WO2024032009A1 - 一种基于模型演进的环境感知方法 - Google Patents

一种基于模型演进的环境感知方法 Download PDF

Info

Publication number
WO2024032009A1
WO2024032009A1 PCT/CN2023/086613 CN2023086613W WO2024032009A1 WO 2024032009 A1 WO2024032009 A1 WO 2024032009A1 CN 2023086613 W CN2023086613 W CN 2023086613W WO 2024032009 A1 WO2024032009 A1 WO 2024032009A1
Authority
WO
WIPO (PCT)
Prior art keywords
matrix
model
channel
environmental
environment
Prior art date
Application number
PCT/CN2023/086613
Other languages
English (en)
French (fr)
Inventor
张朝阳
章一晗
童欣
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
Publication of WO2024032009A1 publication Critical patent/WO2024032009A1/zh

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L25/00—Baseband systems
    • H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
    • H04L25/0202—Channel estimation
    • H04L25/024—Channel estimation channel estimation algorithms
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B17/00—Monitoring; Testing
    • H04B17/30—Monitoring; Testing of propagation channels
    • H04B17/391—Modelling the propagation channel

Definitions

  • the invention belongs to the field of wireless communication technology, and specifically relates to an environment sensing method based on model evolution.
  • Integrated communication and perception is the key technology of the next generation wireless communication system, which requires that while completing normal communication, it can also perceive the physical world, including but not limited to user position, movement speed, human posture, etc.
  • Integrated communication and perception requires the simultaneous realization of communication and perception functions in a system, achieving intensive utilization of resources, and communication results and perception results can cooperate with each other to improve the overall performance of the system, such as autonomous driving and non-destructive security inspection. and key enabling technologies for emerging applications such as smart cities.
  • channel response data also called channel state information CSI
  • integrated communication and perception can be achieved without making major modifications to the hardware structure and software algorithms of the existing communication system. Therefore, it is An extremely important key technology.
  • the most important task is to image environmental targets, because imaging can obtain the largest amount and the most original information about the target, on which various other systems can be developed and implemented. Tasks such as target recognition can be performed based on imaging results.
  • the imaging task is also the most difficult perception task, because it requires an efficient and accurate model to describe the relationship between the target object and electromagnetic waves, and to design an effective algorithm to achieve target imaging.
  • the propagation rules of electromagnetic waves are very complex.
  • an electromagnetic wave encounters an obstacle, its energy will be divided into multiple parts. Simply put, it can be divided into reflected energy, lost heat energy and transmitted energy: part of the energy is reflected back into the space, and part of the energy will enter the object and propagate. , part of the energy will be absorbed by the object and converted into heat energy, but some of the energy will eventually pass through the target object, also called transmission.
  • the imaged target is based on the received channel State information, reversely solve the environmental target.
  • One of the main difficulties is: the mutual influence between environmental targets. For example, the electromagnetic wave propagating to target B has a very weak energy because it is blocked by target A in the middle.
  • the present invention provides an environment sensing method based on model evolution.
  • the environment sensing method based on model evolution is based on the communication system cycle. Perform channel estimation tasks to obtain channel response data to achieve environmental perception. Without making major modifications to the hardware or software of the existing communication system, the integration of perception and communication can be achieved, which solves the problem of how to solve the problem at the base station under the real electromagnetic wave propagation model. The problem of achieving environmental awareness.
  • An environment awareness method based on model evolution includes the following steps:
  • S1 use the base station to periodically receive the pilot sequence signals sent by all active users in a certain space and perform channel estimation on the pilot sequence signals to obtain channel response data;
  • the environmental space is discretized, and combined with the reflection and transmission channel models obtained in step S2, a mathematical model of channel response and environmental targets is constructed.
  • the environmental perception problem is transformed into a generalized Compressed sensing optimization problem;
  • step S4 use the channel response data obtained in step S1, perform model evolution based on the mathematical model of the channel response and environmental target obtained in step S3, and use the evolved model and the mathematical model obtained in step S1
  • the channel response data is used to solve the generalized compressed sensing optimization problem and complete the perception of environmental targets.
  • step S1 while using the base station to perform channel estimation on the pilot sequence signal, the absolute position of the user terminal or the relative position relative to the base station is uploaded to the base station.
  • step S2 is specifically:
  • e is a natural constant
  • j is an imaginary unit
  • is the amplitude of the channel is the phase of the channel
  • ⁇ is the radar reflection cross-section area of the scatterer
  • S22 use the occlusion effect to model the approximate linear propagation characteristics of high-frequency electromagnetic waves, and propose a calculation method for the occlusion effect based on the position distribution of user terminals, base stations and scatterers; set points O A and O C as general points, Point O B represents an obstacle.
  • the main basis for judging whether point O B blocks the direct path between points O A and O C is the following two points:
  • the distance d from the line connecting point O B to points O A and O C is less than a certain threshold ⁇ , that is:
  • Point O B is between points O A and O C , that is:
  • x is the reflection coefficient of the target, is the set of locations of all users and base stations, and g is the above calculation method function;
  • step S3 is specifically:
  • h is the channel response
  • x represents the reflection coefficient vector of the environment
  • y represents the transmission coefficient vector of the environment
  • H is the ideal non-line-of-sight path matrix calculated in step S31
  • B Re and B Tr are calculated in step S32
  • v is random additive white Gaussian noise
  • ⁇ represents the Hadamard product operation
  • a Re represents the measurement matrix of the reflection channel
  • a Tr represents the measurement matrix of the transmission channel
  • step S33 Based on the mathematical model of channel response and environmental target in step S32, the environment sensing problem is transformed into a generalized compressed sensing optimization problem, expressed as follows: st
  • x represents the reflection coefficient vector of the environment
  • y represents the transmission coefficient vector of the environment
  • 2 are the first norm and the second norm of the vector respectively
  • h is the channel response
  • a Re represents the measurement matrix of the reflection channel
  • a Tr represents the measurement matrix of the transmission channel
  • ⁇ is the relaxation variable
  • H is step S21
  • B Re and B Tr are the occlusion effect indication matrices calculated in step S22
  • g( ⁇ , ⁇ ) is the occlusion effect calculation function based on the position relationship, is the location of the base station and all user terminals.
  • step S4 is specifically:
  • h is the channel response
  • x represents the reflection coefficient vector of the environment
  • H is the ideal non-line-of-sight path matrix
  • v is the random additive white Gaussian noise
  • the simple model formula (1) is established as a compressed sensing optimization problem, expressed as follows: st
  • the model can be evolved to obtain the estimated occlusion effect indicator matrix as follows:
  • g( ⁇ , ⁇ ) is the occlusion effect calculation function based on position relationship, is the location of the base station and all user terminals.
  • the mathematical model of channel response and environmental targets can be evolved as:
  • H is the ideal non-line-of-sight path matrix
  • v is random additive white Gaussian noise
  • model formula (2) is established as a compressed sensing optimization problem, expressed as follows:
  • x is the environmental target to be obtained, is the estimation result of step S41, Indicates finding the minimum value of the corresponding expression in the value space of x, ⁇ is the slack variable, h is the channel response data obtained in step S1, Yes press
  • the calculated reflection channel measurement matrix, H is the ideal non-line-of-sight path matrix, Yes press
  • the calculated occlusion indicator matrix, g( ⁇ , ⁇ ), is the occlusion effect calculation function based on the position relationship, is the location of the base station and all user terminals;
  • step S43 for measurement matrix Perform column normalization and transform the compressed sensing optimization problem in step S42 into: st
  • 2 ⁇ xs diag(w) ⁇ x
  • x is the environmental target to be obtained, is the estimation result of step S41, Indicates finding the minimum value of the corresponding expression in the value space of x, h is the channel response obtained in step S1, C Re is the effective measurement matrix after normalization, x s is the normalized environmental reflection coefficient, ⁇ is the slack variable, Yes press The calculated reflection channel measurement matrix, H is the ideal non-line-of-sight path matrix, Yes press The calculated occlusion indicator matrix, g( ⁇ , ⁇ ), is the occlusion effect calculation function based on the position relationship, is the location of the base station and all user terminals.
  • the function diag( ⁇ ) represents the vector elements forming a diagonal matrix, and w represents A vector consisting of the column two norm of , symbol represents matrix
  • nth column of Represents a non-zero constant, N is the number of point clouds in the environment space; after the above preprocessing of the problem, the maximum expectation-generalized approximate message passing algorithm is used to solve it;
  • the model can be evolved to obtain the estimated occlusion effect indicator matrix and as follows:
  • g( ⁇ , ⁇ ) is the occlusion effect calculation function based on position relationship, is the location of the base station and all user terminals.
  • the mathematical model of channel response and environmental targets can be evolved as:
  • x represents the reflection coefficient vector of the environment
  • y represents the transmission coefficient vector of the environment
  • v is random additive white Gaussian noise
  • B represents the mask matrix of the scatterer existence variable
  • Model formula (3) is modeled as an optimization problem: st
  • z is the scatterer existence vector to be found
  • x is the reflection coefficient vector to be found
  • y is the transmission coefficient vector to be found
  • h is the channel response obtained in step S1
  • H is the ideal non-line-of-sight path matrix
  • B represents the mask matrix of the scatterer existence variable
  • ⁇ is the slack variable.
  • step S44 is the location of the base station and all user terminals
  • g( ⁇ , ⁇ ) is the occlusion effect calculation function based on the location relationship
  • Yes press and count The calculated occlusion indicator matrix, and are the average values of the non-zero parts of the reflection coefficient x and transmission coefficient y respectively
  • N is the number of point clouds in the environment space
  • step S45 due to the complexity of the optimization problem in step S44, an alternating optimization method is adopted to iteratively solve the scatterer existence vector z and the modified mask matrix B.
  • the alternating optimization algorithm is described as follows:
  • the beneficial effects of the present invention are: first, based on the electromagnetic wave propagation mechanism, the present invention establishes a reflection and transmission channel model under the occlusion effect, and constructs a mathematical model of channel response and environmental targets, and then transforms the environmental perception problem into a generalized compression
  • the algorithm of model evolution is finally adopted.
  • the necessary information of the advanced model is obtained by solving the goals, realizing the iteration and evolution of the model, realizing the solution of the compressed sensing problem, and completing the environmental goals. perception.
  • the strategy of model evolution it achieves the characteristics of decoupling between environmental targets and minimal need for prior information, and realizes the gradual reconstruction of the environment.
  • the present invention Compared with the existing environment-aware reconstruction algorithm, the present invention
  • the environment perception algorithm based on model evolution has significantly improved the accuracy of environmental target perception, providing an effective method for realizing environment perception in mobile communication systems.
  • the base station under a real electromagnetic wave propagation model, uses a model evolution strategy to realize environment perception, which has the advantages of minimal need for prior information and high accuracy of environment perception.
  • the present invention realizes environment perception based on the communication system periodically performing channel estimation tasks to obtain channel response data, and can realize the integration of perception and communication without making major modifications to the hardware or software of the existing communication system.
  • Figure 1 is a schematic diagram of an environment sensing scene based on electromagnetic wave propagation characteristics provided by an exemplary embodiment
  • Figure 2 is a schematic diagram of an occlusion effect calculation method provided by an exemplary embodiment
  • Figure 3 is a case diagram showing environmental target sensing results comparing the algorithm of the present invention with other reconstruction algorithms provided by an exemplary embodiment
  • Figure 4 is a comparison diagram of the environment sensing performance MSE of the algorithm of the present invention and the MSE of other reconstruction algorithms under different SNR conditions provided by an exemplary embodiment
  • Figure 5 is a diagram showing the relationship between the number of user terminals UE and the environment sensing performance MSE when comparing the algorithm of the present invention with other reconstruction algorithms provided by an exemplary embodiment.
  • the scenario we consider is an uplink communication scenario.
  • a multi-antenna base station access point (AP) is deployed in a certain outdoor area, and there are multiple active single-antenna user terminals at the same time.
  • Equipment (UE) In this scenario, the user sends a signal to the AP, and it is assumed that the user will report his location information from time to time.
  • the AP will perform periodic channel estimation tasks to overcome the random fading of the channel and obtain channel status information. These channel status Information and the user's location are transmitted to a central computing processor for centralized environment awareness.
  • the signal sent by the user will be affected by environmental objects. For example, the signal sent by user 1 reaches the AP after being reflected by target object 1, and the signal sent by user 2 penetrates target object 1 and reaches the AP. Moreover, objects in the environment will also affect each other. From the perspective of user 3, target object 1 is occluded due to the existence of target object 2.
  • the present invention provides an environment perception method based on model evolution, which includes the following steps:
  • S1 use the base station to periodically receive the pilot sequence signals sent by all active users in a certain space and perform channel estimation on the pilot sequence signals to obtain channel response data;
  • step S1 the base station is used to perform channel estimation on the pilot sequence signal and at the same time, the absolute position of the user terminal or the relative position relative to the base station is uploaded to the base station.
  • step S2 is specifically:
  • e is a natural constant
  • j is an imaginary unit
  • is the amplitude of the channel is the phase of the channel
  • ⁇ is the radar reflection cross-section area of the scatterer
  • S22 use the occlusion effect to model the approximate linear propagation characteristics of high-frequency electromagnetic waves, and propose a calculation method for the occlusion effect based on the location distribution of user terminals, base stations and scatterers; as shown in Figure 2, set point O A , O C is a general point, and point O B represents an obstacle.
  • set point O A , O C is a general point, and point O B represents an obstacle.
  • the main basis for judging whether point O B blocks the direct path between points O A and O C is the following two points:
  • the distance d from the line connecting point O B to points O A and O C is less than a certain threshold ⁇ , that is:
  • Point O B is between points O A and O C , that is:
  • x is the reflection coefficient of the target, is the set of locations of all users and base stations, and g is the above calculation method function;
  • the environmental space is discretized, and combined with the reflection and transmission channel models obtained in step S2, a mathematical model of channel response and environmental targets is constructed.
  • the environmental perception problem is transformed into a generalized Compressed sensing optimization problem;
  • step S3 is specifically:
  • h is the channel response
  • x represents the reflection coefficient vector of the environment
  • y represents the transmission coefficient vector of the environment
  • H is the ideal non-line-of-sight path matrix calculated in step S31
  • B Re and B Tr are calculated in step S32
  • v is random additive white Gaussian noise
  • ⁇ represents the Hadamard product operation
  • a Re represents the measurement matrix of the reflection channel
  • a Tr represents the measurement matrix of the transmission channel
  • step S33 Based on the mathematical model of channel response and environmental target in step S32, the environment sensing problem is transformed into a generalized compressed sensing optimization problem, expressed as follows: st
  • x represents the reflection coefficient vector of the environment
  • y represents the transmission coefficient vector of the environment
  • 2 are the first norm and the second norm of the vector respectively
  • h is the channel response
  • a Re represents the measurement matrix of the reflection channel
  • a Tr represents the measurement matrix of the transmission channel
  • ⁇ is the relaxation variable
  • H is step S21
  • B Re and B Tr are the occlusion effect indication matrices calculated in step S22
  • g( ⁇ , ⁇ ) is the occlusion effect calculation function based on the position relationship, is the location of the base station and all user terminals.
  • step S4 use the channel response data obtained in step S1, perform model evolution based on the mathematical model of the channel response and environmental target obtained in step S3, and use the evolved model and the channel response data obtained in step S1 to solve generalized compression Perception optimization problem, completing the perception of environmental goals.
  • step S4 is specifically:
  • h is the channel response
  • x represents the reflection coefficient vector of the environment
  • H is the ideal non-line-of-sight path matrix
  • v is the random additive white Gaussian noise
  • the simple model formula (1) is established as a compressed sensing optimization problem, expressed as follows: st
  • E-GAMP expectation maximum-generalized approximate message passing
  • the model can be evolved to obtain the estimated occlusion effect indicator matrix as follows:
  • g( ⁇ , ⁇ ) is the occlusion effect calculation function based on position relationship, is the location of the base station and all user terminals.
  • the mathematical model of channel response and environmental targets can be evolved as:
  • H is the ideal non-line-of-sight path matrix
  • v is random additive white Gaussian noise
  • model formula (2) is established as a compressed sensing optimization problem, expressed as follows:
  • x is the environmental target to be obtained, is the estimation result of step S41, Indicates finding the minimum value of the corresponding expression in the value space of x, ⁇ is the slack variable, h is the channel response data obtained in step S1, Yes press
  • the calculated reflection channel measurement matrix, H is the ideal non-line-of-sight path matrix, Yes press
  • the calculated occlusion indicator matrix, g( ⁇ , ⁇ ), is the occlusion effect calculation function based on the position relationship, is the location of the base station and all user terminals;
  • step S43 for measurement matrix Perform column normalization and transform the compressed sensing optimization problem in step S42 into: st
  • 2 ⁇ xs diag(w) ⁇ x
  • x is the environmental target to be obtained, is the estimation result of step S41, Indicates finding the minimum value of the corresponding expression in the value space of x, h is the channel response obtained in step S1, C Re is the effective measurement matrix after normalization, x s is the normalized environmental reflection coefficient, ⁇ is the slack variable, Yes press The calculated reflection channel measurement matrix, H is the ideal non-line-of-sight path matrix, Yes press The calculated occlusion indicator matrix, g( ⁇ , ⁇ ), is the occlusion effect calculation function based on the position relationship, is the location of the base station and all user terminals.
  • the function diag( ⁇ ) represents the vector elements forming a diagonal matrix, and w represents A vector consisting of the column two norm of , symbol represents matrix
  • nth column of Represents a non-zero constant, N is the number of point clouds in the environment space; after the above preprocessing of the problem, the Maximum Expectation-Generalized Approximate Message Passing (EM-GAMP) algorithm is used to solve it;
  • E-GAMP Maximum Expectation-Generalized Approximate Message Passing
  • the model can be evolved to obtain the estimated occlusion effect indicator matrix and as follows:
  • g( ⁇ , ⁇ ) is the occlusion effect calculation function based on position relationship, is the location of the base station and all user terminals.
  • the mathematical model of channel response and environmental targets can be evolved as:
  • x represents the reflection coefficient vector of the environment
  • y represents the transmission coefficient vector of the environment
  • v is random additive white Gaussian noise
  • B represents the mask matrix of the scatterer existence variable
  • Model formula (3) is modeled as an optimization problem: st
  • z is the scatterer existence vector to be found
  • x is the reflection coefficient vector to be found
  • y is the transmission coefficient vector to be found
  • h is the channel response obtained in step S1
  • H is the ideal non-line-of-sight path matrix
  • B represents the mask matrix of the scatterer existence variable
  • ⁇ is the slack variable.
  • step S44 is the location of the base station and all user terminals
  • g( ⁇ , ⁇ ) is the occlusion effect calculation function based on the location relationship
  • Yes press and The calculated occlusion indicator matrix and are the average values of the non-zero parts of the reflection coefficient x and transmission coefficient y respectively
  • N is the number of point clouds in the environment space
  • step S45 due to the complexity of the optimization problem in step S44, an alternating optimization method is adopted to iteratively solve the scatterer existence vector z and the modified mask matrix B.
  • the alternating optimization algorithm is described as follows:
  • the EM-GAMP algorithm completely ignores the occlusion effect and the mutual coupling relationship between target objects, so the imaging effect is poor.
  • the algorithm of the present invention is based on the method of model evolution and iteratively corrects the model through calculation, thereby well solving problems such as occlusion effects and coupling relationships of target objects.
  • the environment perception algorithm based on model evolution of the present invention significantly improves the accuracy of environment perception.
  • Figure 4 shows that the environment sensing performance of the algorithm of the present invention is significantly better than other algorithms, and the leading advantage becomes larger and larger as the SNR increases.
  • Figure 5 shows that the environment sensing performance of the method of the present invention improves as the number of users increases, and its advantage over other algorithms is growing.
  • the base station uses the model evolution strategy to realize environment perception, which has the advantages of minimal need for prior information and high accuracy of environment perception.
  • the present invention realizes environment perception based on the communication system periodically performing channel estimation tasks to obtain channel response data, and can realize the integration of perception and communication without making major modifications to the hardware or software of the existing communication system.
  • this invention is based on the electromagnetic wave propagation mechanism, establishes a reflection and transmission channel model under the occlusion effect, and constructs A mathematical model of channel response and environmental targets was built, and then the environmental sensing problem was transformed into a generalized compressed sensing optimization problem. Finally, a model evolution algorithm was used to start from a simple basic model and obtain the necessary information for the advanced model by solving the target.
  • model evolution realizes the iteration and evolution of the model, solve the compressed sensing problem, and complete the perception of environmental targets.
  • model evolution Through the strategy of model evolution, it achieves the characteristics of decoupling between environmental targets and minimal need for prior information, and realizes the gradual reconstruction of the environment.
  • the environment perception algorithm based on model evolution has significantly improved the accuracy of environmental target perception, providing an effective method for realizing environment perception in mobile communication systems.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Power Engineering (AREA)
  • Physics & Mathematics (AREA)
  • Electromagnetism (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本发明公开了一种基于模型演进的环境感知方法，利用现有通信系统中周期性进行信道估计任务得到信道响应数据来对环境进行感知，首先，将电磁波与环境目标的交互机制分为反射与透射，随后构建了信道响应与环境目标的数学模型，并将环境感知问题建模为压缩感知优化问题，最后为了解决该压缩感知问题，设计了从基本模型出发，并通过对目标的求解完成模型的迭代与演进，最终实现了对于环境的感知。相比现有的环境感知重构方法，本发明所考虑的模型更加贴近真实电磁波传播特性，且对环境先验信息要求极少，并且显著的提升了环境感知的精度，为在移动通信系统中实现环境感知提供了一种有效的方法。

Description

一种基于模型演进的环境感知方法 技术领域
本发明属于无线通信技术领域，具体涉及一种基于模型演进的环境感知方法。
背景技术
在一体化通信与感知是下一代无线通信系统的关键技术，其要求在完成正常通信的同时，也能感知物理世界，包括但不限于用户位置、运动速度、人体姿态等。一体化通信与感知要求在一套系统中同时实现通信与感知功能，实现了资源的集约化利用，并且通信结果和感知结果可以互相协作，以提升系统的整体性能，是诸如自动驾驶、无损安检和智慧城市等新兴应用的关键使能技术。
使用来源于常规通信过程中的信道响应数据(也称信道状态信息CSI)，可以在不对现有通信系统的硬件结构和软件算法做较大修改的条件下，实现一体化通信与感知，因此是一种极度重要的关键技术。在诸多的感知任务和目标中，最具重要性的任务就是对环境目标进行成像，因为成像能得到关于目标的数量最大、也是最原始的信息，可以在其之上开发和实现其他各种的任务，如可以基于成像结果进行目标识别等。同时，成像任务也是难度最大的感知任务，因为需要一个高效且准确的模型和描述目标物体与电磁波之间的作用关系，并且设计出有效的算法来实现目标的成像。
电磁波的传播规则，尤其是电磁波与障碍物体之间的相互作用，是很复杂的。当电磁波遇到障碍物体时，其能量会分为多个部分，简单来说可以分为反射的能量、损耗的热能和透射的能量：一部分能量反射回到空间中，一部分能量会进入物体中传播，其中部分能量会被物体吸收并转化为热能，但是还有一部分能量最终穿过目标物体，也称为透射。成像的目标时根据接收得到的信道 状态信息，逆向求解出环境目标，其中的一个主要难点是：环境目标之间的互相影响，如传播到目标B的电磁波由于中间被目标A所遮挡而导致其能量很微弱，这是目标A与B之间的耦合。由于环境中具有众多的目标，因此他们可能具有及其复杂的耦合关系，该耦合关系导致了求解问题的非线性特征，急剧增大了求解的难度。目前的环境感知算法没有充分研究环境目标之间的耦合关系，因此他们在求解真实电磁波传播特性下的成像问题时性能较差。
综上所述，综合考虑环境感知问题与真实电磁波传播模型，一种需要较少先验信息就能实现环境目标的成像的研究具有较高的难度和重要的现实意义。
发明内容
针对上述现有技术的不足，考虑到真实电磁波传播模型下环境目标之间存在着复杂的耦合关系，本发明提供一种基于模型演进的环境感知方法，基于模型演进的环境感知方法基于通信系统周期性进行信道估计任务得到信道响应数据，来实现环境感知，无需对现有通信系统的硬件或软件做出重大修改，即可实现感知通信一体化，解决了在真实电磁波传播模型下，基站端如何实现环境感知的问题。
本发明的目的是通过以下技术方案实现的：
提供一种基于模型演进的环境感知方法，该方法包括以下步骤：
S1，使用基站周期性地接收一定空间内的所有活跃用户发送的导频序列信号并将导频序列信号进行信道估计，得到信道响应数据；
S2，使用电磁波传播机制，基于位置关系的遮挡效应计算方法，建立遮挡效应下的反射与透射信道模型；
S3，将环境空间进行离散化处理，并结合步骤S2中得到的反射与透射信道模型，构建了信道响应与环境目标的数学模型，通过信道响应与环境目标的数学模型将环境感知问题转化为广义的压缩感知优化问题；
S4，使用步骤S1中得到的信道响应数据，基于步骤S3中得到的信道响应与环境目标的数学模型，进行模型演进，并利用演进后的模型和步骤S1中得到 的信道响应数据求解广义的压缩感知优化问题，完成对环境目标的感知。
进一步地，所述步骤S1中使用基站将导频序列信号进行信道估计的同时，往基站上传用户端的绝对位置或对于基站的相对位置。
进一步地，所述步骤S2具体为：
S21，使用电磁波传播机制，在第mr个载波频段，从第mu个用户，经过第n个散射体，到达基站的第mk根天线的理想非视距信道响应可以表达为：
其中，e是自然常数，j是虚数单位，是信道的幅值，是信道的相位；

其中，是第mr个载波的波长，是第mu个用户到第n个散射体的距离，是基站的第mk根天线到第n个散射体的距离，σ是散射体的雷达反射截面积；
S22，对高频电磁波的近似直线传播特性采用遮挡效应进行建模，并提出了一种基于用户终端、基站以及散射体位置分布的遮挡效应计算方法；设点OA,OC是一般点，而点OB代表障碍物，则判断点OB是否对点OA和OC之间的直射路径产生了遮挡的主要判断依据为以下两点：
点OB到点OA,OC连线的距离d小于一定的阈值γ，即：
点OB处于点OA,OC之间，即：
当以上两个条件都满足时，则从OA视角出发，OB被OC所遮挡，对所有目标应用该计算方法，则得到遮挡指示矩阵B：
其中，x是目标的反射系数，是所有用户和基站的位置的集合，g是上述的计算方法函数；
S23，基于步骤S21的理想非视距信道和步骤S22的基于位置关系的遮挡效应计算方法，考虑环境目标的反射系数x，则遮挡效应下的反射信道模型可以表达为：
其中，是反射信道的遮挡指示矩阵，xn是第n个散射体的反射系数；考虑环境目标的透射系数y，则遮挡效应下的透射信道模型可以表达为：
其中，是透射信道的遮挡指示矩阵，yn是第n个散射体的透射系数。
进一步地，所述步骤S3具体为：
S31，将环境空间进行离散化处理，将整个环境空间划分为一个个小立方体，每个小立方体中所包含的环境目标可以等效集中于小立方体中心的一个点，也称为点云划分；设待感知的环境空间大小为LL,LW,LH，小立方体的尺寸大小为ll,lw,lh，则整个环境中一共有N＝LL/ll×LW/lw×LH/lh个点云；使用xn来表示第n个点云的反射系数，如果该点云中不包含散射体，则其反射系数为0，如该点云中包含散射体，则其反射系数是一个非零的数；同时，使用yn来表示第n个点云的透射系数；因此，使用N维的反射系数向量x＝[x1,...,xN]T以及N维的透射系数向量y＝[y1,...,yN]T来表示待感知的环境空间；
S32，将步骤S31中对环境空间进行离散化处理得到的反射系数向量x和透射系数向量y，与步骤S2得到的反射与透射信道模型结合，构建信道响应与环境目标的数学模型如下：
h＝(H⊙BRe)x+(H⊙BTr)y+v
＝ARex+ATry+v
其中，h是信道响应，x表示环境的反射系数向量，y表示环境的透射系数向量，H是按步骤S31所计算的理想非视距路径矩阵，BRe和BTr是按步骤S32所计算的遮挡效应指示矩阵，v是随机的加性白高斯噪声，⊙表示哈达玛积操作，ARe表示反射信道的测量矩阵，ATr表示透射信道的测量矩阵；
S33，基于步骤S32中的信道响应与环境目标的数学模型，将环境感知问题转化为广义压缩感知优化问题，表达如下：

s.t.||h-ARex-ATry||2≤ε
ARe＝H⊙BRe
ATr＝H⊙BTr

其中x表示环境的反射系数向量，y表示环境的透射系数向量，||·||1,||·||2分别是向量的一范数、二范数，表示在x、y的取值空间中寻找对应表达式的最小值，h是信道响应，ARe表示反射信道的测量矩阵，ATr表示透射信道的测量矩阵，ε是松弛变量，H是步骤S21所计算的理想非视距路径矩阵，BRe和BTr是按步骤S22所计算的遮挡效应指示矩阵，g(·,·)是基于位置关系的遮挡效应计算函数，是基站和所有用户终端的位置。
进一步地，所述步骤S4具体为：
S41，采用模型演进算法，将基本模型表示步骤S1中得到的信道响应数据，具体如下：
h＝Hx+v   (1)
其中，h是信道响应，x表示环境的反射系数向量，H是理想非视距路径矩阵，v是随机的加性白高斯噪声，将该简单模型公式(1)建立为压缩感知优化问题，表示如下：

s.t.||h-Hx||2≤ε
其中表示在x的取值空间中寻找对应表达式的最小值，ε是松弛变量；对于该问题的求解，采用最大期望-广义近似消息传递算法；
S42，在得到步骤S41中关于环境目标x的初步求解结果之后，可以对模型进行演进，得到估计的遮挡效应指示矩阵如下：
其中，g(·,·)是基于位置关系的遮挡效应计算函数，是基站和所有用户终端的位置，此时，可以将信道响应与环境目标的数学模型演进为：
其中，H是理想非视距路径矩阵，v是随机的加性白高斯噪声，将模型公式(2)建立为压缩感知优化问题，表示如下：



其中，x是待求的环境目标，是步骤S41的估计结果，表示在x的取值空间中寻找对应表达式的最小值，ε是松弛变量，h是步骤S1得到的信道响应数据，是按计算得到的反射信道测量矩阵，H是理想非视距路径矩阵，是按计算得到的遮挡指示矩阵，g(·,·)是基于位置关系的遮挡效应计算函数，是基站和所有用户终端的位置；
S43，对测量矩阵进列归一化，并将步骤S42中的压缩感知优化问题转化为：

s.t.||h-CRexs||2≤ε



xs＝diag(w)·x
其中，x是待求的环境目标，是步骤S41的估计结果，表示在x的取值空间中寻找对应表达式的最小值，h是步骤S1得到的信道响应，CRe是经过归一化之后的有效测量矩阵，xs是经过归一化的环境反射系数，ε是松弛变量，是按计算得到的反射信道测量矩阵，H是理想非视距路径矩阵，是按计算得到的遮挡指示矩阵，g(·,·)是基于位置关系的遮挡效应计算函数，是基站和所有用户终端的位置，函数diag(·)表示取向量元素组成对角阵，w表示的列二范数组成的向量，符号表示矩阵的第n列，表示一个非零常数，N是环境空间的点云个数；对问题进行上述预处理之后，采用最大期望-广义近似消息传递算法对其进行求解；
S44，在得到步骤S43中关于环境目标x的求解结果之后，可以对模型进行演进，得到估计的遮挡效应指示矩阵和如下：

其中，g(·,·)是基于位置关系的遮挡效应计算函数，是基站和所有用户终端的位置，此时，可以将信道响应与环境目标的数学模型演进为：
其中，x表示环境的反射系数向量，y表示环境的透射系数向量，是按计算得到的反射信道测量矩阵，v是随机的加性白高斯噪声；
对该模型进行观察，可以发现无论是反射系数x还是透射系数y，都依赖于环境中散射体的存在性向量z，因此将建立统一的新模型
h＝(H⊙B)z+v   (3)
其中，B表示散射体存在性变量的掩膜矩阵，计算方式如下
其中与分别是反射系数x以及透射系数y的非零部分的均值，将模型公式(3)建模为一个优化问题：

s.t.||h-(H⊙B)z||2≤ε




其中，z是待求的散射体存在性向量，x是待求的反射系数向量，y是待求的透射系数向量，表示在z的取值空间中寻找对应表达式的最小值，h是步骤S1得到的信道响应，H是理想非视距路径矩阵，B表示散射体存在性变量的掩膜矩阵，ε是松弛变量，和是步骤S44的估计结果，是基站和所有用户终端的位置，g(·,·)是基于位置关系的遮挡效应计算函数，和是按和计 算得到的遮挡指示矩阵，与分别是反射系数x以及透射系数y的非零部分的均值，N是环境空间的点云个数；
S45，由于步骤S44中优化问题的复杂性，采取迭代求解散射体存在性向量z和修正掩膜矩阵B的交替优化方法，对该交替优化算法叙述如下：
在求解散射体存在性向量z时，固定掩膜矩阵B，采用最大期望-广义近似消息传递算法进行z的求解；
在修正掩膜矩阵B时，先固定散射体存在性向量z，并计算得到反射系数和透射系数然后依次计算遮挡效应矩阵和BTr，并合并成为新的掩膜矩阵B；
S46，在求解散射体存在性向量z和修正掩膜矩阵B的过程之间进行交替迭代优化，该算法的结束条件有两个：达到最大迭代次数Tmax或者算法收敛到门限τ以下，算法结束即完成了对环境目标的感知。
本发明的有益效果是：首先，本发明基于电磁波传播机制，建立在遮挡效应下的反射与透射信道模型，并构建了信道响应与环境目标的数学模型，然后将环境感知问题转化为广义的压缩感知优化问题，最后采用模型演进的算法，从简单的基本模型出发，通过对目标的求解获得高级模型的必要信息，实现模型的迭代与演进，实现了对压缩感知问题的求解，完成对于环境目标的感知。其通过模型演进的策略，达到了对环境目标之间的去耦合以及对先验信息需求极少等特点，实现对于环境的逐步重构，相比现有的环境感知重构算法，本发明的基于模型演进的环境感知算法在较为显著地提升了环境目标感知的准确度，为在移动通信系统中实现环境感知提供了一种有效的方法。本发明在真实电磁波传播模型下，基站端利用模型演进的策略实现环境感知，具有对先验信息需求极少，以及环境感知准确度高等优点。本发明基于通信系统周期性进行信道估计任务得到信道响应数据，来实现环境感知，无需对现有通信系统的硬件或软件做出重大修改，即可实现感知通信一体化。
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基于电磁波传播特性的环境感知场景示意图；
图2为一示例性实施例提供的遮挡效应计算方式示意图；
图3为一示例性实施例提供的将本发明的算法与其他重构算法相比较的环境目标感知结果的案例展示图；
图4为一示例性实施例提供的在不同SNR条件下，本发明算法的环境感知性能MSE与其他重构算法的MSE比较图；
图5为一示例性实施例提供的本发明的算法与其他重构算法相比较时，用户终端UE数量与环境感知性能MSE之间的关系图。
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。
如图1所示，我们考虑的场景是上行通信场景，在户外的某一区域中部署有一个多天线的基站接入点(AP)，且同时存在着多个活跃的单天线的用户终端 设备(UE)。在该场景中，用户发送信号至AP，并且假设用户会不定时上报自己的位置信息，AP端则会进行周期性的信道估计任务以克服信道的随机衰落，并得到信道状态信息，这些信道状态信息和用户的位置会被传输到中心的计算处理器上，进行集中式的环境感知。
用户发送的信号会被环境物体所影响，如用户1的发送信号经过目标物体1的反射抵达AP，用户2发射的信号穿透目标物体1而达到AP。并且环境中的物体之间也会互相影响，从用户3的视角出发，目标物体1就由于目标物体2的存在而被遮挡了。
本发明提供一种基于模型演进的环境感知方法，包括以下步骤：
S1，使用基站周期性地接收一定空间内的所有活跃用户发送的导频序列信号并将导频序列信号进行信道估计，得到信道响应数据；
在一实施例中，步骤S1中使用基站将导频序列信号进行信道估计的同时，往基站上传用户端的绝对位置或对于基站的相对位置。
S2，使用电磁波传播机制，基于位置关系的遮挡效应计算方法，建立遮挡效应下的反射与透射信道模型；
在一实施例中，步骤S2具体为：
S21，使用电磁波传播机制，在第mr个载波频段，从第mu个用户，经过第n个散射体，到达基站的第mk根天线的理想非视距信道响应可以表达为：
其中，e是自然常数，j是虚数单位，是信道的幅值，是信道的相位；

其中，是第mr个载波的波长，是第mu个用户到第n个散射体的距离，是基站的第mk根天线到第n个散射体的距离，σ是散射体的雷达反射截面积；
S22，对高频电磁波的近似直线传播特性采用遮挡效应进行建模，并提出了一种基于用户终端、基站以及散射体位置分布的遮挡效应计算方法；如图2所示，设点OA,OC是一般点，而点OB代表障碍物，则判断点OB是否对点OA和OC之间的直射路径产生了遮挡的主要判断依据为以下两点：
点OB到点OA,OC连线的距离d小于一定的阈值γ，即：
点OB处于点OA,OC之间，即：
当以上两个条件都满足时，则从OA视角出发，OB被OC所遮挡，对所有目标应用该计算方法，则得到遮挡指示矩阵B：
其中，x是目标的反射系数，是所有用户和基站的位置的集合，g是上述的计算方法函数；
S23，基于步骤S21的理想非视距信道和步骤S22的基于位置关系的遮挡效应计算方法，考虑环境目标的反射系数x，则遮挡效应下的反射信道模型可以表达为：
其中，是反射信道的遮挡指示矩阵，xn是第n个散射体的反射系数；考虑环境目标的透射系数y，则遮挡效应下的透射信道模型可以表达为：
其中，是透射信道的遮挡指示矩阵，yn是第n个散射体的透射系数。
S3，将环境空间进行离散化处理，并结合步骤S2中得到的反射与透射信道模型，构建了信道响应与环境目标的数学模型，通过信道响应与环境目标的数学模型将环境感知问题转化为广义的压缩感知优化问题；
在一实施例中，步骤S3具体为：
S31，将环境空间进行离散化处理，将整个环境空间划分为一个个小立方体，每个小立方体中所包含的环境目标可以等效集中于小立方体中心的一个点，也称为点云划分；设待感知的环境空间大小为LL,LW,LH，小立方体的尺寸大小为ll,lw,lh，则整个环境中一共有N＝LL/ll×LW/lw×LH/lh个点云；使用xn来表示第n个点云的反射系数，如果该点云中不包含散射体，则其反射系数为0，如该点云中包含散射体，则其反射系数是一个非零的数；同时，使用yn来表示第n个点云的透射系数；因此，使用N维的反射系数向量x＝[x1,...,xN]T以及N维的透射系数向量y＝[y1,...,yN]T来表示待感知的环境空间；
S32，将步骤S31中对环境空间进行离散化处理得到的反射系数向量x和透射系数向量y，与步骤S2得到的反射与透射信道模型结合，构建信道响应与环境目标的数学模型如下：
h＝(H⊙BRe)x+(H⊙BTr)y+v
＝ARex+ATry+v
其中，h是信道响应，x表示环境的反射系数向量，y表示环境的透射系数向量，H是按步骤S31所计算的理想非视距路径矩阵，BRe和BTr是按步骤S32所计算的遮挡效应指示矩阵，v是随机的加性白高斯噪声，⊙表示哈达玛积操作，ARe表示反射信道的测量矩阵，ATr表示透射信道的测量矩阵；
S33，基于步骤S32中的信道响应与环境目标的数学模型，将环境感知问题转化为广义压缩感知优化问题，表达如下：

s.t.||h-ARex-ATry||2≤ε
ARe＝H⊙BRe
ATr＝H⊙BTr

其中x表示环境的反射系数向量，y表示环境的透射系数向量，||·||1,||·||2分别是向量的一范数、二范数，表示在x、y的取值空间中寻找对应表达式的最小值，h是信道响应，ARe表示反射信道的测量矩阵，ATr表示透射信道的测量矩阵，ε是松弛变量，H是步骤S21所计算的理想非视距路径矩阵，BRe和BTr是按步骤S22所计算的遮挡效应指示矩阵，g(·,·)是基于位置关系的遮挡效应计算函数，是基站和所有用户终端的位置。
S4，使用步骤S1中得到的信道响应数据，基于步骤S3中得到的信道响应与环境目标的数学模型，进行模型演进，并利用演进后的模型和步骤S1中得到的信道响应数据求解广义的压缩感知优化问题，完成对环境目标的感知。
在一实施例中，步骤S4具体为：
S41，采用模型演进算法，将基本模型表示步骤S1中得到的信道响应数据，具体如下：
h＝Hx+v    (1)
其中，h是信道响应，x表示环境的反射系数向量，H是理想非视距路径矩阵，v是随机的加性白高斯噪声，将该简单模型公式(1)建立为压缩感知优化问题，表示如下：

s.t.||h-Hx||2≤ε
其中表示在x的取值空间中寻找对应表达式的最小值，ε是松弛变量；对于该问题的求解，采用最大期望-广义近似消息传递(EM-GAMP)算法；
S42，在得到步骤S41中关于环境目标x的初步求解结果之后，可以对模型进行演进，得到估计的遮挡效应指示矩阵如下：
其中，g(·,·)是基于位置关系的遮挡效应计算函数，是基站和所有用户终端的位置，此时，可以将信道响应与环境目标的数学模型演进为：
其中，H是理想非视距路径矩阵，v是随机的加性白高斯噪声，将模型公式(2)建立为压缩感知优化问题，表示如下：



其中，x是待求的环境目标，是步骤S41的估计结果，表示在x的取值空间中寻找对应表达式的最小值，ε是松弛变量，h是步骤S1得到的信道响应数据，是按计算得到的反射信道测量矩阵，H是理想非视距路径矩阵，是按计算得到的遮挡指示矩阵，g(·,·)是基于位置关系的遮挡效应计算函数，是基站和所有用户终端的位置；
S43，对测量矩阵进列归一化，并将步骤S42中的压缩感知优化问题转化为：

s.t.||h-CRexs||2≤ε



xs＝diag(w)·x
其中，x是待求的环境目标，是步骤S41的估计结果，表示在x的取值空间中寻找对应表达式的最小值，h是步骤S1得到的信道响应，CRe是经过归一化之后的有效测量矩阵，xs是经过归一化的环境反射系数，ε是松弛变量，是按计算得到的反射信道测量矩阵，H是理想非视距路径矩阵，是按计算得到的遮挡指示矩阵，g(·,·)是基于位置关系的遮挡效应计算函数，是基站和所有用户终端的位置，函数diag(·)表示取向量元素组成对角阵，w表示的列二范数组成的向量，符号表示矩阵的第n列，表示一个非零常数，N是环境空间的点云个数；对问题进行上述预处理之后，采用最大期望-广义近似消息传递(EM-GAMP)算法对其进行求解；
S44，在得到步骤S43中关于环境目标x的求解结果之后，可以对模型进行演进，得到估计的遮挡效应指示矩阵和如下：

其中，g(·,·)是基于位置关系的遮挡效应计算函数，是基站和所有用户终端的位置，此时，可以将信道响应与环境目标的数学模型演进为：
其中，x表示环境的反射系数向量，y表示环境的透射系数向量，是按计算得到的反射信道测量矩阵，v是随机的加性白高斯噪声；
对该模型进行观察，可以发现无论是反射系数x还是透射系数y，都依赖于环境中散射体的存在性向量z，因此将建立统一的新模型
h＝(H⊙B)z+v   (3)
其中，B表示散射体存在性变量的掩膜矩阵，计算方式如下
其中与分别是反射系数x以及透射系数y的非零部分的均值，将模型公式(3)建模为一个优化问题：

s.t.||h-(H⊙B)z||2≤ε




其中，z是待求的散射体存在性向量，x是待求的反射系数向量，y是待求的透射系数向量，表示在z的取值空间中寻找对应表达式的最小值，h是步骤S1得到的信道响应，H是理想非视距路径矩阵，B表示散射体存在性变量的掩膜矩阵，ε是松弛变量，和是步骤S44的估计结果，是基站和所有用户终端的位置，g(·,·)是基于位置关系的遮挡效应计算函数，和是按和计算得到的遮挡指示矩阵，与分别是反射系数x以及透射系数y的非零部分的均值，N是环境空间的点云个数；
S45，由于步骤S44中优化问题的复杂性，采取迭代求解散射体存在性向量z和修正掩膜矩阵B的交替优化方法，对该交替优化算法叙述如下：
在求解散射体存在性向量z时，固定掩膜矩阵B，采用最大期望-广义近似消息传递(EM-GAMP)算法进行z的求解；
在修正掩膜矩阵B时，先固定散射体存在性向量z，并计算得到反射系数和透射系数然后依次计算遮挡效应矩阵和BTr，并合并成为新的掩膜矩阵B；
S46，在求解散射体存在性向量z和修正掩膜矩阵B的过程之间进行交替迭代优化，该算法的结束条件有两个：达到最大迭代次数Tmax或者算法收敛到门限τ以下，算法结束即完成了对环境目标的感知。
通过计算机仿真可以看出：如图3所示，其中EM-GAMP算法完全忽视了遮挡效应以及目标物体之间的相互耦合关系，因而成像效果较差。而本发明算法基于模型演进的方法，通过计算迭代地修正模型，从而很好地解决了遮挡效应以及目标物体的耦合关系等难题。相比EM-AGMP算法，本发明的基于模型演进的环境感知算法较为显著提升了环境感知的准确度。图4表明本发明算法的环境感知性能显著优于其他算法，且领先优势随着SNR的增大而越来越大。图5表明本发明的方法的环境感知性能随着用户数量的增加而提升，且领先于其他算法的优势越来越大。
综上所述，在真实电磁波传播模型下，基站端利用模型演进的策略实现环境感知，具有对先验信息需求极少，以及环境感知准确度高等优点。本发明基于通信系统周期性进行信道估计任务得到信道响应数据，来实现环境感知，无需对现有通信系统的硬件或软件做出重大修改，即可实现感知通信一体化。首先，本发明基于电磁波传播机制，建立在遮挡效应下的反射与透射信道模型，并构 建了信道响应与环境目标的数学模型，然后将环境感知问题转化为广义的压缩感知优化问题，最后采用模型演进的算法，从简单的基本模型出发，通过对目标的求解获得高级模型的必要信息，实现模型的迭代与演进，实现了对压缩感知问题的求解，完成对于环境目标的感知。其通过模型演进的策略，达到了对环境目标之间的去耦合以及对先验信息需求极少等特点，实现对于环境的逐步重构，相比现有的环境感知重构算法，本发明的基于模型演进的环境感知算法在较为显著地提升了环境目标感知的准确度，为在移动通信系统中实现环境感知提供了一种有效的方法。
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. 一种基于模型演进的环境感知方法，其特征在于，包括以下步骤：
    S1，使用基站周期性地接收一定空间内的所有活跃用户发送的导频序列信号并将导频序列信号进行信道估计，得到信道响应数据；
    S2，使用电磁波传播机制，基于位置关系的遮挡效应计算方法，建立遮挡效应下的反射与透射信道模型；
    S3，将环境空间进行离散化处理，并结合步骤S2中得到的反射与透射信道模型，构建了信道响应与环境目标的数学模型，通过信道响应与环境目标的数学模型将环境感知问题转化为广义的压缩感知优化问题；
    S4，使用步骤S1中得到的信道响应数据，基于步骤S3中得到的信道响应与环境目标的数学模型，进行模型演进，并利用演进后的模型和步骤S1中得到的信道响应数据求解广义的压缩感知优化问题，完成对环境目标的感知。
  2. 根据权利要求1所述的基于模型演进的环境感知方法，其特征在于，所述步骤S1中使用基站将导频序列信号进行信道估计的同时，往基站上传用户端的绝对位置或对于基站的相对位置。
  3. 根据权利要求1所述的方法，其特征在于，所述步骤S2具体为：
    S21，使用电磁波传播机制，在第mr个载波频段，从第mu个用户，经过第n个散射体，到达基站的第mk根天线的理想非视距信道响应可以表达为：
    其中，e是自然常数，j是虚数单位，是信道的幅值，是信道的相位；

    其中，是第mr个载波的波长，是第mu个用户到第n个散射体的距离，是基站的第mk根天线到第n个散射体的距离，σ是散射体的雷达反射截面积；
    S22，对高频电磁波的近似直线传播特性采用遮挡效应进行建模，并提出了一种基于用户终端、基站以及散射体位置分布的遮挡效应计算方法；设点OA,OC是一般点，而点OB代表障碍物，则判断点OB是否对点OA和OC之间的直射路径产生了遮挡的主要判断依据为以下两点：
    点OB到点OA,OC连线的距离d小于一定的阈值γ，即：
    点OB处于点OA,OC之间，即：
    当以上两个条件都满足时，则从OA视角出发，OB被OC所遮挡，对所有目标应用该计算方法，则得到遮挡指示矩阵B：
    其中，x是目标的反射系数，是所有用户和基站的位置的集合，g是上述的计算方法函数；
    S23，基于步骤S21的理想非视距信道和步骤S22的基于位置关系的遮挡效应计算方法，考虑环境目标的反射系数x，则遮挡效应下的反射信道模型可以表达为：
    其中，是反射信道的遮挡指示矩阵，xn是第n个散射体的反射系数；考虑环境目标的透射系数y，则遮挡效应下的透射信道模型可以表达为：
    其中，是透射信道的遮挡指示矩阵，yn是第n个散射体的透射系数。
  4. 根据权利要求1所述的方法，其特征在于，所述步骤S3具体为：
    S31，将环境空间进行离散化处理，将整个环境空间划分为一个个小立方体，每个小立方体中所包含的环境目标可以等效集中于小立方体中心的一个点，也称为点云划分；设待感知的环境空间大小为LL,LW,LH，小立方体的尺寸大小为ll,lw,lh，则整个环境中一共有N＝LL/ll×LW/lw×LH/lh个点云；使用xn来表示第n个点云的反射系数，如果该点云中不包含散射体，则其反射系数为0，如该点云中包含散射体，则其反射系数是一个非零的数；同时，使用yn来表示第n个点云的透射系数；因此，使用N维的反射系数向量x＝[x1,...,xN]T以及N维的透射系数向量y＝[y1,...,yN]T来表示待感知的环境空间；
    S32，将步骤S31中对环境空间进行离散化处理得到的反射系数向量x和透射系数向量y，与步骤S2得到的反射与透射信道模型结合，构建信道响应与环境目标的数学模型如下：
    h＝(H⊙BRe)x+(H⊙BTr)y+v
    ＝ARex+ATry+v
    其中，h是信道响应，x表示环境的反射系数向量，y表示环境的透射系数向量，H是按步骤S31所计算的理想非视距路径矩阵，BRe和BTr是按步骤S32所计算的遮挡效应指示矩阵，v是随机的加性白高斯噪声，⊙表示哈达玛积操作，ARe表示反射信道的测量矩阵，ATr表示透射信道的测量矩阵；
    S33，基于步骤S32中的信道响应与环境目标的数学模型，将环境感知问题转化为广义压缩感知优化问题，表达如下：

    s.t.||h-ARex-ATry||2≤ε
    ARe＝H⊙BRe
    ATr＝H⊙BTr

    其中x表示环境的反射系数向量，y表示环境的透射系数向量，||·||1,||·||2分别是向量的一范数、二范数，表示在x、y的取值空间中寻找对应表达式的最小值，h是信道响应，ARe表示反射信道的测量矩阵，ATr表示透射信道的测量矩阵，ε是松弛变量，H是步骤S21所计算的理想非视距路径矩阵，BRe和BTr是按步骤S22所计算的遮挡效应指示矩阵，g(·,·)是基于位置关系的遮挡效应计算函数，是基站和所有用户终端的位置。
  5. 根据权利要求1所述的方法，其特征在于，所述步骤S4具体为：
    S41，采用模型演进算法，将基本模型表示步骤S1中得到的信道响应数据，具体如下：
    h＝Hx+v    (1)
    其中，h是信道响应，x表示环境的反射系数向量，H是理想非视距路径矩阵，v是随机的加性白高斯噪声，将该简单模型公式(1)建立为压缩感知优化问题，表示如下：

    s.t.||h-Hx||2≤ε
    其中表示在x的取值空间中寻找对应表达式的最小值，ε是松弛变量；对于该问题的求解，采用最大期望-广义近似消息传递算法；
    S42，在得到步骤S41中关于环境目标x的初步求解结果之后，可以对模型进行演进，得到估计的遮挡效应指示矩阵如下：
    其中，g(·,·)是基于位置关系的遮挡效应计算函数，是基站和所有用户终端的位置，此时，可以将信道响应与环境目标的数学模型演进为：
    其中，H是理想非视距路径矩阵，v是随机的加性白高斯噪声，将模型公式(2)建立为压缩感知优化问题，表示如下：



    其中，x是待求的环境目标，是步骤S41的估计结果，表示在x的取值空间中寻找对应表达式的最小值，ε是松弛变量，h是步骤S1得到的信道响应数据，是按计算得到的反射信道测量矩阵，H是理想非视距路径矩阵，是按计算得到的遮挡指示矩阵，g(·,·)是基于位置关系的遮挡效应计算函数，是基站和所有用户终端的位置；
    S43，对测量矩阵进列归一化，并将步骤S42中的压缩感知优化问题转化为：

    s.t.||h-CRexs||2≤ε



    xs＝diag(w)·x
    其中，x是待求的环境目标，是步骤S41的估计结果，表示在x的取值空间中寻找对应表达式的最小值，h是步骤S1得到的信道响应，CRe是经过归一化之 后的有效测量矩阵，xs是经过归一化的环境反射系数，ε是松弛变量，是按计算得到的反射信道测量矩阵，H是理想非视距路径矩阵，是按计算得到的遮挡指示矩阵，g(·,·)是基于位置关系的遮挡效应计算函数，是基站和所有用户终端的位置，函数diag(·)表示取向量元素组成对角阵，w表示的列二范数组成的向量，符号表示矩阵的第n列，表示一个非零常数，N是环境空间的点云个数；对问题进行上述预处理之后，采用最大期望-广义近似消息传递算法对其进行求解；
    S44，在得到步骤S43中关于环境目标x的求解结果之后，可以对模型进行演进，得到估计的遮挡效应指示矩阵和如下：

    其中，g(·,·)是基于位置关系的遮挡效应计算函数，是基站和所有用户终端的位置，此时，可以将信道响应与环境目标的数学模型演进为：
    其中，x表示环境的反射系数向量，y表示环境的透射系数向量，是按计算得到的反射信道测量矩阵，v是随机的加性白高斯噪声；
    对该模型进行观察，可以发现无论是反射系数x还是透射系数y，都依赖于环境中散射体的存在性向量z，因此将建立统一的新模型：
    h＝(H⊙B)z+v,    (3)
    其中，B表示散射体存在性变量的掩膜矩阵，计算方式如下
    其中与分别是反射系数x以及透射系数y的非零部分的均值，将模型公式(3)建模为一个优化问题：

    s.t.||h-(H⊙B)z||2≤ε




    其中，z是待求的散射体存在性向量，x是待求的反射系数向量，y是待求的透射系数向量，表示在z的取值空间中寻找对应表达式的最小值，h是步骤S1得到的信道响应，H是理想非视距路径矩阵，B表示散射体存在性变量的掩膜矩阵，ε是松弛变量，和是步骤S44的估计结果，是基站和所有用户终端的位置，g(·,·)是基于位置关系的遮挡效应计算函数，和是按和计算得到的遮挡指示矩阵，与分别是反射系数x以及透射系数y的非零部分的均值，N是环境空间的点云个数；
    S45，由于步骤S44中优化问题的复杂性，采取迭代求解散射体存在性向量z和修正掩膜矩阵B的交替优化方法，对该交替优化算法叙述如下：
    在求解散射体存在性向量z时，固定掩膜矩阵B，采用最大期望-广义近似消息传递算法进行z的求解；
    在修正掩膜矩阵B时，先固定散射体存在性向量z，并计算得到反射系数和透射系数然后依次计算遮挡效应矩阵和BTr，并合并成为新的掩膜矩阵B；
    S46，在求解散射体存在性向量z和修正掩膜矩阵B的过程之间进行交替迭 代优化，该算法的结束条件有两个：达到最大迭代次数Tmax或者算法收敛到门限τ以下，算法结束即完成了对环境目标的感知。
PCT/CN2023/086613 2022-08-10 2023-04-06 一种基于模型演进的环境感知方法 WO2024032009A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210958277.XA CN115473771A (zh) 2022-08-10 2022-08-10 一种基于模型演进的环境感知方法
CN202210958277.X 2022-08-10

Publications (1)

Publication Number Publication Date
WO2024032009A1 true WO2024032009A1 (zh) 2024-02-15

Family

ID=84367589

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/086613 WO2024032009A1 (zh) 2022-08-10 2023-04-06 一种基于模型演进的环境感知方法

Country Status (2)

Country Link
CN (1) CN115473771A (zh)
WO (1) WO2024032009A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117870034A (zh) * 2024-03-11 2024-04-12 深圳市德尼环境技术有限公司 洁净室环境参数的控制方法、装置及系统

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115473771A (zh) * 2022-08-10 2022-12-13 浙江大学 一种基于模型演进的环境感知方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20200404511A1 (en) * 2018-02-14 2020-12-24 Samsung Electronics Co., Ltd. Device and method for generating environment information for network design in wireless communication system
WO2021238887A1 (en) * 2020-05-25 2021-12-02 Qualcomm Incorporated Environment sensing using radio frequencies
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
CN114747257A (zh) * 2019-11-26 2022-07-12 华为技术有限公司 用于估计无线通信网络中信号遮蔽障碍物和信号反射器的位置的系统和方法
CN115473771A (zh) * 2022-08-10 2022-12-13 浙江大学 一种基于模型演进的环境感知方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20200404511A1 (en) * 2018-02-14 2020-12-24 Samsung Electronics Co., Ltd. Device and method for generating environment information for network design in wireless communication system
CN114747257A (zh) * 2019-11-26 2022-07-12 华为技术有限公司 用于估计无线通信网络中信号遮蔽障碍物和信号反射器的位置的系统和方法
WO2021238887A1 (en) * 2020-05-25 2021-12-02 Qualcomm Incorporated Environment sensing using radio frequencies
CN115462112A (zh) * 2020-05-25 2022-12-09 高通股份有限公司 使用射频进行环境感测
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
CN115473771A (zh) * 2022-08-10 2022-12-13 浙江大学 一种基于模型演进的环境感知方法

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117870034A (zh) * 2024-03-11 2024-04-12 深圳市德尼环境技术有限公司 洁净室环境参数的控制方法、装置及系统
CN117870034B (zh) * 2024-03-11 2024-05-07 深圳市德尼环境技术有限公司 洁净室环境参数的控制方法、装置及系统

Also Published As

Publication number Publication date
CN115473771A (zh) 2022-12-13

Similar Documents

Publication Publication Date Title
WO2024032009A1 (zh) 一种基于模型演进的环境感知方法
CN110365612A (zh) 一种基于近似消息传递算法的深度学习波束域信道估计方法
CN104767535B (zh) 一种低复杂度块状稀疏信号重构方法
WO2024021440A1 (zh) 一种迭代聚焦式毫米波一体化通信与感知方法
CN113965881B (zh) 遮挡效应下的毫米波一体化通信与感知方法
Wang et al. A unified framework for guiding generative ai with wireless perception in resource constrained mobile edge networks
CN114675238A (zh) 一种雷达通信一体化波形直接优化方法和系统
CN113825236A (zh) 一种无线网络中感知、计算和通信的融合方法
US20230362039A1 (en) Neural network-based channel estimation method and communication apparatus
CN110113084B (zh) Mimo闭环传输系统的信道预测方法
Chen et al. The importance of data compression for energy efficiency in sensor networks
Xu et al. Over-the-air learning rate optimization for federated learning
CN114554398A (zh) 室内定位方法、第一定位服务器及室内定位系统
Ngo et al. Deep learning-based adaptive beamforming for mmWave wireless body area network
CN110632584B (zh) 基于并联径向基网络的目标外辐射源被动定位方法
Agostini et al. Federated learning for multipoint channel charting
Ayoubi et al. FPGA implementation of generalized maximal ratio combining receiver diversity
Yang et al. Multi-Frequency Based CSI Compression for Vehicle Localization in Intelligent Transportation System
Wang et al. Physical model-driven deep networks for through-the-wall radar imaging
Xing et al. Intelligent reflecting surface aided environment sensing with cnn-based correction
CN117527017A (zh) 信息发送方法、信息发送装置、电子设备及存储介质
CN116367305B (zh) 一种基于掩码自编器和信号引导的无线定位方法及系统
Gao et al. Research on Joint Beamforming of High-Speed Railway Millimeter-wave MIMO Communication with Reconfigurable Intelligent Surface
CN115001548B (zh) 一种基于反射和透射超表面的noma无线传输方法
Yang et al. A Learning-Based Collaborative Wireless Sensing Framework

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: 23851239

Country of ref document: EP

Kind code of ref document: A1