Publication of WO2024021440A1 publication Critical patent/WO2024021440A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01SRADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S7/00Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
    • G01S7/02Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
    • G01S7/41Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
    • G01S7/414Discriminating targets with respect to background clutter
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04WWIRELESS COMMUNICATION NETWORKS
    • H04W24/00Supervisory, monitoring or testing arrangements
    • H04W24/08Testing, supervising or monitoring using real traffic
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01SRADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S7/00Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
    • G01S7/003Transmission of data between radar, sonar or lidar systems and remote stations
    • G01S7/006Transmission of data between radar, sonar or lidar systems and remote stations using shared front-end circuitry, e.g. antennas
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04WWIRELESS COMMUNICATION NETWORKS
    • H04W4/00Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/02Services making use of location information
    • H04W4/021Services related to particular areas, e.g. point of interest [POI] services, venue services or geofences
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04WWIRELESS COMMUNICATION NETWORKS
    • H04W64/00Locating users or terminals or network equipment for network management purposes, e.g. mobility management

Definitions

  • the invention belongs to the field of wireless communication technology, and specifically relates to an iteratively focused millimeter wave integrated communication and sensing method.
  • the uplink wireless communication scenario users send communication signals to the base station for reception.
  • the sent signals are reflected and scattered by objects in the environment, so that the received signals of the base station contain environmental information.
  • a major difficulty in the design of integrated sensing and communication systems is how to deal with the potentially large number of unknown variables in the environment, so the sparsity of the target environment itself should be exploited. For example, in urban wireless communication scenarios, buildings are sparsely distributed within blocks.
  • electromagnetic waves propagate over a wide range, and any environmental scatterers covered by wireless signals will affect the propagation of electromagnetic waves.
  • the resources available for environment sensing such as the number of users, the number of base station receiving antennas, the number of subcarriers, etc., are limited.
  • the purpose of the present invention is to realize that in an uplink wireless communication scenario, a base station uses uplink data sent by multiple users to realize environment awareness.
  • the present invention uses pilot signals or other known data sequences of existing communication systems to perform sensing, and is compatible with existing communication systems to achieve integration of sensing and communication. Considering that the system resources used for sensing are limited and cannot sense the entire environment in a wide range, an iteratively focused millimeter wave environment sensing method for specific targets is proposed.
  • An iteratively focused millimeter wave integrated communication and sensing method which includes the following steps:
  • S1 in any time slot, use the base station to receive pilot sequence signals of a certain length sent by all active users in the environment to obtain the received signal, where the received signal is the signal after the pilot sequence signal is affected by the environment;
  • step S2 using the received signal in step S1, based on the multi-beam multi-carrier millimeter wave channel model, convert the environmental perception problem of the specific target into a compressed sensing reconstruction problem;
  • step S3 solve the compressed sensing reconstruction problem in step S2 based on the approximate message passing method, and obtain a rough initial result of environment sensing;
  • S4 based on the rough initial results of environmental perception, select a predetermined area from the overall environment as the focused area of interest, and use the background judgment method to divide and judge the target objects in the area of interest and remove background scattering outside the area of interest.
  • the influence of the object on the received signal is obtained to obtain the received signal corresponding to the target object;
  • step S5 calculate the environment sensing result based on the received signal corresponding to the target object obtained in step S4;
  • step S2 is specifically:
  • each pixel represents the environmental information in a small square around it with a size of l s ⁇ w s .
  • the size of the entire range of the environment be L s ⁇ W s.
  • the base station receiving antenna receives the received signal as follows:
  • j represents the complex symbol
  • n R is the receiving antenna number
  • d is the uniform linear array antenna spacing deployed by the base station
  • the free space propagation channel from the n u th user to the n s pixel is expressed as:
  • N s pixels to NR receiving antennas free space propagation channel Expressed as:
  • N s pixels is the guidance vector of N s pixels, is the channel gain from N s pixels to the base station;
  • n R is the receiving antenna number, is the angle of arrival of the n s pixel, Expressed as follows:
  • y is the received signal of all subcarriers
  • w is the beamforming vector of the uniform linear array receiving antenna of all subcarriers
  • x ROI is the environmental information in the area of interest
  • H NLOS is the non-line-of-sight channel of all subcarriers
  • H LOS is the line-of-sight channel of all subcarriers
  • s is the non-line-of-sight channel transmission signal of all subcarriers
  • is the slack variable
  • step S3 is specifically:
  • x represents the element in environmental information x, and all parameters are expressed as ⁇ ( ⁇ ) is the impulse function, ⁇ is the sparsity coefficient; ⁇ x ⁇ [0,1] and ⁇ x represent the mean and variance of the environmental information distribution respectively, and N( ⁇ ) represents the standard normal distribution;
  • N c the number of base stations
  • K the number of symbols
  • N f the number of subcarriers
  • y m is the m-th element of the received signal
  • step S4 is specifically:
  • S41 select a predetermined area from the overall environment as the focused area of interest based on the rough initial results of environmental perception and actual needs, and the target object is within the area of interest;
  • ⁇ i is the background scatterer detection threshold, and as the number of iterations increases, the detection threshold ⁇ i should decrease;
  • is a weight variable used to enhance the robustness of the iterative algorithm. As the number of iterations increases, the weight variable ⁇ should increase.
  • step S5 is specifically:
  • ⁇ back, i and ⁇ back represent the mean and variance of the background environment information distribution respectively
  • is the sparse coefficient
  • N( ⁇ ) represents the standard normal distribution
  • x back represents the background scatterer
  • the scatterer distribution in the selected ROI is Gaussian distribution without sparsity
  • ⁇ ROI and ⁇ ROI represent the mean and variance of ROI environmental information distribution respectively;
  • step S52 according to the prior probability formula obtained in step S51, set the prior probability p(x) of the environmental information inside and outside the area of interest in the current i+1th iteration round;
  • the present invention proposes a design method for an integrated system of millimeter wave sensing and communication systems using existing communication equipment, making full use of different systems based on users sending data Resources are used to realize focused environment sensing.
  • the present invention proposes an iterative focused environment sensing method to solve the problem of low accuracy of large-scale environment sensing due to insufficient system resources.
  • the invention improves the defect that the traditional compressed sensing algorithm cannot focus on a specific range of environmental variables.
  • the algorithm iteration process based on the compressed sensing reconstruction results of each step, the prior probability of the environmental variables is gradually estimated to achieve specific goals.
  • An iterative and progressive compressed sensing sparse reconstruction is proposed.
  • the algorithm of the present invention significantly improves the perception accuracy of specific targets and is superior to existing algorithms.
  • Figure 1 is a schematic diagram of a two-dimensional focused millimeter wave environment sensing scene provided by an exemplary embodiment
  • Figure 2 is a flow chart of an iterative algorithm provided by an exemplary embodiment
  • Figure 3 is a diagram illustrating the relationship between the number of users and the environment sensing accuracy MSE when comparing the present invention with other compressed sensing reconstruction algorithms provided by an exemplary embodiment
  • Figure 4 is a diagram illustrating the relationship between the number of subcarriers and the environmental sensing accuracy MSE when comparing the present invention with other compressed sensing reconstruction algorithms provided by an exemplary embodiment
  • the present invention provides an iteratively focused millimeter wave integrated communication and sensing method, which includes the following steps:
  • S1 in any time slot, use the base station to receive pilot sequence signals of a certain length sent by all active users in the environment to obtain the received signal, where the received signal is the signal after the pilot sequence signal is affected by the environment;
  • step S2 using the received signal in step S1, based on the multi-beam multi-carrier millimeter wave channel model, convert the environmental perception problem of the specific target into a compressed sensing reconstruction problem;
  • step S2 is specifically:
  • each pixel represents the environmental information in a small square with a size of l s ⁇ w s around it.
  • N s L/l s ⁇ W/w s ; each pixel may be empty inside, or there may be scatterers; use a scattering coefficient to represent the scattering coefficient of the small cube where the n s point cloud point is located. If the interior of the small cube is empty, then Therefore, the environmental information of the entire room can be expressed as
  • the base station receiving antenna receives the received signal as follows:
  • j represents the complex symbol
  • n R is the receiving antenna number
  • d is the uniform linear array antenna spacing deployed by the base station
  • the free space propagation channel from the n u th user to the n s pixel is expressed as:
  • N s pixels to NR receiving antennas free space propagation channel Expressed as:
  • N s pixels is the guidance vector of N s pixels, is the channel gain from N s pixels to the base station;
  • n R is the receiving antenna number, is the angle of arrival of the n s pixel, Expressed as follows:
  • y is the received signal of all subcarriers
  • w is the beamforming vector of the uniform linear array receiving antenna of all subcarriers
  • x ROI is the environmental information in the area of interest
  • H NLOS is the non-line-of-sight channel of all subcarriers
  • H LOS is the line-of-sight channel of all subcarriers
  • s is the non-line-of-sight channel transmission signal of all subcarriers
  • is the slack variable
  • step S3 solve the compressed sensing reconstruction problem in step S2 based on the approximate message passing method, and obtain a rough initial result of environmental sensing;
  • step S3 is specifically:
  • x represents the element in environmental information x, and all parameters are expressed as ⁇ ( ⁇ ) is the impulse function, ⁇ is the sparsity coefficient; ⁇ x ⁇ [0,1] and ⁇ x represent the mean and variance of the environmental information distribution respectively, and N( ⁇ ) represents the standard normal distribution;
  • N c the number of base stations
  • K the number of symbols
  • N f the number of subcarriers
  • y m is the m-th element of the received signal
  • S4 based on the rough initial results of environmental perception, select a predetermined area from the overall environment as the focused area of interest, and use the background judgment method to divide and judge the target objects in the area of interest and remove background scattering outside the area of interest.
  • the influence of the object on the received signal is obtained to obtain the received signal corresponding to the target object;
  • step S4 is specifically:
  • S41 select a predetermined area from the overall environment as a focused area of interest based on the rough initial results of environmental perception and actual needs, and the target object is within the area of interest;
  • ⁇ i is the background scatterer detection threshold, and as the number of iterations increases, the detection threshold ⁇ i should decrease;
  • is a weight variable used to enhance the robustness of the iterative algorithm. As the number of iterations increases, the weight variable ⁇ should increase.
  • step S5 calculate the environment sensing result based on the received signal corresponding to the target object obtained in step S4;
  • step S5 is specifically:
  • ⁇ back, i and ⁇ back represent the mean and variance of the background environment information distribution respectively
  • is the sparse coefficient
  • N( ⁇ ) represents the standard normal distribution
  • x back represents the background scatterer
  • the scatterer distribution in the selected ROI is Gaussian distribution without sparsity
  • ⁇ ROI and ⁇ ROI represent the mean and variance of ROI environmental information distribution respectively;
  • step S52 according to the prior probability formula obtained in step S51, set the prior probability p(x) of the environmental information inside and outside the area of interest in the current i+1th iteration round;
  • the design method of the integrated system of millimeter wave sensing and communication system using existing communication equipment makes full use of different system resources to achieve focused environment sensing based on the data sent by the user, and solves the environmental sensing problem. Converted to a compressed sensing reconstruction problem, and then based on an approximate message passing algorithm, an initial rough perception of the environment is achieved.
  • the present invention divides and judges the target object and removes the influence of background scatterers on the received signal. Finally, This iteratively removes background scatterers to obtain more accurate focused sensing results of the target object.
  • the proposed iterative focused environment sensing method solves the problem of large-scale problems caused by insufficient system resources.
  • the problem of low environmental perception accuracy improves the traditional compressed sensing algorithm's inability to focus on a specific range of environmental variables, and provides an efficient environment sensing method for the design of future integrated sensing and communication systems.
  • the compressed sensing reconstruction result of each step gradually estimates the prior probability of environmental variables, and achieves an iterative and progressive compressed sensing sparse reconstruction for a specific target.
  • the algorithm of the present invention significantly improves the perception accuracy of specific targets and is superior to existing algorithms.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Radar Systems Or Details Thereof (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本发明公开了一种迭代聚焦式毫米波一体化通信与感知方法,将环境感知问题转换为压缩感知重构问题,随后基于一种近似消息传递算法实现了对环境初始的粗略感知,根据一种背景判断方法,本发明划分判断目标物体并去除背景散射体对接收信号的影响,最后由此反复迭代的去除背景散射体,以获得目标物体更准确的聚焦感知结果。相比现有的环境感知重构算法,本发明的迭代聚焦式毫米波环境感知算法显著提升了环境感知的准确度,解决了由于系统资源有限从而无法准确感知大范围环境的问题,为未来感知通信一体化系统设计提供了一种高效的环境感知方法。

Description

一种迭代聚焦式毫米波一体化通信与感知方法 技术领域
本发明属于无线通信技术领域,具体涉及一种迭代聚焦式毫米波一体化通信与感知方法。
背景技术
在目前的无线通信领域中,超大规模多输入多输出(MIMO)技术、智能反射面(IRS)、无线人工智能(AI)等创新无线通信技术的出现,为未来无线通信系统设计的设计提供了更多可能性。在可预见的未来无线通信应用场景中,自动驾驶、智能机器人导航、无人机控制等技术不仅需要无线宽带连接,还需要准确的环境信息,包括环境中物体的位置、形状、电磁特性等。因此,感知通信一体化(ISAC)技术作为第六代(6G)无线通信系统的研究热点,旨在利用无线通信设备和基础框架实现环境感知。
在上行链路无线通信场景中,用户发送通信信号至基站接收,发送的信号被环境中的物体反射、散射,从而使得基站的接收信号中蕴含着环境信息。感知通信一体化系统设计的一大难点在于如何应对环境中潜在的大量未知变量,因此应该利用目标环境本身的稀疏性。例如,在城市无线通信场景中,建筑物稀疏的分布在街区内。除此之外,在无线通信应用场景中,电磁波的传播范围很广,任何被无线信号覆盖的环境散射体都会影响电磁波的传播。然而,可用于环境感知的资源,如用户数量、基站接收天线数量、子载波数量等是有限的。即使利用了环境信息的稀疏性,环境感知仍然面临由于大量环境信息变量而导致的系统资源开销大的问题。目前,现有的毫米波通信与感知一体化方法没有考虑到有限系统资源带给感知算法的影响。在有限的系统资源条件下,往往只能实现对环境粗略的感知,获得模糊的成像结果,亟需利用有限的资源实现对 特定目标的聚焦,从而获得精确的环境感知结果。综上所述如何使用有限的系统资源实现对特定目标的精准感知,具有较高的研究难度和现实意义。
发明内容
针对上述现有技术的不足,本发明的目的是为了实现在上行链路无线通信场景中,基站利用多个用户发送的上行数据实现环境感知。本发明利用现有通信系统的导频信号或其它已知数据序列来进行感知,可与现有通信系统兼容,实现感知通信一体化。考虑到用于感知的系统资源有限,无法大范围的感知全部的环境,提出了一种对特定目标的迭代聚焦式毫米波环境感知方法。
本发明的目的是通过以下技术方案实现的:
提供一种迭代聚焦式毫米波一体化通信与感知方法,该方法包括以下步骤:
S1,在任意时隙中,利用基站接收环境中的所有活跃用户发送的一定长度的导频序列信号,得到接收信号,其中,接收信号是导频序列信号受环境影响之后的信号;
S2,利用所述步骤S1中的接收信号,基于多波束多载波毫米波信道模型,将对特定目标的环境感知问题转换为压缩感知重构问题;
S3,基于近似消息传递方法求解所述步骤S2中的压缩感知重构问题,得到粗略的环境感知初始结果;
S4,基于粗略的环境感知初始结果,从整体环境中选择预定区域作为聚焦的感兴趣区域,并根据背景判断方法,划分、判断感兴趣区域中的目标物体且去除感兴趣区域之外的背景散射体对接收信号的影响,得到目标物体对应的接收信号;
S5,基于所述步骤S4中获得的目标物体对应的接收信号,计算环境感知结果;
S6,重复按顺序骤S4和步骤S5直至算法收敛,得到最终环境感知结果。
进一步地,所述步骤S2具体为:
S21,将步骤S1中的接收信号中的环境信息离散化为像素,每个像素代表 其周围大小为l s×w s的小正方形内的环境信息,令全部范围环境大小为L s×W s,则像素总数为N s=L/l s×W/w s;每一个像素内部可能是空的,也可能是存在散射体;使用一个散射系数
Figure PCTCN2022138958-appb-000001
来表示第n s个点云点的所在小立方体的散射系数,若小立方体内部是空的,则
Figure PCTCN2022138958-appb-000002
因此,整个房间的环境信息可以表示为
Figure PCTCN2022138958-appb-000003
S22,采用多波束多载波毫米波信道模型,在第n f个子载波频率上,基站接收天线接收接收信号表示如下:
Figure PCTCN2022138958-appb-000004
其中,
Figure PCTCN2022138958-appb-000005
表示N c个基站RF链路的长度为K码元的接收信号,
Figure PCTCN2022138958-appb-000006
是基站N R个均匀线阵接收天线的波束成形矢量,δ是散射系数的归一化系数,根据像素大小l s×w s选择,归一化系数是描述电磁波接收面积与接收功率之间的物理关系,
Figure PCTCN2022138958-appb-000007
表示N u个用户发送长度为K码元的导频,n为噪声;
Figure PCTCN2022138958-appb-000008
表示在第n f个子载波频率上,N u个用户到N R个接收天线自由空间传播信道;
Figure PCTCN2022138958-appb-000009
为在第n f个子载波上的非视距信道;
其中,
Figure PCTCN2022138958-appb-000010
表示为:
Figure PCTCN2022138958-appb-000011
其中,
Figure PCTCN2022138958-appb-000012
为N u个用户的导向矢量,
Figure PCTCN2022138958-appb-000013
为N u个用户至基站的信道增益;
Figure PCTCN2022138958-appb-000014
其中,j表示复数符号,n R为接收天线编号,
Figure PCTCN2022138958-appb-000015
为第n u个用户的到达角,d为基站部署的均匀线阵天线间距,
Figure PCTCN2022138958-appb-000016
为波长;
Figure PCTCN2022138958-appb-000017
表示如下:
Figure PCTCN2022138958-appb-000018
其中,
Figure PCTCN2022138958-appb-000019
Figure PCTCN2022138958-appb-000020
分别为第n u个用户的至基站之间的信道幅度增益和相移;
在第n f个子载波频率上,第n u个用户到第n s个像素的自由空间传播信道
Figure PCTCN2022138958-appb-000021
为表示为:
Figure PCTCN2022138958-appb-000022
其中,
Figure PCTCN2022138958-appb-000023
Figure PCTCN2022138958-appb-000024
分别为第n u个用户的至第n s个像素之间的信道幅度增益和相移;
在第n f个子载波频率上,N s个像素到N R个接收天线自由空间传播信道
Figure PCTCN2022138958-appb-000025
表示为:
Figure PCTCN2022138958-appb-000026
其中,
Figure PCTCN2022138958-appb-000027
为N s个像素的导向矢量,
Figure PCTCN2022138958-appb-000028
为N s个像素至基站的信道增益;
Figure PCTCN2022138958-appb-000029
其中,n R为接收天线编号,
Figure PCTCN2022138958-appb-000030
为第n s个像素的到达角,
Figure PCTCN2022138958-appb-000031
表示如下:
Figure PCTCN2022138958-appb-000032
其中,
Figure PCTCN2022138958-appb-000033
Figure PCTCN2022138958-appb-000034
分别为第n s个像素的至基站之间的信道幅度增益和相移;
S23,环境信息的估计结果表示为
Figure PCTCN2022138958-appb-000035
表示如下:
Figure PCTCN2022138958-appb-000036
其中,y为全部子载波的接收信号,w为全部子载波的均匀线阵接收天线的波束成形矢量,x ROI为感兴趣区域内的环境信息,H NLOS为全部子载波的非视距信道,H LOS为全部子载波的视距信道,s为全部子载波的非视距信道发送信号,ε为松弛变量;
由于直视信道的自由空间信道系数可以通过数值模型来估计,在第n f个子载波频率上,将包含未知的环境信息的部分接收信号
Figure PCTCN2022138958-appb-000037
表示为,
Figure PCTCN2022138958-appb-000038
将联合N f个子载波的数据,将特定目标的迭代聚焦式环境感知问题转化为压缩感知重构问题方程如下:
Figure PCTCN2022138958-appb-000039
进一步地,所述步骤S3具体为:
S31,首先设置初始粗略的环境感知先验概率,令环境信息分布为伯努利-高斯分布,其概率密度函数p x(x|q)表示为:
p x(x|q)=(1-λ)δ(x)+λN(x|θ xx)
其中,x表示环境信息x中的元素,所有参数表示为
Figure PCTCN2022138958-appb-000040
δ(·)是冲激函数,λ是稀疏系数;θ x∈[0,1]和σ x分别表示环境信息分布的均值和方差,N(·)表示标准正态分布;
S32,近似消息传递算法参数初始化,令输入函数g in(·),g′ in(·)和输出函数g out(·),g′ out(·)分别如下
Figure PCTCN2022138958-appb-000041
Figure PCTCN2022138958-appb-000042
Figure PCTCN2022138958-appb-000043
Figure PCTCN2022138958-appb-000044
Figure PCTCN2022138958-appb-000045
其中,
Figure PCTCN2022138958-appb-000046
为输入变量,σ w为噪声方差;
令迭代次数t G=0,残差
Figure PCTCN2022138958-appb-000047
稀疏向量估计均值
Figure PCTCN2022138958-appb-000048
稀疏向量估计方差
Figure PCTCN2022138958-appb-000049
S33,令M=N cN fK,N c是基站数量,K是码元数量,N f是子载波数量,对于m=1,2,…,M,计算变量z m的估计的均值
Figure PCTCN2022138958-appb-000050
和方差
Figure PCTCN2022138958-appb-000051
Figure PCTCN2022138958-appb-000052
Figure PCTCN2022138958-appb-000053
Figure PCTCN2022138958-appb-000054
S34,对于m=1,2,…,M,计算残差的均值
Figure PCTCN2022138958-appb-000055
和方差
Figure PCTCN2022138958-appb-000056
Figure PCTCN2022138958-appb-000057
Figure PCTCN2022138958-appb-000058
其中,y m是接收信号的第m个元素;
S35,对于n s=1,2,…,N s,计算
Figure PCTCN2022138958-appb-000059
的观测均值
Figure PCTCN2022138958-appb-000060
和方差
Figure PCTCN2022138958-appb-000061
Figure PCTCN2022138958-appb-000062
Figure PCTCN2022138958-appb-000063
S36,对于n s=1,2,…,N s,计算
Figure PCTCN2022138958-appb-000064
的观测均值
Figure PCTCN2022138958-appb-000065
和方差
Figure PCTCN2022138958-appb-000066
Figure PCTCN2022138958-appb-000067
Figure PCTCN2022138958-appb-000068
S37,t G=t G+1,重复执行步骤S33至步骤S36直到达到收敛条件
Figure PCTCN2022138958-appb-000069
ε G为误差容限;
S38,将稀疏变量
Figure PCTCN2022138958-appb-000070
作为环境信息x的粗略的环境感知初始结果。
进一步地,所述步骤S4具体为:
S41,根据粗略的环境感知初始结果和实际需求从整体环境中选择预定区域作为聚焦的感兴趣区域,目标物体在感兴趣区域内;
S42,在第i次迭代中,感兴趣区域外的背景散射体
Figure PCTCN2022138958-appb-000071
被检测如下:
Figure PCTCN2022138958-appb-000072
其中,
Figure PCTCN2022138958-appb-000073
代表第i次迭代中的结果,γ i为背景散射体检测门限,随着迭代次数的增加,检测阈值γ i应该降低;
S43,从接收信号中去除背景散射部分,得到第i+1次迭代ROI内的目标物体的接收信号
Figure PCTCN2022138958-appb-000074
Figure PCTCN2022138958-appb-000075
其中α是一个权重变量,用于增强迭代算法的鲁棒性,随着迭代次数的增加,权重变量α应该升高。
进一步地,所述步骤S5具体为:
S51,设置迭代聚焦过程中环境信息的先验概率,在第i次迭代中,假设背景散射体服从伯努利高斯分布,先验概率公式p(x back)如下:
Figure PCTCN2022138958-appb-000076
其中θ back,i和σ back分别表示背景环境信息分布的均值和方差,λ是稀疏系数,N(·)表示标准正态分布,x back表示背景散射体;
所选ROI中的散射体分布为高斯分布,没有稀疏性;
Figure PCTCN2022138958-appb-000077
其中θ ROI和σ ROI分别表示ROI环境信息分布的均值和方差;
S52,根据步骤S51中得到的先验概率公式,设置当前第i+1个迭代轮次的感兴趣区域内外环境信息的先验概率p(x);
S53,近似消息传递算法参数初始化,令输入函数g in(·),g′ in(·)和输出函数g out(·),g′ out(·)分别如下:
Figure PCTCN2022138958-appb-000078
Figure PCTCN2022138958-appb-000079
Figure PCTCN2022138958-appb-000080
Figure PCTCN2022138958-appb-000081
Figure PCTCN2022138958-appb-000082
令迭代次数t G=0,残差
Figure PCTCN2022138958-appb-000083
稀疏向量估计均值
Figure PCTCN2022138958-appb-000084
稀疏向量估计方差
Figure PCTCN2022138958-appb-000085
S54,令M=N cN fK,对于m=1,2,…,M,计算z m的估计的均值
Figure PCTCN2022138958-appb-000086
和方差
Figure PCTCN2022138958-appb-000087
具体如下:
Figure PCTCN2022138958-appb-000088
Figure PCTCN2022138958-appb-000089
Figure PCTCN2022138958-appb-000090
S55,对于m=1,2,…,M,计算残差的均值
Figure PCTCN2022138958-appb-000091
Figure PCTCN2022138958-appb-000092
Figure PCTCN2022138958-appb-000093
具体如下:
Figure PCTCN2022138958-appb-000094
Figure PCTCN2022138958-appb-000095
其中
Figure PCTCN2022138958-appb-000096
为S43中得到的接收信号的第m个元素;
S56,对于n s=1,2,…,N s,计算
Figure PCTCN2022138958-appb-000097
的观测均值
Figure PCTCN2022138958-appb-000098
和方差
Figure PCTCN2022138958-appb-000099
具体如下:
Figure PCTCN2022138958-appb-000100
Figure PCTCN2022138958-appb-000101
S57,对于n s=1,2,…,N s,计算
Figure PCTCN2022138958-appb-000102
的观测均值
Figure PCTCN2022138958-appb-000103
和方差
Figure PCTCN2022138958-appb-000104
具体如下:
Figure PCTCN2022138958-appb-000105
Figure PCTCN2022138958-appb-000106
S58,t G=t G+1,重复执行步骤S54至步骤S57直到达到收敛条件
Figure PCTCN2022138958-appb-000107
S59,将上述步骤估计所得到的稀疏变量
Figure PCTCN2022138958-appb-000108
作为本轮迭代最终环境感知结果。
本发明的有益效果是:在上行链路无线通信场景中,本发明提出了一种利用现有通信设备进行毫米波感知与通信系一体化系统的设计方法,基于用户发送数据充分利用不同的系统资源来实现聚焦式环境感知,本发明提出迭代聚焦式环境感知方法解决了由于系统资源不足而导致大范围环境感知精确度低的问题。本发明改进了传统压缩感知算法无法实现对环境变量特定范围聚焦的缺陷,在算法迭代过程中,根据每一步的压缩感知重构结果,逐步对环境变量的先验概率进行估计,对特定目标实现了一种迭代渐进式的压缩感知稀疏重构。在相 同的系统资源开销的基础上,本发明的算法显著提升了对特定目标的感知准确度,并优于现有的算法。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案,下面将对实施例中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本发明的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动性的前提下,还可以根据这些附图获得其他的附图。
图1为一示例性实施例提供的二维聚焦式毫米波环境感知场景示意图;
图2为一示例性实施例提供的迭代算法流程图;
图3为一示例性实施例提供的将本发明与其他压缩感知重构算法相比较时,用户数量与环境感知准确度MSE的关系图;
图4为一示例性实施例提供的将本发明与其他压缩感知重构算法相比较时,子载波数量与环境感知准确度MSE的关系图;
具体实施方式
为了更好的理解本申请的技术方案,下面结合附图对本申请实施例进行详细描述。
应当明确,所描述的实施例仅仅是本申请一部分实施例,而不是全部的实施例。基于本申请中的实施例,本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其它实施例,都属于本申请保护的范围。
在本申请实施例中使用的术语是仅仅出于描述特定实施例的目的,而非旨在限制本申请。在本申请实施例和所附权利要求书中所使用的单数形式的“一种”、“所述”和“该”也旨在包括多数形式,除非上下文清楚地表示其他含义。
考虑一个单小区上行链路通信系统,其中一个多天线BS服务于多个单天线用户。为了感知特定目标,所有用户同时向基站发送上行通信信号。由于环 境中的大量散射,每个用户的发射信号通过多条路径传播。因此,基站接收到的信号包含丰富的环境散射信息,目的是让基站对接收到的信号进行处理以实现环境感知。如图1所示,考虑一个简化的二维场景模型,同样可以推导出三维场景的分析方法。灰色目标散射体是想要准确成像的散射体,其余的是我们不关心的背景散射体。
本发明提供一种迭代聚焦式毫米波一体化通信与感知方法,该方法包括以下步骤:
S1,在任意时隙中,利用基站接收环境中的所有活跃用户发送的一定长度的导频序列信号,得到接收信号,其中,接收信号是导频序列信号受环境影响之后的信号;
S2,利用步骤S1中的接收信号,基于多波束多载波毫米波信道模型,将对特定目标的环境感知问题转换为压缩感知重构问题;
在一实施例中,步骤S2具体为:
S21,将步骤S1中的接收信号中的环境信息离散化为像素,每个像素代表其周围大小为l s×w s的小正方形内的环境信息,令全部范围环境大小为L s×W s,则像素总数为N s=L/l s×W/w s;每一个像素内部可能是空的,也可能是存在散射体;使用一个散射系数
Figure PCTCN2022138958-appb-000109
来表示第n s个点云点的所在小立方体的散射系数,若小立方体内部是空的,则
Figure PCTCN2022138958-appb-000110
因此,整个房间的环境信息可以表示为
Figure PCTCN2022138958-appb-000111
S22,采用多波束多载波毫米波信道模型,在第n f个子载波频率上,基站接收天线接收接收信号表示如下:
Figure PCTCN2022138958-appb-000112
其中,
Figure PCTCN2022138958-appb-000113
表示N c个基站RF链路的长度为K码元的接收信号,
Figure PCTCN2022138958-appb-000114
是基站N R个均匀线阵接收天线的波束成形矢量,δ是散射系数的归一化系数,根据像素大小l s×w s选择,归一化系数是描述电磁波接收面积与接收功率之间的物理关系,
Figure PCTCN2022138958-appb-000115
表示N u个用户发送长度为K码元的导频,n为噪声;
Figure PCTCN2022138958-appb-000116
表示在第n f个子载波频率上,N u个用户到N R个接收天线自由空间传播信道;
Figure PCTCN2022138958-appb-000117
为在第n f个子载波上的非视距信道;
其中,
Figure PCTCN2022138958-appb-000118
表示为:
Figure PCTCN2022138958-appb-000119
其中,
Figure PCTCN2022138958-appb-000120
为N u个用户的导向矢量,
Figure PCTCN2022138958-appb-000121
为N u个用户至基站的信道增益;
Figure PCTCN2022138958-appb-000122
其中,j表示复数符号,n R为接收天线编号,
Figure PCTCN2022138958-appb-000123
为第n u个用户的到达角,d为基站部署的均匀线阵天线间距,
Figure PCTCN2022138958-appb-000124
为波长;
Figure PCTCN2022138958-appb-000125
表示如下:
Figure PCTCN2022138958-appb-000126
其中,
Figure PCTCN2022138958-appb-000127
Figure PCTCN2022138958-appb-000128
分别为第n u个用户的至基站之间的信道幅度增益和相移;
在第n f个子载波频率上,第n u个用户到第n s个像素的自由空间传播信道
Figure PCTCN2022138958-appb-000129
为表示为:
Figure PCTCN2022138958-appb-000130
其中,
Figure PCTCN2022138958-appb-000131
Figure PCTCN2022138958-appb-000132
分别为第n u个用户的至第n s个像素之间的信道幅度增益和相移;
在第n f个子载波频率上,N s个像素到N R个接收天线自由空间传播信道
Figure PCTCN2022138958-appb-000133
表示为:
Figure PCTCN2022138958-appb-000134
其中,
Figure PCTCN2022138958-appb-000135
为N s个像素的导向矢量,
Figure PCTCN2022138958-appb-000136
为N s个像素至基站的信道增益;
Figure PCTCN2022138958-appb-000137
其中,n R为接收天线编号,
Figure PCTCN2022138958-appb-000138
为第n s个像素的到达角,
Figure PCTCN2022138958-appb-000139
表示如下:
Figure PCTCN2022138958-appb-000140
其中,
Figure PCTCN2022138958-appb-000141
Figure PCTCN2022138958-appb-000142
分别为第n s个像素的至基站之间的信道幅度增益和相移;
S23,环境信息的估计结果表示为
Figure PCTCN2022138958-appb-000143
表示如下:
Figure PCTCN2022138958-appb-000144
其中,y为全部子载波的接收信号,w为全部子载波的均匀线阵接收天线的波束成形矢量,x ROI为感兴趣区域内的环境信息,H NLOS为全部子载波的非视距信道,H LOS为全部子载波的视距信道,s为全部子载波的非视距信道发送信号,ε为松弛变量;
由于直视信道的自由空间信道系数可以通过数值模型来估计,在第n f个子载波频率上,将包含未知的环境信息的部分接收信号
Figure PCTCN2022138958-appb-000145
表示为,
Figure PCTCN2022138958-appb-000146
将联合N f个子载波的数据,将特定目标的迭代聚焦式环境感知问题转化为压缩感知重构问题方程如下:
Figure PCTCN2022138958-appb-000147
S3,基于近似消息传递方法求解步骤S2中的压缩感知重构问题,得到粗略的环境感知初始结果;
在一实施例中,步骤S3具体为:
S31,首先设置初始粗略的环境感知先验概率,令环境信息分布为伯努利-高斯分布,其概率密度函数p x(x|q)表示为:
p x(x|q)=(1-λ)δ(x)+λN(x|θ xx)
其中,x表示环境信息x中的元素,所有参数表示为
Figure PCTCN2022138958-appb-000148
δ(·)是冲激函数,λ是稀疏系数;θ x∈[0,1]和σ x分别表示环境信息分布的均值和方差,N(·)表示标准正态分布;
S32,近似消息传递算法参数初始化,令输入函数g in(·),g′ in(·)和输出函数g out(·),g′ out(·)分别如下
Figure PCTCN2022138958-appb-000149
Figure PCTCN2022138958-appb-000150
Figure PCTCN2022138958-appb-000151
Figure PCTCN2022138958-appb-000152
Figure PCTCN2022138958-appb-000153
其中,
Figure PCTCN2022138958-appb-000154
为输入变量,σ w为噪声方差;
令迭代次数t G=0,残差
Figure PCTCN2022138958-appb-000155
稀疏向量估计均值
Figure PCTCN2022138958-appb-000156
稀疏向量估计方差
Figure PCTCN2022138958-appb-000157
S33,令M=N cN fK,N c是基站数量,K是码元数量,N f是子载波数量,对于m=1,2,…,M,计算变量z m的估计的均值
Figure PCTCN2022138958-appb-000158
和方差
Figure PCTCN2022138958-appb-000159
Figure PCTCN2022138958-appb-000160
Figure PCTCN2022138958-appb-000161
Figure PCTCN2022138958-appb-000162
S34,对于m=1,2,…,M,计算残差的均值
Figure PCTCN2022138958-appb-000163
和方差
Figure PCTCN2022138958-appb-000164
Figure PCTCN2022138958-appb-000165
Figure PCTCN2022138958-appb-000166
其中,y m是接收信号的第m个元素;
S35,对于n s=1,2,…,N s,计算
Figure PCTCN2022138958-appb-000167
的观测均值
Figure PCTCN2022138958-appb-000168
和方差
Figure PCTCN2022138958-appb-000169
Figure PCTCN2022138958-appb-000170
Figure PCTCN2022138958-appb-000171
S36,对于n s=1,2,…,N s,计算
Figure PCTCN2022138958-appb-000172
的观测均值
Figure PCTCN2022138958-appb-000173
和方差
Figure PCTCN2022138958-appb-000174
Figure PCTCN2022138958-appb-000175
Figure PCTCN2022138958-appb-000176
S37,t G=t G+1,重复执行步骤S33至步骤S36直到达到收敛条件
Figure PCTCN2022138958-appb-000177
ε G为误差容限;
S38,将稀疏变量
Figure PCTCN2022138958-appb-000178
作为环境信息x的粗略的环境感知初始结果。
S4,基于粗略的环境感知初始结果,从整体环境中选择预定区域作为聚焦的感兴趣区域,并根据背景判断方法,划分、判断感兴趣区域中的目标物体且去除感兴趣区域之外的背景散射体对接收信号的影响,得到目标物体对应的接收信号;
在一实施例中,步骤S4具体为:
S41,根据粗略的环境感知初始结果和实际需求从整体环境中选择预定区域 作为聚焦的感兴趣区域,目标物体在感兴趣区域内;
S42,在第i次迭代中,感兴趣区域外的背景散射体
Figure PCTCN2022138958-appb-000179
被检测如下:
Figure PCTCN2022138958-appb-000180
其中,
Figure PCTCN2022138958-appb-000181
代表第i次迭代中的结果,γ i为背景散射体检测门限,随着迭代次数的增加,检测阈值γ i应该降低;
S43,从接收信号中去除背景散射部分,得到第i+1次迭代ROI内的目标物体的接收信号
Figure PCTCN2022138958-appb-000182
Figure PCTCN2022138958-appb-000183
其中α是一个权重变量,用于增强迭代算法的鲁棒性,随着迭代次数的增加,权重变量α应该升高。
S5,基于步骤S4中获得的目标物体对应的接收信号,计算环境感知结果;
在一实施例中,步骤S5具体为:
S51,设置迭代聚焦过程中环境信息的先验概率,在第i次迭代中,假设背景散射体服从伯努利高斯分布,先验概率公式p(x back)如下:
Figure PCTCN2022138958-appb-000184
其中θ back,i和σ back分别表示背景环境信息分布的均值和方差,λ是稀疏系数,N(·)表示标准正态分布,x back表示背景散射体;
所选ROI中的散射体分布为高斯分布,没有稀疏性;
Figure PCTCN2022138958-appb-000185
其中θ ROI和σ ROI分别表示ROI环境信息分布的均值和方差;
S52,根据步骤S51中得到的先验概率公式,设置当前第i+1个迭代轮次的感兴趣区域内外环境信息的先验概率p(x);
S53,近似消息传递算法参数初始化,令输入函数g in(·),g′ in(·)和输出函数 g out(·),g′ out(·)分别如下:
Figure PCTCN2022138958-appb-000186
Figure PCTCN2022138958-appb-000187
Figure PCTCN2022138958-appb-000188
Figure PCTCN2022138958-appb-000189
Figure PCTCN2022138958-appb-000190
令迭代次数t G=0,残差
Figure PCTCN2022138958-appb-000191
稀疏向量估计均值
Figure PCTCN2022138958-appb-000192
稀疏向量估计方差
Figure PCTCN2022138958-appb-000193
S54,令M=N cN fK,对于m=1,2,…,M,计算z m的估计的均值
Figure PCTCN2022138958-appb-000194
和方差
Figure PCTCN2022138958-appb-000195
具体如下:
Figure PCTCN2022138958-appb-000196
Figure PCTCN2022138958-appb-000197
Figure PCTCN2022138958-appb-000198
S55,对于m=1,2,…,M,计算残差的均值
Figure PCTCN2022138958-appb-000199
和方差
Figure PCTCN2022138958-appb-000200
具体如下:
Figure PCTCN2022138958-appb-000201
Figure PCTCN2022138958-appb-000202
其中
Figure PCTCN2022138958-appb-000203
中得到的接收信号的第m个元素;
S56,对于n s=1,2,…,N s,计算
Figure PCTCN2022138958-appb-000204
的观测均值
Figure PCTCN2022138958-appb-000205
和方差
Figure PCTCN2022138958-appb-000206
具体如下:
Figure PCTCN2022138958-appb-000207
Figure PCTCN2022138958-appb-000208
S57,对于n s=1,2,…,N s,计算
Figure PCTCN2022138958-appb-000209
的观测均值
Figure PCTCN2022138958-appb-000210
和方差
Figure PCTCN2022138958-appb-000211
具体如下:
Figure PCTCN2022138958-appb-000212
Figure PCTCN2022138958-appb-000213
S58,t G=t G+1,重复执行步骤S54至步骤S57直到达到收敛条件
Figure PCTCN2022138958-appb-000214
S59,将上述步骤估计所得到的稀疏变量
Figure PCTCN2022138958-appb-000215
作为本轮迭代最终环境感知结果。
S6,重复按顺序骤S4和步骤S5直至算法收敛,其中,迭代流程图如图2所示,得到最终环境感知结果。
通过计算机仿真可以看出:如图3和图4所示,我们对比了本发明的聚焦方法和大范围成像算法之间的成像效果。相比于大范围成像,本发明算法显著提高了ROI内物体的成像准确度。图3表明随着用户数量的增加本发明的方法环境感知效果逐渐提升并优于现有的算法。图4表示随着子载波数量的增加,本发明的方法环境感知效果逐渐提升并优于现有的算法。
在上行链路无线通信场景中,该利用现有通信设备进行毫米波感知与通信系一体化系统的设计方法,基于用户发送数据充分利用不同的系统资源来实现聚焦式环境感知,将环境感知问题转换为压缩感知重构问题,随后基于一种近似消息传递算法实现了对环境初始的粗略感知,根据一种背景判断方法,本发明划分判断目标物体并去除背景散射体对接收信号的影响,最后由此反复迭代 的去除背景散射体,以获得目标物体更准确的聚焦感知结果,相比现有的环境感知重构算法,提出的迭代聚焦式环境感知方法解决了由于系统资源不足而导致大范围环境感知精确度低的问题,改进了传统压缩感知算法无法实现对环境变量特定范围聚焦的缺陷,为未来感知通信一体化系统设计提供了一种高效的环境感知方法,在算法迭代过程中,根据每一步的压缩感知重构结果,逐步对环境变量的先验概率进行估计,对特定目标实现了一种迭代渐进式的压缩感知稀疏重构。在相同的系统资源开销的基础上,本发明算法显著提升了对特定目标的感知准确度,并优于现有的算法。
以上所述仅为本说明书一个或多个实施例的较佳实施例而已,并不用以限制本说明书一个或多个实施例,凡在本说明书一个或多个实施例的精神和原则之内,所做的任何修改、等同替换、改进等,均应包含在本说明书一个或多个实施例保护的范围之内。

Claims (5)

  1. 一种迭代聚焦式毫米波一体化通信与感知方法,其特征在于,包括以下步骤:
    S1,在任意时隙中,利用基站接收环境中的所有活跃用户发送的一定长度的导频序列信号,得到接收信号,其中,接收信号是导频序列信号受环境影响之后的信号;
    S2,利用所述步骤S1中的接收信号,基于多波束多载波毫米波信道模型,将对特定目标的环境感知问题转换为压缩感知重构问题;
    S3,基于近似消息传递方法求解所述步骤S2中的压缩感知重构问题,得到粗略的环境感知初始结果;
    S4,基于粗略的环境感知初始结果,从整体环境中选择预定区域作为聚焦的感兴趣区域,并根据背景判断方法,划分、判断感兴趣区域中的目标物体且去除感兴趣区域之外的背景散射体对接收信号的影响,得到目标物体对应的接收信号;
    S5,基于所述步骤S4中获得的目标物体对应的接收信号,计算环境感知结果;
    S6,重复按顺序骤S4和步骤S5直至算法收敛,得到最终环境感知结果。
  2. 根据权利要求1所述的迭代聚焦式毫米波一体化通信与感知方法,其特征在于,所述步骤S2具体为:
    S21,将步骤S1中的接收信号中的环境信息离散化为像素,每个像素代表其周围大小为l s×w s的小正方形内的环境信息,令全部范围环境大小为L s×W s,则像素总数为N s=L/l s×W/w s;每一个像素内部可能是空的,也可能是存在散射体;使用一个散射系数
    Figure PCTCN2022138958-appb-100001
    来表示第n s个点云点的所在小立方体的散射系数,若小立方体内部是空的,则
    Figure PCTCN2022138958-appb-100002
    因此,整个房间的环境信息可以表示为
    Figure PCTCN2022138958-appb-100003
    S22,采用多波束多载波毫米波信道模型,在第n f个子载波频率上,基站接收天线接收接收信号表示如下:
    Figure PCTCN2022138958-appb-100004
    其中,
    Figure PCTCN2022138958-appb-100005
    表示N c个基站RF链路的长度为K码元的接收信号,
    Figure PCTCN2022138958-appb-100006
    是基站N R个均匀线阵接收天线的波束成形矢量,δ是散射系数的归一化系数,根据像素大小l s×w s选择,归一化系数是描述电磁波接收面积与接收功率之间的物理关系,
    Figure PCTCN2022138958-appb-100007
    表示N u个用户发送长度为K码元的导频,n为噪声;
    Figure PCTCN2022138958-appb-100008
    表示在第n f个子载波频率上,N u个用户到N R个接收天线自由空间传播信道;
    Figure PCTCN2022138958-appb-100009
    为在第n f个子载波上的非视距信道;
    其中,
    Figure PCTCN2022138958-appb-100010
    表示为:
    Figure PCTCN2022138958-appb-100011
    其中,
    Figure PCTCN2022138958-appb-100012
    为N u个用户的导向矢量,
    Figure PCTCN2022138958-appb-100013
    为N u个用户至基站的信道增益;
    Figure PCTCN2022138958-appb-100014
    其中,j表示复数符号,n R为接收天线编号,
    Figure PCTCN2022138958-appb-100015
    为第n u个用户的到达角,d为基站部署的均匀线阵天线间距,
    Figure PCTCN2022138958-appb-100016
    为波长;
    Figure PCTCN2022138958-appb-100017
    表示如下:
    Figure PCTCN2022138958-appb-100018
    其中,
    Figure PCTCN2022138958-appb-100019
    Figure PCTCN2022138958-appb-100020
    分别为第n u个用户的至基站之间的信道幅度增益和相移;
    在第n f个子载波频率上,第n u个用户到第n s个像素的自由空间传播信道
    Figure PCTCN2022138958-appb-100021
    为表示为:
    Figure PCTCN2022138958-appb-100022
    其中,
    Figure PCTCN2022138958-appb-100023
    Figure PCTCN2022138958-appb-100024
    分别为第n u个用户的至第n s个像素之间的信道幅度增益和相移;
    在第n f个子载波频率上,N s个像素到N R个接收天线自由空间传播信道
    Figure PCTCN2022138958-appb-100025
    表示为:
    Figure PCTCN2022138958-appb-100026
    其中,
    Figure PCTCN2022138958-appb-100027
    为N s个像素的导向矢量,
    Figure PCTCN2022138958-appb-100028
    为N s个像素至基站的信道增益;
    Figure PCTCN2022138958-appb-100029
    其中,n R为接收天线编号,
    Figure PCTCN2022138958-appb-100030
    为第n s个像素的到达角,
    Figure PCTCN2022138958-appb-100031
    表示如下:
    Figure PCTCN2022138958-appb-100032
    其中,
    Figure PCTCN2022138958-appb-100033
    Figure PCTCN2022138958-appb-100034
    分别为第n s个像素的至基站之间的信道幅度增益和相移;
    S23,环境信息的估计结果表示为
    Figure PCTCN2022138958-appb-100035
    表示如下:
    Figure PCTCN2022138958-appb-100036
    其中,y为全部子载波的接收信号,w为全部子载波的均匀线阵接收天线的波束成形矢量,x ROI为感兴趣区域内的环境信息,H NLOS为全部子载波的非视距信道,H LOS为全部子载波的视距信道,s为全部子载波的非视距信道发送信号,ε为松弛变量;
    由于直视信道的自由空间信道系数可以通过数值模型来估计,在第n f个子载波频率上,将包含未知的环境信息的部分接收信号
    Figure PCTCN2022138958-appb-100037
    表示为,
    Figure PCTCN2022138958-appb-100038
    将联合N f个子载波的数据,将特定目标的迭代聚焦式环境感知问题转化为压缩感知重构问题方程如下:
    Figure PCTCN2022138958-appb-100039
  3. 根据权利要求2所述的迭代聚焦式毫米波一体化通信与感知方法,其特征在于,所述步骤S3具体为:
    S31,首先设置初始粗略的环境感知先验概率,令环境信息分布为伯努利-高斯分布,其概率密度函数p x(x|q)表示为:
    p x(x|q)=(1-λ)δ(x)+λN(x|θ xx)
    其中,x表示环境信息x中的元素,所有参数表示为
    Figure PCTCN2022138958-appb-100040
    δ(·)是冲激函数,λ是稀疏系数;θ x∈[0,1]和σ x分别表示环境信息分布的均值和方差,N(·)表示标准正态分布;
    S32,近似消息传递算法参数初始化,令输入函数g in(·),g′ in(·)和输出函数g out(·),g′ out(·)分别如下
    Figure PCTCN2022138958-appb-100041
    Figure PCTCN2022138958-appb-100042
    Figure PCTCN2022138958-appb-100043
    Figure PCTCN2022138958-appb-100044
    Figure PCTCN2022138958-appb-100045
    其中,
    Figure PCTCN2022138958-appb-100046
    为输入变量,σ w为噪声方差;
    令迭代次数t G=0,残差
    Figure PCTCN2022138958-appb-100047
    稀疏向量估计均值
    Figure PCTCN2022138958-appb-100048
    稀疏向量估计方差
    Figure PCTCN2022138958-appb-100049
    S33,令M=N cN fK,N c是基站数量,K是码元数量,N f是子载波数量,对于m=1,2,...,M,计算变量z m的估计的均值
    Figure PCTCN2022138958-appb-100050
    和方差
    Figure PCTCN2022138958-appb-100051
    Figure PCTCN2022138958-appb-100052
    Figure PCTCN2022138958-appb-100053
    Figure PCTCN2022138958-appb-100054
    S34,对于m=1,2,...,M,计算残差的均值
    Figure PCTCN2022138958-appb-100055
    和方差
    Figure PCTCN2022138958-appb-100056
    Figure PCTCN2022138958-appb-100057
    Figure PCTCN2022138958-appb-100058
    其中,y m是接收信号的第m个元素;
    S35,对于n s=1,2,...,N s,计算
    Figure PCTCN2022138958-appb-100059
    的观测均值
    Figure PCTCN2022138958-appb-100060
    和方差
    Figure PCTCN2022138958-appb-100061
    Figure PCTCN2022138958-appb-100062
    Figure PCTCN2022138958-appb-100063
    S36,对于n s=1,2,...,N s,计算
    Figure PCTCN2022138958-appb-100064
    的观测均值
    Figure PCTCN2022138958-appb-100065
    和方差
    Figure PCTCN2022138958-appb-100066
    Figure PCTCN2022138958-appb-100067
    Figure PCTCN2022138958-appb-100068
    S37,t G=t G+1,重复执行步骤S33至步骤S36直到达到收敛条件
    Figure PCTCN2022138958-appb-100069
    ε G为误差容限;
    S38,将稀疏变量
    Figure PCTCN2022138958-appb-100070
    作为环境信息x的粗略的环境感知初始结果。
  4. 根据权利要求3所述的迭代聚焦式毫米波一体化通信与感知方法,其特征在于,所述步骤S4具体为:
    S41,根据粗略的环境感知初始结果和实际需求从整体环境中选择预定区域作为聚焦的感兴趣区域,目标物体在感兴趣区域内;
    S42,在第i次迭代中,感兴趣区域外的背景散射体
    Figure PCTCN2022138958-appb-100071
    被检测如下:
    Figure PCTCN2022138958-appb-100072
    其中,
    Figure PCTCN2022138958-appb-100073
    代表第i次迭代中的结果,γ i为背景散射体检测门限,随着迭代次数的增加,检测阈值γ i应该降低;
    S43,从接收信号中去除背景散射部分,得到第i+1次迭代ROI内的目标物体的接收信号
    Figure PCTCN2022138958-appb-100074
    Figure PCTCN2022138958-appb-100075
    其中α是一个权重变量,用于增强迭代算法的鲁棒性,随着迭代次数的增加,权重变量α应该升高。
  5. 根据权利要求1所述的迭代聚焦式毫米波一体化通信与感知方法,其特征在于,所述步骤S5具体为:
    S51,设置迭代聚焦过程中环境信息的先验概率,在第i次迭代中,假设背景散射体服从伯努利高斯分布,先验概率公式p(x back)如下:
    Figure PCTCN2022138958-appb-100076
    其中θ back,i和σ back分别表示背景环境信息分布的均值和方差,λ是稀疏系数,N(·)表示标准正态分布,x back表示背景散射体;
    所选ROI中的散射体分布为高斯分布,没有稀疏性;
    Figure PCTCN2022138958-appb-100077
    其中θ ROI和σ ROI分别表示ROI环境信息分布的均值和方差;
    S52,根据步骤S51中得到的先验概率公式,设置当前第i+1个迭代轮次的感兴趣区域内外环境信息的先验概率p(x);
    S53,近似消息传递算法参数初始化,令输入函数g in(·),g′ in(·)和输出函数g out(·),g′ out(·)分别如下:
    Figure PCTCN2022138958-appb-100078
    Figure PCTCN2022138958-appb-100079
    Figure PCTCN2022138958-appb-100080
    Figure PCTCN2022138958-appb-100081
    Figure PCTCN2022138958-appb-100082
    令迭代次数t G=0,残差
    Figure PCTCN2022138958-appb-100083
    稀疏向量估计均值
    Figure PCTCN2022138958-appb-100084
    稀疏向量估计方差
    Figure PCTCN2022138958-appb-100085
    S54,令M=N cN fK,对于m=1,2,...,M,计算z m的估计的均值
    Figure PCTCN2022138958-appb-100086
    和方差
    Figure PCTCN2022138958-appb-100087
    具体如下:
    Figure PCTCN2022138958-appb-100088
    Figure PCTCN2022138958-appb-100089
    Figure PCTCN2022138958-appb-100090
    S55,对于m=1,2,...,M,计算残差的均值
    Figure PCTCN2022138958-appb-100091
    和方差
    Figure PCTCN2022138958-appb-100092
    具体如下:
    Figure PCTCN2022138958-appb-100093
    Figure PCTCN2022138958-appb-100094
    其中
    Figure PCTCN2022138958-appb-100095
    为S43中得到的接收信号的第m个元素;
    S56,对于n s=1,2,...,N s,计算
    Figure PCTCN2022138958-appb-100096
    的观测均值
    Figure PCTCN2022138958-appb-100097
    和方差
    Figure PCTCN2022138958-appb-100098
    具体如下:
    Figure PCTCN2022138958-appb-100099
    Figure PCTCN2022138958-appb-100100
    S57,对于n s=1,2,...,N s,计算
    Figure PCTCN2022138958-appb-100101
    的观测均值
    Figure PCTCN2022138958-appb-100102
    和方差
    Figure PCTCN2022138958-appb-100103
    具体如下:
    Figure PCTCN2022138958-appb-100104
    Figure PCTCN2022138958-appb-100105
    S58,t G=t G+1,重复执行步骤S54至步骤S57直到达到收敛条件
    Figure PCTCN2022138958-appb-100106
    S59,将上述步骤估计所得到的稀疏变量
    Figure PCTCN2022138958-appb-100107
    作为本轮迭代最终环境感知结果
    Figure PCTCN2022138958-appb-100108
PCT/CN2022/138958 2022-07-29 2022-12-14 一种迭代聚焦式毫米波一体化通信与感知方法 WO2024021440A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US18/358,975 US12019181B2 (en) 2022-07-29 2023-07-26 Iterative focused millimeter wave integrated communication and sensing method

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210906702.0A CN115243311A (zh) 2022-07-29 2022-07-29 一种迭代聚焦式毫米波一体化通信与感知方法
CN202210906702.0 2022-07-29

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US18/358,975 Continuation US12019181B2 (en) 2022-07-29 2023-07-26 Iterative focused millimeter wave integrated communication and sensing method

Publications (1)

Publication Number Publication Date
WO2024021440A1 true WO2024021440A1 (zh) 2024-02-01

Family

ID=83678055

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/138958 WO2024021440A1 (zh) 2022-07-29 2022-12-14 一种迭代聚焦式毫米波一体化通信与感知方法

Country Status (3)

Country Link
US (1) US12019181B2 (zh)
CN (1) CN115243311A (zh)
WO (1) WO2024021440A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115243311A (zh) 2022-07-29 2022-10-25 浙江大学 一种迭代聚焦式毫米波一体化通信与感知方法
CN116233863B (zh) * 2023-05-04 2023-07-25 中国电信股份有限公司浙江分公司 针对高精度定位终端的基站部署方法及装置

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111107023A (zh) * 2019-09-23 2020-05-05 南京邮电大学 在大规模mimo中基于光滑范数的压缩感知信道估计方法
US20200356086A1 (en) * 2019-05-07 2020-11-12 Cisco Technology, Inc. Adaptively calibrated spatio-temporal compressive sensing for sensor networks
CN113965881A (zh) * 2021-09-28 2022-01-21 浙江大学 遮挡效应下的毫米波一体化通信与感知方法
CN115243311A (zh) * 2022-07-29 2022-10-25 浙江大学 一种迭代聚焦式毫米波一体化通信与感知方法

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2022544339A (ja) * 2019-07-26 2022-10-17 ヴェロ3ディー,インコーポレーテッド 三次元オブジェクトの形成における品質保証

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20200356086A1 (en) * 2019-05-07 2020-11-12 Cisco Technology, Inc. Adaptively calibrated spatio-temporal compressive sensing for sensor networks
CN111107023A (zh) * 2019-09-23 2020-05-05 南京邮电大学 在大规模mimo中基于光滑范数的压缩感知信道估计方法
CN113965881A (zh) * 2021-09-28 2022-01-21 浙江大学 遮挡效应下的毫米波一体化通信与感知方法
CN115243311A (zh) * 2022-07-29 2022-10-25 浙江大学 一种迭代聚焦式毫米波一体化通信与感知方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
LI BIWEI; WANG XIANBIN: "Rigid Body Localization and Environment Sensing with 5G Millimeter Wave MIMO", 2021 IEEE 94TH VEHICULAR TECHNOLOGY CONFERENCE (VTC2021-FALL), IEEE, 27 September 2021 (2021-09-27), pages 1 - 5, XP034041440, DOI: 10.1109/VTC2021-Fall52928.2021.9625387 *

Also Published As

Publication number Publication date
US20240045026A1 (en) 2024-02-08
CN115243311A (zh) 2022-10-25
US12019181B2 (en) 2024-06-25

Similar Documents

Publication Publication Date Title
Yan et al. Passive beamforming and information transfer design for reconfigurable intelligent surfaces aided multiuser MIMO systems
WO2024021440A1 (zh) 一种迭代聚焦式毫米波一体化通信与感知方法
CN110463147B (zh) 解码符号的方法以及接收并解码符号的接收器
Arnold et al. Enabling FDD massive MIMO through deep learning-based channel prediction
Xia et al. Generative neural network channel modeling for millimeter-wave UAV communication
KR102154481B1 (ko) 딥러닝을 이용한 대규모 mimo 시스템의 빔포밍 장치 및 방법
CN114124623B (zh) 一种无线通信信道估计方法和装置
CN114745232B (zh) 智能可重构表面辅助毫米波mimo系统的信道估计方法
CN114729982A (zh) 用于定位的方法及装置
Zhu et al. Resource allocation for IRS assisted mmWave integrated sensing and communication systems
Bhowal et al. RIS-aided communications in indoor and outdoor environments: Performance analysis with a realistic channel model
CN113965881B (zh) 遮挡效应下的毫米波一体化通信与感知方法
Wymeersch et al. Adaptive detection probability for mmWave 5G SLAM
Zhou et al. Path loss model based on cluster at 28 GHz in the indoor and outdoor environments
WO2022151900A1 (zh) 一种基于神经网络的信道估计方法及通信装置
CN115473771A (zh) 一种基于模型演进的环境感知方法
Mao et al. Roar-fed: Ris-assisted over-the-air adaptive resource allocation for federated learning
Krunz et al. Online reinforcement learning for beam tracking and rate adaptation in millimeter-wave systems
Liu et al. Distributed beamforming for joint radar-communications
Liu et al. Joint task offloading and resource allocation for RIS-assisted UAV for mobile edge computing networks
CN116743222A (zh) 复用波束赋形方法、通信感知计算一体化系统及相关装置
Hashida et al. IRS-aided communications without channel state information relying on deep reinforcement learning
CN116634358A (zh) 终端定位方法、装置及非易失性存储介质
Wu et al. Learning from UAV experimental results for performance modeling of reconfigurable intelligent surface flying platform
Wang et al. Spaceborne cooperative detection for distributed sensing: Overcoming inter-satellite link limitations via deep information bottleneck

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22952873

Country of ref document: EP

Kind code of ref document: A1