Method, apparatus, device, and storage medium for terminal position estimation

Abstract

An embodiment of the present invention provides a method, device, equipment and storage medium for terminal position estimation. The method includes: determining the same number of candidate values at the next moment and updating the corresponding weight coefficients based on the first number of samples representing the overall state at the current moment and the state space model; determining the resampled first candidate values based on the first candidate values and the first weight coefficients, and using them as the first number of samples adopted when determining the first candidate values at the next-to-next moment; determining the estimated value representing the overall state at the next moment based on the resampled first candidate values, correcting the TDOA deviation between RRU, and determining the position information of the terminal. The method provided by the present invention accurately estimates the TDOA deviation between RRU at any moment through the constructed state space model, corrects the actual measurement values, reduces the cost of TDOA deviation correction, can synchronously determine the terminal position, has high implementation efficiency, and does not require manual maintenance.

Description

Technical Field

[0001] The present invention relates to the field of wireless communication technologies, and in particular, to a method, apparatus, device, and storage medium for estimating the position of a terminal. Background Art

[0002] Distributed 5G base stations (also known as extended 5G base stations) have been widely used in positioning systems, and multiple remote radio units (RRUs) included in the distributed 5G base stations are used to independently measure the arrival timestamps of positioning signals sent by a terminal. Then, differences are respectively calculated to obtain the propagation distance differences of the signals arriving at different RRUs, and the combination values of the distance differences between different RRUs are used to uniquely determine the position of the terminal.

[0003] In existing distributed 5G base station systems, there are time deviations between different RRU devices, resulting in an inherent deviation between the time difference of arrival (TDOA) data measured by the 5G base stations and the true TDOA. This inherent deviation remains unchanged during the normal operation of the distributed 5G base station system. However, if one or more devices in the distributed 5G base station system are restarted, then this inherent deviation generally changes after the restart. If the TDOA measured by the 5G base stations is not subjected to deviation elimination or suppression, significant terminal positioning errors will be introduced.

[0004] In related technologies, the global positioning system (GPS) or Beidou is used for clock synchronization between RRUs, and the obtained clock synchronization accuracy is about 10 - 50 ns, and the corresponding ranging error is 3 meters - 15 meters, which cannot meet the sub-meter positioning requirements. Or by deploying reference terminals with known positions near the base station, and then using the reference terminals to calculate the TDOA deviation between multiple RRUs and compensating it in the TDOA information measured by the 5G base stations. The corresponding clock synchronization accuracy is relatively high and can meet the sub-meter positioning requirements, but this solution has problems such as high deployment cost, high implementation complexity, and difficult later maintenance. Summary of the Invention

[0005] In view of the technical problems existing in the prior art, the present invention provides a method, apparatus, device, and storage medium for estimating the position of a terminal.

[0006] In a first aspect, the present invention provides a method for estimating the position of a terminal, including:

[0007] Based on the first number of samples used to represent the overall state at the current moment and a state space model, determine the first number of candidate samples at the next moment as the first candidate samples;

[0008] Update the weight coefficient corresponding to the first candidate sample based on the weight coefficients corresponding to the first number of samples at the current moment and the state space model, and use it as the first weight coefficient;

[0009] Based on the first candidate sample and the first weight coefficient, determine the resampled first candidate sample as the first number of samples at the current moment used to represent the overall state when determining the first candidate sample at the next next moment;

[0010] Based on the resampled first candidate sample, determine the estimated value used to represent the overall state at the next moment;

[0011] Based on the estimated value used to represent the overall state at the next moment, correct the time difference of arrival (TDOA) deviation between the first remote radio unit (RRU) and the reference remote radio unit (RRU), and determine the location information of the terminal;

[0012] The overall state includes the location information of the terminal and the states of multiple remote radio units (RRUs); the states of the multiple remote radio units (RRUs) include the TDOA deviation between the first remote radio unit (RRU) and the reference remote radio unit (RRU); the first remote radio unit (RRU) and the reference remote radio unit (RRU) are any one of the multiple remote radio units (RRUs), and the first remote radio unit (RRU) is different from the reference remote radio unit (RRU);

[0013] The state space model is used to represent the first correlation relationship between the overall states at adjacent moments and the second correlation relationship between the overall state at the next moment, the angle of arrival (AOA) information, and the TDOA deviation.

[0014] Optionally, the state space model is used to represent the first correlation relationship between the overall states at adjacent moments, and the method for determining the first correlation relationship includes:

[0015] Based on the multiple remote radio units (RRUs) using a one-dimensional linear array or a two-dimensional planar array, determine the dimension corresponding to the terminal state included in the overall state, where the terminal state includes the location information of the terminal in each dimension and the speed of the terminal in each dimension;

[0016] Based on the overall state, state transition matrix, noise input matrix, and state noise vector at the current moment, establish a correlation relationship with the overall state at the next moment.

[0017] Optionally, the state space model is used to represent a second correlation relationship between the overall state at the next moment, the angle of arrival (AOA) information, and the time difference of arrival (TDOA) deviation. The method for determining the second correlation relationship includes:

[0018] Based on the attitude angle and position information corresponding to each remote radio unit (RRU) among the multiple remote radio units (RRUs) at the next moment, the position information of the terminal at the next moment, and the AOA observation noise at the next moment, establish a correlation relationship with the AOA observation vector at the next moment;

[0019] Based on the position information of the terminal at the next moment, the position information of the first RRU, the position information of the reference RRU, the TDOA deviation corresponding to the first RRU and the reference RRU, and the TDOA observation noise corresponding to the first RRU and the reference RRU, establish a correlation relationship with the TDOA observation vector at the next moment;

[0020] The AOA observation vector at the next moment includes the AOA information measured by the multiple remote radio units (RRUs) at the next moment;

[0021] The TDOA observation vector at the next moment includes the TDOA observation values corresponding to multiple first remote radio units (RRUs) and the reference RRU.

[0022] Optionally, the method for obtaining the AOA observation vector at the next moment and the TDOA observation vector at the next moment includes:

[0023] Measure the AOA information corresponding to the reference terminal moving on the first motion trajectory at the next moment, and the TDOA information corresponding to the first RRU and the reference RRU, respectively, as the AOA observation vector at the next moment and the TDOA observation vector at the next moment;

[0024] The first motion trajectory is the trajectory of the reference terminal moving within the coverage area of the service area provided by the multiple remote radio units (RRUs).

[0025] Optionally, determining the first number of candidate samples at the next moment based on the first number of samples at the current moment and the state space model includes:

[0026] Determine the first number of candidate samples at the next moment based on the first number of samples at the current moment and the first association relationship.

[0027] Optionally, updating the weight coefficient corresponding to the first candidate sample based on the weight coefficient corresponding to the first number of samples at the current moment and the state space model includes:

[0028] Determine that the multiple remote radio units (RRUs) adopt a one-dimensional linear array or a two-dimensional planar array;

[0029] If the multiple remote radio units (RRUs) adopt a one-dimensional linear array, update the weight corresponding to the first candidate sample at the next moment based on the weight coefficient corresponding to the first number of samples at the current moment, the angle of arrival (AOA) information of the multiple remote radio units (RRUs) at the next moment, the covariance matrix of the time difference of arrival (TDOA) observation noise, and a preset constraint condition; the preset constraint condition is used to limit the uniqueness of the determined position information of the terminal;

[0030] If the multiple remote radio units (RRUs) adopt a two-dimensional planar array, update the weight corresponding to the first candidate sample at the next moment based on the weight coefficient corresponding to the first number of samples at the current moment, the angle of arrival (AOA) information of the multiple remote radio units (RRUs) at the next moment, and the covariance matrix of the time difference of arrival (TDOA) observation noise to obtain the first weight coefficient.

[0031] Optionally, updating the weight corresponding to the first candidate sample at the next moment based on the weight coefficient corresponding to the first number of samples at the current moment, the angle of arrival (AOA) information of the multiple remote radio units (RRUs) at the next moment, the covariance matrix of the time difference of arrival (TDOA) observation noise, and a preset constraint condition to obtain the first weight coefficient includes:

[0032] Based on the first number of samples at the current moment and the state space model, determine the position information of the terminal at the next moment;

[0033] If the position information of the terminal at the next moment meets the preset constraint condition, update the weight corresponding to the first candidate sample at the next moment based on the weight coefficient corresponding to the first number of samples at the current moment, the angle of arrival (AOA) information of the multiple remote radio units (RRUs) at the next moment, and the covariance matrix of the time difference of arrival (TDOA) observation noise to obtain the first weight coefficient.

[0034] In a second aspect, the present invention further provides a device for estimating the position of a terminal, including:

[0035] A first determination module, configured to determine, based on the first number of samples used to represent the overall state at the current moment and a state space model, the first number of candidate samples at the next moment as first candidate samples;

[0036] A second determination module, configured to update the weight coefficients corresponding to the first candidate samples based on the weight coefficients corresponding to the first number of samples at the current moment and the state space model as first weight coefficients;

[0037] A third determination module, configured to determine the resampled first candidate samples based on the first candidate samples and the first weight coefficients as the first number of samples used when determining the first candidate samples at the next next moment;

[0038] A fourth determination module, configured to determine an estimated value used to represent the overall state at the next moment based on the resampled first candidate samples;

[0039] A joint estimation module, configured to correct the time difference of arrival (TDOA) deviation between a first remote radio unit (RRU) and a reference remote radio unit (RRU) based on the estimated value used to represent the overall state at the next moment, and determine the location information of the terminal;

[0040] The overall state includes the location information of the terminal and the states of multiple remote radio units (RRUs); the states of the multiple remote radio units (RRUs) include the TDOA deviation between the first remote radio unit (RRU) and the reference remote radio unit (RRU); the first remote radio unit (RRU) and the reference remote radio unit (RRU) are any of the multiple remote radio units (RRUs), and the first remote radio unit (RRU) is different from the reference remote radio unit (RRU);

[0041] The state space model is used to represent a first correlation relationship between the overall states at adjacent moments and a second correlation relationship between the overall state at the next moment, the angle of arrival (AOA) information, and the TDOA deviation.

[0042] In a third aspect, the present invention further provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, where when the processor executes the program, the method for estimating the terminal location as described in the first aspect above is implemented.

[0043] In a fourth aspect, the present invention further provides a non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the method for estimating the terminal location as described in the first aspect above is implemented.

[0044] In a fifth aspect, the present invention further provides a computer program product, including a computer program which, when executed by a processor, implements the method for estimating the terminal position as described in the first aspect above.

[0045] The method, apparatus, device, and storage medium for estimating the terminal position provided by the present invention accurately estimate the TDOA deviation between a first RRU and a reference RRU at any moment through the first correlation relationship between the overall states at adjacent moments in the constructed state space model, and the second correlation relationship between the overall state at the next moment, the angle of arrival (AOA) information, and the time difference of arrival (TDOA) deviation. Based on this estimated value, the measured value of the TDOA deviation between the first RRU and the reference RRU is corrected, avoiding the high cost of deploying reference terminals and the high difficulty of maintenance, reducing the cost of TDOA deviation correction, and simultaneously enabling the estimation of the terminal position. The implementation process is simple, efficient, and does not require manual maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0047] Figure 1 is a flowchart of the method for estimating the terminal position provided by an embodiment of the present invention;

[0048] Figure 2 is a structural diagram of a distributed 5G base station provided by an embodiment of the present invention;

[0049] Figure 3 is one of the implementation diagrams of the method for estimating the terminal position provided by an embodiment of the present invention;

[0050] Figure 4 is another implementation diagram of the method for estimating the terminal position provided by an embodiment of the present invention;

[0051] Figure 5 is a comparison diagram of the positioning errors corresponding to two methods provided by an embodiment of the present invention;

[0052] Figure 6 is a structural diagram of the apparatus for estimating the terminal position provided by an embodiment of the present invention;

[0053] Figure 7 is a schematic physical structure diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0054] In the embodiments of the present invention, the term "and / or" describes the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone. The character " / " generally indicates that the associated objects before and after are in an "or" relationship.

[0055] In the embodiments of the present invention, the term "plurality" refers to two or more, and other quantifiers are similar.

[0056] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0057] Figure 1 is a schematic flowchart of a method for estimating the terminal position provided by an embodiment of the present invention. As Figure 1 shown, the method includes:

[0058] Step 101: Based on the first number of samples used to represent the overall state at the current moment and the state space model, determine the first number of candidate samples at the next moment as the first candidate samples; both the first number of samples and the first number of candidate samples are used to represent the overall state; the overall state includes the position information of the terminal and the states of multiple RRUs; the states of the multiple RRUs include the time difference of arrival (TDOA) deviation between the first RRU and the reference RRU; the first RRU and the reference RRU are any one of the multiple RRUs, and the first RRU is different from the reference RRU;

[0059] The state space model is used to represent the first association relationship between the overall states at adjacent moments and the second association relationship between the overall state at the current moment and the angle of arrival (AOA) information at the current moment.

[0060] Specifically, a distributed 5G base station generally includes a baseband processing unit (Building Base band Unit, BBU), multiple extension units (HUB), and multiple remote radio units (RRU), as Figure 2As shown in the figure, an RRU generally includes an Active Antenna Unit (AAU), which is responsible for transmitting and receiving 5G radio frequency signals. The extension unit HUB is responsible for combining the radio frequency signals of multiple RRUs into a single signal after appropriate processing and sending it to the Baseband Processing Unit (BBU). The Baseband Processing Unit (BBU) is responsible for processing 5G baseband signals and related protocol interaction processes. The Baseband Processing Unit (BBU) is connected to the extension unit HUB, and the extension unit HUB is connected to the Remote Radio Unit (RRU) through the Common Public Radio Interface (CPRI) and the enhanced Common Public Radio Interface (eCPRI).

[0061] The research on measured data shows that in an actual distributed 5G base station system, there is an inherent deviation between the TDOA data measured by the 5G base station and the true TDOA. It can be understood that the synchronization deviation between multiple RRUs of the base station can remain stable for a long time after power-on, that is, within a certain period of time, it can be considered that there is a fixed deviation value in the TDOA values between multiple RRUs of the distributed 5G base station. This inherent deviation remains unchanged during the normal operation of the distributed 5G base station system. However, if one or more RRU devices in the distributed 5G base station system are restarted, then this inherent deviation generally changes after the restart. If the TDOA deviation measured by the 5G base station is not eliminated or suppressed, significant terminal positioning errors will be introduced. During the process of eliminating or suppressing the TDOA deviation, different signals measured need to be processed. Due to the influence of factors such as multipath when measuring different signals, there are different degrees of measurement deviations in the Angle of Arrival (AOA) of the signals measured by the RRU at different test points. If the terminal is in a stationary state, this measurement deviation will significantly affect the estimation accuracy of the TDOA deviation and the positioning accuracy of the terminal. The traditional deviation elimination method is to set several reference points with known positions in the deployment area of the positioning system, place reference terminals for measurement, and then calibrate the TDOA deviation, which has relatively high time and labor costs. Therefore, the present invention uses a terminal in a moving state to eliminate the influence of this measurement deviation on the estimation accuracy of the TDOA deviation and the positioning accuracy through the AOA information of all points on the movement trajectory of the terminal within a certain period of time. Here, the points can be understood as the positions where the terminal is located.

[0062] When constructing the state space model provided in the present invention, the position information of the terminal at different times is obtained, and each RRU in the distributed 5G base station estimates the time of arrival (TOA) and the angle of arrival (AOA). That is, after each RRU receives the uplink reference signal of the terminal, it processes the signal and sends the processed signal to the BBU. The BBU processes the signal sent by the RRU to obtain the channel state information and sends it to the positioning server. The positioning server uses the channel state information to estimate the TOA and AOA of each RRU respectively, and then calculates the TDOA between the uplink reference signal arriving at different RRUs. However, there may be a deviation between the measured TDOA value between different RRUs and the actual TDOA value. The first correlation relationship between the overall states of multiple RRUs included in the distributed 5G base station at adjacent times, and the second correlation information between the AOA information determined according to the position information of the terminal and the position information of multiple RRUs and the overall state are established, that is, the state space model is constructed.

[0063] After constructing the above state space model, obtain the first number of samples used to represent the overall state at the current moment, and the above state space model, and determine the first number of candidate samples at the next moment as the first candidate samples, where the first number is a positive integer; each of the samples and the first candidate samples is used to represent the overall state, and the overall state includes the states of multiple RRUs, and the states of the multiple RRUs include the TDOA deviation between the first RRU and the reference RRU. Here, both the first RRU and the reference RRU are any one of the above multiple RRUs, and the first RRU is different from the reference RRU. The TDOA deviation between the first RRU and the reference RRU represents the difference between the measured TDOA value of the first RRU and the reference RRU and the actual TDOA value of the first RRU and the reference RRU. The actual TDOA value of the first RRU and the reference RRU can be understood as the fixed value of the TDOA between the first RRU and the reference RRU when the base station is in a stable working state. Because in the case of no unexpected power failure, the synchronization deviation between multiple RRUs in the base station can remain stable for a long time after power-on. In the following embodiments, in order to illustrate the TDOA deviation between the first RRU and the reference RRU, the reference RRU is mainly taken as the first RRU for illustration, but it is not limited that the reference RRU is only the first RRU, and it can be any one of multiple RRUs, as long as it is different from the first RRU.

[0064] Step 102: Update the weight coefficient corresponding to the first candidate sample based on the weight coefficient corresponding to the first number of samples at the current moment and the state space model, as the first weight coefficient;

[0065] After determining the first candidate sample for representing the overall state at the next moment, it is also necessary to update the weight corresponding to the first candidate sample. Specifically, the weight coefficient corresponding to the first number of samples at the current moment and the state space model can be used to update the weight corresponding to the first candidate sample at the next moment, so as to obtain the weight coefficient corresponding to each first candidate sample, that is, the first weight coefficient corresponding to the first candidate sample.

[0066] Step 103: Based on the first candidate sample and the first weight coefficient, determine the resampled first candidate sample as the first number of samples used when determining the first candidate sample at the next-to-next moment;

[0067] After determining the first candidate sample and the weight coefficient corresponding to each first candidate sample, resample these first candidate samples according to the weight from high to low. For example, the first candidate sample is represented as The first weight coefficient is represented as Assume the magnitude relationship of the first weight coefficients is Then resample the first candidate samples according to the weight from high to low first to obtain the resampled first candidate samples for representing the overall state at the next moment. The corresponding result includes Λ values, which can be specifically represented as Or Or And use these resampled first candidate samples as the first number of samples for representing the overall state at the current moment in the next iteration, that is, The corresponding weight is The above is only for illustrative purposes and does not specifically limit the possible representation methods of the estimated value for representing the overall state at the next moment. Use the above resampled first candidate samples as the first number of samples used when determining the first candidate sample at the next-to-next moment.

[0068] Step 104: Based on the resampled first candidate sample, determine the estimated value for representing the overall state at the next moment;

[0069] After determining the resampled first candidate sample and the corresponding first weight coefficient, the estimated value for representing the overall state at the next moment can be determined according to the method of weighted summation or the method of taking the average.

[0070] Step 105: Based on the estimated value for representing the overall state at the next moment, correct the time difference of arrival (TDOA) deviation between the first remote radio unit (RRU) and the reference remote radio unit (RRU), and determine the position information of the terminal.

[0071] After determining the estimated value for representing the overall state at the next moment through the above steps, the TDOA deviation between the first RRU and the reference RRU among the multiple RRUs represented by this estimated value can be used to correct the existing TDOA deviation between the first RRU and the reference RRU. In addition, the estimated value also includes the position information of the terminal, that is, the position information of the terminal can be estimated based on this estimated value.

[0072] The method for estimating the terminal position provided by the present invention accurately estimates the TDOA deviation between the first RRU and the reference RRU at any moment through the first correlation relationship between the overall states at adjacent moments in the constructed state space model, and the second correlation relationship between the overall state at the next moment, the angle of arrival AOA information, and the time difference of arrival TDOA deviation. And based on this estimated value, the measured value of the TDOA deviation between the first RRU and the reference RRU is corrected, avoiding the high cost of deploying reference terminals and the high difficulty of maintenance, reducing the cost of TDOA deviation correction, and can simultaneously estimate the position of the terminal. The implementation process is simple, efficient, and does not require manual maintenance.

[0073] Optionally, the state space model is used to represent the first correlation relationship between the overall states at adjacent moments, and the determination method of the first correlation relationship includes:

[0074] Based on the multiple remote radio units RRU using a one-dimensional linear array or a two-dimensional planar array, determine the dimension corresponding to the terminal state included in the overall state. The terminal state includes the position information of the terminal in each dimension and the speed of the terminal in each dimension.

[0075] Based on the overall state at the current moment, the state transition matrix, the noise input matrix, and the state noise vector, establish a correlation relationship with the overall state at the next moment.

[0076] Specifically, the determination method corresponding to the first correlation relationship between the overall states at adjacent moments in the constructed state space model includes:

[0077] Determine that the multiple RRUs use a one-dimensional linear array or a two-dimensional planar array. Both the linear array and the planar array refer to array antennas. The active antenna unit AAU is a component of the RRU and is used to transmit and receive radio signals. Determining that the RRU uses a one-dimensional linear array or a two-dimensional planar array is to determine whether the AAU in the RRU is a one-dimensional linear array or a two-dimensional planar array. Among them, the one-dimensional linear array can be understood as being able to measure the angle of arrival AOA of signals in a two-dimensional space, that is, one-dimensional AOA information, and the two-dimensional planar array can be understood as being able to measure the angle of arrival AOA of signals in a three-dimensional space, that is, two-dimensional AOA information. Among them, the one-dimensional AOA information mainly includes the incident angle, and the two-dimensional AOA information mainly includes the azimuth angle and the elevation angle.

[0078] Determining that multiple RRU adopt a one-dimensional linear array means determining that the terminal state is two-dimensional information, specifically including that the position information of the terminal is two-dimensional and the speed of the terminal is also two-dimensional.

[0079] Determining that multiple RRU adopt a two-dimensional planar array means determining that the terminal state is three-dimensional information, specifically including that the position information of the terminal is three-dimensional and the speed of the terminal is also three-dimensional.

[0080] Based on the dimension corresponding to the terminal state and the one-dimensional linear array or two-dimensional planar array adopted by the multiple remote radio units (RRU), respectively determine the state transition matrix and the noise input matrix.

[0081] Based on the overall state, state transition matrix, noise input matrix, and state noise vector at the current moment, establish the correlation relationship with the overall state at the next moment.

[0082] The time difference between adjacent moments is the tracking sampling interval. According to the tracking sampling interval and the speed information of the terminal, the position change of the terminal within the tracking sampling interval can be determined. Then, according to the position change and the position information of the terminal at the previous moment, the candidate position information of the terminal at the next moment can be determined. In addition, in combination with the TDOA deviation between the first RRU and the reference RRU among the multiple RRU, the original state transition matrix that only represents the change of the terminal state is extended, and the state transition elements corresponding to the TDOA deviation are added to obtain the state transition matrix F in the present invention. k-1 In order to make the finally estimated terminal position and TDOA deviation more accurate, external noise needs to be considered. Therefore, a noise input matrix and a state noise vector are introduced. The noise input matrix and the state noise vector reflect the external noise interference that may exist in different terminal states and different TDOA deviations between the first RRU and the reference RRU.

[0083] Therefore, according to the overall state, state transition matrix, noise input matrix, and state noise vector at the current moment, establish the correlation relationship with the overall state at the next moment, that is, establish the state equation for TDOA deviation calibration, which can be specifically expressed by the formula: x k = F k-1 x k-1 + G k-1 w k-1 ;

[0084] Among them, x k represents the overall state at the kth moment, x k-1denotes the overall state at time k - 1, which includes the state of the terminal and the states of multiple RRUs. The state of the terminal includes the position and velocity of the terminal. The states of multiple RRUs include the TDOA deviation between the first RRU and the reference RRU. The first RRU and the reference RRU are any one of the multiple RRUs mentioned above, and the first RRU is different from the reference RRU; F k-1 is the state transition matrix corresponding to the overall state at time k - 1; G k-1 is the noise input matrix at time k - 1, w k-1 is the state noise vector at time k - 1.

[0085] According to whether multiple RRUs adopt a one - dimensional linear array or a two - dimensional planar array, the dimensions corresponding to the state of the terminal are different. When multiple RRUs adopt a one - dimensional linear array, that is, when the antennas assembled by the multiple RRUs belong to a one - dimensional linear array, the dimension corresponding to the state of the terminal is two - dimensional, and the rectangular coordinate system established with the terminal and multiple RRUs is also two - dimensional. Then the overall state at time k can be specifically expressed as and denotes the two - dimensional position coordinates of the phase center of the terminal antenna in the local rectangular coordinate system l at time k, and respectively denote the velocity components of the phase center of the terminal antenna on the x - axis and y - axis in the local rectangular coordinate system l at time k, b i,k denotes the TDOA deviation between the (i + 1) - th RRU and the reference RRU, that is, the 1 - st RRU at time k.

[0086] Correspondingly, when multiple RRUs adopt a one - dimensional linear array, F k-1 is the state transition matrix corresponding to the overall state at time k - 1; G k-1 is the noise input matrix at time k - 1, and can be expressed by the formulas respectively as:

[0087]

[0088] where T represents the tracking sampling interval, that is, the time difference between adjacent times.

[0089] When multiple RRUs adopt a two - dimensional planar array, that is, when the antennas assembled by the multiple RRUs belong to a two - dimensional planar array, the dimension corresponding to the state of the terminal is three - dimensional, and the rectangular coordinate system established with the terminal and multiple RRUs is also three - dimensional. Then the overall state at time k can be specifically expressed as and denotes the three - dimensional position coordinates of the phase center of the terminal antenna in the local rectangular coordinate system l at time k, and respectively denote the velocity components of the phase center of the terminal antenna on the x - axis, y - axis and z - axis in the local rectangular coordinate system l at time k, bi,k The TDOA deviation between the (i + 1)-th RRU and the reference RRU (i.e., the 1st RRU) at time k, and the (i + 1)-th RRU is different from the 1st RRU.

[0090] Correspondingly, when a two-dimensional planar array is adopted for multiple RRUs, F k-1 is the state transition matrix corresponding to the overall state at time k - 1; G k-1 is the noise input matrix at time k - 1, which can be respectively expressed by the following formulas:

[0091]

[0092] By the above method, the first correlation relationship between the overall states representing adjacent moments in the state space model is established, and the overall state at the next moment can be inferred based on the overall state at the current moment.

[0093] Optionally, the state space model is used to represent the second correlation relationship between the overall state at the next moment, the angle of arrival (AOA) information, and the time difference of arrival (TDOA) deviation. The determination method of the second correlation relationship includes:

[0094] Based on the attitude angle and position information corresponding to each remote radio unit (RRU) among the multiple RRUs at the next moment, the position information of the terminal at the next moment, and the AOA observation noise at the next moment, establish the correlation relationship with the AOA observation vector at the next moment;

[0095] Based on the position information of the terminal at the next moment, the position information of the first RRU, the position information of the reference RRU, the TDOA deviation corresponding to the first RRU and the reference RRU, and the TDOA observation noise corresponding to the first RRU and the reference RRU, establish the correlation relationship with the TDOA observation vector at the next moment;

[0096] The AOA observation vector at the next moment includes the AOA information measured by the multiple RRUs at the next moment;

[0097] The TDOA observation vector at the next moment includes the TDOA observation values corresponding to the multiple first RRUs and the reference RRU.

[0098] Specifically, each of the above-mentioned multiple RRUs also has corresponding status information, specifically including the attitude angle of the RRU in the local rectangular coordinate system and the position information of the RRU. The attitude angle of the RRU in the local rectangular coordinate system includes the azimuth angle, the elevation angle, and the roll angle. For example, the azimuth angle, the elevation angle, and the roll angle of the i-th RRU antenna in the local rectangular coordinate system can be respectively expressed as ψ i , θ i , and γ i . The position information of the RRU can be represented in the form of three-dimensional position coordinates. For example, the position coordinates of the i-th RRU antenna in the local rectangular coordinate system l are and . Among them, represents the x-axis coordinate of the i-th RRU in the current rectangular coordinate system l, represents the y-axis coordinate of the i-th RRU in the current rectangular coordinate system l, represents the height coordinate (z-axis coordinate) of the i-th RRU in the current rectangular coordinate system l.

[0099] The second correlation relationship between the overall state, the angle of arrival AOA information, and the time difference of arrival TDOA deviation can be expressed as: ξ k = ψ(x k ) + υ k ;

[0100] Among them, ξ k represents the angle of arrival AOA observation vector and the time difference of arrival TDOA observation vector of the above-mentioned multiple RRUs at time k, and ψ(x k ) represents the result after processing the overall state x k through a non-linear function; υ k represents the total observation noise vector at time k.

[0101] Among them, the angle of arrival AOA information depends on whether the above-mentioned multiple RRUs use a one-dimensional linear array or a two-dimensional planar array, and the dimension of this angle of arrival AOA information is also different. For example, if the above-mentioned multiple RRUs use a one-dimensional linear array, the angle of arrival AOA information is one-dimensional AOA information, and the position information of the corresponding terminal is two-dimensional, and can be specifically expressed as:

[0102]

[0103] ψ(x k ) = [h1(x k )…h N (x k ) g1(x k )…g N-1 (x k )] T ;

[0104]

[0105]

[0106] α i = cosψ i cosγ i - sinψ i sinθ i sinγ i ;

[0107] β i = sinψ i cosγ i + cosψ i sinθ i sinγ i ; κi i = - cosθ i sinγ i ;

[0108] υ k = [v 1,k …v N,k τ 1,k …τ N-1,k T ;

[0109] where ξ k represents the arrival angle of arrival (AOA) observation vector and the time difference of arrival (TDOA) observation vector of the multiple RRU at time k, represents the incident angle of the phase center of the terminal T antenna in the antenna coordinate system a of the i-th RRU at time k, that is, the AOA information of the terminal measured by the i-th RRU at time k, represents the TDOA information between the phase center of the terminal T antenna and the (i + 1)-th RRU antenna and between the phase center of the terminal T antenna and the 1st RRU antenna. The (i + 1)-th RRU and the 1st RRU are different.

[0110] h i (x k ) represents the incident angle in the one-dimensional AOA information of the terminal at time k, and also represents the calculated value of the incident angle of the phase center of the terminal antenna in the antenna coordinate system a of the i-th RRU at time k, which is determined based on the position information of the terminal, the position information of the RRU, and the attitude angle of the RRU. Among them, the position information of the terminal also includes the third-dimensional position information represents the height coordinate of the phase center of the terminal T antenna in the local rectangular coordinate system l, which is preset as a known fixed value. α i , β i and κ i ​Indicates the correlation coefficient, determined based on the attitude angle of the i-th RRU.

[0111] g i (x k ) is the calculated value of the TDOA information between the antenna phase center of the terminal T at time k and the (i + 1)-th RRU antenna, and between the antenna phase center of the terminal T and the 1st RRU antenna. It is determined based on the distance between the position of the terminal in the local rectangular coordinate system and the position of the (i + 1)-th RRU in the local rectangular coordinate system, the difference in distance between the position of the terminal in the local rectangular coordinate system and the position of the reference RRU (the 1st RRU) in the local rectangular coordinate system, and the TDOA deviation between the (i + 1)-th RRU and the reference RRU (the 1st RRU).

[0112] υ k Is the total observation noise vector at time k, where v i,k Is the observation noise of the incident angle (one-dimensional AOA information) of the i-th RRU at time k, and τ i,k Is the observation noise of the TDOA between the (i + 1)-th RRU and the 1st RRU at time k.

[0113] If the above multiple RRUs use a two-dimensional planar array, then the angle of arrival AOA information is two-dimensional AOA information, and the corresponding position information of the terminal is three-dimensional, which can be specifically expressed as:

[0114]

[0115] ψ(x k ) = (f1(x k )…f N (x k ) h1(x k )…h N (x k ) g1(x k )…g N-1 (x k )] T ;

[0116]

[0117]

[0118]

[0119] υ k = [ε 1,k …ε N,k v 1,k …v N,k τ 1,k …τ N-1,k T ;​

[0120] Among them, ξ k represents the arrival angle of arrival (AOA) observation vector and time difference of arrival (TDOA) observation vector of the above-mentioned multiple remote radio units (RRUs) at time k, and represents the two-dimensional AOA information of the terminal measured by the i-th RRU at time k, and the two-dimensional AOA information includes azimuth angle and elevation angle; represents the azimuth angle of the phase center of the terminal T antenna in the antenna coordinate system a of the i-th RRU at time k; represents the elevation angle of the phase center of the terminal antenna in the antenna coordinate system a of the i-th RRU at time k; represents the TDOA information between the phase center of the terminal T antenna and the (i + 1)-th RRU antenna and between the phase center of the terminal T antenna and the 1st RRU antenna at time k; and the position coordinates of the terminal in the local rectangular coordinate system l, and are the position coordinates of the phase center of the i-th RRU antenna in the local rectangular coordinate system l, ψ i , θ i and γ i are the attitude angles of the i-th RRU antenna in the local rectangular coordinate system, which are azimuth angle, elevation angle and roll angle respectively.

[0121] f i (x k ) and h i (x k ) represent the azimuth angle and elevation angle in the two-dimensional AOA information of the terminal respectively, and are determined according to the position coordinates of the terminal in the local rectangular coordinate system, the position coordinates of the phase center of the i-th RRU antenna in the local rectangular coordinate system, and the attitude angle of the i-th RRU antenna in the local rectangular coordinate system.

[0122] g i (x k ) is the calculated value of the TDOA information between the phase center of the terminal T antenna and the (i + 1)-th RRU antenna and between the phase center of the terminal T antenna and the 1st RRU antenna at time k, and is based on the distance between the position of the terminal in the local rectangular coordinate system and the position of the (i + 1)-th RRU in the local rectangular coordinate system, the distance difference between the position of the terminal in the local rectangular coordinate system and the position of the reference RRU (the 1st RRU) in the local rectangular coordinate system, and the TDOA deviation between the (i + 1)-th RRU and the reference RRU (the 1st RRU).

[0123] υ k is the total observation noise vector at time k, ε i,k and v i,kis the observation noise of the two-dimensional AOA information, ε i,k is the azimuth observation noise of the i-th RRU at time k, v i,k is the elevation observation noise of the i-th RRU at time k, τ i,k is the TDOA observation noise between the (i + 1)-th RRU and the reference RRU (the 1st RRU) at time k.

[0124] By establishing a second correlation relationship between the overall state, the angle of arrival AOA information, and the time difference of arrival TDOA deviation for representing the next moment in the state space model, the overall state (the first candidate sample) at the next moment inferred according to the first correlation relationship is input, and the calculated value of the corresponding AOA information and the calculated value of the TDOA are obtained. Then, using the AOA information and the TDOA information actually measured by the reference terminal under the motion state at this moment, the deviation between the two is determined, preparing for the next correction of the first candidate sample.

[0125] The above-mentioned multiple RRUs can be a one-dimensional linear array or a two-dimensional planar array. Based on the above description, the state space model proposed by the present invention can be expressed as:

[0126]

[0127] Among them, the first expression of the state space model represents the first correlation relationship between the overall states at adjacent moments, and the second expression represents the second correlation relationship between the overall state at the current moment and the angle of arrival AOA information at the current moment.

[0128] Optionally, the method for obtaining the angle of arrival AOA observation vector and the time difference of arrival TDOA observation vector at the next moment includes:

[0129] Measure the angle of arrival AOA information corresponding to the reference terminal moving on the first motion trajectory at the next moment and the time difference of arrival TDOA information corresponding to the first remote radio unit RRU and the reference remote radio unit RRU, and respectively use them as the angle of arrival AOA observation vector and the time difference of arrival TDOA observation vector at the next moment;

[0130] The first motion trajectory is the trajectory of the reference terminal moving within the coverage area that can be served by the multiple remote radio units RRU.

[0131] Specifically, the angle of arrival (AOA) observation vector at the next moment and the time difference of arrival (TDOA) observation vector at the next moment can be obtained by holding a reference terminal or placing the reference terminal on a moving carrier near each of the above-mentioned multiple remote radio units (RRUs). The above-mentioned multiple RRUs are RRUs that can provide services to the terminal. It can be understood that the movement range of the terminal is within the coverage range of the service areas provided by the above-mentioned respective RRUs. The above-mentioned multiple RRUs will measure the AOA information and TDOA values corresponding to different moments during the movement of the terminal, and then determine the TDOA deviation value corresponding to the first remote radio unit (RRU) and the reference remote radio unit (RRU).

[0132] The method for estimating the terminal position provided by the present invention does not require setting a number of reference points with known positions in the deployment area of the positioning system, placing reference terminals for measurement and then calibrating the TDOA deviation. Instead, it uses the terminal in a moving state to obtain the AOA information and TDOA information measured at different moments when it moves in the coverage area of the RRU, and uses this information to correct the estimated position information of the terminal. Moreover, the AOA information obtained by the terminal in a moving state can eliminate the influence of measurement deviation on the TDOA deviation estimation and positioning accuracy.

[0133] Optionally, determining the first number of candidate samples at the next moment based on the first number of samples at the current moment and the state space model includes:

[0134] Determining the first number of candidate samples at the next moment based on the first number of samples at the current moment and the first correlation relationship.

[0135] Specifically, according to the first correlation relationship represented by the above state space model, the corresponding prior probability density function is determined. And based on this prior probability density function, the first number of particles at the initial moment and the initial weights corresponding to each particle are extracted to form the first set of samples, which can be specifically expressed as The initial particle weight is According to the first number of samples and the first correlation relationship, importance sampling is performed to obtain the first number of candidate samples used to represent the overall state at the next moment, and the corresponding formula can be expressed as:

[0136] Wherein, represents the first number of samples used to represent the overall state at the current moment k - 1, F k-1 represents the state transition matrix at the current moment k - 1, G k-1 represents the noise input matrix at the current moment k - 1, w k-1 represents the state noise vector at the current moment k - 1, Represent the first number of candidate samples for representing the overall state at the next moment k.

[0137] Through the state space model provided by the present invention, for RRU using a one-dimensional linear array or a two-dimensional planar array, the correlation relationship between the overall states at different moments can be established through a unified model, and the overall state at the next moment can be inferred using the overall state at the current moment, where the overall state includes the position, speed, and TDOA deviation of the terminal.

[0138] Optionally, updating the weight coefficient corresponding to the first candidate sample based on the weight coefficients corresponding to the first number of samples at the current moment and the state space model includes:

[0139] Determine that the multiple remote radio units RRU use a one-dimensional linear array or a two-dimensional planar array;

[0140] If the multiple remote radio units RRU use a one-dimensional linear array, then based on the weight coefficients corresponding to the first number of samples at the current moment, the angle of arrival AOA information of the multiple remote radio units RRU at the next moment, the covariance matrix of the time difference of arrival TDOA observation noise, and a preset constraint condition, update the weight corresponding to the first candidate sample at the next moment to obtain the first weight coefficient; the preset constraint condition is used to limit the uniqueness of the determined position information of the terminal;

[0141] If the multiple remote radio units RRU use a two-dimensional planar array, then based on the weight coefficients corresponding to the first number of samples at the current moment, the angle of arrival AOA information of the multiple remote radio units RRU at the next moment, and the covariance matrix of the time difference of arrival TDOA observation noise, update the weight corresponding to the first candidate sample at the next moment to obtain the first weight coefficient.

[0142] Specifically, when the above-mentioned multiple RRU use a one-dimensional linear array, that is, when the antennas assembled by the above-mentioned multiple RRU belong to a one-dimensional linear array, the determined terminal position may not be unique through the state space model. Therefore, it is necessary to determine whether the preset constraint condition is satisfied. The preset constraint condition can be expressed as:

[0143]

[0144] Where, Ψ i 、θ i and γ i represent the attitude angles of the i-th RRU antenna in the local rectangular coordinate system, which are the azimuth angle, elevation angle, and roll angle respectively; and represent the position coordinates of the terminal antenna phase center in the local rectangular coordinate system l at time k; and is the position coordinate of the phase center of the i-th RRU antenna in the local rectangular coordinate system l.

[0145] According to the second correlation relationship in the above state space model, determine the AOA information of the above multiple RRUs at the next moment k and the covariance matrix R of the TDOA observation noise k , and determine the likelihood function corresponding to any candidate sample according to the following formula Specifically expressed as:

[0146]

[0147] where is the likelihood function of the first candidate sample at the next moment ; R k represents the covariance matrix of the AOA information of the above multiple RRUs at the next moment k and the TDOA observation noise; represents the result after processing the first candidate sample at the next moment through a nonlinear function; ξ k represents the one-dimensional AOA and TDOA observation vectors of multiple RRUs at the next moment k.

[0148] Then, using this likelihood function and the weights corresponding to the first number of samples at the current moment k - 1, update the weights of the candidate samples at the next moment and perform normalization processing to obtain the weights corresponding to the first number of candidate samples at the next moment, that is, the first weight coefficient, which is specifically divided into two cases:

[0149] (1) When the above multiple RRUs adopt a one-dimensional linear array, that is, when it is determined that the antennas assembled by the above multiple RRUs belong to a one-dimensional linear array, the formula for determining the first weight coefficient can be specifically expressed as:

[0150]

[0151] where is the first weight coefficient corresponding to the first candidate sample at the next moment k, and the first candidate sample is used to represent the overall state, is an intermediate variable, represents the weights corresponding to the first number of sample data used to represent the overall state at the current moment k - 1.

[0152] (2) When the above multiple RRUs adopt a two-dimensional planar array, that is, when the antennas assembled by the above multiple RRUs belong to a two-dimensional planar array, the formula for determining the first weight coefficient can be specifically expressed as:

[0153]

[0154] where For the first weight coefficient corresponding to the first candidate sample at the next moment k, the first candidate sample is used to represent the overall state. is an intermediate variable. It represents the weights corresponding to the first number of sample data used to represent the overall state at the current moment k - 1.

[0155] That is, in the case where the RRU adopts a one-dimensional linear array, the judgment process of the preset constraint conditions is added. In the case where the RRU adopts a two-dimensional planar array, it is not necessary to judge the preset constraint conditions.

[0156] After updating the weight coefficient of the first candidate sample in the above manner, after obtaining the first weight coefficient, then according to the first candidate sample and the first weight coefficient corresponding to this first candidate sample resample the first candidate samples in the order of the first weight coefficient from high to low, obtain the first number of samples corresponding to the next moment, and take their average value to obtain the state estimation value at the next moment, which is expressed as:

[0157] For example, the first candidate sample the first weight coefficient and then resample the first candidate samples in the order of the first weight coefficient from high to low, and the resampling result is {c, c, c, c,..., Λ, Λ, Λ,..., b, b,..., 1, a} or {c, c, c, c,..., Λ, Λ, Λ,..., b, b,..., 1} or {c, c, c, c,..., v, Λ, Λ,..., b, b,..., 1} or {c, c, c, c,..., Λ, v, Λ,..., b}, where the total number of samples in any resampling result is the first number Λ. Through the above resampling, the resampled samples better conform to the actual distribution, so that the finally determined state estimation value at the next moment k is more accurate.

[0158] The method for estimating the terminal position provided by the present invention, through the first correlation relationship between the overall states at adjacent moments in the constructed state space model, and the second correlation relationship between the overall state at the next moment, the angle of arrival AOA information, and the time difference of arrival TDOA deviation, accurately estimates the TDOA deviation between any moment's first RRU and the reference RRU, and corrects the measured value of the TDOA deviation between the first RRU and the reference RRU based on this estimated value, avoiding the high cost of deploying reference terminals and the high difficulty of maintenance, reducing the cost of realizing TDOA deviation correction, and can synchronously realize the estimation of the terminal position, with a simple implementation process, high efficiency, and no need for manual maintenance.

[0159] To more clearly illustrate the method for estimating the terminal position provided by the embodiments of the present invention, the following uses a specific example for illustration.

[0160] Figure 3 It is one of the implementation schematic diagrams of the method for estimating the terminal position provided by the embodiments of the present invention. As Figure 3 shown, it includes:

[0161] Multiple RRUs of a distributed 5G base station adopt a one-dimensional linear array. This RRU can measure the one-dimensional AOA information of the terminal uplink reference signal and perform terminal positioning based on this one-dimensional AOA information. However, in actual RRU deployment, the RRU position is usually higher than the position of the 5G terminal, that is, the two are not in the same plane. The present invention models the one-dimensional AOA information measured by the RRU through the Incidence Angle Observation Model (IAOM), so as to obtain a more accurate TDOA deviation and terminal position estimate.

[0162] (1) Establish a state equation for TDOA deviation calibration, that is, establish a first correlation relationship represented by a state space model. This first correlation relationship represents the correlation relationship between the overall states at adjacent times: x k = F k -1x k -1 + G k -1w k -1;

[0163] Where x k represents the overall state, and represent the two-dimensional position coordinates of the terminal antenna phase center in the local rectangular coordinate system l at time k, and respectively represent the velocity components of the terminal antenna phase center on the x-axis and y-axis in the local rectangular coordinate system l at time k, b i,k represents the TDOA between RRUi + 1 and the reference RRU, that is, RRU1 at time k, F k-1 is the state transition matrix at time k - 1, G k -1 is the noise input matrix at time k - 1. The T in the state transition matrix and the noise input matrix is the tracking sampling interval, and w k -1 is the state noise vector at time k - 1.

[0164] (2) Establish an observation equation for TDOA deviation calibration, that is, establish a second correlation relationship represented by a state space model. This second correlation relationship represents the correlation relationship between the overall state at the current time and the arrival angle AOA information at the current time: ξ k = ψ(x k) + υ k ;

[0165] where ξ k represents the one-dimensional AOA and TDOA observation vectors of N RRU at time k, and ψ(x k ) represents the result after processing the overall state x k through a non-linear function, including the one-dimensional AOA information of the terminal measured by the i-th RRU at time k, and the TDOA information between the phase center of the terminal antenna and the (i + 1)-th RRU antenna and between the phase center of the terminal antenna and the 1st RRU antenna at time k; υ k is the total observation noise vector at time k, where v i,k is the observation noise of the incident angle (one-dimensional AOA information) of the i-th RRU at time k, and τ i,k is the TDOA observation noise between the (i + 1)-th RRU and the 1st RRU at time k. And the height coordinate of the terminal in the local rectangular coordinate system l is set to a known fixed value.

[0166] In addition, to ensure the uniqueness of the positioning solution, the following constraint conditions need to be satisfied, that is:

[0167]

[0168] (3) After completing the construction of the state space model according to the above method, based on the filtering algorithm, determine the state estimate value at the next moment. Here, the particle filter algorithm is mainly used as an example for illustration, and other filtering algorithms such as the Kalman filter algorithm can also implement the method for estimating the terminal position of the present invention.

[0169] According to the constructed state space model, initialize the particle filter, that is, extract Λ particles from the prior probability density function p(x0) determined by the state equation calibrated according to the TDOA deviation The initial particle weights are Subsequently, perform importance sampling, and use the first correlation relationship of the state space model to obtain Λ candidate particle samples That is

[0170] Then update the weights of the Λ candidate particle samples, and the specific corresponding formula is:

[0171]

[0172]

[0173] where is the likelihood function of the Λ candidate particle samples , and ξ kDenote the one - dimensional AOA and TDOA observation vectors of N RRU at time k. Denote the one - dimensional AOA and TDOA observation model of N RRU at time k, R k is the covariance matrix of the one - dimensional AOA and TDOA observation noise vector υ k at time k, is the normalized weight.

[0174] Subsequently, resampling is performed to obtain a new particle set The corresponding particle weights are

[0175] Finally, the state estimate value at time k can be expressed as:

[0176] According to the state estimate value at the above - mentioned time K, compensate the TDOA deviation between the first RRU and the reference RRU, and determine the position information of the terminal.

[0177] Figure 4 is the second implementation schematic diagram of the terminal position estimation method provided by the embodiments of the present invention, as Figure 4 shown, including:

[0178] Multiple RRUs of the distributed 5G base station adopt a two - dimensional planar array, which can measure the two - dimensional AOA information of the 5G terminal uplink reference signal, including the azimuth angle and the elevation angle. Therefore, three - dimensional positioning of the terminal can be realized without knowing the height coordinate of the terminal.

[0179] (1) Establish a state equation for TDOA deviation calibration, that is, establish the first correlation relationship represented by the state - space model, and this first correlation relationship represents the correlation relationship between the overall states at adjacent times: x k = F k-1 x k-1 + G k- 1w k-1 ;

[0180] where and represent the three - dimensional position coordinates of the phase center of the terminal T antenna in the local rectangular coordinate system l at time k, and represent the velocity components of the phase center of the terminal antenna on the x - axis, y - axis, and z - axis in the local rectangular coordinate system l at time k, respectively. b i,k represents the TDOA deviation between RRUi + 1 and the reference RRU, that is, RRU1 at time k. F k-1 is the state transition matrix at time k - 1, G k-1 is the noise input matrix at time k - 1, T is the tracking sampling interval, wk-1 is the state noise vector at time k-1.

[0181] (2) Establish an observation equation for TDOA deviation calibration, that is, establish the second correlation relationship represented by the state space model, and this second correlation relationship represents the correlation relationship between the overall state at the current moment and the angle of arrival (AOA) information at the current moment: ξ k = ψ(x k ) + υ k ;

[0182] where ξ k represents the two-dimensional AOA and TDOA observation vector of N RRU at time k, and ψ(x k ) represents the result after processing the overall state x k through a non-linear function, including the two-dimensional AOA information measured by the i-th RRU at time k, and the TDOA information between the phase center of the terminal antenna and the (i + 1)-th RRU antenna and the first RRU antenna at time k. The two-dimensional AOA information b includes the azimuth angle of the phase center of the terminal antenna in the coordinate system a of the i-th RRU antenna and the elevation angle of the phase center of the terminal antenna in the coordinate system a of the i-th RRU antenna; υ k is the total observation noise vector at time k, including the observation noise of the two-dimensional AOA information and the TDOA observation noise between the (i + 1)-th RRU and the first RRU at time k, and the observation noise of the two-dimensional AOA information includes the azimuth angle observation noise and the elevation angle observation noise of the i-th RRU at time k.

[0183] The position coordinates of the terminal in the local rectangular coordinate system l are three-dimensional coordinates, and the velocity of the terminal is also represented as three-dimensional; the position coordinates of the phase center of the i-th RRU antenna in the local rectangular coordinate system l are three-dimensional, and the attitude angles of the i-th RRU antenna in the local rectangular coordinate system are the azimuth angle, elevation angle, and roll angle respectively.

[0184] (3) After completing the construction of the state space model according to the above method, based on the filtering algorithm, determine the state estimation value at the next moment. Here, the particle filtering algorithm is mainly used as an example for illustration, and other filtering algorithms such as the Kalman filtering algorithm can also implement the method for estimating the terminal position of the present invention in the same way.

[0185] According to the constructed state space model, initialize the particle filter, that is, extract Λ particles from the prior probability density function p(x0) determined by the state equation of TDOA deviation calibration The initial particle weight is Subsequently, perform importance sampling, and use the first correlation relationship of the state space model to obtain Λ particle candidate samples That is

[0186] Then, the weights of Λ particle candidate samples are updated, and the corresponding formula is as follows:

[0187]

[0188] Among them, the meanings of the parameters have been provided in the relevant descriptions of Figure 3 .

[0189] Subsequently, resampling is performed to obtain a new particle set The corresponding particle weights are

[0190] Finally, the state estimation value at time k can be expressed as:

[0191] According to the state estimation value at the above-mentioned time K, the TDOA deviation between the first RRU and the reference RRU is compensated, and the position information of the terminal is determined.

[0192] Through actual measurement experiments, the actual measurement data of the terminal position estimation method proposed by the present invention are collected, indicating that as the number of laps of the terminal running increases, the positioning performance of the terminal position estimation method proposed by the present invention gradually improves and better positioning performance can be obtained. In the third lap, compared with the reference terminal with a known position arranged near the RRU, the root mean squared error (RMSE) of using the terminal position estimation method proposed by the present invention is reduced by 28.2%. For the scenario with a 90% positioning error, the error rate is reduced by 36.4%. For details, please refer to Table 1 and Table 2.

[0193] Table 1 RMSE of positioning by two methods

[0194] Root Mean Square Error RMSE (m) Deploy reference terminal Method for estimating terminal position proposed by the present invention The 1st circle 0.39 1.02 The 2nd circle 0.43 0.31 The 3rd circle 0.39 0.28

[0195] Table 2 90% positioning error of two methods

[0196] 90% Error (m) Deploy reference terminal Method for estimating terminal position proposed by the present invention The 1st circle 0.68 1.04 The 2nd circle 0.74 0.50 The 3rd circle 0.66 0.42

[0197] In addition, Figure 5 is a comparison schematic diagram of the positioning errors corresponding to the two methods provided by the embodiments of the present invention. Figure 5 In (a), it represents the probability distribution of the positioning error corresponding to the arrangement of the reference terminal. Figure 5 In (b), it represents the probability distribution of the positioning error corresponding to the terminal position estimation method proposed by the present invention. Figure 5 In the figure, the horizontal axis represents different positioning errors, and the vertical axis represents the proportion (probability) of the number of samples corresponding to a certain positioning error in the total number of all samples. From Figure 5It can be seen that under the same positioning error, the sample probability of the method for estimating the terminal position proposed by the present invention conforming to the positioning error is greater than the corresponding sample probability when determining the terminal position by the method of deploying reference terminals. It can be understood that the positioning error corresponding to the method for estimating the terminal position proposed by the present invention is smaller.

[0198] Figure 6 is a schematic structural diagram of a device for estimating a terminal position provided by an embodiment of the present invention. As Figure 6 shown, the device includes:

[0199] A first determination module 601, configured to determine, based on a first number of samples used to represent an overall state at a current moment and a state space model, a first number of candidate samples at a next moment as first candidate samples;

[0200] A second determination module 602, configured to update weight coefficients corresponding to the first candidate samples based on weight coefficients corresponding to the first number of samples at the current moment and the state space model as first weight coefficients;

[0201] A third determination module 603, configured to determine the first candidate samples after resampling based on the first candidate samples and the first weight coefficients as the first number of samples used when determining the first candidate samples at a next-to-next moment;

[0202] A fourth determination module 604, configured to determine an estimated value used to represent an overall state at a next moment based on the first candidate samples after resampling;

[0203] A joint estimation module 605, configured to correct a time difference of arrival (TDOA) deviation between a first remote radio unit (RRU) and a reference remote radio unit (RRU) based on the estimated value used to represent an overall state at a next moment and determine position information of a terminal;

[0204] The first number of samples and the first number of candidate samples are both used to represent an overall state; the overall state includes position information of a terminal and states of a plurality of remote radio units (RRUs); the states of the plurality of remote radio units (RRUs) include a TDOA deviation between the first remote radio unit (RRU) and the reference remote radio unit (RRU); the first remote radio unit (RRU) and the reference remote radio unit (RRU) are any one of the plurality of remote radio units (RRUs), and the first remote radio unit (RRU) is different from the reference remote radio unit (RRU); the state space model is used to represent a first correlation relationship between the overall states at adjacent moments and a second correlation relationship between the overall state at a next moment, an angle of arrival (AOA) information, and a TDOA deviation.

[0205] The apparatus for estimating the terminal position provided by the embodiments of the present invention can implement the technical solution of the method for estimating the terminal position in any of the above embodiments. The implementation principle and the beneficial effects are similar to those of the method for estimating the terminal position. For details, refer to the implementation principle and the beneficial effects of the method for estimating the terminal position, which will not be elaborated here.

[0206] Figure 7 is a schematic structural diagram of an electronic device provided by an embodiment of the present invention. As Figure 7 shown, the electronic device may include: a processor 710, a communication interface 720, a memory 730, and a communication bus 740. Among them, the processor 710, the communication interface 720, and the memory 730 communicate with each other through the communication bus 740. The processor 710 can call the logical instructions in the memory 730 to execute the method for estimating the terminal position. The method includes: determining, based on the first number of samples used to represent the overall state at the current moment and the state space model, the first number of candidate samples at the next moment as the first candidate samples; updating, based on the weight coefficients corresponding to the first number of samples at the current moment and the state space model, the weight coefficients corresponding to the first candidate samples as the first weight coefficients; determining, based on the first candidate samples and the first weight coefficients, the resampled first candidate samples as the first number of samples used when determining the first candidate samples at the next next moment; determining, based on the resampled first candidate samples, the estimated value used to represent the overall state at the next moment; correcting, based on the estimated value used to represent the overall state at the next moment, the time difference of arrival (TDOA) deviation between the first remote radio unit (RRU) and the reference remote radio unit (RRU), and determining the position information of the terminal; the overall state includes the position information of the terminal and the states of multiple remote radio units (RRUs); the states of the multiple remote radio units (RRUs) include the TDOA deviation between the first remote radio unit (RRU) and the reference remote radio unit (RRU); the first remote radio unit (RRU) and the reference remote radio unit (RRU) are any one of the multiple remote radio units (RRUs), and the first remote radio unit (RRU) is different from the reference remote radio unit (RRU); the state space model is used to represent the first correlation relationship between the overall states at adjacent moments and the second correlation relationship between the overall state at the next moment, the angle of arrival (AOA) information, and the TDOA deviation.

[0207] In addition, when the logical instructions in the above-mentioned memory 730 are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0208] On the other hand, the present invention also provides a computer program product. The computer program product includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the method for estimating the terminal position provided by the above-mentioned various methods. The method includes: based on the first number of samples used to represent the overall state at the current moment and the state space model, determining the first number of candidate samples at the next moment as the first candidate samples; based on the weight coefficients corresponding to the first number of samples at the current moment and the state space model, updating the weight coefficients corresponding to the first candidate samples as the first weight coefficients; based on the first candidate samples and the first weight coefficients, determining the resampled first candidate samples as the first number of samples used when determining the first candidate samples at the next next moment; based on the resampled first candidate samples, determining the estimated value used to represent the overall state at the next moment; based on the estimated value used to represent the overall state at the next moment, correcting the time difference of arrival (TDOA) deviation between the first remote radio unit (RRU) and the reference remote radio unit (RRU), and determining the position information of the terminal; the overall state includes the position information of the terminal and the states of multiple remote radio units (RRUs); the states of the multiple remote radio units (RRUs) include the TDOA deviation between the first remote radio unit (RRU) and the reference remote radio unit (RRU); the first remote radio unit (RRU) and the reference remote radio unit (RRU) are any one of the multiple remote radio units (RRUs), and the first remote radio unit (RRU) is different from the reference remote radio unit (RRU); the state space model is used to represent the first correlation relationship between the overall states at adjacent moments and the second correlation relationship between the overall state at the next moment, the angle of arrival (AOA) information, and the TDOA deviation.

[0209] In another aspect, the present invention further provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements a method for estimating the position of a terminal provided by the above-mentioned various methods. The method includes: determining, based on the first number of samples representing the overall state at the current moment and a state space model, the first number of candidate samples at the next moment as the first candidate samples; updating, based on the weight coefficients corresponding to the first number of samples at the current moment and the state space model, the weight coefficients corresponding to the first candidate samples as the first weight coefficients; determining, based on the first candidate samples and the first weight coefficients, the first candidate samples after resampling as the first number of samples used when determining the first candidate samples at the next next moment; determining, based on the first candidate samples after resampling, an estimated value representing the overall state at the next moment; correcting, based on the estimated value representing the overall state at the next moment, the time difference of arrival (TDOA) deviation between the first remote radio unit (RRU) and the reference remote radio unit (RRU), and determining the position information of the terminal; the overall state includes the position information of the terminal and the states of multiple remote radio units (RRUs); the states of the multiple remote radio units (RRUs) include the TDOA deviation between the first remote radio unit (RRU) and the reference remote radio unit (RRU); the first remote radio unit (RRU) and the reference remote radio unit (RRU) are any of the multiple remote radio units (RRUs), and the first remote radio unit (RRU) is different from the reference remote radio unit (RRU); the state space model is used to represent the first correlation relationship between the overall states at adjacent moments and the second correlation relationship between the overall state at the next moment, the angle of arrival (AOA) information, and the TDOA deviation.

[0210] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative labor.

[0211] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0212] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. And these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for estimating the position of a terminal, characterized in that Including: Based on the first number of samples used to represent the overall state at the current moment and the state space model, determining the first number of candidate samples at the next moment as the first candidate samples; Based on the weight coefficients corresponding to the first number of samples at the current moment and the state space model, updating the weight coefficients corresponding to the first candidate samples as the first weight coefficients; Based on the first candidate samples and the first weight coefficients, determining the resampled first candidate samples as the first number of samples used when determining the first candidate samples at the next next moment; Based on the resampled first candidate samples, determining the estimated value used to represent the overall state at the next moment; Based on the estimated value used to represent the overall state at the next moment, correcting the time difference of arrival (TDOA) deviation between the first remote radio unit (RRU) and the reference remote radio unit (RRU), and determining the location information of the terminal; The overall state includes the location information of the terminal and the states of multiple remote radio units (RRUs); the states of the multiple remote radio units (RRUs) include the TDOA deviation between the first remote radio unit (RRU) and the reference remote radio unit (RRU); the first remote radio unit (RRU) and the reference remote radio unit (RRU) are any of the multiple remote radio units (RRUs), and the first remote radio unit (RRU) is different from the reference remote radio unit (RRU); The state space model is used to represent the first correlation relationship between the overall states at adjacent moments, and the second correlation relationship between the overall state at the next moment, the angle of arrival (AOA) information, and the TDOA deviation; 2. The method for estimating the terminal position according to claim 1, characterized in that The state space model is used to represent the first correlation relationship between the overall states at adjacent moments, and the method for determining the first correlation relationship includes: Based on the multiple remote radio units (RRUs) using a one-dimensional linear array or a two-dimensional planar array, determining the dimension corresponding to the terminal state included in the overall state, where the terminal state includes the location information of the terminal in each dimension and the velocity of the terminal in each dimension; Based on the overall state, the state transition matrix, the noise input matrix, and the state noise vector at the current moment, establishing the correlation relationship with the overall state at the next moment; 3. The method for estimating the terminal position according to claim 2, characterized in that, The state space model is used to represent the second correlation relationship between the overall state at the next moment, the AOA information, and the TDOA deviation, and the method for determining the second correlation relationship includes: Based on the attitude angle and location information corresponding to each remote radio unit (RRU) among the multiple remote radio units (RRUs) at the next moment, the location information of the terminal at the next moment, and the AOA observation noise at the next moment, establishing the correlation relationship with the AOA observation vector at the next moment; Establish an association relationship with the time difference of arrival (TDOA) observation vector at the next moment based on the position information of the terminal at the next moment, the position information of the first remote radio unit (RRU), the position information of the reference RRU, the deviation of the time difference of arrival (TDOA) corresponding to the first RRU and the reference RRU, and the observation noise of the time difference of arrival (TDOA) corresponding to the first RRU and the reference RRU; The angle of arrival (AOA) observation vector at the next moment includes the AOA information measured by the multiple RRUs at the next moment; The time difference of arrival (TDOA) observation vector at the next moment includes the TDOA observation values corresponding to multiple first RRUs and the reference RRU; 4. The method for estimating the terminal position according to claim 3, wherein The method for obtaining the AOA observation vector and the TDOA observation vector at the next moment includes: Measure the AOA information corresponding to the reference terminal moving on the first movement trajectory at the next moment and the TDOA information corresponding to the first RRU and the reference RRU, and use them as the AOA observation vector and the TDOA observation vector at the next moment respectively; The first movement trajectory is the trajectory of the reference terminal moving within the coverage area of the service area that the multiple RRUs can provide; 5. The method for estimating the terminal position according to claim 4, characterized in that, Determining the first number of candidate samples at the next moment based on the first number of samples at the current moment and the state space model includes: Determine the first number of candidate samples at the next moment based on the first number of samples at the current moment and the first association relationship; 6. The method for estimating the terminal position according to claim 5, wherein Updating the weight coefficients corresponding to the first candidate samples based on the weight coefficients corresponding to the first number of samples at the current moment and the state space model includes: Determine that the multiple RRUs use a one-dimensional linear array or a two-dimensional planar array; If the multiple RRUs use a one-dimensional linear array, then based on the weight coefficients corresponding to the first number of samples at the current moment, the AOA information of the multiple RRUs at the next moment, the covariance matrix of the TDOA observation noise, and a preset constraint condition, update the weight of the first candidate sample at the next moment to obtain the first weight coefficient; the preset constraint condition is used to limit the uniqueness of the determined position information of the terminal; If the multiple RRUs use a two-dimensional planar array, then based on the weight coefficients corresponding to the first number of samples at the current moment, the AOA information of the multiple RRUs at the next moment, and the covariance matrix of the TDOA observation noise, update the weight of the first candidate sample at the next moment to obtain the first weight coefficient.

7. The method for estimating the terminal position according to claim 6, wherein Updating the weight corresponding to the first candidate sample at the next moment based on the weight coefficients corresponding to the first number of samples at the current moment, the angle of arrival (AOA) information of the multiple remote radio units (RRUs) at the next moment, the covariance matrix of the time difference of arrival (TDOA) observation noise, and a preset constraint condition to obtain the first weight coefficient, includes: Determining the position information of the terminal at the next moment based on the first number of samples at the current moment and the state space model; If the position information of the terminal at the next moment satisfies the preset constraint condition, updating the weight corresponding to the first candidate sample at the next moment based on the weight coefficients corresponding to the first number of samples at the current moment, the AOA information of the multiple RRUs at the next moment, and the covariance matrix of the TDOA observation noise to obtain the first weight coefficient.

8. An apparatus for estimating the position of a terminal, characterized in that, Including: A first determination module, configured to determine the first number of candidate samples at the next moment as the first candidate samples based on the first number of samples representing the overall state at the current moment and the state space model; A second determination module, configured to update the weight coefficient corresponding to the first candidate sample as the first weight coefficient based on the weight coefficients corresponding to the first number of samples at the current moment and the state space model; A third determination module, configured to determine the resampled first candidate samples as the first number of samples representing the overall state at the current moment used when determining the first candidate samples at the next next moment based on the first candidate samples and the first weight coefficient; A fourth determination module, configured to determine the estimated value representing the overall state at the next moment based on the resampled first candidate samples; A joint estimation module, configured to correct the TDOA deviation between the first RRU and the reference RRU based on the estimated value representing the overall state at the next moment and determine the position information of the terminal; The overall state includes the position information of the terminal and the states of the multiple RRUs; the states of the multiple RRUs include the TDOA deviation between the first RRU and the reference RRU; the first RRU and the reference RRU are any of the multiple RRUs, and the first RRU is different from the reference RRU; The state space model is used to represent the first correlation relationship between the overall states at adjacent moments and the second correlation relationship between the overall state at the next moment, the AOA information, and the TDOA deviation.

9. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method for estimating the terminal position according to any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the method for estimating the terminal position according to any one of claims 1 to 7.

Images