CN109362083B - High-speed railway wireless channel database construction and data preprocessing method - Google Patents
High-speed railway wireless channel database construction and data preprocessing method Download PDFInfo
- Publication number
- CN109362083B CN109362083B CN201811287153.3A CN201811287153A CN109362083B CN 109362083 B CN109362083 B CN 109362083B CN 201811287153 A CN201811287153 A CN 201811287153A CN 109362083 B CN109362083 B CN 109362083B
- Authority
- CN
- China
- Prior art keywords
- channel
- data
- wireless channel
- speed railway
- speed
- 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.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
- H04W16/22—Traffic simulation tools or models
Abstract
The invention provides a method for constructing a wireless channel database of a high-speed railway and preprocessing data, belonging to the technical field of wireless communication. The method comprises the steps of obtaining channel actual measurement and simulation original data by utilizing high-speed rail channel measurement based on a channel detector, high-speed rail channel measurement based on a TD-LTE network and high-speed rail channel simulation based on a theoretical method; carrying out missing value processing and logic cleaning on the actually measured original data, and forming wireless channel data of the high-speed railway together with the simulated original data; constructing a Hankel matrix of channel frequency response data, and denoising the data by a principal component analysis method; selecting the data after noise reduction, training by using a neural network to obtain a convergent neural network model, and intelligently extracting effective multipath components of channel impulse response in wireless channel data. The method can be used for constructing the high-speed railway wireless channel original database of the complete scene, provides pure and effective channel data, and is more accurate and credible for subsequent high-speed railway wireless channel characteristic mining and channel model construction.
Description
Technical Field
The invention relates to the technical field of wireless communication, in particular to a method for constructing a wireless channel database of a high-speed railway and preprocessing data.
Background
In recent years, the rapid development of the high-speed railway in China, and with the development of modern information technology, people also expect that broadband data services enjoyable on the ground can be realized in a high-speed railway carriage. However, the fast fading channel of the high-speed railway always restricts the effectiveness and reliability of information transmission of the high-speed railway wireless access system. The research on the high-speed railway broadband wireless communication system and the key technology thereof has very important significance, and the accurate cognition of the high-speed railway wireless channel is the first problem of the research on the high-speed railway broadband wireless communication.
The most direct and efficient method for studying the characteristics of a wireless channel is the field channel measurement. The channel data obtained in the measurement can accurately reflect the propagation characteristics of the channel, and an accurate channel model can be further extracted. The high-speed rail wireless access scene has a great difference from the traditional land cellular scene in the aspect of wireless channel propagation, the high-speed rail needs to pass through terrains such as plains, mountains, canyons, gobi, U-shaped grooves, bridges, hills, lakes, stations, tunnels, viaducts and the like, and the wireless channel characteristics of the high-speed rail wireless access scene show scene diversification. In addition, due to the difference of access modes (direct access or relay access) and the particularity of networking modes (strip coverage + cell merging), the high-speed rail wireless channel scenarios include in-car, out-of-car to in-car and multi-link scenarios.
Although some basic characteristics (space-time-frequency dispersion characteristics) of the high-speed rail channel have been explored to some extent so far, the high-speed rail channel is a multi-scene frequently-switched channel, and the channel correlation characteristics and the non-stationary characteristics between scenes need to be further mined. In addition, due to the diversification of high-speed rail scenes, multipath clusters corresponding to scatterers in different scenes also present different characteristics, and no relevant research is available at present. More importantly, the domestic and foreign standards organizations have not yet provided a high-speed rail channel model capable of covering a complete scene. In fact, the root cause of these gaps is the lack of sufficient high-speed rail channel data. Therefore, a set of high-speed railway wireless channel databases is urgently needed to be constructed.
In the channel measurement process, the acquired channel data has noise interference and invalid signal components, which can affect the accuracy of subsequent channel characteristic mining and modeling. Therefore, it is necessary to preprocess the channel raw data. How to denoise the data and how to identify the effective multipath components from the data is a problem that needs to be solved for preprocessing. To solve these two problems, it can be converted into two types of problems that can be solved by machine learning methods: and (5) reducing dimensions and classifying.
Disclosure of Invention
The invention aims to provide a method for constructing a complete high-speed railway wireless channel database, effectively reducing noise of data and effectively extracting effective multipath components, so as to solve the technical problems in the background technology.
In order to achieve the purpose, the invention adopts the following technical scheme:
the invention provides a method for constructing a wireless channel database of a high-speed railway and preprocessing data, which comprises the following steps:
step S110: acquiring channel actual measurement original data of a common high-speed rail scene by utilizing high-speed rail channel measurement based on a channel detector and high-speed rail channel measurement based on a TD-LTE network, and acquiring channel simulation original data of a special high-speed rail scene by utilizing high-speed rail channel simulation based on a theoretical method;
step S120: carrying out missing value processing and logic cleaning on the acquired wireless channel actual measurement original data, and forming wireless channel data of the high-speed railway together with the simulation original data;
step S130: constructing a Hankel matrix of high-speed railway wireless channel frequency response data, and denoising the high-speed railway wireless channel data according to a principal component analysis method;
step S140: selecting the data of the high-speed railway wireless channel after partial noise reduction to form a data set capable of covering all wireless channel scenes of the high-speed railway;
step S150; training the data set by using a neural network, and learning and distinguishing effective multipath components and ineffective multipath components of the wireless channel impulse response;
step S160: and extracting effective multipath components of the channel impulse response in the wireless channel data according to the neural network model trained in the step S150.
Further, the obtaining of channel actual measurement raw data by measuring the high-speed rail channel based on the TD-LTE network includes: and estimating the channel frequency response by carrying out frequency domain correlation processing on the reference signals of the receiving cell and the local transmitting cell of the downlink in the TD-LTE network.
Further, the acquiring of the channel simulation original data by using the high-speed rail channel simulation based on the theoretical method includes: and modeling a certain determined wireless channel scene according to one of a ray tracing theory, a random propagation diagram theory or a geometric random scattering theory to obtain channel simulation data.
Further, the missing value processing on the acquired wireless channel actual measurement raw data includes: and interpolating missing data in the uplink subframe time by adopting a Lagrange interpolation method, a polynomial interpolation method or a spline interpolation method.
Further, the missing value processing on the acquired wireless channel actual measurement raw data includes: and setting a certain initial position, and extracting data at equal intervals to avoid data loss caused by the blank of the uplink subframe.
Further, the logic cleaning of the acquired wireless channel measured raw data includes: and performing logic cleaning on data with the response of the measuring equipment system by adopting a frequency domain processing method to eliminate the response of the measuring equipment system:
wherein H (f) is the cleaned channel frequency response,
for measuring the resulting raw channel frequency response, Hsys(f) The device system response is measured.
Further, constructing a Hankel matrix of the frequency response data of the high-speed railway wireless channel comprises the following steps:
a Hankel matrix of channel frequency response data constructed as follows:
where m + N-1 is N, N is the number of channel taps, and m and N are set equal so that the Hankel matrix is a square matrix.
Further, the denoising the high-speed railway wireless channel data according to a principal component analysis method comprises: and carrying out zero equalization processing on the constructed channel frequency response matrix, then adopting a principal component analysis method to reduce the dimension of the matrix and reconstruct signals, wherein the characteristic size reflects the projection energy size of the original signals in each principal component direction, and taking the characteristic vector corresponding to a larger characteristic value for analysis to reduce noise interference.
Further, the step S150 specifically includes:
and judging whether a tap of a channel impulse response is an effective multipath component or not by adopting a backward propagation neural network model based on the influence of the amplitude of adjacent taps, and taking the power value of the tap to be judged, the power values of the front M taps and the power values of the rear M taps, which are 2M +1 power points in total, as the input of the neural network model to train and obtain the converged neural network model.
Further, common high-speed rail scenes comprise in-train, out-train to in-train and multi-link scenes of a high-speed train; the special high-speed rail scene comprises a tunnel scene.
The invention achieves the following beneficial effects: the method can be used for constructing the high-speed railway wireless channel original database of a complete scene, provides pure and effective channel data, and is more accurate and credible for subsequent high-speed railway wireless channel characteristic mining and channel model construction.
Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a flow chart of a method for constructing a wireless channel database of a high speed railway and preprocessing data according to an embodiment of the present invention.
Fig. 2 is a flow chart of a method for constructing a wireless channel database of a high speed railway and preprocessing data according to a second embodiment of the present invention.
Fig. 3 is a schematic diagram of the noise reduction and effective multipath component extraction results of the measured channel impulse response according to the second embodiment of the present invention.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or modules having the same or similar functionality throughout. The embodiments described below with reference to the accompanying drawings are illustrative only for the purpose of explaining the present invention, and are not to be construed as limiting the present invention.
As used herein, the singular forms "a", "an", "the" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or modules, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, modules, and/or groups thereof.
It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.
For the convenience of understanding of the embodiments of the present invention, the following description will be further explained by taking specific embodiments as examples with reference to the drawings, and the embodiments are not to be construed as limiting the embodiments of the present invention.
It will be understood by those of ordinary skill in the art that the figures are merely schematic representations of one embodiment and that the elements or devices in the figures are not necessarily required to practice the present invention.
Example one
As shown in fig. 1, an embodiment of the present invention provides a method for constructing a wireless channel database of a high-speed railway and preprocessing data, including the following steps:
step S110: acquiring channel actual measurement original data of a common high-speed rail scene by utilizing high-speed rail channel measurement based on a channel detector and high-speed rail channel measurement based on a TD-LTE network, and acquiring channel simulation original data of a special high-speed rail scene by utilizing high-speed rail channel simulation based on a theoretical method;
step S120: carrying out missing value processing and logic cleaning on the acquired wireless channel actual measurement original data, and forming wireless channel data of the high-speed railway together with the simulation original data;
step S130: constructing a Hankel matrix of high-speed railway wireless channel frequency response data, and denoising the high-speed railway wireless channel data according to a principal component analysis method;
step S140: selecting the data of the high-speed railway wireless channel after partial noise reduction to form a data set capable of covering all wireless channel scenes of the high-speed railway;
step S150; training the data set by using a neural network, and learning and distinguishing effective multipath components and ineffective multipath components of the wireless channel impulse response;
step S160: and extracting effective multipath components of the channel impulse response in the wireless channel data according to the neural network model trained in the step S150.
In a first embodiment of the present invention, the obtaining of channel actual measurement raw data by measuring a high-speed rail channel based on a TD-LTE network includes: and estimating the channel frequency response by carrying out frequency domain correlation processing on the reference signals of the receiving cell and the local transmitting cell of the downlink in the TD-LTE network.
In a first specific embodiment of the present invention, the acquiring of the channel simulation original data by using the high-speed rail channel simulation based on the theoretical method includes: and modeling a certain determined wireless channel scene according to one of a ray tracing theory, a random propagation diagram theory or a geometric random scattering theory to obtain channel simulation data.
In a first embodiment of the present invention, the performing missing value processing on the acquired wireless channel actual measurement raw data includes: and interpolating missing data in the uplink subframe time by adopting a Lagrange interpolation method, a polynomial interpolation method or a spline interpolation method.
In a first embodiment of the present invention, the performing missing value processing on the acquired wireless channel actual measurement raw data includes: and setting a certain initial position, and extracting data at equal intervals to avoid data loss caused by the blank of the uplink subframe.
In a first embodiment of the present invention, the logic cleaning of the acquired wireless channel measured raw data includes: and performing logic cleaning on data with the response of the measuring equipment system by adopting a frequency domain processing method to eliminate the response of the measuring equipment system:
wherein H (f) is the cleaned channel frequency response,
for measuring the resulting raw channel frequency response, Hsys(f) The device system response is measured.
In a first embodiment of the invention, constructing a Hankel matrix of the high-speed railway wireless channel frequency response data comprises:
a Hankel matrix of channel frequency response data constructed as follows:
where m + N-1 is N, N is the number of channel taps, and m and N are set equal so that the Hankel matrix is a square matrix.
In a first embodiment of the present invention, the denoising the high speed railway wireless channel data according to a principal component analysis method includes: and carrying out zero equalization processing on the constructed channel frequency response matrix, then adopting a principal component analysis method to reduce the dimension of the matrix and reconstruct signals, wherein the characteristic size reflects the projection energy size of the original signals in each principal component direction, and taking the characteristic vector corresponding to a larger characteristic value for analysis to reduce noise interference.
In a first embodiment of the present invention, the step S150 specifically includes:
and judging whether a tap of a channel impulse response is an effective multipath component or not by adopting a backward propagation neural network model based on the influence of the amplitude of adjacent taps, and taking the power value of the tap to be judged, the power values of the front M taps and the power values of the rear M taps, which are 2M +1 power points in total, as the input of the neural network model to train and obtain the converged neural network model.
In the first embodiment of the invention, the common high-speed rail scenes comprise in-train, out-train to in-train and multilink scenes of a high-speed train; the special high-speed rail scene comprises a tunnel scene.
Example two
As shown in fig. 2, a second embodiment of the present invention provides a method for constructing a wireless channel database of a high-speed railway and preprocessing data based on machine learning, including:
acquiring original data of a high-speed rail channel from three aspects, including high-speed rail channel measurement based on a channel detector, high-speed rail channel measurement based on a TD-LTE network and high-speed rail channel simulation based on a theoretical method;
carrying out missing value processing and logic cleaning on the obtained measured data of various high-speed rail scenes and forming a high-speed railway wireless channel database together with simulation data of certain special high-speed rail scenes;
constructing a Hankel matrix of channel frequency response data, and denoising the data according to a principal component analysis method;
selecting representative channel data to train a neural network model, and learning and distinguishing effective and ineffective multipath components of channel impulse response;
and according to the neural network model converged by training, intelligently extracting effective multipath components from the channel impulse response in the database.
In a second embodiment of the present invention, the measuring of the high-speed rail channel based on the TD-LTE network refers to estimating a channel frequency response by performing frequency domain correlation processing on a receiving cell reference signal and a local transmitting cell reference signal of a downlink in the TD-LTE network.
In the second embodiment of the present invention, the theoretical method refers to a ray tracing theory, a random propagation diagram theory, or a geometric random scattering theory.
In a second specific embodiment of the present invention, the multiple high-speed rail scenes and the special high-speed rail scenes refer to multiple scenes such as inside and outside of a high-speed train (e.g., a plain, a city, a mountain area, etc.), outside-to-inside of a train, multiple links, etc., and special scenes such as a tunnel.
In a second embodiment of the present invention, the missing value processing means interpolating missing data in an uplink subframe time or setting an initial position by using an interpolation method for high-speed rail channel measurement based on the TD-LTE network, and extracting the data at equal intervals.
In a second embodiment of the present invention, the logical flush is to eliminate the measurement device system response present in the channel data.
In a second embodiment of the present invention, the constructing the Hankel matrix of the channel frequency response data is to construct a Hankel square matrix with rows and columns equal to each other.
In the second embodiment of the present invention, the selecting of representative channel data means that these data should cover all scenes in the constructed channel database.
EXAMPLE III
The embodiment of the invention provides a high-speed railway wireless channel database construction and machine learning-based data preprocessing method, which comprises the following steps:
the method comprises the steps of obtaining original data of a high-speed rail channel from three aspects, including high-speed rail channel measurement based on a channel detector, high-speed rail channel measurement based on a TD-LTE network and high-speed rail channel simulation based on a theoretical method.
The high-speed rail channel measurement based on the channel detector adopts commercial channel detectors (such as Propsound, RUSK and the like) or channel detectors developed by colleges and universities (such as Chirp channel detector of the university of Dulun, England) to carry out channel measurement activities aiming at a certain determined scene of the high-speed rail. The high-speed rail channel measurement based on the TD-LTE network is to perform frequency domain correlation processing on a receiving cell reference signal and a local transmitting cell reference signal by acquiring downlink signals in the TD-LTE network in different scenes along a high-speed rail, and estimate channel frequency response. The high-speed rail channel simulation based on the theoretical method is to model a certain scene according to a ray tracing theory, a random propagation diagram theory or a geometric random scattering theory to obtain channel simulation data.
The high-speed rail channel measurement based on the channel detector and the high-speed rail channel measurement based on the TD-LTE network can be used for acquiring channel data of various high-speed rail scenes, such as scenes of an inside train, an outside train (such as plains, cities, mountainous areas and the like), an outside train to the inside train, a multilink and the like of a high-speed train. The high-speed rail channel simulation based on the theoretical method can be used for acquiring channel data of special high-speed rail scenes, such as tunnels and other scenes.
And carrying out missing value processing and logic cleaning on the acquired measured data of various high-speed rail scenes and forming a high-speed railway wireless channel database together with simulation data of certain special high-speed rail scenes.
Because a downlink subframe and an uplink subframe exist in a signal in the TD-LTE network at the same time, for a downlink, the uplink subframe does not transmit data, and therefore missing value processing needs to be performed on the acquired data. The missing value processing can adopt interpolation methods such as Lagrange interpolation, polynomial interpolation, spline interpolation and the like to interpolate missing data in the uplink subframe time. It is also possible to set an initial position and extract data at equal intervals to avoid data loss due to uplink subframe blanking, but this approach would reduce the channel sampling rate.
Logical flushing is the elimination of the measurement device system response present in the channel data. The raw channel measurements include the objective physical channel and the measurement device system response, and the non-ideal characteristics of the measurement device system response may affect the measurement accuracy of the objective physical channel. Therefore, there is a need for a logical flush of data for which there is a measurement device system response. The system response can be eliminated by adopting a frequency domain processing method, which comprises the following steps:
wherein H (f) is the cleaned channel frequency response,
for measuring the resulting raw channel frequency response, Hsys(f) The device system response is measured.
And constructing a Hankel matrix of the channel frequency response data, and denoising the data according to a principal component analysis method. A Hankel matrix of channel frequency response data constructed as follows:
where m + N-1 is N, N being the number of channel taps. Typically, m and n are set equal so that the Hankel matrix is a square matrix. The constructed channel frequency response matrix is subjected to zero-mean processing, then the principal component analysis method is adopted to reduce the dimension of the matrix and reconstruct signals, wherein the characteristic size reflects the projection energy size of the original signals in each principal component direction, and the characteristic vector corresponding to a larger characteristic value is taken for analysis, so that the noise interference can be reduced. Generally, the principal component with the accumulated contribution rate exceeding 90% is selected for signal reconstruction.
And selecting representative channel data to train the neural network model, and learning and distinguishing effective and ineffective multipath components of channel impulse response. Such a dichotomy problem can be effectively solved by using a back propagation neural network model. Generally, effective multipath components exist in a cluster form, so that whether a tap of a channel impulse response is an effective multipath component needs to be judged, and the influence of the amplitude of an adjacent tap needs to be considered, for example, when a certain tap is judged, the power value of the tap, the power values of the first M taps and the power values of the last M taps, and a total of 2M +1 power points are used as the input of a neural network model. The selected training data should cover all scenes in the constructed channel database to ensure that the model can be applied to all data. And according to the neural network model converged by training, intelligently extracting effective multipath components from the channel impulse response in the database.
Fig. 3 shows the result of the method of the present application for preprocessing the measured data of the high-speed rail wireless channel. The black dotted line (i) is the power result of the actually measured channel impulse response, the black dotted line (ii) is the result of noise reduction by principal component analysis (the cumulative contribution rate reaches 95%), and the multipath tap included in the black solid line (iii) is the effective multipath component extracted from the neural network model based on the training convergence. It can be seen from the figure that the method performs noise reduction to a certain extent on the actually measured channel impulse response, and performs better effective multipath component extraction on the signal impulse response after noise reduction.
In summary, in the embodiment of the invention, the original data of the high-speed railway channel is obtained by utilizing three aspects of high-speed railway channel measurement based on the channel detector, high-speed railway channel measurement based on the TD-LTE network and high-speed railway channel simulation based on a theoretical method, so that an original database of the wireless channel of the high-speed railway with a complete scene is constructed; the method comprises the steps of carrying out deletion value processing and logic cleaning on acquired measured data of various high-speed rail scenes, forming a high-speed railway wireless channel database together with simulation data of certain special high-speed rail scenes, constructing a Hankel matrix of channel frequency response data, carrying out noise reduction on the data according to a principal component analysis method, selecting representative channel data to train a neural network model, learning and distinguishing effective and ineffective multipath components of channel impulse response, and carrying out intelligent extraction on effective multipath components of the channel impulse response in the database according to a neural network model converged by training, so that pure and credible wireless channel data information is provided, and the subsequent high-speed railway wireless channel characteristic mining and channel model construction are more accurate and credible.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (4)
1. A high-speed railway wireless channel database construction and data preprocessing method is characterized by comprising the following flow steps:
step S110: acquiring channel actual measurement original data of a common high-speed rail scene by utilizing high-speed rail channel measurement based on a channel detector and high-speed rail channel measurement based on a TD-LTE network, and acquiring channel simulation original data of a special high-speed rail scene by utilizing high-speed rail channel simulation based on a theoretical method;
step S120: carrying out missing value processing and logic cleaning on the acquired wireless channel actual measurement original data, and forming wireless channel data of the high-speed railway together with the simulation original data;
step S130: constructing a Hankel matrix of high-speed railway wireless channel frequency response data, and denoising the high-speed railway wireless channel data according to a principal component analysis method;
step S140: selecting the data of the high-speed railway wireless channel after partial noise reduction to form a data set capable of covering all wireless channel scenes of the high-speed railway;
step S150; training the data set by using a neural network, and learning and distinguishing effective multipath components and ineffective multipath components of the wireless channel impulse response;
step S160: extracting effective multipath components of channel impulse response in the wireless channel data according to the neural network model trained in the step S150;
the acquiring of the channel simulation original data by utilizing the high-speed rail channel simulation based on the theoretical method comprises the following steps: modeling a certain wireless channel scene according to one of a ray tracking theory, a random propagation diagram theory or a geometric random scattering theory to obtain channel simulation data;
the logic cleaning of the acquired wireless channel actual measurement original data comprises the following steps: and performing logic cleaning on data with the response of the measuring equipment system by adopting a frequency domain processing method to eliminate the response of the measuring equipment system:
wherein H (f) is the cleaned channel frequency response,
for measuring the resulting raw channel frequency response, Hsys(f) A system response for the measurement device;
the Hankel matrix for constructing the frequency response data of the high-speed railway wireless channel comprises the following steps:
a Hankel matrix of channel frequency response data constructed as follows:
wherein m + N-1 is N, N is the number of channel taps, and m and N are set equal to each other, so that the Hankel matrix is a square matrix;
the denoising of the high-speed railway wireless channel data according to a principal component analysis method comprises: carrying out zero-mean processing on the constructed channel frequency response matrix, then adopting a principal component analysis method to reduce the dimension of the matrix and reconstruct signals, wherein the characteristic size reflects the projection energy size of the original signals in each principal component direction, and taking a characteristic vector corresponding to a larger characteristic value for analysis to reduce noise interference;
the step S150 specifically includes:
judging whether a tap of a channel impulse response is an effective multipath component or not by adopting a backward propagation neural network model based on the influence of the amplitude of adjacent taps, and taking the power value of the tap to be judged, the power values of the front M taps and the power values of the rear M taps, which are 2M +1 power points in total, as the input of the neural network model to train and obtain a converged neural network model;
common high-speed rail scenes comprise in-train, out-train to in-train and multi-link scenes of a high-speed train;
the special high-speed rail scene comprises a tunnel scene.
2. The method for constructing the wireless channel database of the high-speed railway and preprocessing data as claimed in claim 1, wherein the obtaining of the channel actual measurement raw data by the high-speed railway channel measurement based on the TD-LTE network comprises: and estimating the channel frequency response by carrying out frequency domain correlation processing on the reference signals of the receiving cell and the local transmitting cell of the downlink in the TD-LTE network.
3. The method for constructing the wireless channel database and preprocessing the data of the high-speed railway according to any one of claims 1-2, wherein the processing of the missing values of the acquired raw data of the wireless channel measured comprises: and interpolating missing data in the uplink subframe time by adopting a Lagrange interpolation method, a polynomial interpolation method or a spline interpolation method.
4. The method for constructing the wireless channel database and preprocessing the data of the high-speed railway according to any one of claims 1-2, wherein the processing of the missing values of the acquired raw data of the wireless channel measured comprises: and setting a certain initial position, and extracting data at equal intervals to avoid data loss caused by the blank of the uplink subframe.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811287153.3A CN109362083B (en) | 2018-10-31 | 2018-10-31 | High-speed railway wireless channel database construction and data preprocessing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811287153.3A CN109362083B (en) | 2018-10-31 | 2018-10-31 | High-speed railway wireless channel database construction and data preprocessing method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109362083A CN109362083A (en) | 2019-02-19 |
| CN109362083B true CN109362083B (en) | 2021-09-07 |
Family
ID=65347481
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811287153.3A Active CN109362083B (en) | 2018-10-31 | 2018-10-31 | High-speed railway wireless channel database construction and data preprocessing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109362083B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110210054A (en) * | 2019-04-25 | 2019-09-06 | 中国电力科学研究院有限公司 | A kind of sampled data preprocess method |
| CN111026744A (en) * | 2019-12-11 | 2020-04-17 | 新奥数能科技有限公司 | Data management method and device based on energy station system model framework |
| CN111132181B (en) * | 2020-03-27 | 2020-07-21 | 北京中铁建电气化设计研究院有限公司 | Ray tracing technology method and device applied to wireless communication network |
| CN114422061B (en) * | 2022-03-31 | 2022-07-19 | 中铁第四勘察设计院集团有限公司 | Adaptive prediction method for wireless signal propagation in railway environment |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102811104A (en) * | 2012-07-18 | 2012-12-05 | 北京交通大学 | Wireless signal channel simulation transmitter, receiver, system and method |
| CN103873395A (en) * | 2014-03-18 | 2014-06-18 | 北京交通大学 | Intelligent mobile communication method based on rail transit wireless environment diagram |
| CN105429922A (en) * | 2015-11-10 | 2016-03-23 | 杭州电子科技大学 | Channel estimation method based on comb-type pilot frequency for DDO-OFDM system |
| CN105530660A (en) * | 2015-12-15 | 2016-04-27 | 厦门大学 | Channel modeling method and device based on principal component analysis |
| CN107241696A (en) * | 2017-06-28 | 2017-10-10 | 中国科学院计算技术研究所 | Multipath effect discriminating conduct and method for estimating distance based on channel condition information |
| CN107276933A (en) * | 2017-06-14 | 2017-10-20 | 杭州电子科技大学 | For the channel estimation methods based on second-order statistic in uplink multi-users mimo system |
-
2018
- 2018-10-31 CN CN201811287153.3A patent/CN109362083B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102811104A (en) * | 2012-07-18 | 2012-12-05 | 北京交通大学 | Wireless signal channel simulation transmitter, receiver, system and method |
| CN103873395A (en) * | 2014-03-18 | 2014-06-18 | 北京交通大学 | Intelligent mobile communication method based on rail transit wireless environment diagram |
| CN105429922A (en) * | 2015-11-10 | 2016-03-23 | 杭州电子科技大学 | Channel estimation method based on comb-type pilot frequency for DDO-OFDM system |
| CN105530660A (en) * | 2015-12-15 | 2016-04-27 | 厦门大学 | Channel modeling method and device based on principal component analysis |
| CN107276933A (en) * | 2017-06-14 | 2017-10-20 | 杭州电子科技大学 | For the channel estimation methods based on second-order statistic in uplink multi-users mimo system |
| CN107241696A (en) * | 2017-06-28 | 2017-10-10 | 中国科学院计算技术研究所 | Multipath effect discriminating conduct and method for estimating distance based on channel condition information |
Non-Patent Citations (3)
| Title |
|---|
| 基于信道数据的特征挖掘与建模研究;马啸川;《中国优秀硕士学位论文数据库》;20180316;第3章-第6章 * |
| 基于图论算法的无线信道特征提取与场景识别研究;姚碧圆;《中国优秀硕士学位论文数据库》;20170501;第3节-第4节 * |
| 高速铁路无线信道传播特性、建模与测量方法研究;周涛;《中国博士学位论文数据库》;20160601;第1节-第4节 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109362083A (en) | 2019-02-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109362083B (en) | High-speed railway wireless channel database construction and data preprocessing method | |
| CN111314166B (en) | Multi-node comprehensive simulation system and method | |
| CN102118199B (en) | Implementation method of multi-antenna spectrum sensing scheme based on space-time diversity | |
| Alvarez et al. | New channel impulse response model for UWB indoor system simulations | |
| CN101316115B (en) | Detection method, equipment and system of pilot frequency sequence signal | |
| CN101330358A (en) | Method and system for restraining interference and combining diversity | |
| Shi et al. | Modeling of wireless channel between UAV and vessel using the FDTD method | |
| US10972141B2 (en) | Method for estimating arrival time based on noise cancellation | |
| CN106375045A (en) | Multilink channel detection system and method in high-speed railway scene | |
| CN105044734B (en) | High-precision carrier-to-noise ratio estimation method | |
| CN103346984B (en) | Method for estimating local clustering sparse channel based on BSL0 | |
| CN105848245A (en) | Multi-user energy collection relay system information transmission method | |
| CN103001714B (en) | Fast global system for mobile communications for railway (GSM-R) interference identification method | |
| CN113391329A (en) | Beidou satellite navigation signal distortion adaptive compensation method | |
| CN107733464A (en) | A kind of associated detecting method and system of Chirp spread spectrum communication systems | |
| CN112929141A (en) | Unmanned aerial vehicle detection and identification method and system based on graph-borne signal matching | |
| Gentile et al. | A channel propagation model for the 700 MHz band | |
| CN102404044A (en) | Frame synchronization detecting method for uplink signals in digital wireless trunking communication system based on TDMA (Time Division Multiple Access) technology and frame synchronization detecting device | |
| CN203243335U (en) | Satellite-to-ground time service testing system under spread spectrum channel | |
| CN112364845B (en) | Signal-to-noise ratio evaluation method and device, electronic equipment and storage medium | |
| CN115913291A (en) | Non-line-of-sight channel modeling method for underground coal mine intelligent super-surface wireless communication | |
| CN1996974A (en) | Self-adapted channel estimate device and its method based on CDMA frequency expansion sequence | |
| CN101605118B (en) | HF-VHF communication frame synchronization system and method | |
| CN114936976A (en) | Restoration method for generating anti-network haze image based on memory perception module | |
| CN101132213B (en) | Offset method for ascending interference of base station in TDD system |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |