CN115622839A - Channel propagation state identification method and device and electronic equipment - Google Patents
Channel propagation state identification method and device and electronic equipment Download PDFInfo
- Publication number
- CN115622839A CN115622839A CN202110783840.XA CN202110783840A CN115622839A CN 115622839 A CN115622839 A CN 115622839A CN 202110783840 A CN202110783840 A CN 202110783840A CN 115622839 A CN115622839 A CN 115622839A
- Authority
- CN
- China
- Prior art keywords
- cir
- input data
- propagation state
- receiving end
- components
- Prior art date
- Legal status (The legal status 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 status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 50
- 230000010287 polarization Effects 0.000 claims abstract description 148
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 114
- 238000012545 processing Methods 0.000 claims abstract description 32
- 238000007781 pre-processing Methods 0.000 claims abstract description 29
- 230000004044 response Effects 0.000 claims abstract description 10
- 238000004590 computer program Methods 0.000 claims description 15
- 241000209094 Oryza Species 0.000 claims description 12
- 235000007164 Oryza sativa Nutrition 0.000 claims description 12
- 235000009566 rice Nutrition 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 11
- 238000003860 storage Methods 0.000 claims description 10
- 230000008569 process Effects 0.000 claims description 8
- 238000000605 extraction Methods 0.000 claims description 3
- 230000006870 function Effects 0.000 description 27
- 238000011176 pooling Methods 0.000 description 25
- 238000010586 diagram Methods 0.000 description 14
- 210000002569 neuron Anatomy 0.000 description 14
- 238000012360 testing method Methods 0.000 description 10
- 238000012549 training Methods 0.000 description 10
- 230000004913 activation Effects 0.000 description 8
- 230000000694 effects Effects 0.000 description 7
- 239000000284 extract Substances 0.000 description 6
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 3
- 230000009977 dual effect Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 238000012805 post-processing Methods 0.000 description 3
- 238000012163 sequencing technique Methods 0.000 description 3
- 238000004088 simulation Methods 0.000 description 3
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 241000196324 Embryophyta Species 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000002407 reforming Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000007493 shaping process Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
Images
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
- H04L25/0212—Channel estimation of impulse response
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/0001—Systems modifying transmission characteristics according to link quality, e.g. power backoff
- H04L1/0023—Systems modifying transmission characteristics according to link quality, e.g. power backoff characterised by the signalling
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/02—Arrangements for detecting or preventing errors in the information received by diversity reception
- H04L1/06—Arrangements for detecting or preventing errors in the information received by diversity reception using space diversity
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/02—Arrangements for detecting or preventing errors in the information received by diversity reception
- H04L1/06—Arrangements for detecting or preventing errors in the information received by diversity reception using space diversity
- H04L1/0618—Space-time coding
- H04L1/0675—Space-time coding characterised by the signaling
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Power Engineering (AREA)
- Quality & Reliability (AREA)
- Radio Transmission System (AREA)
Abstract
The invention provides a channel propagation state identification method and device and electronic equipment. The method comprises the following steps: extracting CIR components of channel impulse response CIR received by a signal receiving end in different polarization directions; obtaining the statistical characteristics among the multipath according to the CIR component; performing data preprocessing on the CIR component and the statistical characteristic to obtain input data; and processing the input data based on a Convolutional Neural Network (CNN) model to acquire a channel propagation state. In the embodiment of the application, CIR components among transmitting and receiving antennas in different polarization directions and statistical characteristics among multiple paths are used as input data of a CNN model, and a channel propagation state is obtained based on the CNN model. The CIR characteristic and the CNN of the dual-polarized antenna are combined, and the channel propagation state can be accurately distinguished without relying on prior information.
Description
Technical Field
The present invention relates to the field of communications technologies, and in particular, to a method and an apparatus for identifying a channel propagation state, and an electronic device.
Background
In the field Of indoor wireless positioning, many wireless positioning technologies are based on Time Of Arrival (ToA), time Difference Of Arrival (TDoA), angle Of Arrival (DoA), and Received Signal Strength (RSS). The technology for positioning based on parameters such as ToA and DoA is very sensitive to reliability of Line of Sight (LoS) signals, and when LoS signals are unavailable, namely, a system is in a non-Line of Sight (NLoS) state, a large number of positioning errors can be introduced into the whole system. For such applications in complex multipath propagation scenarios, correlation studies are needed to mitigate the impact of NLoS propagation, where a key step is to identify the propagation state of the channel.
The existing identification methods of channel propagation states can be divided into two categories: one is a statistical information identification-based method, which is mainly based on the change of signal statistical characteristics caused by two different propagation states; the other type is a method based on propagation path loss, which realizes channel propagation state identification by extracting channel characteristic parameters and combining hypothesis testing according to the distribution of the channel characteristic parameters. Both of the above two methods have high requirements for threshold division, and the threshold is sensitive to the environment, and a suitable threshold needs to be determined by enough prior information, so that a high accuracy rate of the division can be achieved.
Disclosure of Invention
The invention aims to provide a channel propagation state identification method, a channel propagation state identification device and electronic equipment, which are used for solving the problem that the existing channel propagation state identification method needs to depend on prior information.
The embodiment of the invention provides a method for identifying a channel propagation state, which comprises the following steps:
extracting CIR components of Channel Impulse Response (CIR) received by a receiving end in different polarization directions;
obtaining the statistical characteristics among the multipath according to the CIR component;
performing data preprocessing on the CIR component and the statistical characteristic to obtain input data;
and processing the input data based on a Convolutional Neural Network (CNN) model to obtain a channel propagation state.
Optionally, the CIR component comprises:
a first CIR component generated by array elements of the vertical polarization of the sending end and the vertical polarization of the receiving end;
a second CIR component generated by array elements of vertical polarization of the sending end and horizontal polarization of the receiving end;
a third CIR component generated by array elements of horizontal polarization of the sending end and vertical polarization of the receiving end;
and a fourth CIR component generated by the array elements of the horizontal polarization of the transmitting end and the horizontal polarization of the receiving end.
Optionally, the statistical characteristics include: mean and variance of the CIR components.
Optionally, the obtaining statistical characteristics between multipaths according to the CIR component includes:
calculating the mean value of CIR components among L pieces of multipath, wherein the calculation formula of the mean value is as follows:
wherein,representing a mean of the CIR components; | H rxu,txv I represents the CIR component generated by the array element with the receiving end antenna polarization direction u and the transmitting end antenna polarization direction v; l represents the number of multi-paths, and L represents the first path;
calculating the variance of CIR components among L pieces of multipath, wherein the variance is calculated according to the following formula:
wherein,representing a variance of the CIR component; | H rxu,txv I represents the CIR component generated by the array element with the receiving end antenna polarization direction u and the transmitting end antenna polarization direction v; l represents the number of multipaths, and L represents the ith path.
Optionally, the performing data preprocessing on the CIR component and the statistical characteristic to obtain input data includes:
and dividing the CIR components in different polarization directions and the statistical characteristics into dimensions to obtain M × N dimensional input data.
Optionally, the performing data preprocessing on the CIR component and the statistical characteristic to obtain input data includes:
and dividing the CIR components in different polarization directions, the statistical characteristics and the auxiliary characteristics of the signals into dimensions to obtain M × N dimension input data.
Optionally, the assist features comprise at least one of:
a rice factor;
a signal rise time;
root mean square delay spread;
kurtosis of the received power.
Optionally, the processing the input data based on the CNN model to obtain a channel propagation state includes:
inputting the input data into a Convolutional Neural Network (CNN) model to obtain the probability of the channel propagation state;
wherein the channel propagation state comprises: a line-of-sight propagation state or a non-line-of-sight propagation state.
An embodiment of the present invention further provides an electronic device, including: memory, transceiver, processor:
a memory for storing a computer program; a transceiver for transceiving data under the control of the processor; a processor for reading the computer program in the memory and performing the following operations:
extracting CIR components of channel impulse response CIR received by a signal receiving end in different polarization directions;
obtaining the statistical characteristics among the multipath according to the CIR component;
performing data preprocessing on the CIR component and the statistical characteristic to obtain input data;
and processing the input data based on a Convolutional Neural Network (CNN) model to acquire a channel propagation state.
Optionally, the CIR component comprises:
a first CIR component generated by array elements of the vertical polarization of the sending end and the vertical polarization of the receiving end;
a second CIR component generated by array elements of vertical polarization of the sending end and horizontal polarization of the receiving end;
a third CIR component generated by array elements of horizontal polarization of the sending end and vertical polarization of the receiving end;
and a fourth CIR component generated by the array elements of the horizontal polarization of the transmitting end and the horizontal polarization of the receiving end.
Optionally, the statistical characteristics include: mean and variance of the CIR components.
Optionally, the processor is configured to read the computer program in the memory and perform the following operations:
calculating the mean value of CIR components among L multipath, wherein the calculation formula of the mean value is as follows:
wherein,representing a mean of the CIR components; | H rxu,txv I represents the CIR component generated by the array element with the receiving end antenna polarization direction u and the transmitting end antenna polarization direction v; l represents the number of multi-paths, and L represents the first path;
calculating the variance of CIR components among L pieces of multipath, wherein the variance is calculated according to the following formula:
wherein,representing a variance of the CIR component; | H rxu,txv I represents a CIR component generated by an array element with the receiving end antenna polarization direction u and the transmitting end antenna polarization direction v; l represents the number of multipaths, and L represents the ith path.
Optionally, the processor performs data preprocessing on the CIR component and the statistical characteristic to obtain input data, including:
and dividing the CIR components in different polarization directions and the statistical characteristics into dimensions to obtain M × N dimensional input data.
Optionally, the processor performs data preprocessing on the CIR component and the statistical characteristic to obtain input data, including:
and dividing the CIR components in different polarization directions, the statistical characteristics and the auxiliary characteristics of the signals into dimensions to obtain M × N dimension input data.
Optionally, the assist feature comprises at least one of:
a rice factor;
a signal rise time;
root mean square delay spread;
kurtosis of the received power.
Optionally, the processing, by the processor, the input data based on a convolutional neural network CNN model to obtain a channel propagation state includes:
inputting the input data into a Convolutional Neural Network (CNN) model to obtain the probability of the channel propagation state;
wherein the channel propagation state comprises: a line-of-sight propagation state or a non-line-of-sight propagation state.
The embodiment of the present invention further provides an apparatus for identifying a channel propagation status, including:
the information extraction unit is used for extracting CIR components of channel impact response CIRs received by the signal receiving end in different polarization directions;
a first processing unit, configured to obtain statistical characteristics between multipaths according to the CIR component;
the second processing unit is used for carrying out data preprocessing on the CIR component and the statistical characteristic to obtain input data;
and the state identification unit is used for processing the input data based on a Convolutional Neural Network (CNN) model to acquire a channel propagation state.
Embodiments of the present invention also provide a processor-readable storage medium, on which a computer program is stored, which, when being executed by a processor, implements the steps of the above-mentioned method for identifying a channel propagation status.
The technical scheme of the invention has the beneficial effects that:
according to the embodiment of the application, CIR components among transmitting and receiving antennas in different polarization directions and the statistical characteristics among multiple paths are used as input data of a CNN model, and a channel propagation state is obtained based on the CNN model. The CIR characteristic and the CNN of the dual-polarized antenna are combined, and the channel propagation state can be accurately distinguished without relying on prior information.
Drawings
Fig. 1 is a flow chart illustrating a method for identifying a channel propagation status according to an embodiment of the present invention;
FIG. 2 shows one of the CNN model diagrams of an embodiment of the invention;
fig. 3 is a schematic diagram illustrating a location configuration of a base station according to an embodiment of the present invention;
FIG. 4 is a second schematic diagram of a CNN model according to an embodiment of the invention;
FIG. 5 is a second schematic diagram of a location configuration of a base station according to an embodiment of the invention;
FIG. 6 is a third schematic diagram of a CNN model according to an embodiment of the present invention;
FIG. 7 is a third exemplary location allocation diagram of a base station according to the present invention;
FIG. 8 is a fourth illustration of a CNN model according to an embodiment of the invention;
fig. 9 is a schematic structural diagram of an apparatus for identifying a propagation state of a channel according to an embodiment of the present invention;
fig. 10 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
In order to make the technical problems, technical solutions and advantages of the present invention more apparent, the following detailed description is given with reference to the accompanying drawings and specific embodiments. In the following description, specific details such as specific configurations and components are provided only to help the full understanding of the embodiments of the present invention. Thus, it will be apparent to those skilled in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the invention. In addition, descriptions of well-known functions and constructions are omitted for clarity and conciseness.
It should be appreciated that reference throughout this specification to "one embodiment" or "an embodiment" means that a particular feature, structure or characteristic described in connection with the embodiment is included in at least one embodiment of the present invention. Thus, the appearances of the phrases "in one embodiment" or "in an embodiment" in various places throughout this specification are not necessarily all referring to the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
In various embodiments of the present invention, it should be understood that the sequence numbers of the following processes do not mean the execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.
The term "and/or" in the embodiments of the present invention describes an association relationship of associated objects, and indicates that three relationships may exist, for example, a and/or B may indicate: a exists alone, A and B exist simultaneously, and B exists alone. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship.
In the embodiments of the present application, the term "plurality" means two or more, and other terms are similar thereto.
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
In making the description of the embodiments of the present invention, some concepts used in the following description will first be explained.
Specifically, embodiments of the present invention provide a method and an apparatus for identifying a channel propagation state, and an electronic device, so as to solve a problem that an existing method for identifying a channel propagation state needs to rely on prior information.
As shown in fig. 1, an embodiment of the present invention provides a method for identifying a channel propagation state, which specifically includes the following steps:
The receiving end is a signal receiving end, can be a base station or a terminal, the receiving end and the signal sending end are configured to be dual-polarized antennas, and the dual-polarized antennas comprise horizontal polarized antennas and vertical polarized antennas. The horizontal polarized antenna and the vertical polarized antenna of the sending end send the same information in sequence, and the receiving end extracts the CIR components between the receiving and sending antennas in different polarization directions according to the CIRs received by the two polarized antennas.
And step 102, obtaining the statistical characteristics among the multipath according to the CIR component.
The statistical properties may include mean and variance of the CIR components, i.e., inter-multipath statistical properties calculated for extracted CIR components between the transceiving antennas of different polarization directions. The number of multipaths can be estimated, for example: the number of multipaths can be detected by using Minimum Description Length (MDL), toplitz matrix (Toeplitz) decomposition, and the like.
And 103, carrying out data preprocessing on the CIR component and the statistical characteristic to obtain input data. The data preprocessing may be a data shaping process of the CIR component and the statistical characteristic to convert the original data into a matrix form as input data.
And 104, processing the input data based on a Convolutional Neural Network (CNN) model to acquire a channel propagation state.
The channel propagation states include LoS and NLoS states. And (4) automatically extracting input data characteristics by using a CNN model, and further identifying the LoS \ NLoS propagation state. The CNN model is a training model obtained by continuously iterating a large amount of M multiplied by N dimensional data with real LoS/NLoS labels as a training set to enable a final loss function to converge to a minimum value. Inputting the input data obtained according to the CIR component and the statistical characteristic into the CNN model, and obtaining the recognition result of the channel propagation state between the receiving end and the transmitting end.
In the prior art, when channel propagation state identification is performed, sample features need to be extracted from multiple dimensions, corresponding thresholds are set for different features, channel propagation state judgment is performed based on the thresholds, and the thresholds need to be set depending on enough prior information to determine a proper threshold value range, so that high distinguishing accuracy is achieved. According to the embodiment of the application, CIR components among transmitting and receiving antennas in different polarization directions and the statistical characteristics among multiple paths are used as input data of a CNN model, and a channel propagation state is obtained based on the CNN model. The CIR characteristic and the CNN of the dual-polarized antenna are combined, and the channel propagation state can be accurately distinguished without relying on prior information.
As an alternative embodiment, the CIR components include CIR components of transceiver antennas of four different polarization directions:
(1) The first CIR component generated by the array elements of the transmitting end vertical polarization and the receiving end vertical polarization can be recorded as | H rxV,txV |;
(2) The second CIR component generated by the array elements of the vertical polarization of the transmitting end and the horizontal polarization of the receiving end can be recorded as | H rxH,txV |;
(3) The third CIR component generated by array elements of the transmitting end horizontal polarization and the receiving end vertical polarization can be recorded as | H rxV,txH |;
(4) The fourth CIR component generated by the array elements of the transmitting end horizontal polarization and the receiving end horizontal polarization can be recorded as | H rxH,txH |。
Calculating statistical characteristics among the multipaths by using the extracted CIR components among the transmitting and receiving antennas in the four different polarization directions, wherein the statistical characteristics may include: mean and variance of the CIR components.
Specifically, the obtaining the statistical characteristics among the multipaths according to the CIR component may include: calculating the mean value of CIR components among L multipath, wherein the calculation formula of the mean value is as follows:
wherein,representing a mean of the CIR components; | H rxu,txv I represents the CIR component generated by the array element with the receiving end antenna polarization direction u and the transmitting end antenna polarization direction v; l is a radical of an alcoholRepresents the number of multipaths, and l represents the ith path.
Calculating the variance of CIR components among L pieces of multipath, wherein the variance is calculated according to the following formula:
wherein,representing a variance of the CIR component; | H rxu,txv I represents the CIR component generated by the array element with the receiving end antenna polarization direction u and the transmitting end antenna polarization direction v; l represents the number of multipaths, and L represents the ith path.
As an alternative embodiment, the pre-processing the CIR components and the statistical characteristics to obtain input data includes: and dividing the CIR components in different polarization directions and the statistical characteristics into dimensions to obtain M × N dimension input data.
In this embodiment, the data preprocessing is to make the CIR components between the transmitting and receiving antennas of four different polarization directions and the statistical characteristics between two multipaths (respectively, the CIR components are the statistical characteristics between the transmitting and receiving antennas of four different polarization directionsAnd) And sequentially forming M multiplied by N dimensional data as the input of the CNN model.
As an alternative embodiment, the pre-processing the CIR component and the statistical characteristic to obtain the input data includes: and dividing the CIR components in different polarization directions, the statistical characteristics and the auxiliary characteristics of the signals into dimensions to obtain M x N dimensional input data.
This embodiment adds assist features when dividing the data dimension, which may include at least one of:
a rice factor;
a signal rise time;
root mean square delay spread;
kurtosis of the received power.
In this embodiment, the input data of the CNN model may be divided into dimensions based on the CIR components in different polarization directions and the statistical characteristics, or may be divided into dimensions based on the CIR components in different polarization directions, the statistical characteristics, and the auxiliary features. The channel information that can reflect the propagation state of the channel includes not only polarization information (i.e., the CIR component and the statistical characteristic), but also auxiliary information such as the rice factor, the signal rise time, the root mean square delay spread, and the kurtosis of the received power. The auxiliary information can reflect the difference of LoS and NLoS path statistical characteristics to a certain extent, so that when the channel propagation state identification is carried out, the auxiliary characteristic is combined to carry out data processing and then the data processing is used as input data, the input data containing the auxiliary characteristic is processed based on a CNN model, and a more accurate channel propagation state identification result can be obtained. In a simulation experiment, the auxiliary information is added, and the characteristics are automatically extracted and identified through a CNN model, so that LoS/NLoS identification accuracy higher than that of the single polarization information can be achieved.
When the dimensionality of the data is divided, the dimensionality division is carried out after the multiple data features are integrated. The way of data partitioning into dimensions is for example: the obtained CIR components a and b are integrated into 1 x (a and b) dimension, and then sequentially arranged in a row according to a certain sequence such as the sequence of H11 → H12 → H21 → H22 to obtain data of 1 x (4 a and b) dimension, wherein H11, H12, H21 and H22 are CIR components in different polarization directions respectively. When data preprocessing is carried out, if other auxiliary identification feature information (namely the auxiliary features) exists, the auxiliary identification feature information is shaped into a 1*c dimensional form and is supplemented to the obtained 1 x (4 x a b) dimensional data. Finally, the data with dimension of M multiplied by N is formed through reforming, and the value of M, N can be adjusted through the results of multiple experiments.
As an optional embodiment, the processing the input data based on the CNN model to obtain the channel propagation state includes: inputting the input data into a Convolutional Neural Network (CNN) model to obtain the probability of the channel propagation state; wherein the channel propagation state comprises: a line-of-sight propagation state or a non-line-of-sight propagation state.
The embodiment utilizes the CNN model to automatically extract the data characteristics to identify the LoS/NLoS propagation state. And inputting the input data into the trained CNN model to obtain the recognition result of the channel propagation state, namely the probability of the LoS \ NLoS propagation state.
The CNN model is shown in fig. 2, for example, and the CNN structure includes a structure formed by combining a plurality of convolutional layer stacking pooling layers, full link layers, and SoftMax layers. The convolution layer performs convolution operation with input data through a convolution kernel, and extracts a feature generation feature graph as the input of the next layer. Can be expressed by the following formula:
wherein M is j An input feature map representing a jth neuron, w is a convolution kernel connecting the (g-1) th layer to the g-th layer,is the bias of the jth neuron at layer g,represents the characteristics of the ith neuron of the (g-1) th layer,representing the characteristics of the jth neuron at the g-th level. f is a linear rectification function (ReLU) activation function, the formula of which is as follows:
when x is greater than 0, the gradient is constant to 1, the problem of gradient dissipation of a back propagation algorithm in the process of optimizing a deep neural network is solved, and the convergence speed is high.
The convolutional layer is followed by a pooling layer, the pooling layer has the function of carrying out aggregation statistics on the feature information of adjacent regions, the probability statistics is used for reducing the dimension and the calculation amount, and meanwhile, the use of pooling reduces the overfitting and the propagation of noise. After the convolution layer and the pooling layer are processed, deeper features are extracted from the input data.
The full-connection layer carries out weighted summation on the characteristics of the previous layer (namely the pooling layer) and converts the output into a one-dimensional vector; the last fully-connected layer uses the SoftMax activation function at the output. The method has the main function of combining the extracted features of the previous layers together for classification, the output result can be interpreted as the probability of each type, and the probability formula is as follows:
where X is the output of the previous layer (i.e., the output of the pooling layer), T represents the transpose, and ω represents the weight of the neuron; r is the number of labels, R is the R-th label, and in this embodiment, R =2,2 labels are "LoS" and "NLoS", respectively; s denotes the subscript of the predicted class, s =1,2; p represents the probability of LoS or NLoS classification.
In this embodiment, the CNN model may employ: (1) dropout optimization, wherein an overfitt phenomenon can be reduced after a dropout function is introduced into a full connection layer; (2) Adam optimization, the setting of the learning rate determines whether the neural network can converge to a global minimum. Choosing a higher learning rate does not converge to a global minimum, and choosing a lower learning rate helps the neural network converge to a global minimum, but it takes longer to converge. The Adam optimization algorithm is an extension of a random gradient descent method, and independent adaptive learning rates are designed for different parameters by calculating first moment estimation and second moment estimation of a gradient, so that a CNN can reach a good convergence result in a relatively short time. In addition, because a standard network architecture which can bring high accuracy in all test sets does not exist, different architectures can be tried to achieve better classification performance of the CNN by adjusting the number of convolution layers and pooling layers and changing the size and the number of convolution cores through experiments.
The CNN model training needs to collect a large amount of M × N dimensional data with real LoS/NLoS labels, and part of the collected M × N dimensional data is used as a training set and is iterated continuously, so that the final loss function can be converged to the minimum value; the remaining portion of the data serves as a test set to test whether the trained model achieves the desired effect. And storing the weight values of each layer of the trained CNN network to obtain the required CNN model.
Optionally, after the input data is processed based on the CNN model to obtain the channel propagation state, the accuracy of the recognition result may be determined. For example: and taking the CIR of the receiving end with the known actual channel propagation state as input data of the CNN model, and comparing the obtained identification result of the channel propagation state with the actual channel propagation state to determine the accuracy. Taking the identification result of the channel propagation state as LoS as an example, B CIRs with the maximum power can be selected from the CIRs of A Base stations, the probability of the CNN model predicted as LoS is sequenced, and C Base Stations (BS) with the maximum probability of LoS are selected to be judged as BS containing LoS, wherein A is more than or equal to B and more than or equal to C.
The following describes an implementation process of the channel propagation status identification method according to an embodiment.
Example one:
(1) Taking a signal sending end as a base station as an example, configuring the position of the base station, wherein the position of the base station is configured to be an Indoor factory (IOO) scene with a sparse cluster and a high base station height; IOO: l =120m, W =50m, H =3m, as shown in fig. 3.
(2) The receiving end and the transmitting end antennas are configured as follows:
the transmitting end is configured to be a dual-polarized antenna with a polarization angle of 4 multiplied by 4+/-45 degrees;
the receiving end is configured as a dual polarized antenna with 2 x 2 vertical/horizontal polarization angles.
(3) The structure of the CNN model is shown in fig. 4:
a 5-layer CNN structure is used, which includes 2 convolutional-layer stacked pooling layers, 2 fully-connected layers, and a SoftMax function output layer, as shown in fig. 4. The convolution layer uses convolution kernel with convolution kernel of 5 × 5 size step size 1, the pooling kernel uses maximum pooling kernel with 2 × 2 size step size 2, and both adopt zero-filling strategy. A ReLu nonlinear activation function was used after each convolutional layer. The first convolutional layer comprises 32 convolutional kernels, the input data dimension is 8 x 9, the output dimension is 8 x 9 x 32, and the dimension after the maximum pooling layer is 4 x 5 x 32; the second convolutional layer contains 64 convolutional kernels, the output dimension is 4 × 5 × 64, and the output dimension after the maximum pooling is 2 × 3 × 64; the fully-connected layer includes 384 neurons in the first layer, 64 neurons in the second layer, and 2 neurons in the output layer, and the Softmax activation function normalizes the output to range between (0,1).
(4) The step of identifying the propagation status of the channel comprises the following steps:
the method comprises the following steps: the sending end horizontal polarization antenna and the vertical polarization antenna send the same information in sequence, and the receiving end extracts CIR components in four different polarization directions according to the information received by the two polarization antennas: | H rxV,txV |、|H rxH,txV |、|H rxV,txH |、|H rxH,txH L, |; calculating the statistic characteristics among the multi-paths for the extracted CIR components among the transmitting and receiving antennas in the four different polarization directions: mean valueSum varianceThe calculation formula is not described in detail.
Step two: data preprocessing: statistical characteristics of CIR components and two kinds of multi-paths between transmitting and receiving antennas in four different polarization directions: (And) And sequentially forming 8 × 9 dimensions as input of the CNN model.
Step three: and inputting the data into a trained CNN model to identify the propagation state of LoS \ NLoS.
Wherein the training of the middle CNN model comprises: and the terminal randomly scatters 600 points in the arrangement range of 12 base stations, collects 7200 pieces of 8 × 9-dimensional data obtained in the second step, takes the corresponding real LoS/NLoS propagation state information as a data label, and inputs 6000 pieces of collected data as a training set into the CNN network for continuous iteration, so that the final loss function can converge to a minimum value. The remaining 1200 data are used as a test set to test whether the trained model achieves the desired effect. And (4) storing the weight of each layer of the trained CNN network to obtain the CNN model required by the third step.
Step four (optional): and (3) data post-processing: selecting 10 CIRs with the maximum power from the CIRs of 12 base stations, sequencing probability results of CNN predicted as LOS, selecting 3 BSs with the maximum LoS probability to judge as the BSs containing LoS, and calculating the accuracy.
Example two:
(1) Taking a signal sending end as a base station as an example, firstly, configuring the position of the base station, wherein the position of the base station is configured to be an Indoor factory (InF-DH) scene with dense clusters and high base station height; inF-DH: l = 120mw =60m d =20m, as shown in fig. 5.
(2) The receiving end antenna and the transmitting end antenna are configured as follows:
the transmitting end is configured to be a dual-polarized antenna with a polarization angle of 4 multiplied by 4+/-45 degrees;
the receiving end is configured as a dual polarized antenna with 2 x 2 vertical/horizontal polarization angles.
(3) CNN model structure, as shown in FIG. 6.
This example employs a 6-layer CNN structure, which contains 3 convolutional layer stack pooling layers, 2 fully-connected layers, and a SoftMax function output layer, as shown in fig. 6. The convolution kernel used by the convolution layer is a convolution kernel with the size of 3 multiplied by 3 and the step size of 1, the pooling kernel is a maximum pooling kernel with the size of 2 multiplied by 2 and the step size of 2, and zero padding strategies are adopted. A ReLu nonlinear activation function was used after each convolutional layer. The first convolution layer comprises 32 convolution kernels, the input data dimension is 9 x 9, the output dimension is 9 x 32, and the dimension after the maximum pooling layer is 5 x 32; the second convolutional layer contains 64 convolutional kernels, the output dimension is 5 × 5 × 64, and the output dimension after the maximum pooling is 3 × 3 × 64; the third layer of convolutional layer comprises 64 convolutional kernels, and the output dimensionality of pooling is 2 multiplied by 64; the first layer of the fully-connected layer comprises 256 neurons, the second layer comprises 64 neurons, the output layer comprises 2 neurons, and the output is normalized by the Softmax activation function to be within the range of (0,1).
(4) And identifying the propagation state of the channel:
the method comprises the following steps: the sending end horizontal polarization antenna and the vertical polarization antenna send the same information in sequence, and the receiving end extracts the CIR components among the receiving and sending antennas in four different polarization directions according to the information received by the two polarization antennas: | H rxV,txV |、|H rxH,txV |、|H rxV,txH |、|H rxH,txH L; then calculating the statistic characteristics among the multi-paths for the extracted CIR components among the transmitting and receiving antennas in the four different polarization directions: mean valueSum varianceThe calculation formula is not described in detail.
Step two: preprocessing of data: the CIR components between the transmitting and receiving antennas in four different polarization directions and the statistical characteristics between two kinds of multi-paths are calculated (And) And the Rice factor is sequentially integrated into 9 x 9 dimensional data as input to the CNN model.
Step three: and inputting the data into a trained CNN model to identify the propagation state of LoS \ NLoS.
Wherein the training of the middle CNN model comprises: the terminal randomly scatters 600 points within the arrangement range of 18 base stations, collects 10800 pieces of 9 × 9-dimensional data described in step two, takes corresponding real LoS/NLoS propagation state information as a label of the data, and then inputs 9000 pieces of the collected data as a training set into the CNN network for continuous iteration, so that the final loss function can converge to a minimum value. The remaining 1800 data were used as a test set to test whether the trained model achieved the desired effect. And (4) storing the weight of each layer of the trained CNN network to obtain the CNN model required by the third step.
Step four (optional): and (3) data post-processing: selecting 10 CIRs with the maximum power from the CIRs of 18 base stations, sequencing probability results of CNN predicted as LoS, selecting 3 BSs with the maximum probability of LoS to judge as the BSs containing LoS, and calculating the accuracy.
The simulation result shows that the accuracy rate of the BS including the LoS path can reach 98.3% when 3 BSs with the highest LoS probability are selected in the embodiment.
Example three:
(1) Taking a signal sending end as a base station as an example, firstly configuring the position of the base station, wherein the position of the base station is configured into an indoor plant InF-SH scene with a sparse cluster and a higher base station height; inF-SH: l =300mw =150m d =50m, as shown in fig. 7.
(2) The receiving end antenna and the transmitting end antenna are configured as follows:
the transmitting end is configured to be a dual-polarized antenna with a polarization angle of 4 multiplied by 4+/-45 degrees;
the receiving end is configured as a dual polarized antenna with 2 x 2 vertical/horizontal polarization angles.
(3) CNN model structure, as shown in fig. 8.
This example employs a 6-layer CNN structure, which contains 3 convolutional layer stack pooling layers, 2 fully-connected layers, and a SoftMax function output layer, as shown in fig. 8. The convolution kernel used by the convolution layer is a convolution kernel with the size of 5 multiplied by 5 and the step size of 1, the pooling kernel is a maximum pooling kernel with the size of 2 multiplied by 2 and the step size of 2, and zero padding strategies are adopted. A ReLu nonlinear activation function was used after each convolutional layer. The first convolutional layer comprises 64 convolutional kernels, the input data dimension is 9 x 9, the output dimension is 9 x 64, the dimension after the maximum pooling layer is 5 x 64, the second convolutional layer comprises 64 convolutional kernels, the output dimension is 5 x 64, and the output dimension after the maximum pooling is 3 x 64; the third layer of convolutional layer comprises 64 convolutional kernels, and the output dimensionality of pooling is 2 multiplied by 64; the fully connected layer has 256 neurons in the first layer, 64 neurons in the second layer, 2 neurons in the output layer, and the output is normalized by the Softmax activation function to be within the range of (0,1).
(4) And identifying the propagation state of the channel:
the method comprises the following steps: the horizontal polarized antenna and the vertical polarized antenna of the sender send the same information in sequence, and the receiver extracts four CIR components in different polarization directions according to the information received by the two polarized antennas: | H rxV,txV |、|H rxH,txV |、|H rxV,txH |、|H rxH,txH L; then calculating the statistic characteristics among the multi-paths for the extracted CIR components among the transmitting and receiving antennas in the four different polarization directions:sum varianceThe calculation formula is not described in detail.
Step two: preprocessing of data: the CIR components between the transmitting and receiving antennas in four different polarization directions and the statistical characteristics between two kinds of multi-paths are calculated (And) The rice factor and the root mean square delay spread are sequentially integrated into 9 x 9 dimensional data as input to the CNN model.
Step three: and inputting the data into a trained CNN model to identify the propagation state of LoS \ NLoS.
Wherein the training of the middle CNN model comprises: the terminal randomly scatters 600 points within the arrangement range of 18 base stations, collects 10800 pieces of 9 × 9-dimensional data described in step two, takes corresponding real LoS/NLoS propagation state information as a label of the data, and then inputs 9000 pieces of the collected data as a training set into the CNN network for continuous iteration, so that the final loss function can converge to a minimum value. The remaining 1800 data were used as a test set to test whether the trained model achieved the desired effect. And D, storing the weight of each layer of the trained CNN network to obtain the CNN model required by the step three.
Step four (optional): and (3) data post-processing: selecting 10 CIRs with the maximum power from the CIRs of 18 base stations, sequencing probability results of CNN predicted as LoS, selecting 3 BSs with the maximum probability of LoS to judge as the BSs containing LoS, and calculating the accuracy.
The result obtained through simulation shows that the accuracy rate of the BS including the LoS path can reach 99% when 3 BSs with the maximum LoS probability are selected in the embodiment.
It should be noted that, when preprocessing data, in addition to the above rice factor and root mean square delay spread, one or more of other auxiliary features may be added to the CIR component and the statistical characteristic and processed as input data of the CNN model, for example: data dimension division is performed by using CIR components of four different polarization directions, two statistical characteristics, a rice factor, signal rise time, root mean square delay spread and kurtosis of received power, and a data dimension division mode is not described herein. Because the auxiliary characteristics can reflect the difference of the LoS and NLoS path statistical characteristics to a certain extent, when the channel propagation state is identified, the auxiliary information is added, the characteristics are automatically extracted and identified through a CNN model, and the LoS/NLoS identification accuracy rate higher than that of the LoS/NLoS identification rate which only uses polarization information can be achieved.
According to the embodiment of the application, CIR components among transmitting and receiving antennas in different polarization directions and the statistical characteristics among multiple paths are used as input data of a CNN model, and a channel propagation state is obtained based on the CNN model. The CIR characteristic and the CNN of the dual-polarized antenna are combined, and the channel propagation state can be accurately distinguished without relying on prior information.
The above embodiments are described with respect to the positioning method of the present invention, and the embodiments will be further described with reference to the accompanying drawings.
Specifically, as shown in fig. 9, an embodiment of the present invention provides an apparatus 900 for identifying a channel propagation state, including:
an information extraction unit 910, configured to extract CIR components of channel impulse response CIR received by a signal receiving end in different polarization directions;
a first processing unit 920, configured to obtain statistical characteristics between multipaths according to the CIR component;
a second processing unit 930, configured to perform data preprocessing on the CIR component and the statistical characteristic to obtain input data;
and a state identification unit 940, configured to process the input data based on a convolutional neural network CNN model, and acquire a channel propagation state.
Optionally, the CIR component comprises:
a first CIR component generated by array elements of vertical polarization of a sending end and vertical polarization of a receiving end;
a second CIR component generated by array elements of vertical polarization of the sending end and horizontal polarization of the receiving end;
a third CIR component generated by array elements of horizontal polarization of the sending end and vertical polarization of the receiving end;
and a fourth CIR component generated by the array elements of the horizontal polarization of the transmitting end and the horizontal polarization of the receiving end.
Optionally, the statistical characteristics include: mean and variance of the CIR components.
Optionally, the first processing unit 920 is specifically configured to:
calculating the mean value of CIR components among L multipath, wherein the calculation formula of the mean value is as follows:
wherein,representing a mean of the CIR components; | H rxu,txv I represents the CIR component generated by the array element with the receiving end antenna polarization direction u and the transmitting end antenna polarization direction v; l represents the number of multi-paths, and L represents the first path;
calculating the variance of CIR components among L pieces of multipath, wherein the variance is calculated according to the following formula:
wherein,representing a variance of the CIR component; | H rxu,txv I represents the CIR component generated by the array element with the receiving end antenna polarization direction u and the transmitting end antenna polarization direction v; l represents the number of multipaths, and L represents the ith path.
Optionally, the second processing unit 930 is specifically configured to: and dividing the CIR components in different polarization directions and the statistical characteristics into dimensions to obtain M × N dimension input data.
Optionally, the second processing unit 930 is specifically configured to: and dividing the CIR components in different polarization directions, the statistical characteristics and the auxiliary characteristics of the signals into dimensions to obtain M × N dimension input data.
Optionally, the assist features comprise at least one of:
a rice factor;
a signal rise time;
root mean square delay spread;
kurtosis of the received power.
Optionally, the state identification unit is specifically configured to: inputting the input data into a Convolutional Neural Network (CNN) model to obtain the probability of the channel propagation state;
wherein the channel propagation state comprises: a line-of-sight propagation state or a non-line-of-sight propagation state.
According to the embodiment of the application, CIR components among transmitting and receiving antennas in different polarization directions and the statistical characteristics among multiple paths are used as input data of a CNN model, and a channel propagation state is obtained based on the CNN model. The CIR characteristic and the CNN of the dual-polarized antenna are combined, and the channel propagation state can be accurately distinguished without relying on prior information.
It should be noted that, the apparatus provided in the embodiment of the present invention can implement all the method steps implemented by the method embodiment and achieve the same technical effect, and detailed descriptions of the same parts and beneficial effects as the method embodiment in this embodiment are omitted here.
It should be noted that, in the embodiment of the present application, the division of the unit is schematic, and is only one logic function division, and when the actual implementation is realized, another division manner may be provided. In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented as a software functional unit and sold or used as a stand-alone product, may be stored in a processor readable storage medium. Based on such understanding, the technical solution of the present application may be substantially implemented or contributed by the prior art, or all or part of the technical solution may be embodied in a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, a network device, or the like) or a processor (processor) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
As shown in fig. 10, an embodiment of the present invention further provides an electronic device, including: memory 1020, transceiver 1000, processor 1010; a memory 1020 for storing a computer program; a transceiver 1000 for transceiving data under the control of the processor 1010; a processor 1010 for reading the computer program in the memory and performing the following operations:
extracting CIR components of channel impulse response CIR received by a signal receiving end in different polarization directions;
obtaining the statistical characteristics among the multipath according to the CIR component;
performing data preprocessing on the CIR component and the statistical characteristic to obtain input data;
and processing the input data based on a Convolutional Neural Network (CNN) model to acquire a channel propagation state.
Optionally, the CIR component comprises:
a first CIR component generated by array elements of the vertical polarization of the sending end and the vertical polarization of the receiving end;
a second CIR component generated by array elements of vertical polarization of the sending end and horizontal polarization of the receiving end;
a third CIR component generated by array elements of horizontal polarization of the sending end and vertical polarization of the receiving end;
and a fourth CIR component generated by the array elements of the horizontal polarization of the sending end and the horizontal polarization of the receiving end.
Optionally, the statistical characteristics include: mean and variance of the CIR components.
Optionally, the processor 1010 is configured to read the computer program in the memory and execute the following operations:
calculating the mean value of CIR components among L multipath, wherein the calculation formula of the mean value is as follows:
wherein,represents the aboveMean of the CIR components; | H rxu,txv I represents the CIR component generated by the array element with the receiving end antenna polarization direction u and the transmitting end antenna polarization direction v; l represents the number of multi-paths, and L represents the first path;
calculating the variance of CIR components among L pieces of multipath, wherein the variance is calculated according to the following formula:
wherein,representing a variance of the CIR component; | H rxu,txv I represents the CIR component generated by the array element with the receiving end antenna polarization direction u and the transmitting end antenna polarization direction v; l represents the number of multipaths, and L represents the ith path.
Optionally, the processor 1010 performs data preprocessing on the CIR component and the statistical characteristic to obtain input data, including:
and dividing the CIR components in different polarization directions and the statistical characteristics into dimensions to obtain M × N dimensional input data.
Optionally, the processor 1010 performs data preprocessing on the CIR component and the statistical characteristic to obtain input data, including:
and dividing the CIR components in different polarization directions, the statistical characteristics and the auxiliary characteristics of the signals into dimensions to obtain M × N dimension input data.
Optionally, the assist features comprise at least one of:
a rice factor;
a signal rise time;
root mean square delay spread;
kurtosis of the received power.
Optionally, the processing, by the processor, the input data based on a convolutional neural network CNN model to obtain a channel propagation state includes:
inputting the input data into a Convolutional Neural Network (CNN) model to obtain the probability of the channel propagation state;
wherein the channel propagation state comprises: a line-of-sight propagation state or a non-line-of-sight propagation state.
According to the embodiment of the application, CIR components among transmitting and receiving antennas in different polarization directions and the statistical characteristics among multiple paths are used as input data of a CNN model, and a channel propagation state is obtained based on the CNN model. The CIR characteristic and the CNN of the dual-polarized antenna are combined, and the channel propagation state can be accurately distinguished without relying on prior information.
Where in fig. 10, the bus architecture may include any number of interconnected buses and bridges, with various circuits being linked together, particularly one or more processors represented by processor 1000 and memory represented by memory 1020. The bus architecture may also link together various other circuits such as peripherals, voltage regulators, power management circuits, and the like, which are well known in the art, and therefore, will not be described any further herein. The bus interface provides an interface. The transceiver 1010 may be a number of elements including a transmitter and a transceiver providing a means for communicating with various other apparatus over a transmission medium. The processor 1000 is responsible for managing the bus architecture and general processing, and the memory 1020 may store data used by the processor 1000 in performing operations.
The processor 1010 may be a Central Processing Unit (CPU), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA), or a Complex Programmable Logic Device (CPLD), and may also have a multi-core architecture.
It should be noted that, the electronic device provided in the embodiment of the present invention can implement all the method steps implemented by the method embodiment and achieve the same technical effect, and detailed descriptions of the same parts and beneficial effects as the method embodiment in this embodiment are omitted here.
In addition, the embodiment of the present invention further provides a processor-readable storage medium, on which a computer program is stored, wherein the program, when executed by a processor, implements the steps of the method for identifying a channel propagation status as described above. And the same technical effect can be achieved, and in order to avoid repetition, the description is omitted. The readable storage medium can be any available medium or data storage device that can be accessed by a processor, including but not limited to magnetic memory (e.g., floppy disks, hard disks, magnetic tapes, magneto-optical disks (MOs), etc.), optical memory (e.g., CDs, DVDs, BDs, HVDs, etc.), and semiconductor memory (e.g., ROMs, EPROMs, EEPROMs, non-volatile memory (NAND FLASH), solid State Disks (SSDs)), etc.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer-executable instructions. These computer-executable instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These processor-executable instructions may also be stored in a processor-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the processor-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These processor-executable instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present application without departing from the spirit and scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is intended to include such modifications and variations as well.
Claims (18)
1. A method for identifying a propagation status of a channel, comprising:
extracting CIR components of channel impulse response CIR received by a signal receiving end in different polarization directions;
obtaining the statistical characteristics among the multipath according to the CIR component;
performing data preprocessing on the CIR component and the statistical characteristic to obtain input data;
and processing the input data based on a Convolutional Neural Network (CNN) model to acquire a channel propagation state.
2. The method of claim 1, wherein the CIR component comprises:
a first CIR component generated by array elements of the vertical polarization of the sending end and the vertical polarization of the receiving end;
a second CIR component generated by array elements of vertical polarization of the sending end and horizontal polarization of the receiving end;
a third CIR component generated by array elements of horizontal polarization of the sending end and vertical polarization of the receiving end;
and a fourth CIR component generated by the array elements of the horizontal polarization of the transmitting end and the horizontal polarization of the receiving end.
3. The method of claim 1, wherein the statistical characteristic comprises: mean and variance of the CIR components.
4. The method of claim 3, wherein obtaining statistical properties between multipaths from the CIR component comprises:
calculating the mean value of CIR components among L multipath, wherein the calculation formula of the mean value is as follows:
wherein,representing a mean of the CIR components; | H rxu,txv I represents a CIR component generated by an array element with the receiving end antenna polarization direction u and the transmitting end antenna polarization direction v; l represents the number of multi-paths, and L represents the first path;
calculating the variance of CIR components among L pieces of multipath, wherein the variance is calculated according to the following formula:
wherein,representing a variance of the CIR component; | H rxu,txv I represents a CIR component generated by an array element with the receiving end antenna polarization direction u and the transmitting end antenna polarization direction v; l represents the number of multipaths, and L represents the ith path.
5. The method of claim 1, wherein the pre-processing the CIR component and the statistical characteristic to obtain input data comprises:
and dividing the CIR components in different polarization directions and the statistical characteristics into dimensions to obtain M × N dimension input data.
6. The method of claim 1, wherein the pre-processing the CIR component and the statistical characteristic to obtain input data comprises:
and dividing the CIR components in different polarization directions, the statistical characteristics and the auxiliary characteristics of the signals into dimensions to obtain M × N dimension input data.
7. The method of claim 6, wherein the assist features comprise at least one of:
a rice factor;
a signal rise time;
root mean square delay spread;
kurtosis of the received power.
8. The method according to claim 1, wherein the processing the input data based on the CNN model to obtain the channel propagation state comprises:
inputting the input data into a Convolutional Neural Network (CNN) model to obtain the probability of the channel propagation state;
wherein the channel propagation state comprises: a line-of-sight propagation state or a non-line-of-sight propagation state.
9. An electronic device, comprising: memory, transceiver, processor:
a memory for storing a computer program; a transceiver for transceiving data under control of the processor; a processor for reading the computer program in the memory and performing the following operations:
extracting CIR components of channel impulse response CIR received by a signal receiving end in different polarization directions;
obtaining the statistical characteristics among the multipath according to the CIR component;
performing data preprocessing on the CIR component and the statistical characteristic to obtain input data;
and processing the input data based on a Convolutional Neural Network (CNN) model to acquire a channel propagation state.
10. The electronic device of claim 9, wherein the CIR component comprises:
a first CIR component generated by array elements of the vertical polarization of the sending end and the vertical polarization of the receiving end;
a second CIR component generated by array elements of vertical polarization of the sending end and horizontal polarization of the receiving end;
a third CIR component generated by array elements of horizontal polarization of the sending end and vertical polarization of the receiving end;
and a fourth CIR component generated by the array elements of the horizontal polarization of the transmitting end and the horizontal polarization of the receiving end.
11. The electronic device of claim 9, wherein the statistical characteristic comprises: mean and variance of the CIR components.
12. The electronic device of claim 11, wherein the processor is configured to read the computer program in the memory and perform the following operations:
calculating the mean value of CIR components among L multipath, wherein the calculation formula of the mean value is as follows:
wherein,represents the aboveMean of the CIR components; | H rxu,txv I represents the CIR component generated by the array element with the receiving end antenna polarization direction u and the transmitting end antenna polarization direction v; l represents the number of multi-paths, and L represents the ith path;
calculating the variance of CIR components among L pieces of multipath, wherein the variance is calculated according to the following formula:
wherein,representing a variance of the CIR component; | H rxu,txv I represents the CIR component generated by the array element with the receiving end antenna polarization direction u and the transmitting end antenna polarization direction v; l represents the number of multipaths, and L represents the ith path.
13. The electronic device of claim 9, wherein the processor performs data pre-processing on the CIR component and the statistical characteristic to obtain input data, comprising:
and dividing the CIR components in different polarization directions and the statistical characteristics into dimensions to obtain M × N dimension input data.
14. The electronic device of claim 9, wherein the processor performs data pre-processing on the CIR component and the statistical characteristic to obtain input data, comprising:
and dividing the CIR components in different polarization directions, the statistical characteristics and the auxiliary characteristics of the signals into dimensions to obtain M × N dimension input data.
15. The electronic device of claim 14, wherein the assist features comprise at least one of:
a rice factor;
a signal rise time;
root mean square delay spread;
kurtosis of the received power.
16. The electronic device of claim 9, wherein the processor processes the input data based on a Convolutional Neural Network (CNN) model to obtain a channel propagation status, and comprises:
inputting the input data into a Convolutional Neural Network (CNN) model to obtain the probability of the channel propagation state;
wherein the channel propagation state comprises: a line-of-sight propagation state or a non-line-of-sight propagation state.
17. An apparatus for identifying a propagation status of a channel, comprising:
the information extraction unit is used for extracting CIR components of channel impact response CIRs received by the signal receiving end in different polarization directions;
a first processing unit, configured to obtain statistical characteristics between multipaths according to the CIR component;
the second processing unit is used for carrying out data preprocessing on the CIR component and the statistical characteristic to obtain input data;
and the state identification unit is used for processing the input data based on a Convolutional Neural Network (CNN) model to acquire a channel propagation state.
18. A processor-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method for identifying a propagation state of a channel according to any one of claims 1 to 8.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110783840.XA CN115622839A (en) | 2021-07-12 | 2021-07-12 | Channel propagation state identification method and device and electronic equipment |
PCT/CN2022/095223 WO2023284427A1 (en) | 2021-07-12 | 2022-05-26 | Method and apparatus for identifying channel propagation state, and electronic device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110783840.XA CN115622839A (en) | 2021-07-12 | 2021-07-12 | Channel propagation state identification method and device and electronic equipment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN115622839A true CN115622839A (en) | 2023-01-17 |
Family
ID=84854444
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110783840.XA Pending CN115622839A (en) | 2021-07-12 | 2021-07-12 | Channel propagation state identification method and device and electronic equipment |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN115622839A (en) |
WO (1) | WO2023284427A1 (en) |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2008017033A2 (en) * | 2006-08-03 | 2008-02-07 | Ntt Docomo Inc. | Line-of-sight (los) or non-los (nlos) identification method using multipath channel statistics |
CN103532647A (en) * | 2013-10-14 | 2014-01-22 | 无锡清华信息科学与技术国家实验室物联网技术中心 | Sight distance propagation path judging method and system on basis of time domain features of WiFi (Wireless Fidelity) physical layer |
CN107370533A (en) * | 2016-05-12 | 2017-11-21 | 上海贝尔股份有限公司 | It is a kind of to be used to carry out the methods, devices and systems for simulating CSI feedback |
CN108650628A (en) * | 2018-03-24 | 2018-10-12 | 西安电子科技大学 | The indoor orientation method that ranging and fingerprint based on Wi-Fi network are combined |
CN109151707A (en) * | 2018-01-22 | 2019-01-04 | 中南大学 | A kind of sighting distance/obstructed path recognition methods under moving condition |
CN111770527A (en) * | 2020-06-24 | 2020-10-13 | 长安大学 | Visible and invisible channel identification method and device based on two-dimensional features |
CN112039807A (en) * | 2020-08-31 | 2020-12-04 | 中兴通讯股份有限公司 | Downlink channel estimation method, device, communication equipment and storage medium |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070047678A1 (en) * | 2005-08-30 | 2007-03-01 | Motorola, Inc. | Method and system for combined polarimetric and coherent processing for a wireless system |
-
2021
- 2021-07-12 CN CN202110783840.XA patent/CN115622839A/en active Pending
-
2022
- 2022-05-26 WO PCT/CN2022/095223 patent/WO2023284427A1/en active Application Filing
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2008017033A2 (en) * | 2006-08-03 | 2008-02-07 | Ntt Docomo Inc. | Line-of-sight (los) or non-los (nlos) identification method using multipath channel statistics |
CN103532647A (en) * | 2013-10-14 | 2014-01-22 | 无锡清华信息科学与技术国家实验室物联网技术中心 | Sight distance propagation path judging method and system on basis of time domain features of WiFi (Wireless Fidelity) physical layer |
CN107370533A (en) * | 2016-05-12 | 2017-11-21 | 上海贝尔股份有限公司 | It is a kind of to be used to carry out the methods, devices and systems for simulating CSI feedback |
CN109151707A (en) * | 2018-01-22 | 2019-01-04 | 中南大学 | A kind of sighting distance/obstructed path recognition methods under moving condition |
CN108650628A (en) * | 2018-03-24 | 2018-10-12 | 西安电子科技大学 | The indoor orientation method that ranging and fingerprint based on Wi-Fi network are combined |
CN111770527A (en) * | 2020-06-24 | 2020-10-13 | 长安大学 | Visible and invisible channel identification method and device based on two-dimensional features |
CN112039807A (en) * | 2020-08-31 | 2020-12-04 | 中兴通讯股份有限公司 | Downlink channel estimation method, device, communication equipment and storage medium |
Non-Patent Citations (3)
Title |
---|
CHANGHUI JIANG等: "An UWB Channel Impulse Response De-Noising Method for NLOS/LOS Classification Boosting", 《 IEEE COMMUNICATIONS LETTERS ( VOLUME: 24, ISSUE: 11, NOVEMBER 2020)》, 16 July 2020 (2020-07-16) * |
ELEMENT BREGAR 等: "Channel impulse response based vehicle analysis in tunnels", 《 2019 13TH EUROPEAN CONFERENCE ON ANTENNAS AND PROPAGATION (EUCAP)》, 5 April 2019 (2019-04-05) * |
MAXIMILIAN STAHLKE: "NLOS Detection using UWB Channel Impulse Responses and Convolutional Neural Networks", 《 2020 INTERNATIONAL CONFERENCE ON LOCALIZATION AND GNSS (ICL-GNSS)》, 4 June 2020 (2020-06-04) * |
Also Published As
Publication number | Publication date |
---|---|
WO2023284427A1 (en) | 2023-01-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109617584B (en) | MIMO system beam forming matrix design method based on deep learning | |
Wu et al. | A deep neural network-based indoor positioning method using channel state information | |
CN108768567B (en) | Multipath clustering method and device, electronic equipment and readable storage medium | |
Engel et al. | Incremental learning of multivariate gaussian mixture models | |
CN108510982A (en) | Audio event detection method, device and computer readable storage medium | |
CN110224771B (en) | Spectrum sensing method and device based on BP neural network and information geometry | |
Adjemov et al. | The use of artificial neural networks for classification of signal sources in cognitive radio systems | |
Geng et al. | Multipoint channel charting with multiple-input multiple-output convolutional autoencoder | |
Liu et al. | Location-based visible region recognition in extra-large massive MIMO systems | |
CN118018367A (en) | Model-driven S-CSI estimation method, equipment and base station for large-scale MIMO | |
Li et al. | A variational learning approach for concurrent distance estimation and environmental identification | |
CN111340057A (en) | Classification model training method and device | |
CN115622839A (en) | Channel propagation state identification method and device and electronic equipment | |
Chatterjee et al. | Parameter training in manet using artificial neural network | |
CN115567131A (en) | 6G wireless channel characteristic extraction method based on dimensionality reduction complex convolution network | |
CN116663643A (en) | Training method and data classification method of quantum neural network | |
Lee et al. | Indoor cell-level localization based on RSSI classification | |
CN113030853B (en) | RSS and AOA combined measurement-based multi-radiation source passive positioning method | |
CN112996106B (en) | Honeycomb-removing large-scale MIMO system positioning method | |
CN112087275B (en) | Cooperative spectrum sensing method based on birth and death process and viscous hidden Markov model | |
CN114337883A (en) | CNN cooperative spectrum sensing method and system based on covariance matrix Cholesky decomposition | |
JP2020091813A (en) | Learning method for neural network, computer program and computer device | |
Li et al. | Joint channel parameter estimation using evolutionary algorithm | |
CN115499278B (en) | MIMO signal modulation identification method based on lightweight neural network | |
Lima et al. | Least-squares support vector machines for DOA estimation: a step-by-step description and sensitivity analysis |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |