WO2021155714A1 - 到达时间测量值的非直射径消除方法、装置及终端 - Google Patents

到达时间测量值的非直射径消除方法、装置及终端 Download PDF

Info

Publication number
WO2021155714A1
WO2021155714A1 PCT/CN2020/136564 CN2020136564W WO2021155714A1 WO 2021155714 A1 WO2021155714 A1 WO 2021155714A1 CN 2020136564 W CN2020136564 W CN 2020136564W WO 2021155714 A1 WO2021155714 A1 WO 2021155714A1
Authority
WO
WIPO (PCT)
Prior art keywords
measurement value
toa measurement
toa
value
nlos
Prior art date
Application number
PCT/CN2020/136564
Other languages
English (en)
French (fr)
Inventor
张振宇
任斌
达人
李刚
于大飞
郑占旗
孙韶辉
Original Assignee
大唐移动通信设备有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 大唐移动通信设备有限公司 filed Critical 大唐移动通信设备有限公司
Priority to EP20917676.7A priority Critical patent/EP4102900A4/en
Priority to US17/793,923 priority patent/US20230046671A1/en
Publication of WO2021155714A1 publication Critical patent/WO2021155714A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S5/00—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations
    • G01S5/02—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations using radio waves
    • G01S5/0205—Details
    • G01S5/0218—Multipath in signal reception
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/02—Services making use of location information
    • H04W4/023—Services making use of location information using mutual or relative location information between multiple location based services [LBS] targets or of distance thresholds
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S5/00—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations
    • G01S5/02—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations using radio waves
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S5/00—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations
    • G01S5/02—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations using radio waves
    • G01S5/06—Position of source determined by co-ordinating a plurality of position lines defined by path-difference measurements
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W64/00—Locating users or terminals or network equipment for network management purposes, e.g. mobility management
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S11/00—Systems for determining distance or velocity not using reflection or reradiation
    • G01S11/02—Systems for determining distance or velocity not using reflection or reradiation using radio waves
    • G01S11/08—Systems for determining distance or velocity not using reflection or reradiation using radio waves using synchronised clocks
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S2205/00—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations
    • G01S2205/001—Transmission of position information to remote stations
    • G01S2205/008—Transmission of position information to remote stations using a mobile telephone network
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S5/00—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations
    • G01S5/02—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations using radio waves
    • G01S5/0205—Details

Definitions

  • This application relates to the field of communication technology, and in particular to a method, a device and a terminal for eliminating indirect rays of a measured value of arrival time.
  • Wireless cellular network positioning can be considered as a process in which the base station (or receiver) receives radio waves between the base station and the receiver and determines the location of the receiver based on the characteristics of the radio waves.
  • errors that affect positioning accuracy are mainly divided into two categories.
  • the first category is the measurement error caused by the receiver design due to the thermal noise at the receiver. Its statistical characteristic is Gaussian white noise, which is inevitable in actual operation, and its influence on positioning accuracy is limited.
  • the second type of error is caused by the severe destruction of radio wave characteristics. Take TOA (Time of Arrival) measurement as an example. Due to multipath and non-line-of-sight propagation, the TOA measurement value obtained by the receiver is inaccurate. In this case, the usual positioning algorithm will have a large positioning deviation.
  • NLOS Non-Line of Sight
  • the algorithm can be used to make the TOA value obtained by the receiver are LOS (Line of LOS). sight, direct radius), and then use a positioning algorithm to determine the user's position, so as to obtain an accurate mobile station position.
  • LOS Line of LOS
  • the base station or receiver can perform NLOS identification or correction through advanced positioning algorithms.
  • the TOA measurement value received by the base station is the case where the LOS path and the NLOS path are aliased. Taking 6 base stations and a single user as an example, that is, there are 6 TOA measurement values.
  • One possible situation is that there are 3 TOA measurement values as LOS paths, and three TOA measurement values are NLOS paths. In the above situation, how to identify the NLOS path, discard the NLOS measurement result, or make further corrections to the identified NLOS result is a key issue that needs to be studied in wireless communication positioning.
  • the measurement result of a single TOA cannot be further analyzed, and it is necessary to carry out multiple ranging of the target within a period of time.
  • the first type is non-Bayesian detection.
  • the NLOS identification is performed by analyzing the results of multiple TOA measurements and using the characteristics of the result distribution. For example, the variance of the TOA measurement value of NLOS is often greater than the LOS measurement value, and this information is used for NLOS identification.
  • the TOA results belonging to the LOS measurement value obey the Gaussian distribution.
  • NLOS propagation can be identified by checking whether the measured value obeys the Gaussian distribution.
  • K-S, A-D, Chi-Square, Gross test, skewness and kurtosis tests have appeared successively.
  • the second category is Bayesian detection.
  • the prior probability of LOS/NLOS propagation can be used to identify the LOS/NLOS propagation path based on the different probability statistical distributions of its error.
  • the above methods all need to predict certain prior information, such as the prior probability of LOS/NLOS propagation, the statistical characteristics of NLOS errors, and the predetermined false alarm probability.
  • the information is often very different, and it is often impossible to know it without actual measurement in advance, so the scope of application of the above method is limited.
  • the identification of NLOS requires a period of time to perform multiple ranging on the target. Due to the time-varying characteristics of the channel, it is difficult for the TOA results of multiple measurements to obey a specific distribution assumption. Even the LOS path TOA measurement value does not obey the Gaussian distribution during long-term measurement; in addition, LOS/NLOS in the actual situation The a priori probability of propagation is also difficult to predict, which makes it difficult to eliminate the NLOS of the TOA measurement value.
  • the embodiments of the present application propose a method, device and terminal for eliminating the indirect radiation path of the measured value of the arrival time.
  • an embodiment of the present application proposes a method for eliminating the indirect radiation path of the arrival time measurement value, which includes:
  • the identification label is used to indicate whether the TOA measurement value after the screening is NLOS;
  • the TOA measurement value after screening is corrected according to the identification tag to eliminate the error caused by NLOS in the TOA measurement value after screening.
  • the embodiment of the present application also proposes a device for eliminating the indirect radiation path of the measured value of the arrival time, including:
  • the measurement value modeling module is used to perform mixed Gaussian modeling on the probability density of the TOA measurement value of the arrival time of each base station to the terminal UE, and to screen the TOA measurement value after the mixed Gaussian modeling;
  • the NLOS identification module is used to perform non-direct-radial NLOS identification on the TOA measured value after screening to obtain an identification label; the identification label is used to indicate whether the TOA measured value after the screening is NLOS;
  • the measurement value correction module is used to correct the screened TOA measurement value according to the identification tag, so as to eliminate the error caused by NLOS in the screened TOA measurement value.
  • an embodiment of the present application also proposes a terminal, including:
  • At least one processor At least one processor
  • At least one memory communicatively connected with the processor, wherein:
  • the memory stores program instructions that can be executed by the processor, and the processor invokes the program instructions to execute the following methods:
  • the identification label is used to indicate whether the TOA measurement value after the screening is NLOS;
  • the TOA measurement value after screening is corrected according to the identification tag to eliminate the error caused by NLOS in the TOA measurement value after screening.
  • an embodiment of the present application also proposes a non-transitory computer-readable storage medium, the non-transitory computer-readable storage medium stores a computer program, and the computer program causes the computer to execute the following method:
  • the identification label is used to indicate whether the TOA measurement value after the screening is NLOS;
  • the TOA measurement value after screening is corrected according to the identification tag to eliminate the error caused by NLOS in the TOA measurement value after screening.
  • the embodiment of the present application performs mixed Gaussian modeling and screening on the probability density of each TOA measurement value to ensure that the TOA measurement value belonging to the LOS path is accurately found in the case of LOS and NLOS aliasing; at the same time, it passes Correct the TOA measurement value after screening, eliminate the error caused by NLOS in the TOA measurement value after screening, and improve the positioning accuracy of the user.
  • FIG. 1 is a schematic flow chart of a method for eliminating indirect rays of a measured value of arrival time according to an embodiment of the application;
  • FIG. 2 is a schematic diagram of a UE positioning scenario provided by an embodiment of this application.
  • FIG. 3 is a schematic diagram of the probability distribution of a Gaussian mixture model provided by an embodiment of the application.
  • FIG. 4 is a schematic diagram of the TOA error distribution curve when there is a direct beam provided by an embodiment of the application
  • FIG. 5 is a schematic diagram of a TOA error distribution curve when there is no direct radiation diameter provided by an embodiment of the application.
  • FIG. 6 is a schematic diagram of the TOA error distribution curve when the NLOS variance is large according to an embodiment of the application;
  • FIG. 7 is a schematic diagram of a curve of the final positioning error CDF distribution provided by an embodiment of the application.
  • FIG. 8 is a schematic structural diagram of a device for eliminating indirect beams of measured arrival time provided by an embodiment of the application.
  • Fig. 9 is a logical block diagram of an electronic device provided by an embodiment of the application.
  • FIG. 1 shows a schematic flow chart of a method for eliminating indirect rays of a measured value of arrival time provided by this embodiment, including:
  • the TOA measurement value is the time from the base station to the UE of the PRS signal obtained by the UE measuring the downlink PRS (Positioning Reference Signal, positioning reference signal).
  • the probability density of the TOA measurement value is the probability density distribution obtained by statistical analysis using multiple TOA measurement values as random variables, and the TOA measurement value is modeled as a random variable that obeys the mixed Gaussian.
  • Gaussian mixture modeling is the process of establishing a Gaussian mixture model; among them, Gaussian mixture model uses Gaussian probability density distribution (normal distribution curve) to accurately quantify things, and is a model that decomposes things into several models based on Gaussian probability density distribution .
  • Gaussian probability density distribution normal distribution curve
  • S102 Perform NLOS identification on the TOA measurement value after screening to obtain an identification label.
  • the identification tag is used to indicate whether the TOA measurement value after screening is NLOS.
  • the TOA measurement values after screening are identified, and the TOA measurement values with NLOS errors are determined.
  • S103 Correct the TOA measurement value after screening according to the identification tag, so as to eliminate the error caused by NLOS in the TOA measurement value after screening.
  • the TOA measurement value carrying the NLOS tag is corrected to obtain the TOA measurement value that eliminates the NLOS error.
  • the mixed Gaussian modeling of the probability density of the TOA measurement value of the arrival time of each base station to the terminal UE in S101 specifically includes:
  • the mixture Gaussian modeling is performed according to the probability density of the first TOA measurement value, and the second TOA measurement value represented by the mixture Gaussian model is obtained.
  • the first TOA measurement value is the TOA measurement value of each base station that is measured by the UE and reaches the UE.
  • the second TOA measurement value is a TOA measurement value obtained by performing Gaussian mixture modeling on the probability density of the first TOA measurement value.
  • the number of channel changes is the number of times an additive deviation greater than zero changes with the channel.
  • the PRS configuration information sent by each base station is received, and the downlink PRS signal is measured according to each PRS configuration information, and the TOA measurement value of each base station arriving at the UE is obtained.
  • the PRS configuration information is the configuration information of the PRS signal sent by the base station to the UE.
  • the following steps may be included:
  • Step1 The UE receives the PRS configuration information notified by the network, measures the downlink PRS signal and obtains the multiple TOA measurement values of each base station arriving at the UE.
  • the multiple TOA measurement values measured by the receiving end are called the first TOA measurement value;
  • the first TOA measurement value measured many times has no obvious change. Due to the influence of multipath and non-direct radiation diameter, the TOA measurement value measured each time has a big difference.
  • the first TOA measurement value measured multiple times is used as a random variable, and through the analysis of its probability density, the first TOA measurement value is modeled as a random variable obeying the mixed Gaussian distribution. On the basis of this modeling, Identify the NLOS path. According to modeling assumptions, use the first TOA measurement value from each base station to the user to analyze the probability density (for example: the variational Dirichlet process) to obtain the probability density parameter of the first TOA measurement value, record and record it as The second TOA measurement.
  • the probability density for example: the variational Dirichlet process
  • the probability density of the TOA measurement value obeys the mixed Gaussian distribution, and the probability density analysis is performed on all the measured values to obtain the various parameters of the mixed Gaussian model.
  • a data preprocessing method based on a Gaussian mixture model is proposed.
  • the LOS/NLOS path can be further detected, and the NLOS path can be corrected or eliminated to further improve the positioning accuracy.
  • the screening of TOA measurement values after Gaussian mixture modeling in S101 specifically includes:
  • the following steps may also be included:
  • Step3 Take the second TOA measurement value in Step2, and use the second TOA measurement value with the smallest average value as the third TOA measurement value; that is, analyze the second TOA measurement value obtained in Step2, filter out the useless TOA measurement value, and obtain the first TOA measurement value.
  • Three TOA measurement values Take the second TOA measurement value in Step2, and use the second TOA measurement value with the smallest average value as the third TOA measurement value; that is, analyze the second TOA measurement value obtained in Step2, filter out the useless TOA measurement value, and obtain the first TOA measurement value.
  • S102 specifically includes:
  • S103 specifically includes:
  • the fourth TOA measurement value is a TOA measurement value obtained after NLOS discrimination is performed on the third TOA measurement value.
  • the fifth TOA measurement value is the fifth TOA measurement value obtained after removing the TOA measurement value with the largest NLOS error in the fourth TOA measurement value.
  • the following steps may also be included:
  • Step4 Perform NLOS identification for the third TOA measurement value and label it to obtain the fourth TOA measurement value.
  • the NLOS identification method includes the standard deviation NLOS identification method or the skewness and kurtosis identification method.
  • Step5. remove the TOA measurement value with the largest NLOS error between the base station and the user to obtain the fifth TOA measurement value, where the removal method can use leave-one-out method or other methods such as gradient descent method.
  • the design of error compensation for the TOA measurement value belonging to the NLOS path includes: combining the leave-one method to remove the TOA measurement value with the largest NLOS error between the base station and the user, and traversing the data for NLOS correction.
  • the TOA positioning error is within 10 meters, which can be effectively corrected, thereby improving the positioning accuracy.
  • This embodiment combines the leave-one-out method, Range Residuals Test (RRT, Range Residuals Test), and traversal data to perform NLOS repair, so that the positioning accuracy is greatly improved.
  • RRT Range Residuals Test
  • NLOS identification based on the Gaussian mixture model improves the probability of successful NLOS identification and provides a basis for NLOS error correction, which greatly reduces the complexity of the overall positioning algorithm.
  • the method further includes:
  • the user positioning result is calculated according to the corrected sixth TOA measurement value, the user positioning algorithm and the predefined criterion, and the UE is positioned according to the user positioning result.
  • the predefined criterion includes selecting the user position with the smallest distance residual.
  • the sixth TOA measurement value is the TOA measurement value after NLOS elimination is performed on the fifth TOA measurement value.
  • the following steps may also be included:
  • the fifth TOA measurement value processed by Step6 and Step5 is combined with the tag in Step4 operation, the NLOS measurement is read, and the data is traversed and corrected.
  • the NLOS error is corrected to obtain the sixth TOA measurement value, and the end user positioning result is obtained through the user positioning algorithm and predefined criteria based on the sixth TOA measurement.
  • the standard deviation NLOS identification method or the skewness and kurtosis NLOS identification method in Step 4 the leave-one method in Step 5, the gradient descent method, and the user positioning algorithm in Step 6 are not the only calculation methods and can be substituted
  • the two methods of processing Step4 and Step5 can be used interchangeably, and the positioning algorithm in Step6 can use Chan algorithm or least squares method and so on.
  • the positioning accuracy can be greatly improved.
  • the performing Gaussian mixture modeling according to the probability density of the first TOA measurement value to obtain the second TOA measurement value represented by the Gaussian mixture model specifically includes:
  • the probability density of each first TOA measurement value is mixed Gaussian modeling to obtain the second TOA measurement value
  • r Bias is an additive deviation greater than zero, Is the measurement error of the first TOA measurement value; n is the number of measurements, n ⁇ N; k is the number of channel changes of the value of r Bias as the channel changes; ⁇ i is the mean value corresponding to base station i; ⁇ i is the corresponding value of base station i Standard deviation, ⁇ i is the amplitude corresponding to the Gaussian distribution of base station i, Express The probability density.
  • the acquiring the TOA measurement value with the smallest mean value among the second TOA measurement values as the third TOA measurement value specifically includes:
  • z is the number corresponding to the smallest mean among the mean values of each second TOA measurement ⁇ 1 , ⁇ 2 ,... ⁇ k ,; ⁇ z is the mean value corresponding to the Gaussian distribution with the smallest mean in the Gaussian mixture distribution; j It is the number when traversing the number of channel changes.
  • Step1-Step6 in the process of removing the NLOS of the TOA measurement value includes the following steps:
  • Step 1 The UE receives the PRS configuration information notified by the network, measures the downlink PRS signal, and obtains the TOA measurement value of each base station to the UE.
  • FIG. 2 is a schematic diagram of the TOA positioning process.
  • the base station participating in the positioning sends the PRS to the terminal, and the UE side generates the PRS sequence according to the PRS information notified by the network side.
  • the user receives the PRS signal on the base station side, the received PRS signal and the locally generated PRS signal are processed, and the TOA measurement value from the user to the corresponding base station can be measured, and it is the first TOA measurement value.
  • the TOA measurement reflects the distance between the user and the base station. When the UE is not moving, repeat this step to obtain the measured distances from multiple base stations to the UE.
  • the problem to be solved in this embodiment is how to distinguish the measured value of the direct radiation path between the user and the base station from the multiple measured values (the user and the base station are not blocked).
  • the other case is when the distance between the user and the base station is measured as When the diameter is not direct, how to correct the error of the measured value so that the corrected measurement value is close to the measurement result in the case of the direct diameter.
  • Step2 the establishment of the Gaussian mixture model of TOA measurement value.
  • a TOA measurement from base station i to user a can be described as:
  • r Bias is an additive deviation greater than zero, which represents the influence of multipath and NLOS caused by channel changes. Its size changes as the terminal environment changes. Is the TOA measurement error, modeled as the mean value is 0, and the variance is Gaussian noise.
  • the time-varying characteristics of the channel between the user and the base station are modeled by formula (3), in which the influence of the multipath and the NLOS path is the main reason for the time-varying of the channel.
  • a total of 1000 frames are used to measure the positions of the base station and the terminal.
  • r Bias C 1
  • r Bias C 2
  • C 1 and C 2 are constants. Due to noise It is a Gaussian distribution, then according to the knowledge of probability theory, the first measured TOA value can be known It is a random variable, that is, it is measured multiple times and the values obtained are different. Random variables can be described by a probability density function (PDF, Probability Density Function).
  • PDF Probability Density Function
  • the value of r Bias in different measurement periods is not the same, so in N measurements, It can be considered as a superposition of multiple Gaussian distributions with different means.
  • the first 500 frames, 500 Mean of 500 after 500 frames Mean of Therefore it can be considered that the probability density function of the first TOA measurement value obeys the mixed Gaussian distribution.
  • n the TOA value of the nth measurement
  • the probability density of the first TOA measurement value can be expressed as formula (1).
  • p(*) represents the probability density.
  • k is the total number of categories, which represents the number of times the value of r Bias changes with the channel. In a measurement time of tens of minutes, it is considered that k ⁇ 4.
  • ⁇ i is the mean corresponding to each category
  • ⁇ i is the standard deviation
  • ⁇ i the amplitude corresponding to each Gaussian distribution, which is called the mixed Gaussian coefficient, and the mixed coefficient should satisfy:
  • determining the distribution of the first TOA measurement value determines the following parameters: k is the total number of categories, ⁇ i is the amplitude corresponding to each Gaussian distribution, and ⁇ i is the mean value ⁇ corresponding to each category. i is the standard deviation.
  • the probability density is estimated, and the probability density distribution of the first TOA measurement value can be obtained: GMM(k, ⁇ i , ⁇ i , ⁇ i ),i ⁇ 1,2,...,k.
  • the probability density function of the data can reflect the characteristics of the data distribution, such as the mean value and variance of the first TOA measurement value.
  • the following analyzes the parameters of the probability density distribution.
  • Corresponding mean Record the mean and variance N( ⁇ i ,( ⁇ i ) 2 ),i ⁇ 1,2,...,k as the second TOA measurement value for subsequent processing. If 1000 frames are measured, there are 1000 first TOA measurement values But after probability density analysis, the second TOA measurement value is only 2k (k means and k variances).
  • the TOA measurement essentially reflects the change of the channel.
  • the hypothetical modeling of the TOA measurement value (Gaussian mixture model) is used to reflect whether the transmission path from the user to the base station is blocked at different times.
  • the advantage of this step is that through the analysis of the Gaussian mixture model, it provides the possibility for the user to identify the NLOS when there is no prior information at the receiving end and only a few TOA measurements.
  • Step3 Take Step2 to obtain the Gaussian distribution with the smallest mean ⁇ i in the mixed Gaussian model as the third TOA measurement value.
  • the N TOA measurement values form a Gaussian mixture distribution. After the variational Dirichlet process, the parameters of the mixture Gaussian distribution can be obtained.
  • Figure 3 shows the Gaussian mixture model obtained after the TOA measurement value between a certain user and the base station is processed.
  • the TOA measurement value belonging to this Gaussian distribution is the third TOA measurement value.
  • this step filters out most of the TOA measurement values of NLOS by analyzing the mean value of the mixed Gaussian distribution. Based on model assumptions, through the analysis of the relationship between the mean value and the LOS diameter, most of the TOA measurement values belonging to NLOS can be quickly and accurately discharged.
  • Step4 Identify NLOS based on Gaussian mixture model.
  • the identification of NLOS based on Gaussian mixture model includes NLOS identification method based on standard deviation and NLOS identification method based on skewness and kurtosis.
  • the NLOS identification method based on standard deviation is as follows:
  • r Bias In the absence of NLOS diameter, r Bias is approximately 0, so the measured TOA measurement results should obey a Gaussian distribution with a mean of 0 and a variance of ⁇ . In the case of the presence of the NLOS path, since the NLOS error and the measurement error are independent of each other, for the third TOA measurement value, a Gaussian component with a larger variance can be considered as the presence of the NLOS path. Therefore, a binary detection model can be constructed:
  • ⁇ Th is the threshold standard deviation, which can be set according to different actual environments.
  • Figures 4 and 5 respectively show the graphs of the Gaussian mixture distribution in the presence of LOS paths and only NLOS paths. It can be seen from the figure that the standard deviation of NLOS is much larger than the standard deviation of LOS diameter.
  • the NLOS recognition method based on skewness and kurtosis is as follows:
  • the measured value of LOS diameter should obey the Gaussian distribution, while the measured value of NLOS will not obey the standard Gaussian distribution.
  • Slope and skewness are an important tool for testing whether the sample conforms to the Gaussian distribution. Therefore, skewness and skewness can be used.
  • the kurtosis is used to identify NLOS.
  • the n-th first TOA measurement value belongs to the k-th category.
  • Skewness can detect whether the sample is symmetrically distributed, and the kurtosis of symmetrically distributed samples is 0. Assuming that there are m TOA measurements belonging to the Gaussian distribution with the smallest mean, you can use the skewness to test it:
  • the Kurtosis of the Gaussian distribution is 3, and the kurtosis can be used to verify whether the sample is Gaussian:
  • the second binary detection model can be constructed:
  • the standard deviation NLOS identification method or the skewness and kurtosis NLOS identification method is used in the hypothetical model of this embodiment, which can accurately and quickly identify the unidentified NLOS measurement values in Step 3, which is a follow-up NLOS processing, user positioning provides accurate data.
  • Step5. Remove the NLOS measurement value with the largest error.
  • the average value of the NLOS error between a certain base station and the user is about 8 meters, which has seriously affected the positioning accuracy and is difficult to be effectively corrected, so this base station should be eliminated.
  • the measurement result of this base station is regarded as invalid data.
  • the method to identify the maximum NLOS error is as follows:
  • Different combinations will calculate different user positions (x o , y o ), and then have different distance residuals. Select the combination with the smallest distance residual. Taking 6 base stations as an example, suppose the combination with the smallest distance residual is 1, 2,3,4,6, then it can be judged that the position error of base station 5 is the largest, then base station 5 is removed, not participating in subsequent positioning, and the distance residual d min at this time is recorded, and the TOA measurement combination obtained at this time is the fifth TOA measurement value.
  • this step can speed up the operation of subsequent steps, and the subsequent calculation amount can be reduced by eliminating the maximum NLOS error measurement.
  • one maximum NLOS error measurement can be eliminated, or multiple measurements can be eliminated. The elimination of inaccurate measurement speeds up the calculation on the one hand, and improves the positioning accuracy on the other hand.
  • Step6 Traverse the data to correct the NLOS error to obtain the sixth TOA measurement value, and then obtain the end user positioning result based on the predefined criteria.
  • the predefined criterion includes selecting the user position with the smallest distance residual.
  • the small step interval [0,2,3,4]
  • the large step interval [0,5,10,15]. Since the NLOS error is unknown and positive, it is judged to be the TOA measurement value of the NLOS error Need to subtract all the values in the interval in turn to get the corrected TOA measurement
  • the TOA correction value obtained at this time is the sixth TOA measurement value.
  • the fifth TOA measurement value is used to perform a user location determination based on the Chan algorithm ((x new , y new )). To determine the user's location once, a distance residual detection is required: If there are J NLOS errors, there are J 4 NLOS error corrections, accompanied by J 4 user location positioning results and J 4 distance residuals.
  • the value with the smallest distance residual is selected, and the user position at this time is determined as the final user position.
  • the user position is determined based on the minimum distance residual, thereby greatly improving the positioning accuracy of the user in the NLOS scenario.
  • the user has no more information about the wireless environment between the user and the base station.
  • the base station side includes the following steps:
  • A2 and 6 base stations respectively send downlink PRS signals to the UE.
  • Lasts for a period of time such as 1000 frames.
  • the UE side includes the following steps:
  • the UE receives the PRS configuration information notified by the network, measures the downlink PRS signal and obtains the TOA measurement value of each base station to the UE, and counts the TOA measurement results of multiple frames.
  • the first TOA measurement value from each base station to the user for processing treat the first TOA measurement value as a random variable, analyze its probability density (for example, use a variational Dirichlet process for analysis) to obtain a Gaussian mixture model. Assuming that the channel between base station i and user a changes twice, there are two Gaussian distributions for mixing, N( ⁇ 1 ,( ⁇ 1 ) 2 ) and N( ⁇ 2 ,( ⁇ 2 ) 2 ) will be ⁇ 1 , ⁇ 1 , ⁇ 2 , and ⁇ 2 are recorded as the second TOA measurement value.
  • N( ⁇ 1 ,( ⁇ 1 ) 2 ) and N( ⁇ 2 ,( ⁇ 2 ) 2 ) will be ⁇ 1 , ⁇ 1 , ⁇ 2 , and ⁇ 2 are recorded as the second TOA measurement value.
  • the fourth TOA measurement is N( ⁇ 1 ,( ⁇ 1 ) 2 ),LOS/NLOS.
  • B5 B2 ⁇ B4 all take base station i to user a as an example. Assuming that there are 6 base stations participating in positioning, there are 6 fourth TOA measurement values, and the mean value of each measurement value ⁇ i,i ⁇ [16] is taken out to remove The TOA measurement value with the largest NLOS error between the base station and the user, and five fifth TOA measurement values N( ⁇ i ,( ⁇ i ) 2 ),LOS/NLOS,i ⁇ [1,2,3,5,6], Among them, the removal method can use the leave-one-out method. Assuming that 6 base stations are involved in positioning, the TOA result of the 4th base station is exhausted using B5.
  • the predefined criterion includes selecting the user position with the smallest distance residual.
  • Figure 7 shows the simulation results in Indoor3 scenarios at a bandwidth of 50MHZ.
  • the unimproved Chan algorithm was simulated under the same conditions; the input of the improved algorithm and the channel are the same TOA measurement data. It can be seen from the results that the TOA measurement value after NLOS recognition and correction greatly improves the final The accuracy of positioning.
  • This embodiment solves the problem of error identification and compensation of the radio communication positioning system based on the TOA measurement value under the NLOS channel condition, and further improves the accuracy of user position calculation by reducing the error of the TOA measurement value.
  • FIG. 8 shows a schematic structural diagram of an indirect path elimination device for time-of-arrival measurement values provided by this embodiment.
  • the device includes a measurement value modeling module 801, an NLOS identification module 802, and a measurement value correction module 803, wherein :
  • the measured value modeling module 801 is configured to perform mixed Gaussian modeling on the probability density of the TOA measurement value of the arrival time of each base station to the terminal UE, and screen the TOA measurement value after the mixed Gaussian modeling;
  • the NLOS identification module 802 is used to perform non-direct-radial NLOS identification on the screened TOA measurement value to obtain an identification label; the identification label is used to indicate whether the screened TOA measurement value is NLOS;
  • the measurement value correction module 803 is configured to correct the screened TOA measurement value according to the identification tag, so as to eliminate the error caused by NLOS in the screened TOA measurement value.
  • the measurement value modeling module 801 performs mixed Gaussian modeling on the probability density of the TOA measurement value of the arrival time of each base station to the terminal UE, and screens the TOA measurement value after the mixed Gaussian modeling; the NLOS identification The module 802 performs non-direct-radial NLOS discrimination on the TOA measurement value after screening to obtain an authentication label; the authentication label is used to indicate whether the TOA measurement value after the screening is NLOS; the measured value correction module 803 pairs according to the authentication label The TOA measurement value after the screening is corrected to eliminate the error caused by the NLOS in the TOA measurement value after the screening.
  • the device for eliminating the indirect radiation path of the measured value of arrival time described in this embodiment can be used to implement the above method embodiment, and its principle and technical effect are similar, and will not be repeated here.
  • the electronic device includes: a processor (processor) 901, a memory (memory) 902 and a bus 903;
  • the processor 901 and the memory 902 communicate with each other through the bus 903;
  • the processor 901 is configured to call program instructions in the memory 902 to execute the following steps:
  • the identification label is used to indicate whether the TOA measurement value after the screening is NLOS;
  • the TOA measurement value after screening is corrected according to the identification tag to eliminate the error caused by NLOS in the TOA measurement value after screening.
  • the hybrid Gaussian modeling of the probability density of the TOA measurement value of the arrival time of each base station to the terminal UE specifically includes:
  • the first TOA measurement value is the TOA measurement value of each base station that is measured by the UE and reaches the UE.
  • the performing Gaussian mixture modeling according to the probability density of the first TOA measurement value to obtain the second TOA measurement value represented by the Gaussian mixture model specifically includes:
  • the probability density of each first TOA measurement value is mixed Gaussian modeling to obtain the second TOA measurement value
  • r Bias is an additive deviation greater than zero, Is the measurement error of the first TOA measurement value; n is the number of measurements, n ⁇ N; k is the number of channel changes of the value of r Bias as the channel changes; ⁇ i is the mean value corresponding to base station i; ⁇ i is the corresponding value of base station i Standard deviation, ⁇ i is the amplitude corresponding to the Gaussian distribution of base station i, Express The probability density.
  • the screening of the TOA measurement value after the Gaussian mixture modeling includes:
  • the acquiring the TOA measurement value with the smallest mean value among the second TOA measurement values as the third TOA measurement value specifically includes:
  • z is the number corresponding to the smallest mean among the mean values of the second TOA measurement values ⁇ 1 , ⁇ 2 ,... ⁇ k ,; ⁇ z is the mean value corresponding to the Gaussian distribution with the smallest mean in the Gaussian mixture distribution; j It is the number when traversing the number of channel changes.
  • the non-direct diameter NLOS identification is performed on the TOA measurement value after screening to obtain an identification label, which specifically includes:
  • the correcting the TOA measurement value after screening according to the identification tag specifically includes:
  • Correction processing is performed on the corresponding fifth TOA measurement value according to each identification tag, and the corrected TOA measurement value is obtained.
  • the method further includes:
  • the predefined criterion includes selecting the user position with the smallest distance residual.
  • the method before performing Gaussian mixture modeling on the probability density of TOA measurement values at each arrival time, and screening the TOA measurement values after Gaussian mixture modeling, the method further includes:
  • Receive positioning reference signal PRS configuration information sent by each base station and measure the downlink PRS signal according to each PRS configuration information, to obtain the TOA measurement value of each base station arriving at the UE.
  • the device for eliminating the indirect radiation path of the measured value of arrival time described in this embodiment can be used to implement the above method embodiment, and its principle and technical effect are similar, and will not be repeated here.
  • the computer program product includes a computer program stored on a non-transitory computer-readable storage medium.
  • the computer program includes program instructions. When the program instructions are executed by a computer, the computer The following methods can be performed:
  • the identification label is used to indicate whether the TOA measurement value after the screening is NLOS;
  • the TOA measurement value after screening is corrected according to the identification tag to eliminate the error caused by NLOS in the TOA measurement value after screening.
  • the identification label is used to indicate whether the TOA measurement value after the screening is NLOS;
  • the TOA measurement value after screening is corrected according to the identification tag to eliminate the error caused by NLOS in the TOA measurement value after screening.
  • the device embodiments described above are merely illustrative.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in One place, or it can be distributed to multiple network units.
  • Some or all of the modules can be selected according to actual needs to achieve the objectives of the solutions of the embodiments. Those of ordinary skill in the art can understand and implement it without creative work.
  • each implementation manner can be implemented by software plus a necessary general hardware platform, and of course, it can also be implemented by hardware.
  • the above technical solution essentially or the part that contributes to the existing technology can be embodied in the form of a computer software product, which can be stored in a computer-readable storage medium, such as ROM/RAM, A magnetic disk, an optical disk, etc., include several instructions to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some parts of the embodiment.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Pure & Applied Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Mathematical Optimization (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Remote Sensing (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Evolutionary Biology (AREA)
  • Operations Research (AREA)
  • Probability & Statistics with Applications (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Algebra (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Databases & Information Systems (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Mobile Radio Communication Systems (AREA)
  • Position Fixing By Use Of Radio Waves (AREA)

Abstract

本申请实施例公开了一种到达时间测量值的非直射径消除方法、装置及终端，方法包括：对各基站到达终端UE的TOA测量值的概率密度进行混合高斯建模，并对混合高斯建模后的TOA测量值进行筛选和NLOS鉴别，得到鉴别标签；鉴别标签用于指示筛选后的TOA测量值是否为NLOS；根据鉴别标签对筛选后的TOA测量值进行修正，以消除筛选后的TOA测量值中的NLOS所带来的误差。通过对各TOA测量值的概率密度进行混合高斯建模和筛选，确保在LOS与NLOS混叠的情况下准确找出属于LOS径的TOA测量值；同时通过对筛选后的TOA测量值进行修正，消除筛选后的TOA测量值中的NLOS所带来的误差，提高用户的定位精度。

Description

到达时间测量值的非直射径消除方法、装置及终端
相关申请的交叉引用
本申请要求于2020年02月05日提交的申请号为2020100807374，发明名称为“到达时间测量值的非直射径消除方法、装置及终端”的中国专利申请的优先权，其通过引用方式全部并入本文。
技术领域
本申请涉及通信技术领域，具体涉及一种到达时间测量值的非直射径消除方法、装置及终端。
背景技术
无线蜂窝网定位可以认为是基站(或接收机端)接收到基站与接收机之间的无线电波，并根据无线电波的特性，从而确定接收机位置的过程。在移动通信定位中，对定位精度产生影响的误差主要分为两类。第一类是由于接收机端的热噪声，接收机设计导致的测量误差。其统计特点为高斯白噪声，在实际操作中不可避免，其对于定位精度的影响有限。第二类误差是由于无线电波特性被严重破坏所引起的。以TOA(Time of Arrival，到达时间)测量为例，由于多径和非视距传播，导致接收机获得的TOA测量值不准确，通常的定位算法在此情况下会出现较大的定位偏差，甚至导致无法定位的情况。在实际通信环境中，此类由于信道环境而产生的误差无法避免，这些误差会严重影响定位精度。因此如何在复杂的无线通信环境中克服由于信道造成的影响，是无线定位算法研究的关键性问题。
关于如何克服NLOS(Non-Line of Sight，非直射径)所带来的定位误差，有两种方式可以进行NLOS误差抑制：首先可以通过算法，使得接收机得到的TOA值均为LOS(Line of sight，直射径)，再经过 定位算法进行用户位置确定，从而获得准确的移动台位置。但是在接收机侧，在无其他先验知识的前提下，很难进行LOS径或NLOS径的判定。对于复杂时变的信道情况，即便有对于信道的先验知识，也往往难以进行有效实时的TOA测量值识别。其次，基站(或接收机)可以通过先进的定位算法进行NLOS识别或者修正。即基站(或接收机)接收到的TOA测量值是LOS径与NLOS径混叠的情况。以6基站，单用户为例，即存在6个TOA测量值，一种可能的情况为有3个TOA测量值为LOS径，有三个TOA测量值为NLOS径。在上述情况下，如何对NLOS径进行识别，抛弃NLOS测量的结果，或者对于识别出的NLOS结果进行进一步修正，是无线通信定位中需要研究的关键问题。
对于NLOS的识别，单次TOA的测量结果无法进行进一步分析，需要进行一段时间内对目标进行多次测距。主要分为两种方式：第一类为非贝叶斯检测，在LOS/NLOS先验概率位置的情况，通过分析多次TOA测量的结果，利用结果分布的特性进行NLOS鉴别。例如NLOS的TOA测量值方差往往大于LOS测量值，利用此信息进行NLOS鉴别。其次，由上述描述可知，属于LOS测量值的TOA结果服从于高斯分布，因此，在未知LOS/NLOS传播先验概率的情况下，可以通过检验测量值是否服从高斯分布来识别NLOS传播。近年来先后出现了K-S、A-D、Chi-Square、格鲁斯检验、偏斜度和峭度检验等检验方法。第二类为贝叶斯检测，LOS/NLOS传播先验概率，则可以根据其误差具有的不同概率统计分布，采用广义似然比检验等方法来识别LOS/NLOS传播路径。上述方法都需要预知一定的先验信息，例如LOS/NLOS传播的先验概率，NLOS误差统计特性、预定的虚警概率等。然而在不同的实际定位场景下，这些信息往往差别很大，不事先进行实际测量往往无法知晓，故上述方法的适用范围有限。
无论采用上述哪种方式，对于NLOS的鉴别均需要一段时间对于目标进行多次测距。由于信道的时变特性，多次测量的TOA结果难以 服从特定的分布假设，即使是LOS径TOA测量值，在长时间测量时也存在不服从高斯分布的情况；另外，实际情况中LOS/NLOS传播先验概率也难以预知，造成了TOA测量值的NLOS难以消除。
发明内容
由于现有方法存在上述问题，本申请实施例提出一种到达时间测量值的非直射径消除方法、装置及终端。
第一方面，本申请实施例提出一种到达时间测量值的非直射径消除方法，包括：
对各基站到达终端UE的到达时间TOA测量值的概率密度进行混合高斯建模，并对混合高斯建模后的TOA测量值进行筛选；
对筛选后的TOA测量值进行非直射径NLOS鉴别，得到鉴别标签；所述鉴别标签用于指示筛选后的TOA测量值是否为NLOS；
根据所述鉴别标签对所述筛选后的TOA测量值进行修正，以消除所述筛选后的TOA测量值中的NLOS所带来的误差。
第二方面，本申请实施例还提出一种到达时间测量值的非直射径消除装置，包括：
测量值建模模块，用于对各基站到达终端UE的到达时间TOA测量值的概率密度进行混合高斯建模，并对混合高斯建模后的TOA测量值进行筛选；
NLOS鉴别模块，用于对筛选后的TOA测量值进行非直射径NLOS鉴别，得到鉴别标签；所述鉴别标签用于指示筛选后的TOA测量值是否为NLOS；
测量值修正模块，用于根据所述鉴别标签对所述筛选后的TOA测量值进行修正，以消除所述筛选后的TOA测量值中的NLOS所带来的误差。
第三方面，本申请实施例还提出一种终端，包括：
至少一个处理器；以及
与所述处理器通信连接的至少一个存储器，其中：
所述存储器存储有可被所述处理器执行的程序指令，所述处理器调用所述程序指令执行以下方法：
对各基站到达终端UE的到达时间TOA测量值的概率密度进行混合高斯建模，并对混合高斯建模后的TOA测量值进行筛选；
对筛选后的TOA测量值进行非直射径NLOS鉴别，得到鉴别标签；所述鉴别标签用于指示筛选后的TOA测量值是否为NLOS；
根据所述鉴别标签对所述筛选后的TOA测量值进行修正，以消除所述筛选后的TOA测量值中的NLOS所带来的误差。
第四方面，本申请实施例还提出一种非暂态计算机可读存储介质，所述非暂态计算机可读存储介质存储计算机程序，所述计算机程序使所述计算机执行以下方法：
对各基站到达终端UE的到达时间TOA测量值的概率密度进行混合高斯建模，并对混合高斯建模后的TOA测量值进行筛选；
对筛选后的TOA测量值进行非直射径NLOS鉴别，得到鉴别标签；所述鉴别标签用于指示筛选后的TOA测量值是否为NLOS；
根据所述鉴别标签对所述筛选后的TOA测量值进行修正，以消除所述筛选后的TOA测量值中的NLOS所带来的误差。
由上述技术方案可知，本申请实施例通过对各TOA测量值的概率密度进行混合高斯建模和筛选，确保在LOS与NLOS混叠的情况下准确找出属于LOS径的TOA测量值；同时通过对筛选后的TOA测量值进行修正，消除筛选后的TOA测量值中的NLOS所带来的误差，提高用户的定位精度。
附图说明
为了更清楚地说明本申请实施例或现有技术中的技术方案，下面 将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些图获得其他的附图。
图1为本申请一实施例提供的一种到达时间测量值的非直射径消除方法的流程示意图；
图2为本申请一实施例提供的一种UE定位的场景示意图；
图3为本申请一实施例提供的高斯混合模型的概率分布示意图；
图4为本申请一实施例提供的存在直射径时的TOA误差分布的曲线示意图；
图5为本申请一实施例提供的不存在直射径时的TOA误差分布的曲线示意图；
图6为本申请一实施例提供的NLOS方差较大时的TOA误差分布的曲线示意图；
图7为本申请一实施例提供的最终定位的误差CDF分布的曲线示意图；
图8为本申请一实施例提供的一种到达时间测量值的非直射径消除装置的结构示意图；
图9为本申请一实施例提供的电子设备的逻辑框图。
具体实施方式
下面结合附图，对本申请的具体实施方式作进一步描述。以下实施例仅用于更加清楚地说明本申请的技术方案，而不能以此来限制本申请的保护范围。
图1示出了本实施例提供的一种到达时间测量值的非直射径消除方法的流程示意图，包括：
S101、对各基站到达UE的TOA测量值的概率密度进行混合高斯建模，并对混合高斯建模后的TOA测量值进行筛选。
其中，TOA测量值为UE对下行PRS(Positioning Reference Signal，定位参考信号)进行测量得到的PRS信号从基站到UE的时间。
TOA测量值的概率密度是将多个TOA测量值作为随机变量，通过统计分析得到的概率密度分布，将TOA测量值建模为服从混合高斯的随机变量。
混合高斯建模为建立混合高斯模型的过程；其中，混合高斯模型是用高斯概率密度分布(正态分布曲线)精确地量化事物，是一个将事物分解为若干个基于高斯概率密度分布形成的模型。
具体地，由于获取的各TOA测量值中存在部分无用的TOA测量值，为了提供后续处理的效率，需要对混合高斯建模后的TOA测量值进行筛选，去除无用的TOA测量值。
S102、对筛选后的TOA测量值进行NLOS鉴别，得到鉴别标签。
其中，所述鉴别标签用于指示筛选后的TOA测量值是否为NLOS。
具体地，根据标准差以及偏斜度与峭度等鉴别方法，对筛选后的TOA测量值进行鉴别，确定存在NLOS误差的TOA测量值。
S103、根据所述鉴别标签对所述筛选后的TOA测量值进行修正，以消除所述筛选后的TOA测量值中的NLOS所带来的误差。
具体地，对携带NLOS标签的TOA测量值进行修正，得到消除NLOS误差的TOA测量值。通过对消除NLOS误差的TOA测量值进行定位处理，能够得到更为精确的用户定位结果。
本实施例通过对各TOA测量值的概率密度进行混合高斯建模和筛选，确保在LOS与NLOS混叠的情况下准确找出属于LOS径的TOA测量值；同时通过对筛选后的TOA测量值进行修正，消除筛选后的 TOA测量值中的NLOS所带来的误差，提高用户的定位精度。
进一步地，在上述方法实施例的基础上，S101中所述对各基站到达终端UE的到达时间TOA测量值的概率密度进行混合高斯建模，具体包括：
根据第一TOA测量值的概率密度进行混合高斯建模，得到混合高斯模型表示的第二TOA测量值。
其中，所述第一TOA测量值为UE测量得到的各基站到达UE的TOA测量值。
所述第二TOA测量值为对第一TOA测量值的概率密度进行混合高斯建模后得到的TOA测量值。
所述信道变化次数为大于零的加性偏差随着信道变化的次数。
S101之前还包括：
接收各基站发送的PRS配置信息，并根据各PRS配置信息对下行PRS信号进行测量，得到各基站到达UE的TOA测量值。
其中，所述PRS配置信息为基站向UE发送的PRS信号的配置信息。
具体地，在进行TOA测量值的NLOS消除的过程中，可以包括如下步骤：
Step1、UE接收网络通知的PRS配置信息，测量下行PRS信号并且获取各个基站到达本UE的多次TOA测量值，将接收端多次测得的TOA测量值称为第一TOA测量值；
Step2、在直射径条件下，多次测得的第一TOA测量值无明显变化，由于多径以及非直射径的影响，使得每次测得的TOA测量值有较大的差别。本实施例将多次测得的第一TOA测量值作为随机变量，通过对于其概率密度的分析，将第一TOA测量值建模为服从混合高斯分布的随机变量，在此建模基础上，进行NLOS径的识别。根据建模假设，利用每个基站到用户的第一TOA测量值进行概率密度的分 析(例如：变分狄利克雷过程)，得到第一TOA测量值的概率密度参数，将其记录并记为第二TOA测量值。
需要说明的是，根据本实施例的假设，TOA测量值的概率密度服从混合高斯分布，并对所有测量值进行概率密度分析，得到混合高斯模型的各个参数。
对于基站侧：包括如下步骤：
通过高层信令向UE发送下行PRS配置信息；
向UE发送下行PRS信号。
本实施例针对于长时间测距，存在信道变化从而影响TOA测量结果的场景，提出基于高斯混合模型的数据预处理方式。通过使用高斯混合模型进行数据分析，可以进一步检测LOS/NLOS径，从而对NLOS径进行数据修正或者剔除，进一步提高定位精度。
进一步地，在上述方法实施例的基础上，S101中所述对混合高斯建模后的TOA测量值进行筛选，具体包括：
获取所述第二TOA测量值中均值最小的TOA测量值，作为第三TOA测量值。
具体地，在进行TOA测量值的NLOS消除的过程中，还可以包括如下步骤：
Step3、取Step2中的第二TOA测量值，将均值最小的第二TOA测量值作为第三TOA测量值；即将Step2得到的第二TOA测量值进行分析，过滤掉无用的TOA测量值，得到第三TOA测量值。
通过对第二TOA测量值进行筛选，取均值最小的TOA测量值，能够过滤掉部分无用的TOA测量值，提高后续处理的效率。
进一步地，在上述方法实施例的基础上，S102具体包括：
根据标准差NLOS鉴别方法或偏斜度与峭度NLOS鉴别方法对所述第三TOA测量值进行NLOS鉴别，得到第四TOA测量值和对应的鉴别标签；
S103具体包括：
根据留一法或梯度下降法去除所述第四TOA测量值中NLOS误差最大的TOA测量值，得到第五TOA测量值；
根据各鉴别标签对对应的第五TOA测量值进行遍历修正处理，得到经过修正后的TOA测量值。
其中，所述第四TOA测量值为对第三TOA测量值进行NLOS鉴别后得到的TOA测量值。
所述第五TOA测量值为去除第四TOA测量值中NLOS误差最大的TOA测量值后得到的第五TOA测量值。
具体地，在进行TOA测量值的NLOS消除的过程中，还可以包括如下步骤：
Step4、针对第三TOA测量值进行NLOS鉴别并打上标签，得到第四TOA测量值，其中，NLOS鉴别方法包括标准差NLOS鉴别方法或偏斜度与峭度鉴别NLOS的方法。
Step5、在第四TOA测量值中，去除基站与用户间NLOS误差最大的TOA测量值，得到第五TOA测量值，其中，去除的方法可以使用留一法或其他方法如梯度下降法。
经过NLOS鉴别后，对属于NLOS径的TOA测量值进行误差补偿的设计包括：结合留一法去除基站与用户间NLOS误差最大的TOA测量值，遍历数据进行NLOS修正。
具体地，基于留一法以及遍历数据进行NLOS误差修正的定位方式，TOA定位误差在10米以内，可以对其进行有效修正，从而提高了定位的精度。
本实施例结合留一法、位置残差检测(RRT，Range Residuals Test)、遍历数据进行NLOS修复，使得定位精度得到大幅度提升。同时，基于高斯混合模型对NLOS鉴别，提高了NLOS鉴别成功的 概率，并且为NLOS误差修正提供依据，使得整体定位算法的复杂度大大降低。
进一步地，在上述方法实施例的基础上，S103之后，还包括：
根据修正得到的第六TOA测量值、用户定位算法和预定义准则计算得到用户定位结果，并根据所述用户定位结果对UE进行定位。
其中，所述预定义准则包括选取距离残差最小的用户位置。
所述第六TOA测量值为对第五TOA测量值进行NLOS消除后的TOA测量值。
具体地，在进行TOA测量值的NLOS消除的过程中，还可以包括如下步骤：
Step6、Step5处理后的第五TOA测量值与Step4操作中的标签结合，读出NLOS测量，并对数据进行遍历修正处理。对NLOS误差进行修正得到第六TOA测量值，基于第六TOA测量通过用户定位算法和预定义准则获得最终用户定位结果。
需要说明的是，Step4中的标准差NLOS鉴别方法或偏斜度与峭度NLOS鉴别法，Step5中的留一法，梯度下降法，Step6中的用户定位算法，为非唯一计算方式，可以替代处理Step4、Step5中两种方法均可交替使用，Step6中定位算法可以使用Chan算法或最小二乘法等等。
本实施例通过对NLOS误差进行消除，能够大大提高定位精度。
进一步地，在上述方法实施例的基础上，所述根据第一TOA测量值的概率密度进行混合高斯建模，得到混合高斯模型表示的第二TOA测量值，具体包括：
通过下述公式，对各第一TOA测量值的概率密度进行混合高斯建模，得到第二TOA测量值
Figure PCTCN2020136564-appb-000001
Figure PCTCN2020136564-appb-000002
其中，
Figure PCTCN2020136564-appb-000003
为第n次测量得到的用户a与基站i之间的第一TOA测量值，
Figure PCTCN2020136564-appb-000004
为第n次测量得到的用户a到基站i之间的TOA真实值，r Bias为大于零的加性偏差，
Figure PCTCN2020136564-appb-000005
为第一TOA测量值的测量误差；n为测量次数，n≤N；k为r Bias的值随着信道变化的信道变化次数；μ i为基站i对应的均值；σ i为基站i对应的标准差，α i为基站i的高斯分布对应的幅值，
Figure PCTCN2020136564-appb-000006
表示
Figure PCTCN2020136564-appb-000007
的概率密度。
进一步地，所述获取所述第二TOA测量值中均值最小的TOA测量值，作为第三TOA测量值，具体包括：
获取所述第二TOA测量值中均值最小的TOA测量值，并将所述均值最小的TOA测量值作为用户与基站i之间的第三TOA测量值：
Figure PCTCN2020136564-appb-000008
其中，z为各第二TOA测量值的均值μ 1,μ 2,…μ k,，中最小均值所对应的序号；μ z为高斯混合分布中，均值最小的高斯分布所对应的均值；j为遍历信道变化次数时的编号。
具体来说，在进行TOA测量值的NLOS消除的过程中的Step1-Step6，具体包括如下步骤：
Step1、UE接收网络通知的PRS配置信息，测量下行PRS信号并且获取各个基站到达本UE的TOA测量值。
图2为TOA定位过程示意图，参与定位的基站向终端发送PRS，UE侧根据网络侧通知的PRS信息，进行PRS序列生成。当用户接收到基站侧的PRS信号时，将接收PRS信号与本地生成的PRS信号进行处理，可测量到用户到相应基站的TOA测量值，并令其为第一TOA测量值。TOA测量值反映了用户与基站之间的距离。在UE不进行移 动的情况下，重复此步骤，可以得到多个基站到UE之间的距离测量值。若测量时用户与基站之间的无线信道环境发生变化，或者用户处于非直射径的情况下(即用户与基站之间有遮挡)，则多个测量值之间会有较大的不同。本实施例需要解决的问题为如何在多个测量值中分辩出用户与基站为直射径的测量值(用户与基站无遮挡)，另一种情况为当用户与基站之间的距离测量全为非直射径时，如何对测量值进行误差修正，使得修正后的测量值接近于直射径情况下的测量结果。
Step2、TOA测量值混合高斯模型的建立。
由于多径以及NLOS径的影响，TOA测量值准确度将会严重下降。基站i到用户a的一次TOA测量可以被描述为：
Figure PCTCN2020136564-appb-000009
Figure PCTCN2020136564-appb-000010
为用户a到基站i的TOA测量值，单位为米。
Figure PCTCN2020136564-appb-000011
为用户a到基站i的TOA真实值，单位为米。在二维定位时，假设用户坐标可以表示为(x,y)，基站坐标表示为(x i,y i)则
Figure PCTCN2020136564-appb-000012
r Bias为大于零的加性偏差，其表示了信道的变化引起的多径以及NLOS的影响。其大小随着终端环境的变化而改变。
Figure PCTCN2020136564-appb-000013
为TOA测量误差，建模为均值为0，方差为
Figure PCTCN2020136564-appb-000014
的高斯噪声。
本实施例将用户与基站之间信道的时变特点通过公式(3)进行建模，其中多径以及NLOS径的影响是造成信道时变的主要原因。对公式(3)进行简要分析，r Bias表示了时变信道对于第一TOA测量值的影响。若r Bias＝0，则表示用户与基站之间的无线环境是稳定不变的。r Bias＞0则表示了用户与基站之间的路径出现了遮挡，例如信号被墙体折射后被用户接收。在一次TOA测量时，r Bias被视为定值，但是随着测量时间的增加，r Bias的值会改变。例如共使用1000帧测量基站与终端的位置，前0～500帧，r Bias＝C 1，后500～1000帧，r Bias＝C 2，C 1与C 2为常数。由于噪声
Figure PCTCN2020136564-appb-000015
是高斯分布，则根据概率论知识，可以得知第一 测TOA量值
Figure PCTCN2020136564-appb-000016
是一个随机变量，即多次对其测量，得到的值不同。随机变量可以由概率密度函数(PDF，Probability Density Function)来描述。通过上述分析，
Figure PCTCN2020136564-appb-000017
为高斯分布，r Bias在不同测量时期的值不尽相同，因此在N次测量中，
Figure PCTCN2020136564-appb-000018
可以认为是多个均值不同的高斯分布的叠加。继续以上个例子说明，前500帧，500个
Figure PCTCN2020136564-appb-000019
的均值为
Figure PCTCN2020136564-appb-000020
后500帧，500个
Figure PCTCN2020136564-appb-000021
的均值为
Figure PCTCN2020136564-appb-000022
因此可以认为第一TOA测量值的概率密度函数服从混合高斯分布。
以用户a与基站i之间的第一TOA测量值为例，将
Figure PCTCN2020136564-appb-000023
写为
Figure PCTCN2020136564-appb-000024
简写为
Figure PCTCN2020136564-appb-000025
n表示第n次测量的TOA值，第一TOA测量值的概率密度可以表示为公式(1)。其中p(*)表示概率密度。k为类别总数，表示r Bias的值随着信道变化的次数，在数十分钟的测量时间内，认为k≤4。μ i为每个类别对应的均值σ i为标准差，α i为每个高斯分布对应的幅值，称为混合高斯系数，且混合系数应满足：
Figure PCTCN2020136564-appb-000026
通过公式，我们可以得知，确定第一TOA测量值的分布确定下述几个参数：k为类别总数，α i为每个高斯分布对应的幅值，μ i为每个类别对应的均值σ i为标准差。
Figure PCTCN2020136564-appb-000027
通过变分狄利克雷过程，对N个
Figure PCTCN2020136564-appb-000028
结果进行概率密度的估计，则可以得到第一TOA测量值的概率密度分布：GMM(k,α i,μ i,σ i),i∈1,2,...,k。
数据的概率密度函数可以反映出数据分布的特性，比如第一TOA测量值的均值，方差等分布特征。下面对概率密度分布的参数进行分析。k为混合模型的类别数，在本实施例中表示测量时信道发生了几次变化，以测量1000帧，前500帧r Bias＝C 1，后500帧r Bias＝C 2为例， 则此时k＝2。对应的均值
Figure PCTCN2020136564-appb-000029
将均值和方差N(μ i,(σ i) 2),i∈1,2,...,k记录下来，作为第二TOA测量值进行后续处理。若测量1000帧，则有1000个第一TOA测量值
Figure PCTCN2020136564-appb-000030
但是经过概率密度分析，第二TOA测量值只有2k个(k个均值和k个方差)。
需要说明的是，TOA的测量本质上反映了信道的变化，本步骤通过TOA测量值的假设建模(高斯混合模型)，来反映不同时刻，用户到基站的传输路径是否被遮挡。此步骤的优点在于通过对高斯混合模型的分析，为用户在接收端无任何先验信息，仅有若干TOA测量值的情况下的NLOS鉴别提供了可能性。
Step3、取Step2获得混合高斯模型中均值μ i最小的高斯分布作为第三TOA测量值。
N个TOA测量值形成了一个高斯混合分布，经过变分狄利克雷过程，可以获得混合高斯分布的参数。图3给出了某个用户与基站之间的TOA测量值经过处理后，得到的高斯混合模型。
根据公式(3)，用户TOA测量值
Figure PCTCN2020136564-appb-000031
由用户位置真实值
Figure PCTCN2020136564-appb-000032
和正向偏差r Bias以及噪声
Figure PCTCN2020136564-appb-000033
组成，因此
Figure PCTCN2020136564-appb-000034
的均值应大于等于用户位置真实值
Figure PCTCN2020136564-appb-000035
即：
Figure PCTCN2020136564-appb-000036
对应于第二TOA测量值，应有：
Figure PCTCN2020136564-appb-000037
故对应第二TOA测量值，由于所有高斯分布的均值均超过真实值，故只需要取均值最小的一组高斯分布进行NLOS鉴别，如公式(2)所示，其余高斯分布直接可以判定为NLOS径。
取出最小均值的所对应的高斯分布N(μ z,(σ z) 2)，进行后续NLOS检验。属于此高斯分布的TOA测量值为第三TOA测量值。
需要说明的是，本步骤通过对混合高斯分布均值的分析，过滤掉大部分NLOS的TOA测量值。能够基于模型假设，通过均值与LOS径关系的分析，快速准确的排出大部分属于NLOS的TOA测量值。
Step4、基于高斯混合模型对NLOS进行鉴别。
基于高斯混合模型对NLOS进行鉴别包括基于标准差的NLOS鉴别方法和基于偏斜度与峭度的NLOS识别方法。
基于标准差的NLOS鉴别方法如下：
由公式(3)可知，引起TOA测量误差的原因有两项：
1)NLOS引起的正向偏差r Bias；
2)测量噪声
Figure PCTCN2020136564-appb-000038
在不存在NLOS径的情况下r Bias近似为0，因此测得的TOA测量结果应服从均值为0，方差为σ的高斯分布。在存在NLOS径的情况下，由于NLOS误差与测量误差相互独立，因此对于第三TOA测量值，具有较大方差的高斯分量可以认为是存在NLOS径。因此可以构建二元检测模型：
Figure PCTCN2020136564-appb-000039
其中σ Th为门限标准差，可以根据实际环境的不同进行设定，本实施例参考门限为σ Th＝1，认为标准差小于1米的高斯分布是LOS径，其余为NLOS径。将鉴别完毕的第三TOA测量值打入标签(LOS/NLOS)，作为后续NLOS修正以及最终定位的输入值，即有
Figure PCTCN2020136564-appb-000040
打标签具体方式为：在每个测量值后新增一个变量j，j∈[0,1]，当测量值为LOS径时，j＝1，否则j＝0。将打好标签的TOA测量值记做第四TOA测量值。
图4和图5分别示出了存在LOS径以及只有NLOS径情况下混合高斯分布的图示。由图可以看出NLOS的标准差远大于LOS径的标准差。
经过N＝1000帧测量后TOA的误差概率分布图。即：
Figure PCTCN2020136564-appb-000041
由图5可以分析得知，在此例中，k＝3，μ 1＝0,μ 2＝9.3,μ 3＝14.2且σ 1＝0.9＜σ 2＜σ 3。
基于偏斜度与峭度的NLOS识别方法如下：
由上述分析可知LOS径的测量值应服从高斯分布，而NLOS测量值大概率不服从标准高斯分布，斜度和偏度是检验样本是否符合高斯分布的一个重要工具，因此可以用偏斜度和峭度进行NLOS的鉴别。
对于斜度和偏度的检验需要用到属于均值最小的高斯分布的所有TOA测量值。因此需要对TOA测量值进行简单分类。利用马氏距离(Mahalanobis distance)，对第一TOA值进行检测，找出属于最小均值高斯分布的TOA测量值。
Figure PCTCN2020136564-appb-000042
对于第n个数据
Figure PCTCN2020136564-appb-000043
遍历j∈{1,...,k}，若j＝k时
Figure PCTCN2020136564-appb-000044
最小，则第n个第一TOA测量值归属为第k类。
Figure PCTCN2020136564-appb-000045
偏斜度(Skewness)可以检测样本是否对称分布，对称分布的样本峭度为0。假设有m个TOA测量值属于最小均值的高斯分布，则可以使用偏斜度对其进行检验：
Figure PCTCN2020136564-appb-000046
高斯分布的峭度(Kurtosis)为3，可以通过峭度验证样本是否为高斯分布：
Figure PCTCN2020136564-appb-000047
因此可以构建第二种二元检测模型：
Figure PCTCN2020136564-appb-000048
将鉴别完毕的高斯分布的均值打入标签(LOS/NLOS)，作为后续NLOS修正以及最终定位的输入值，即有
Figure PCTCN2020136564-appb-000049
将打好标签的TOA测量值记做第四TOA测量值。
需要说明的是，将标准差NLOS鉴别方法或偏斜度与峭度NLOS鉴别方法用在本实施例的假设模型中，可以准确快速的将Step3中未鉴别出的NLOS测量值鉴别出来，为后续NLOS处理，用户定位提供准确数据。
Step5、除去误差最大的NLOS测量值。
对鉴别出用户与基站之间存在NLOS的路径进行遍历处理，误差修正。为了进一步减小运算量，在修正前需要进行基于留一法的距离残差检测，以m基站单用户进行定位为例，检测的目的在于找出NLOS误差最大的基站进行剔除，保留其余m-1基站进行定位。由于遍历处理无法修正较大的NLOS误差，将最大NLOS误差所属的基站剔除，一方面保证了算法的有效性，另一方面减少了运算的复杂度。如图6所示，某基站与用户之间NLOS误差均值在8米左右，已经严重影响定位精度，且难以被有效修正，故应将此基站剔除。此基站的测量结果视为无效数据。鉴别最大NLOS误差的方法如下：
假设m基站参与用户定位(m＞3)，取m-1个基站进行定位，共有
Figure PCTCN2020136564-appb-000050
种组合。(留一法)
将每种组合分别用Chan算法进行定位，输入为m-1第四TOA测量值
Figure PCTCN2020136564-appb-000051
结合输入利用Chan算法进行用户位置确定，得到用户位置初始值(x o,y o)。
定义距离残差为：
Figure PCTCN2020136564-appb-000052
(x i,y i)为基站坐标，m为基站数。
不同的组合将会计算出不同的用户位置(x o,y o)，进而有不同的距离残差，选取距离残差最小的组合，以6基站为例，假设距离残差最小的组合为1,2,3,4,6，那么可以判断基站5的位置误差最大，则将基站5剔除，不参与后续定位，且记录此时的距离残差d min，此时得到的TOA测量组合为第五TOA测量值。
需要说明的是，采用本步骤可以加快后续步骤的运行，通过剔除最大NLOS误差测量可以减少后续的计算量。通过本步骤可以剔除一个最大NLOS误差测量，也可以剔除多个测量。不准确测量的剔除一方面加快运算，另一方面可以提高定位精度。
Step6、遍历数据对NLOS误差进行修正得到第六TOA测量值，然后基于预定义准则获得最终用户定位结果。
其中，预定义准则包括选取距离残差最小用户位置。
取出经过Step5大NLOS误差剔除的组合的TOA测量值
Figure PCTCN2020136564-appb-000053
对其进行LOS/NLOS标签进行读取，LOS测量值不做处理，得到J(J≤m-1)个NLOS测量值。
判断最小距离残差d min的大小，若d min≤8*(m-1)，进行小步长误差修正，否则进行大步长修正。本专利中小步长取为1米，大步长取为5米。
设定小步长区间：[0,2,3,4]，大步长区间：[0,5,10,15]，由于NLOS误差未知且为正数，因此判断为NLOS误差的TOA测量值需要依次减去区间中所有取值，得到修正的TOA测量
Figure PCTCN2020136564-appb-000054
以小步长为例，
Figure PCTCN2020136564-appb-000055
此时得到的TOA修正值为第六TOA测量值。并且每进行一次NLOS误差修正，便使用第五TOA测量值进行一次基于Chan算法的用户位置确定((x new,y new))。确定一次用户位置，需要进行一次距离残差检测：
Figure PCTCN2020136564-appb-000056
若有J个NLOS误差，则有J 4个NLOS误差修正，伴随有J 4个用户位置定位结果和J 4个距离残差。
选出距离残差最小的值，此时的用户位置确定为最终的用户位置。
需要说明的是，通过对第五TOA值进行遍历法NLOS修正，基于距离残差最小进行用户位置确定，从而大大地提高了NLOS场景下用户的定位精度。
本实施例通过将TOA测量值的概率密度建模为高斯混合模型，并使用变分狄利克雷过程求得混合模型参数的方式，在用户对于用户与基站之间的无线环境没有更多信息的情况下，可以使用此方法等效建模信道变化对于接受信号的影响。针对于TOA测量时用户信道时变的情况，本方法确保在LOS与NLOS混叠的情况下可以准确找出属于LOS径的TOA测量值。针对TOA测量值全为NLOS的情况，本专利可以对其进行鉴别，并加以修正。综上使得用户最终定位的结果更加精确。
在具体执行过程中，基站侧包括以下步骤：
A1、通过高层信令向UE发送下行PRS配置信息例如：BW＝50MHz，PRS资源包括6个OFDM符号，Comb因子为6；
A2、6个基站分别向UE发送下行PRS信号。持续一段时间，例如1000帧。
UE侧包括以下步骤：
B1、UE接收网络通知的PRS配置信息，测量下行PRS信号并且获取各个基站到达本UE的TOA测量值，统计多帧的TOA测量结果，每个基站到UE之间均有1000个TOA测量值，以基站i到用户a到为例，将这1000个TOA测量值成为第一TOA测量值。
B2、利用每个基站到用户的第一TOA测量值进行处理，将第一TOA测量值视为随机变量，分析其概率密度(如使用变分狄利克雷过程进行分析)，得到混合高斯模型。假设基站i到与用户a之间的信道变化2次，则有两个高斯分布进行混合，N(μ 1,(σ 1) 2)和N(μ 2,(σ 2) 2)将μ 1，σ 1，μ 2，σ 2记录下来作为第二TOA测量值。
B3、取B2获得的各个高斯分布的均值μ 1＜μ 2，μ 2，将最小的均值作为第三TOA测量值；假设μ 1＜μ 2，则第三TOA测量值为N(μ 1,(σ 1) 2)。
B4、针对第三TOA测量值(N(μ 1,(σ 1) 2))进行NLOS鉴别并打上标签，得到第四TOA测量值，其中，NLOS鉴别方法包括标准差NLOS鉴别方法或偏斜度与峭度鉴别NLOS。此时第四TOA测量为N(μ 1,(σ 1) 2),LOS/NLOS。
B5、B2～B4均以基站i到与用户a为例，假设有6个基站参与定位，则有6个第四TOA测量值，取出每个测量值的均值μi,i∈[16]进行去除基站与用户间NLOS误差最大的TOA测量值，得到5个第五TOA测量值N(μ i,(σ i) 2),LOS/NLOS,i∈[1,2,3,5,6]，其中，去除的方法可 以使用留一法。假设6个基站参与定位，采用B5排出了第4个基站的TOA结果。
B6、读取第五测量值的LOS/NLOS标签(假设基站1、2、3为LOS，5、6为NLOS)，遍历属于NLOS标签的第五测量值的均值数据μ j,j∈[5,6]，对NLOS误差进行修正得到第六TOA测量值(包含未修正的基站1、2、3的第五TOA测量值，以及修正过的基站5、6的第六TOA测量值)，然后基于μ q,q∈[1,2,3,5,6]和预定义准则获得最终用户定位结果，其中，预定义准则包括选取距离残差最小用户位置。
图7示出了在50MHZ的带宽下，在Indoor3种景下的仿真结果。作为对比，未经改进的Chan算法在同样条件下被仿真；改进算法和Channel的输入均为同样的TOA测量数据，由结果可见，经过NLOS识别与修正的TOA测量值，极大的提高了最终定位的精度。
本实施例在NLOS信道条件下，解决基于TOA测量值的无线电通信定位系统的误差鉴别和补偿问题，并且通过减少TOA测量值的误差，进一步的提高用户位置计算的精度。
图8示出了本实施例提供的一种到达时间测量值的非直射径消除装置的结构示意图，所述装置包括：测量值建模模块801、NLOS鉴别模块802和测量值修正模块803，其中：
所述测量值建模模块801用于对各基站到达终端UE的到达时间TOA测量值的概率密度进行混合高斯建模，并对混合高斯建模后的TOA测量值进行筛选；
所述NLOS鉴别模块802用于对筛选后的TOA测量值进行非直射径NLOS鉴别，得到鉴别标签；所述鉴别标签用于指示筛选后的TOA测量值是否为NLOS；
所述测量值修正模块803用于根据所述鉴别标签对所述筛选后的TOA测量值进行修正，以消除所述筛选后的TOA测量值中的NLOS所带来的误差。
具体地，所述测量值建模模块801对各基站到达终端UE的到达时间TOA测量值的概率密度进行混合高斯建模，并对混合高斯建模后的TOA测量值进行筛选；所述NLOS鉴别模块802对筛选后的TOA测量值进行非直射径NLOS鉴别，得到鉴别标签；所述鉴别标签用于指示筛选后的TOA测量值是否为NLOS；所述测量值修正模块803根据所述鉴别标签对所述筛选后的TOA测量值进行修正，以消除所述筛选后的TOA测量值中的NLOS所带来的误差。
本实施例通过对各TOA测量值的概率密度进行混合高斯建模和筛选，确保在LOS与NLOS混叠的情况下准确找出属于LOS径的TOA测量值；同时通过对筛选后的TOA测量值进行修正，消除筛选后的TOA测量值中的NLOS所带来的误差，提高用户的定位精度。
本实施例所述的到达时间测量值的非直射径消除装置可以用于执行上述方法实施例，其原理和技术效果类似，此处不再赘述。
参照图9，所述电子设备，包括：处理器(processor)901、存储器(memory)902和总线903；
其中，
所述处理器901和存储器902通过所述总线903完成相互间的通信；
所述处理器901用于调用所述存储器902中的程序指令，以执行如下步骤：
对各基站到达终端UE的到达时间TOA测量值的概率密度进行混合高斯建模，并对混合高斯建模后的TOA测量值进行筛选；
对筛选后的TOA测量值进行非直射径NLOS鉴别，得到鉴别标签；所述鉴别标签用于指示筛选后的TOA测量值是否为NLOS；
根据所述鉴别标签对所述筛选后的TOA测量值进行修正，以消除所述筛选后的TOA测量值中的NLOS所带来的误差。
本实施例通过对各TOA测量值的概率密度进行混合高斯建模和筛选，确保在LOS与NLOS混叠的情况下准确找出属于LOS径的TOA测量值；同时通过对筛选后的TOA测量值进行修正，消除筛选后的TOA测量值中的NLOS所带来的误差，提高用户的定位精度。
进一步地，在上述实施例的基础上，所述对各基站到达终端UE的到达时间TOA测量值的概率密度进行混合高斯建模，具体包括：
根据第一TOA测量值的概率密度进行混合高斯建模，得到混合高斯模型表示的第二TOA测量值；
其中，所述第一TOA测量值为UE测量得到的各基站到达UE的TOA测量值。
进一步地，在上述实施例的基础上，所述根据第一TOA测量值的概率密度进行混合高斯建模，得到混合高斯模型表示的第二TOA测量值，具体包括：
通过下述公式，对各第一TOA测量值的概率密度进行混合高斯建模，得到第二TOA测量值
Figure PCTCN2020136564-appb-000057
Figure PCTCN2020136564-appb-000058
其中，
Figure PCTCN2020136564-appb-000059
为第n次测量得到的用户a与基站i之间的第一TOA测量值，
Figure PCTCN2020136564-appb-000060
为第n次测量得到的用户a到基站i之间的TOA真实值，r Bias为大于零的加性偏差，
Figure PCTCN2020136564-appb-000061
为第一TOA测量值的测量误差；n为测量次数，n≤N；k为r Bias的值随着信道变化的信道变化次数；μ i为基站i对应的均值；σ i为基站i对应的标准差，α i为基站i的高斯分布对应的幅值，
Figure PCTCN2020136564-appb-000062
表示
Figure PCTCN2020136564-appb-000063
的概率密度。
进一步地，在上述实施例的基础上，所述对混合高斯建模后的TOA测量值进行筛选，具体包括：
获取所述第二TOA测量值中均值最小的TOA测量值，作为第三TOA测量值。
进一步地，在上述实施例的基础上，所述获取所述第二TOA测量值中均值最小的TOA测量值，作为第三TOA测量值，具体包括：
获取所述第二TOA测量值中均值最小的TOA测量值，并将所述均值最小的TOA测量值作为用户与基站i之间的第三TOA测量值：
Figure PCTCN2020136564-appb-000064
其中，z为各第二TOA测量值的均值μ 1,μ 2,…μ k,，中最小均值所对应的序号；μ z为高斯混合分布中，均值最小的高斯分布所对应的均值；j为遍历信道变化次数时的编号。
进一步地，在上述实施例的基础上，所述对筛选后的TOA测量值进行非直射径NLOS鉴别，得到鉴别标签，具体包括：
根据标准差NLOS鉴别方法或偏斜度与峭度NLOS鉴别方法对所述第三TOA测量值进行NLOS鉴别，得到第四TOA测量值和对应的鉴别标签；
所述根据所述鉴别标签对所述筛选后的TOA测量值进行修正，具体包括：
根据留一法或梯度下降法去除所述第四TOA测量值中NLOS误差最大的TOA测量值，得到第五TOA测量值；
根据各鉴别标签对对应的第五TOA测量值进行修正处理，得到经过修正后的TOA测量值。
进一步地，在上述实施例的基础上，所述根据所述鉴别标签对所述筛选后的TOA测量值进行修正之后，还包括：
根据修正得到的第六TOA测量值、用户定位算法和预定义准则计算得到用户定位结果，并根据所述用户定位结果对UE进行定位；
其中，所述预定义准则包括选取距离残差最小的用户位置。
进一步地，在上述实施例的基础上，所述对各到达时间TOA测量值的概率密度进行混合高斯建模，并对混合高斯建模后的TOA测量值进行筛选之前，还包括：
接收各基站发送的定位参考信号PRS配置信息，并根据各PRS配置信息对下行PRS信号进行测量，得到各基站到达UE的TOA测量值。
本实施例所述的到达时间测量值的非直射径消除装置可以用于执行上述方法实施例，其原理和技术效果类似，此处不再赘述。
本实施例公开一种计算机程序产品，所述计算机程序产品包括存储在非暂态计算机可读存储介质上的计算机程序，所述计算机程序包括程序指令，当所述程序指令被计算机执行时，计算机能够执行如下方法：
对各到达时间TOA测量值的概率密度进行混合高斯建模，并对混合高斯建模后的TOA测量值进行筛选；
对筛选后的TOA测量值进行非直射径NLOS鉴别，得到鉴别标签；所述鉴别标签用于指示筛选后的TOA测量值是否为NLOS；
根据所述鉴别标签对所述筛选后的TOA测量值进行修正，以消除所述筛选后的TOA测量值中的NLOS所带来的误差。
本实施例提供一种非暂态计算机可读存储介质，所述非暂态计算机可读存储介质存储计算机指令，所述计算机指令使所述计算机执行如下方法：
对各到达时间TOA测量值的概率密度进行混合高斯建模，并对混合高斯建模后的TOA测量值进行筛选；
对筛选后的TOA测量值进行非直射径NLOS鉴别，得到鉴别标签；所述鉴别标签用于指示筛选后的TOA测量值是否为NLOS；
根据所述鉴别标签对所述筛选后的TOA测量值进行修正，以消除所述筛选后的TOA测量值中的NLOS所带来的误差。
以上所描述的装置实施例仅仅是示意性的，其中所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部模块来实现本实施例方案的目的。本领域普通技术人员在不付出创造性的劳动的情况下，即可以理解并实施。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到各实施方式可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件来实现。基于这样的理解，上述技术方案本质上或者说对现有技术做出贡献的部分可以以计算机软件产品的形式体现出来，该计算机软件产品可以存储在计算机可读存储介质中，如ROM/RAM、磁碟、光盘等，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行各个实施例或者实施例的某些部分所述的方法。
应说明的是：以上实施例仅用以说明本申请的技术方案，而非对其限制；尽管参照前述实施例对本申请进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本申请各实施例技术方案的精神和范围。

Claims (25)

  1. 一种到达时间测量值的非直射径误差消除方法，其特征在于，包括：
    对各基站到达终端UE的到达时间TOA测量值的概率密度进行混合高斯建模，并对混合高斯建模后的TOA测量值进行筛选；
    对筛选后的TOA测量值进行非直射径NLOS鉴别，得到鉴别标签；所述鉴别标签用于指示筛选后的TOA测量值是否为NLOS；
    根据所述鉴别标签对所述筛选后的TOA测量值进行修正，以消除所述筛选后的TOA测量值中的NLOS所带来的误差。
  2. 根据权利要求1所述的到达时间测量值的非直射径误差消除方法，其特征在于，所述对各基站到达终端UE的到达时间TOA测量值的概率密度进行混合高斯建模，具体包括：
    根据第一TOA测量值的概率密度进行混合高斯建模，得到混合高斯模型表示的第二TOA测量值；
    其中，所述第一TOA测量值为UE测量得到的各基站到达UE的TOA测量值。
  3. 根据权利要求2所述的到达时间测量值的非直射径误差消除方法，其特征在于，所述根据第一TOA测量值的概率密度进行混合高斯建模，得到混合高斯模型表示的第二TOA测量值，具体包括：
    通过下述公式，对各第一TOA测量值的概率密度进行混合高斯建模，得到第二TOA测量值
    Figure PCTCN2020136564-appb-100001
    Figure PCTCN2020136564-appb-100002
    其中，
    Figure PCTCN2020136564-appb-100003
    为第n次测量得到的用户a与基站i之间的第一TOA测量值，
    Figure PCTCN2020136564-appb-100004
    Figure PCTCN2020136564-appb-100005
    为第n次测量得到的用户a到基站i之间的TOA真实值，r Bias为大于零的加性偏差，
    Figure PCTCN2020136564-appb-100006
    为第一TOA测量值的测量误差；n为测量次数，n≤N；k为r Bias的值随着信道变化 的信道变化次数；μ i为基站i对应的均值；σ i为基站i对应的标准差，α i为基站i的高斯分布对应的幅值，
    Figure PCTCN2020136564-appb-100007
    Figure PCTCN2020136564-appb-100008
    表示
    Figure PCTCN2020136564-appb-100009
    的概率密度。
  4. 根据权利要求2或3所述的到达时间测量值的非直射径误差消除方法，其特征在于，所述对混合高斯建模后的TOA测量值进行筛选，具体包括：
    获取所述第二TOA测量值中均值最小的TOA测量值，作为第三TOA测量值。
  5. 根据权利要求4所述的到达时间测量值的非直射径误差消除方法，其特征在于，所述获取所述第二TOA测量值中均值最小的TOA测量值，作为第三TOA测量值，具体包括：
    获取所述第二TOA测量值中均值最小的TOA测量值，并将所述均值最小的TOA测量值作为用户与基站i之间的第三TOA测量值：
    Figure PCTCN2020136564-appb-100010
    其中，z为各第二TOA测量值的均值μ 1,μ 2,…μ k,，中最小均值所对应的序号；μ z为高斯混合分布中，均值最小的高斯分布所对应的均值；j为遍历信道变化次数时的编号。
  6. 根据权利要求5所述的到达时间测量值的非直射径误差消除方法，其特征在于，所述对筛选后的TOA测量值进行非直射径NLOS鉴别，得到鉴别标签，具体包括：
    根据标准差NLOS鉴别方法或偏斜度与峭度NLOS鉴别方法对所述第三TOA测量值进行NLOS鉴别，得到第四TOA测量值和对应的鉴别标签；
    所述根据所述鉴别标签对所述筛选后的TOA测量值进行修正，具体包括：
    根据留一法或梯度下降法去除所述第四TOA测量值中NLOS误差最大的TOA测量值，得到第五TOA测量值；
    根据各鉴别标签对对应的第五TOA测量值进行修正处理，得到经过修正后的TOA测量值。
  7. 根据权利要求6所述的到达时间测量值的非直射径误差消除方法，其特征在于，所述根据所述鉴别标签对所述筛选后的TOA测量值进行修正之后，还包括：
    根据修正得到的第六TOA测量值、用户定位算法和预定义准则计算得到用户定位结果，并根据所述用户定位结果对UE进行定位；
    其中，所述预定义准则包括选取距离残差最小的用户位置。
  8. 根据权利要求1-3任一项或5-7任一项所述的到达时间测量值的非直射径误差消除方法，其特征在于，所述对各基站到达终端UE的到达时间TOA测量值的概率密度进行混合高斯建模，并对混合高斯建模后的TOA测量值进行筛选之前，还包括：
    接收各基站发送的定位参考信号PRS配置信息，并根据各PRS配置信息对下行PRS信号进行测量，得到各基站到达UE的TOA测量值。
  9. 一种到达时间测量值的非直射径误差消除装置，其特征在于，包括：
    测量值建模模块，用于对各基站到达终端UE的到达时间TOA测量值的概率密度进行混合高斯建模，并对混合高斯建模后的TOA测量值进行筛选；
    NLOS鉴别模块，用于对筛选后的TOA测量值进行非直射径NLOS鉴别，得到鉴别标签；所述鉴别标签用于指示筛选后的TOA测量值是否为NLOS；
    测量值修正模块，用于根据所述鉴别标签对所述筛选后的TOA测量值进行修正，以消除所述筛选后的TOA测量值中的NLOS所带来的误差。
  10. 根据权利要求9所述的到达时间测量值的非直射径误差消除装置，其特征在于，所述测量值建模模块具体用于：
    根据第一TOA测量值的概率密度进行混合高斯建模，得到混合高斯模型表示的第二TOA测量值；
    其中，所述第一TOA测量值为UE测量得到的各基站到达UE的TOA测量值。
  11. 根据权利要求10所述的到达时间测量值的非直射径误差消除装置，其特征在于，所述根据第一TOA测量值的概率密度进行混合高斯建模，得到混合高斯模型表示的第二TOA测量值，具体包括：
    通过下述公式，对各第一TOA测量值的概率密度进行混合高斯建模，得到第二TOA测量值
    Figure PCTCN2020136564-appb-100011
    Figure PCTCN2020136564-appb-100012
    其中，
    Figure PCTCN2020136564-appb-100013
    为第n次测量得到的用户a与基站i之间的第一TOA测量值，
    Figure PCTCN2020136564-appb-100014
    Figure PCTCN2020136564-appb-100015
    为第n次测量得到的用户a到基站i之间的TOA真实值，r Bias为大于零的加性偏差，
    Figure PCTCN2020136564-appb-100016
    为第一TOA测量值的测量误差；n为测量次数，n≤N；k为r Bias的值随着信道变化的信道变化次数；μ i为基站i对应的均值；σ i为基站i对应的标准差，α i为基站i的高斯分布对应的幅值，
    Figure PCTCN2020136564-appb-100017
    Figure PCTCN2020136564-appb-100018
    表示
    Figure PCTCN2020136564-appb-100019
    的概率密度。
  12. 根据权利要求10或11所述的到达时间测量值的非直射径误差消除装置，其特征在于，所述测量值建模模块具体用于：
    获取所述第二TOA测量值中均值最小的TOA测量值，作为第三TOA测量值。
  13. 根据权利要求12所述的到达时间测量值的非直射径误差消除装置，其特征在于，所述获取所述第二TOA测量值中均值最小的TOA测量值，作为第三TOA测量值，具体包括：
    获取所述第二TOA测量值中均值最小的TOA测量值，并将所述均值最小的TOA测量值作为用户与基站i之间的第三TOA测量值：
    Figure PCTCN2020136564-appb-100020
    其中，z为各第二TOA测量值的均值μ 1,μ 2,…μ k,，中最小均值所对应的序号；μ z为高斯混合分布中，均值最小的高斯分布所对应的均值；j为遍历信道变化次数时的编号。
  14. 根据权利要求13所述的到达时间测量值的非直射径误差消除装置，其特征在于，所述NLOS鉴别模块具体用于：
    根据标准差NLOS鉴别方法或偏斜度与峭度NLOS鉴别方法对所述第三TOA测量值进行NLOS鉴别，得到第四TOA测量值和对应的鉴别标签；
    所述测量值修正模块具体用于：
    根据留一法或梯度下降法去除所述第四TOA测量值中NLOS误差最大的TOA测量值，得到第五TOA测量值；
    根据各鉴别标签对对应的第五TOA测量值进行修正处理，得到经过修正后的TOA测量值。
  15. 根据权利要求14所述的到达时间测量值的非直射径误差消除装置，其特征在于，所述根据所述鉴别标签对所述筛选后的TOA测量值进行修正之后，还包括获取模块，用于：
    根据修正得到的第六TOA测量值、用户定位算法和预定义准则计算得到用户定位结果，并根据所述用户定位结果对UE进行定位；
    其中，所述预定义准则包括选取距离残差最小的用户位置。
  16. 根据权利要求9-11任一项或13-15任一项所述的到达时间测量值的非直射径误差消除装置，其特征在于，所述对各基站到达终端UE的到达时间TOA测量值的概率密度进行混合高斯建模，并对混合高斯建模后的TOA测量值进行筛选之前，还包括测量值获取模块，用于：
    接收各基站发送的定位参考信号PRS配置信息，并根据各PRS配置信息对下行PRS信号进行测量，得到各基站到达UE的TOA测量值。
  17. 一种终端，包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，其特征在于，所述处理器执行所述程序时执行如下步骤：
    对各基站到达终端UE的到达时间TOA测量值的概率密度进行混合高斯建模，并对混合高斯建模后的TOA测量值进行筛选；
    对筛选后的TOA测量值进行非直射径NLOS鉴别，得到鉴别标签；所述鉴别标签用于指示筛选后的TOA测量值是否为NLOS；
    根据所述鉴别标签对所述筛选后的TOA测量值进行修正，以消除所述筛选后的TOA测量值中的NLOS所带来的误差。
  18. 根据权利要求17所述的终端，其特征在于，所述对各基站到达终端UE的到达时间TOA测量值的概率密度进行混合高斯建模，具体包括：
    根据第一TOA测量值的概率密度进行混合高斯建模，得到混合高斯模型表示的第二TOA测量值；
    其中，所述第一TOA测量值为UE测量得到的各基站到达UE的TOA测量值。
  19. 根据权利要求18所述的终端，其特征在于，所述根据第一TOA测量值的概率密度进行混合高斯建模，得到混合高斯模型表示的第二TOA测量值，具体包括：
    通过下述公式，对各第一TOA测量值的概率密度进行混合高斯建模，得到第二TOA测量值
    Figure PCTCN2020136564-appb-100021
    Figure PCTCN2020136564-appb-100022
    其中，
    Figure PCTCN2020136564-appb-100023
    为第n次测量得到的用户a与基站i之间的第一TOA测量值，
    Figure PCTCN2020136564-appb-100024
    Figure PCTCN2020136564-appb-100025
    为第n次测量得到的用户a到基站i之间的TOA真实值，r Bias为大于零的加性偏差，
    Figure PCTCN2020136564-appb-100026
    为第一TOA测量值的测量误差；n为测量次数，n≤N；k为r Bias的值随着信道变化的信道变化次数；μ i为基站i对应的均值；σ i为基站i对应的标准差，α i为基站i的高斯分布对应的幅值，
    Figure PCTCN2020136564-appb-100027
    Figure PCTCN2020136564-appb-100028
    表示
    Figure PCTCN2020136564-appb-100029
    的概率密度。
  20. 根据权利要求18或19所述的终端，其特征在于，所述对混合高斯建模后的TOA测量值进行筛选，具体包括：
    获取所述第二TOA测量值中均值最小的TOA测量值，作为第三TOA测量值。
  21. 根据权利要求20所述的终端，其特征在于，所述获取所述第二TOA测量值中均值最小的TOA测量值，作为第三TOA测量值，具体包括：
    获取所述第二TOA测量值中均值最小的TOA测量值，并将所述均值最小的TOA测量值作为用户与基站i之间的第三TOA测量值：
    Figure PCTCN2020136564-appb-100030
    其中，z为各第二TOA测量值的均值μ 1,μ 2,…μ k,，中最小均值所对应的序号；μ z为高斯混合分布中，均值最小的高斯分布所对应的均值；j为遍历信道变化次数时的编号。
  22. 根据权利要求21所述的终端，其特征在于，所述对筛选后的TOA测量值进行非直射径NLOS鉴别，得到鉴别标签，具体包括：
    根据标准差NLOS鉴别方法或偏斜度与峭度NLOS鉴别方法对所述第三TOA测量值进行NLOS鉴别，得到第四TOA测量值和对应的鉴别标签；
    所述根据所述鉴别标签对所述筛选后的TOA测量值进行修正，具体包括：
    根据留一法或梯度下降法去除所述第四TOA测量值中NLOS误差最大的TOA测量值，得到第五TOA测量值；
    根据各鉴别标签对对应的第五TOA测量值进行修正处理，得到经过修正后的TOA测量值。
  23. 根据权利要求22所述的终端，其特征在于，所述根据所述鉴别标签对所述筛选后的TOA测量值进行修正之后，还包括：
    根据修正得到的第六TOA测量值、用户定位算法和预定义准则计算得到用户定位结果，并根据所述用户定位结果对UE进行定位；
    其中，所述预定义准则包括选取距离残差最小的用户位置。
  24. 根据权利要求17-19任一项或21-23任一项所述的终端，其特征在于，所述对各到达时间TOA测量值的概率密度进行混合高斯建模，并对混合高斯建模后的TOA测量值进行筛选之前，还包括：
    接收各基站发送的定位参考信号PRS配置信息，并根据各PRS配置信息对下行PRS信号进行测量，得到各基站到达UE的TOA测量值。
  25. 一种非暂态计算机可读存储介质，其上存储有计算机程序，其特征在于，该计算机程序被处理器执行时实现如权利要求1至8任一所述的到达时间测量值的非直射径误差消除方法。
PCT/CN2020/136564 2020-02-05 2020-12-15 到达时间测量值的非直射径消除方法、装置及终端 WO2021155714A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP20917676.7A EP4102900A4 (en) 2020-02-05 2020-12-15 INDIRECT OBSERVATION ELIMINATION METHOD AND APPARATUS FOR ARRIVAL TIME MEASUREMENT VALUE, AND TERMINAL
US17/793,923 US20230046671A1 (en) 2020-02-05 2020-12-15 Method and device for eliminating non-line of sight errors of time of arrival measurement values, and terminal

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010080737.4 2020-02-05
CN202010080737.4A CN113225667B (zh) 2020-02-05 2020-02-05 到达时间测量值的非直射径消除方法、装置及终端

Publications (1)

Publication Number Publication Date
WO2021155714A1 true WO2021155714A1 (zh) 2021-08-12

Family

ID=77085530

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/136564 WO2021155714A1 (zh) 2020-02-05 2020-12-15 到达时间测量值的非直射径消除方法、装置及终端

Country Status (4)

Country Link
US (1) US20230046671A1 (zh)
EP (1) EP4102900A4 (zh)
CN (1) CN113225667B (zh)
WO (1) WO2021155714A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115184859A (zh) * 2022-06-27 2022-10-14 南京森林警察学院 一种构建非视距传播场景下测距和测角误差消除方法

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116170871A (zh) * 2021-11-22 2023-05-26 维沃移动通信有限公司 定位方法、装置、终端及网络侧设备
CN117518076B (zh) * 2023-11-07 2024-04-19 中南大学 一种管道内机器人定位的非视距误差识别和缓解方法
CN117560767B (zh) * 2024-01-09 2024-04-09 上海银基信息安全技术股份有限公司 测距值nlos识别方法及装置、接收装置及存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101282577A (zh) * 2008-04-09 2008-10-08 山东大学 基于Ad-hoc网络技术的NLOS定位误差消除方法
CN102547827A (zh) * 2010-12-30 2012-07-04 展讯通信（上海）有限公司 非可视路径的鉴别方法及移动终端定位方法
CN106102161A (zh) * 2016-05-30 2016-11-09 天津大学 基于聚类算法分析数据优化的室内定位方法
US20160327628A1 (en) * 2015-05-07 2016-11-10 Alcatel-Lucent Usa Inc. Method And Apparatus For Determining Non-Line Of Sight Bias Estimation
CN108337197A (zh) * 2017-01-17 2018-07-27 中兴通讯股份有限公司 一种直射径提取方法及装置

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1235430C (zh) * 2002-11-07 2006-01-04 华为技术有限公司 一种位置估计方法
US7881407B2 (en) * 2004-04-30 2011-02-01 Sirf Technology, Inc. Systems and methods for mitigating multipath signals
CN101202585B (zh) * 2006-12-14 2011-09-28 电信科学技术研究院 无线通信基站、终端及其获取系统信息的方法
CN101466145B (zh) * 2009-01-04 2011-05-04 上海大学 基于神经网络的双基站精确定位方法
CN104093207B (zh) * 2014-07-28 2017-09-15 湘潭大学 一种基于测距误差模型修正的无线定位方法
TWI666947B (zh) * 2018-05-16 2019-07-21 Yuan Ze University 多用戶隨機存取訊號之分析方法
CN110596641B (zh) * 2019-09-29 2023-04-18 中国科学院测量与地球物理研究所 混合los/nlos场景下基于toa定位模式的l1范数方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101282577A (zh) * 2008-04-09 2008-10-08 山东大学 基于Ad-hoc网络技术的NLOS定位误差消除方法
CN102547827A (zh) * 2010-12-30 2012-07-04 展讯通信（上海）有限公司 非可视路径的鉴别方法及移动终端定位方法
US20160327628A1 (en) * 2015-05-07 2016-11-10 Alcatel-Lucent Usa Inc. Method And Apparatus For Determining Non-Line Of Sight Bias Estimation
CN106102161A (zh) * 2016-05-30 2016-11-09 天津大学 基于聚类算法分析数据优化的室内定位方法
CN108337197A (zh) * 2017-01-17 2018-07-27 中兴通讯股份有限公司 一种直射径提取方法及装置

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
ERICSSON: "Evaluations of Richer UE feedback for OTDOA", 3GPP TSG RAN WG1 MEETING #82, R1-154763, 30 August 2015 (2015-08-30), XP051021037 *
See also references of EP4102900A4 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115184859A (zh) * 2022-06-27 2022-10-14 南京森林警察学院 一种构建非视距传播场景下测距和测角误差消除方法

Also Published As

Publication number Publication date
CN113225667B (zh) 2022-03-29
EP4102900A4 (en) 2024-03-13
CN113225667A (zh) 2021-08-06
US20230046671A1 (en) 2023-02-16
EP4102900A1 (en) 2022-12-14

Similar Documents

Publication Publication Date Title
WO2021155714A1 (zh) 到达时间测量值的非直射径消除方法、装置及终端
CN108667856B (zh) 一种网络异常检测方法、装置、设备及存储介质
Wylie et al. The non-line of sight problem in mobile location estimation
US11762396B2 (en) Positioning system and positioning method based on WI-FI fingerprints
US7916814B2 (en) Method and apparatus for real-time pulse parameter estimator
CN110287537A (zh) 用于频标输出跳变检测的抗野值自适应卡尔曼滤波方法
Liguori et al. A preliminary study on the estimation of the uncertainty of traffic noise measurements
CN110494741B (zh) 用于从所制造组件识别的图案缺陷的系统性及随机性表征的系统、方法及计算机程序产品
CN107677290B (zh) 惯性导航系统精度评定的测试方法及装置
CN116719073B (zh) 面向gnss解算域的粗差探测与剔除方法
CN106597122A (zh) 一种雷达及通信信号的脉冲宽度检测算法
CN102484739B (zh) 用于确定采样率差的设备和方法
CN115277165A (zh) 一种车辆网络风险确定方法、装置、设备及存储介质
CN111640096B (zh) 电子产品外观的检测方法、装置及终端
CN112927201A (zh) 一种曲线检测方法及装置
CN113834782A (zh) 一种基于水质检测的设备运行方法及系统
CN113342616A (zh) 异常指标信息的定位方法、装置、电子设备及存储介质
JP2008258486A (ja) 分布解析方法および装置、異常設備推定方法および装置、上記分布解析方法または異常設備推定方法をコンピュータに実行させるためのプログラム、並びに上記プログラムを記録したコンピュータ読み取り可能な記録媒体
CN112763989B (zh) 一种基于cdif的抖动信号分选方法
CN117312635B (zh) 在线检测数据分析处理方法、系统、电子设备及存储介质
US20220374772A1 (en) Substrate treating apparatus and data change determination method
CN113312400A (zh) 一种浮法玻璃等级判定方法及装置
CN117146857A (zh) 一种基于载体位移的重力匹配结果校验方法
Kim Trends on NLOS Error Identification and Mitigation Methods for WiFi FTM-based Indoor Positioning
CN118068309A (zh) 一种超声波雷达测距精度与稳定性测试方法及系统

Legal Events

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

Ref document number: 20917676

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2020917676

Country of ref document: EP

Effective date: 20220905