WO2024050701A1 - 定位方法、电子设备、计算机可读存储介质及芯片系统 - Google Patents

定位方法、电子设备、计算机可读存储介质及芯片系统 Download PDF

Info

Publication number
WO2024050701A1
WO2024050701A1 PCT/CN2022/117371 CN2022117371W WO2024050701A1 WO 2024050701 A1 WO2024050701 A1 WO 2024050701A1 CN 2022117371 W CN2022117371 W CN 2022117371W WO 2024050701 A1 WO2024050701 A1 WO 2024050701A1
Authority
WO
WIPO (PCT)
Prior art keywords
points
information
sparse
residual
position information
Prior art date
Application number
PCT/CN2022/117371
Other languages
English (en)
French (fr)
Inventor
李彦淳
杨刚华
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to PCT/CN2022/117371 priority Critical patent/WO2024050701A1/zh
Publication of WO2024050701A1 publication Critical patent/WO2024050701A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L23/00—Details of semiconductor or other solid state devices
    • H01L23/34—Arrangements for cooling, heating, ventilating or temperature compensation ; Temperature sensing arrangements
    • H01L23/40—Mountings or securing means for detachable cooling or heating arrangements ; fixed by friction, plugs or springs
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W64/00—Locating users or terminals or network equipment for network management purposes, e.g. mobility management

Definitions

  • the present application relates to the field of wireless positioning technology, and in particular, to a positioning method, electronic device, computer-readable storage medium and chip system.
  • wireless signals can also be used in wireless positioning systems on the basis of being used for data communication to locate the transmitter or receiver of the wireless positioning system.
  • the wireless positioning system can collect the channel state information (CSI) corresponding to the channel between the transmitter and the receiver, and then use simultaneous localization and mapping (SLAM) technology to perform CSI Process to get the location of the sender or receiver.
  • CSI channel state information
  • SLAM simultaneous localization and mapping
  • This application provides a positioning method, electronic device, computer-readable storage medium and chip system, which solves the problem of inaccurate positioning of the sending end or receiving end in the prior art.
  • a positioning method is provided, applied to any first device of a wireless communication system.
  • the wireless communication system includes: a plurality of first devices and second devices, and the second device is connected to each of the first devices.
  • the first devices are all connected wirelessly, and any two first devices are connected wirelessly.
  • the method includes:
  • the information corresponding to the sparse points is calculated according to a preset algorithm to obtain the first position information, and the information corresponding to the residual points is calculated through a pre-trained neural network to obtain the second position information. Both the location information and the second location information are used to represent the location of the second device;
  • Actual location information of the second device is determined based on the first location information and the second location information.
  • a sparse point of the second device is determined. and residual points, the method also includes:
  • the channel state information is determined based on the reference signal.
  • the information corresponding to the residual points is calculated through a pre-trained neural network to obtain the second position information, including:
  • the information corresponding to the residual points is calculated to obtain the second position information.
  • the information corresponding to the sparse points is calculated according to a preset algorithm to obtain the first position information, including:
  • the sparse points associate the sparse points with historical feature points, where the historical feature points are the sparse points of the second device that have been extracted;
  • the information corresponding to the sparse points is calculated to obtain the first location information.
  • the method further includes:
  • the information corresponding to the sparse points is calculated according to a preset algorithm to obtain the first position information, including:
  • the information corresponding to the residual points is calculated through the pre-trained neural network to obtain the second position information, including:
  • the second position information is obtained by calculating information corresponding to the residual points and other information corresponding to the residual points sent by the second device through a pre-trained neural network.
  • the channel status of the communication channel between the first device and the second device is Information is extracted to obtain sparse points and residual points, including:
  • Each spectrum is detected, and sparse points and residual points in each spectrum are determined.
  • the method further includes:
  • the neural network is optimized through the loss function of the neural network and combined with the first position information.
  • the embodiment of the present application provides a positioning method that collects the CSI of the communication channel between the first device and the second device and extracts the CSI to obtain the sparse one-dimensional or multi-dimensional spectrum of the CSI.
  • the information corresponding to the points and residual points respectively is calculated through the SLAM algorithm to calculate the information of the associated sparse points to obtain the first position information.
  • the information corresponding to the residual points is input into the pre-trained neural network and output through the neural network.
  • the second location information is then fused and calculated based on the first location information and the second location information to obtain the actual location information of the second device.
  • the residual point information in the one-dimensional or multi-dimensional spectrum of the CSI is obtained, and the second device is positioned again through the trained neural network combined with the residual point information to obtain the second position information, so that the second device can be positioned through the third
  • the second position information corrects the first position information, thereby improving the accuracy of positioning the second device.
  • the above-mentioned neural network is based on the spatial consistency characteristics of each residual point, combined with the mutual information included in the information of each residual point, and generates a message format field that is convenient for describing multiple feature combinations through disentangled representation, thereby The accuracy of the second position information can be improved and the feature set learned by the neural network can be expanded.
  • the second device is positioned through multiple first devices, and during the positioning process, each first device can interact with the location information of the second device, so that the second device can be positioned through more location information. Improved accuracy in locating second devices.
  • a positioning device is provided, applied to any first device of a wireless communication system.
  • the wireless communication system includes: a plurality of first devices and second devices, and the second device is connected to each of the first devices.
  • the first devices are all connected wirelessly, and any two first devices are connected wirelessly.
  • the device includes:
  • An extraction module configured to extract the channel state information of the communication channel between the first device and the second device, and determine the sparse points and residual points of the second device;
  • the calculation module is used to calculate the information corresponding to the sparse points according to a preset algorithm to obtain the first position information, and to calculate the information corresponding to the residual points through a pre-trained neural network to obtain the second position information.
  • the first location information and the second location information are both used to represent the location of the second device;
  • a first determination module configured to determine the actual location information of the second device according to the first location information and the second location information.
  • the device further includes:
  • a first receiving module configured to receive the reference signal sent by the second device
  • the second determination module is configured to determine the channel state information according to the reference signal.
  • the calculation module is specifically configured to input the information corresponding to the residual points into the neural network; according to the pre-trained algorithm in the neural network and the pre-trained parameters, calculate the information corresponding to the residual point, and obtain the second position information.
  • the calculation module is further specifically configured to associate the sparse points with historical feature points according to the information corresponding to the sparse points, and the historical feature points are The sparse points of the second device have been extracted; through a preset algorithm and in combination with historical feature points associated with the sparse points, the information corresponding to the sparse points is calculated to obtain the first location information.
  • the device further includes:
  • a second receiving module configured to receive the sparse points and residual points of the second device sent by other second devices
  • the calculation module is also specifically configured to calculate the information corresponding to the sparse point and other information corresponding to the sparse point sent by the second device according to a preset algorithm to obtain the first location information. ;
  • the calculation module is also specifically configured to calculate the information corresponding to the residual point and other information corresponding to the residual point sent by the second device through a pre-trained neural network to obtain the second position. information.
  • the extraction module is specifically used to extract the channel state information to obtain one-dimensional or multi-dimensional Spectrum; detect each spectrum and determine the sparse points and residual points in each spectrum.
  • the device further includes:
  • An optimization module configured to optimize the neural network based on the second position information, through the loss function of the neural network, and in combination with the first position information.
  • a third aspect provides an electronic device, including: a processor configured to run a computer program stored in a memory, so that the electronic device implements the positioning method according to any one of the first aspects.
  • a fourth aspect provides a computer-readable storage medium that stores a computer program.
  • the computer program is executed by a processor, the positioning method as described in any one of the first aspects is implemented.
  • a fifth aspect provides a chip system.
  • the chip system includes a memory and a processor.
  • the processor executes a computer program stored in the memory to implement the positioning method as described in any one of the first aspects.
  • Figure 1 is a schematic waveform diagram of a channel impulse response provided by an embodiment of the present application.
  • Figure 2A is a system schematic diagram of a wireless communication system provided by an embodiment of the present application.
  • Figure 2B is a system schematic diagram of another wireless communication system provided by an embodiment of the present application.
  • Figure 3A is a schematic flow chart of a positioning method provided by an embodiment of the present application.
  • Figure 3B is a schematic framework diagram of a first device positioning a second device according to an embodiment of the present application
  • Figure 4 is a schematic framework diagram of another first device positioning a second device according to an embodiment of the present application.
  • Figure 5 is a structural block diagram of a positioning device provided by an embodiment of the present application.
  • Figure 6 is a schematic structural diagram of a terminal device provided by this application.
  • Figure 7 is a schematic structural diagram of a network device provided by an embodiment of the present application.
  • positioning can also be performed based on wireless signals, so as to locate the sending end of the wireless signal or the receiving end of the wireless signal to determine the location. The location of the sender or receiver.
  • ultra wide band (UWB) systems can be used for positioning. Since the bandwidth of the channel is significantly increased in the UWB system, the resolution of the UWB system can be improved, thereby achieving more accurate positioning. Moreover, UWB systems can also be combined with oversampling technology to improve the effective signal-to-noise ratio at the receiving end.
  • a UWB system may include multiple UWB base stations.
  • the UWB base station can obtain the corresponding channel impulse response (CIR) for the device to be positioned, and determine it through the leading edge detection (LED) algorithm based on the obtained CIR.
  • CIR channel impulse response
  • LED leading edge detection
  • the rising edge of the first peak in CIR determines the first path delay, and the first path propagation distance can be calculated by multiplying it by the speed of light.
  • the location of the device to be located can be determined.
  • Figure 1 is a schematic waveform diagram of a channel impulse response provided by an embodiment of the present application.
  • the figure shows the noise intensity determined based on the LED algorithm and the preset first path rising slope signal intensity threshold.
  • the waveform corresponding to the channel impulse response it can be determined that the peak value of peak A in the channel impulse response is greater than the first path rising slope signal strength threshold, while the peak value of wave peak B is greater than the noise intensity, but smaller than the first path rising slope signal strength threshold.
  • the distance corresponding to the 0 value of the horizontal axis where the wave peak A is located is the line of sight path (LoS path).
  • the amplitude of the wave peak A is greater than the first path.
  • the delay corresponding to the horizontal axis position at this time can be determined as the first path delay, so that the LoS path distance can be determined based on the first path delay.
  • the peak value of wave peak B is less than the first path rising slope signal strength threshold, which means that wave peak B belongs to the NLoS path situation.
  • the energy of the first path is severely attenuated, and accurate analysis based on the first path delay cannot be performed. position.
  • the time domain sampling interval is 0.2 nanoseconds (ns), and the typical multipath resolution can reach the sub-meter level.
  • the size of many objects can reach decimeter level or even smaller, so the UWB system cannot distinguish the characteristics of these objects, resulting in the failure of multipath to successfully distinguish, or it is subject to the multipath aliasing effect. It is impossible to accurately estimate parameters.
  • the sender or receiver can be positioned based on wireless multipath instant localization and mapping (SLAM) positioning technology.
  • SLAM wireless multipath instant localization and mapping
  • the transmitting end or the receiving end can first obtain the CSI, extract one-dimensional or multi-dimensional spectrum (such as at least one of delay spectrum, angular spectrum and Doppler spectrum) based on the CSI, and then perform the analysis on the one-dimensional or multi-dimensional spectrum.
  • Feature point detection identifying sparse points in one- or multi-dimensional spectra.
  • the sparse points in each feature point can be associated with each marked feature point, and then calculated through the SLAM algorithm to obtain the position information of the sender or receiver, so that when the line-of-sight path is blocked, the observed Each feature point is associated and matched with each marked feature point to achieve positioning of the sender or receiver.
  • positioning can be performed based on CSI collected by multiple transmitters or multiple receivers. Since multiple transmitters or multiple receivers are located at different locations in space, the CSI collected by each transmitter or each receiver has spatial diversity, which can greatly reduce the occurrence of systematic locations at specific locations and specific directions. Probability of accuracy deterioration. Moreover, since the CSI collected by multiple transmitters or multiple receivers has a certain degree of repeatability, positioning accuracy can be improved.
  • the embodiment of the present application proposes a positioning method.
  • the first device obtains the CSI corresponding to the second device and processes the CSI through a preset neural network to obtain a feature descriptor used to describe the second device, and then based on The feature descriptor is combined with the positioning technology based on wireless multipath SLAM to determine the location of the second device.
  • the characteristic descriptor of the second device is obtained.
  • the positioning accuracy of the second device can be improved based on the characteristic descriptor, and the accuracy of positioning the second device can be improved.
  • Figure 2A is a system schematic diagram of a wireless communication system provided by an embodiment of the present application.
  • the The wireless communication system may include: a first device 210 and a second device 220.
  • the above-mentioned wireless communication systems may include: narrowband-internet of things (NB-IoT), global system for mobile communications (GSM), enhanced data rate GSM evolution system (enhanced data rate for GSM evolution, EDGE), wideband code division multiple access system (wideband code division multiple access, WCDMA), code division multiple access 2000 system (code division multiple access, CDMA2000), time division synchronous code division multiple access system (time division-synchronization code Division multiple Access, TD-SCDMA), long term evolution system (long term evolution, LTE) and the three major application scenarios of the next generation 5G mobile communication system: enhanced mobile broadband (eMBB), ultra-high reliability and low latency Communication (ultra-reliable and low latency communications, URLLC) and massive machine type of communication (mMTC), the embodiments of this application do not limit the wireless communication system.
  • eMBB enhanced mobile broadband
  • URLLC ultra-high reliability and low latency Communication
  • mMTC massive machine type of communication
  • wireless communication systems are usually composed of cells. Each cell contains a base station (BS).
  • the base station can provide communication services to multiple mobile stations (MS).
  • MS mobile stations
  • the first device 210 in the above wireless communication system may be a base station, and the second device 220 may be a mobile station; or, the first device 210 in the above wireless communication system may be a mobile station, and the second device 220 may be a base station,
  • the embodiment of the present application does not limit the first device 210 and the second device 220.
  • the base station may include: a baseband unit (BBU) and a remote radio unit (RRU).
  • BBU baseband unit
  • RRU remote radio unit
  • the BBU and RRU can be placed in different places.
  • the RRU can be placed in an area with high traffic volume, and the BBU can be placed in the central computer room.
  • the BBU and the RRU can also be placed in the same area.
  • the BBU and the RRU can be placed in the same computer room.
  • the BBU and the RRU can also be used as different components under the same rack.
  • the embodiments of this application do not specify the locations of the BBU and the RRU. Make limitations.
  • the mobile station may be a terminal device or other wireless communication node (or may also include other auxiliary communication and computing nodes), and the embodiment of the present application does not limit the mobile station.
  • the positioning method provided by the embodiments of the present application can be calculated and implemented by the wireless communication chip in the base station or mobile station, or can also be calculated and implemented by other processing chips in the base station or mobile station. The embodiments of the present application focus on the chip that implements the positioning method in the base station or mobile station. No restrictions.
  • the first device 210 as a base station and the second device 220 as a mobile station as an example, the following describes the positioning method and wireless communication system provided by the embodiment of the present application.
  • Figure 2B is a system schematic diagram of another wireless communication system provided by an embodiment of the present application.
  • the wireless communication system may include: a first device A, a first device B, and a second device C. Communication channels exist between the second device C and both the first device A and the first device B.
  • the second device C can send reference signals (such as reference symbols or long training fields (LTF), etc.) to the first device A and the first device B respectively.
  • the reference signals pass through multiple paths.
  • both the first device A and the first device B can detect the channel status between the second device C and the first device A and the first device B according to the reference signal.
  • get the first CSI and the second CSI get the first CSI and the second CSI.
  • the first CSI is used to represent the channel state between the first device A and the second device C
  • the second CSI is used to represent the channel state between the first device A and the second device C.
  • the first device A may first preprocess the first CSI, and then input the preprocessed CSI into a preset neural network to obtain the first feature descriptor corresponding to the second device C.
  • the first device B may also first perform feature extraction on the second CSI to obtain the second feature descriptor.
  • the first device A shares the first feature descriptor with the first device B as a priori information. Then the first device B can locate the second device C based on the second feature descriptor combined with the first feature descriptor. , determine the location of the second device C.
  • the reference signal may propagate from the second device C to the first device A and the first device B in a straight line, or may propagate to the first device A and the first device B via reflection from regular points and/or irregular points. Other paths may be used to propagate the reference signal from the second device C to the first device A and the first device B.
  • the embodiments of this application do not limit the multiple paths for propagating the reference signal.
  • both regular points and irregular points are objects that can reflect reference signals.
  • Regular points can be objects with regular shapes
  • irregular points can be objects with irregular shapes, or they can be composed of multiple regular points.
  • the embodiments of this application do not limit regular points and irregular points.
  • the a priori information may include: identification information of the device under test, estimated position information of the device under test, target characteristics of regular points, and target characteristics and attributes of irregular points.
  • the device under test is a device that needs to be positioned by the detection device; the identification information of the device under test is used to indicate the device under test; and the estimated location information of the device under test is the location of the device under test estimated by the detection device.
  • the device under test may be the second device C
  • the detection device may be the first device A and the first device B
  • the identification information of the device under test may be the identification information of the second device C
  • the estimated location information of the device under test may be
  • the target feature of the regular point may be the location information of the regular point.
  • the location information of the regular point may be coordinate information (such as two-dimensional coordinates). information, three-dimensional coordinate information, polar coordinate information or parameters of one or more dimensions in a specific coordinate system), and can be described in the form of distribution, such as mean, mean variance or other distribution representations.
  • the target characteristics and attributes of the irregular points can be multiple channel impulse response (CIR) segments (such as delay, angle, and amplitude segments).
  • CIR channel impulse response
  • the target characteristics and attributes of the irregular points can include the CIR.
  • Information such as the relative delay of the segments (peak A and peak B shown in Figure 1), the number of delays of the CIR segment, and the amplitude and phase of each tap in the CIR segment.
  • the following will first describe a positioning method provided by an embodiment of the present application, taking a first device to position a second device as an example.
  • FIG. 3A is a schematic flow chart of a positioning method provided by an embodiment of the present application.
  • FIG. 3B is a schematic framework diagram of a first device positioning a second device provided by an embodiment of the present application.
  • this method can be applied to the first device of the above-mentioned wireless communication system. Referring to Figure 3A and Figure 3B, the method includes:
  • Step 301 Obtain the CSI of the communication channel between the first device and the second device according to the reference signal sent by the second device.
  • the reference signal may be a reference symbol or LTF, etc., and the embodiment of the present application does not limit the reference signal.
  • the first device can send a wireless signal to the second device through the communication channel, and can also receive the wireless signal sent by the second device through the communication channel, thereby realizing communication between the first device and the second device.
  • the first device can not only communicate with the second device, but also position the second device and determine the position of the second device in space.
  • the second device can send a reference signal to the first device through a communication channel, and the first device can receive the reference signal according to the communication channel, detect the received reference signal, and determine the relationship between the first device and the second device.
  • CSI of the communication channel so that in subsequent steps, the first device can locate the second device based on the obtained CSI.
  • Step 302 Extract the CSI to obtain a one-dimensional or multi-dimensional spectrum.
  • the one-dimensional or multi-dimensional spectrum may include: delay spectrum, angular spectrum, Doppler spectrum, etc. or their combination, such as delay-angle spectrum.
  • the embodiment of the present application does not limit the extracted one-dimensional or multi-dimensional spectrum.
  • the first device can preprocess the CSI and obtain the CIR fragment in the CSI, so that the second device can be positioned and determined based on the one-dimensional or multi-dimensional spectrum in the CIR fragment and combined with the pre-trained neural network. The location of the second device in space.
  • the first device can first use low-pass filtering to denoise the CSI, and then separate the phase signal and amplitude signal in the CSI, and linearly transform the phase signal to eliminate synchronization errors. At the same time, it can also Normalize the amplitude signal. The first device can recombine the normalized amplitude signal and the linearly transformed phase signal, and then use the inverse fast Fourier transform method to process the recombined signal to obtain multiple CIR segments, so that the multiple CIR segments can be processed based on the multiple CIR segments. Extract one-dimensional or multi-dimensional spectra including delay spectrum, angular spectrum and Doppler spectrum.
  • Step 303 Detect the one-dimensional or multi-dimensional spectrum and determine the sparse points and residual points in the one-dimensional or multi-dimensional spectrum.
  • the sparse points are points in the spectrum where the energy density is greater than the first threshold, and the residual points are points in the spectrum where the energy density is less than the second threshold, and the second threshold is less than the first threshold. That is to say, sparse points are sparsely arranged points in the spectrum, and residual points are unclear points in the spectrum.
  • the second device may continue to move, causing the wireless propagation path between the first device and the second device to change.
  • the CSI of the communication channel also changes accordingly.
  • the sparse points and residual points in the one-dimensional or multi-dimensional spectrum extracted according to CSI will also change.
  • the first device can identify the sparse points and residual points in the one-dimensional or multi-dimensional spectrum according to the one-dimensional or multi-dimensional spectrum, so that in subsequent steps, the first device can determine the first device based on each sparse point and each residual point. 2. Location of equipment.
  • the first device can calculate the parameters corresponding to each point in the spectrum (such as frequency, energy density, variance and expected value, etc.) according to the preset calculation formula, and then calculate the parameters corresponding to each point in the spectrum according to the preset first calculation formula.
  • Multiple thresholds such as the threshold value and the second threshold value are compared in combination with the corresponding parameters of each point to determine the sparse points and residual points included in each spectrum.
  • the first device may use a compressed sensing detection method to detect one-dimensional or multi-dimensional spectrum, and obtain sparse points and residual points in the one-dimensional or multi-dimensional spectrum.
  • Step 304 For each sparse point, associate the sparse point with the historical feature point.
  • the historical feature points are the sparse points extracted based on the spectrum before the sparse points are extracted by the first device this time.
  • the second device can send the reference signal to the first device multiple times. Then, after the first device receives the reference signal for the first time, it can extract the reference signal based on the reference signal. Sparse points; after that, the first device can receive the reference signal for the second time and extract the sparse points again. At this time, the sparse points extracted based on the first received reference signal are the historical feature points.
  • the first device can associate the sparse points extracted for the second time with the historical feature points.
  • the first device can identify based on each sparse point and determine the location of the second device. Before determining multiple sparse points in step 303, the first device has obtained multiple historical feature points, then the first device can use data such as the position status and motion status estimation information of the second device, as well as the historical status of the second device, etc. Information is used to perform association judgment or association likelihood calculation, thereby associating the multiple sparse points determined in step 303 with the multiple historical feature points that have been obtained.
  • the first device can compare the parameters corresponding to the sparse point (such as distance, angle, speed, etc.) with the parameters corresponding to each historical feature point. Compare the parameters to determine the parameters corresponding to the historical feature points that are the same or similar to the parameters corresponding to the sparse points, so that it can be determined that the historical feature points corresponding to the parameters and the sparse points are the same feature points, and then the sparse points can be compared with the determined Correlate historical feature points.
  • the parameters corresponding to the sparse point such as distance, angle, speed, etc.
  • the first device can determine whether each sparse point corresponds to a historical feature point in the above manner, and associate the sparse points corresponding to the historical feature points to obtain multiple associated sparse points.
  • the historical information of a certain historical feature point includes historical speed and historical position.
  • Step 305 Based on the SLAM algorithm, perform calculation according to each sparse point in the one-dimensional or multi-dimensional spectrum to obtain the first location information of the second device.
  • the first device After the first device associates each sparse point, it can calculate based on each sparse point through the preset SLAM algorithm to obtain the first location information of the second device, so that in subsequent steps, the first device can calculate based on The first location information determines the actual location information of the second device.
  • the first device can calculate the state transition matrix in the SLAM algorithm based on the extracted information of each sparse point, and then combine the preset input control model, control input information and noise to calculate the second device's a location information.
  • Step 306 Input the residual points into the neural network to obtain the second position information of the second device.
  • the first device can not only correlate the extracted sparse points and obtain the first position information, but also input the extracted residual point information into a preset neural network, and use the neural network to calculate the residual point information. The information is calculated to obtain the second position information.
  • the first device can input the message format field composed of the information of each residual point into a preset neural network, and determine the characteristic attributes included in each information in the message format field through the neural network. For each piece of information, the first device can select the two feature attributes with the largest mutual information based on the mutual information included in the associated residual point based on each feature attribute, and combine it with the preset parameters to calculate the second position of the second device. information.
  • the information corresponding to the j-th residual point is S j .
  • the characteristic attribute with the largest mutual information with the j-th residual point can be obtained as The feature attribute with the second largest mutual information with the j-th residual point can be in
  • the neural network can perform calculations based on the preset formula and combined with the preset parameters to obtain the second location information of the second device.
  • d pos is used to represent the second position information
  • C length is used to represent the length of the message format field composed of multiple residual point information
  • H (S j ) is a preset parameter used to represent the jth The entropy corresponding to the residual point.
  • the information of multiple residual points can be represented by different message format fields, that is, the second device can be represented by multiple description methods.
  • the first message can describe that this is an elliptical distribution of scatterers (rather than a rectangular distribution), and the second message can describe that this is an elliptical distribution of scatterers (not rectangular distribution).
  • the first message can describe the length of the major axis of the elliptical distribution of scatterers
  • the third message can describe the length of the minor axis of the elliptical distribution of scatterers
  • the fourth message can describe the density of the elliptical distribution of scatterers, or space. Poisson arrival rate.
  • the obtained residual points may include false residual points.
  • the false residual points do not correspond to actual values in the environment where the second device is located. Feature points. Therefore, when the first device locates the second device through the neural network, the neural network can also identify and generalize false residual points among the multiple residual points to improve the accuracy of the second position information.
  • the first device may be out of order in the process of encoding the information of the residual points according to the signal strength (such as the second device at different times).
  • the information of multiple environmental feature points in the environment where the device is located is exchanged as a whole, or the order of the four messages in the above example is disrupted). Therefore, the neural network can also generalize the order of information of each residual point, thereby improving the accuracy of the second position information.
  • the first device can target the second position information, that is, d pos in the above example, combined with the above mentioned
  • the neural network is trained by generalizing the false residual points and arrangement order, so that the trained neural network can generalize the false residual points and arrangement order, thereby outputting more accurate second position information.
  • Step 307 Determine the actual location information of the second device based on the first location information and the second location information.
  • the first device can further analyze the first location information and the second location information to determine the actual location of the second device, that is, the actual location information of the second device. , complete the positioning of the second device.
  • the first device can calculate based on the first location information and the second location information through a preset Bayesian function to obtain the probabilities of the second device in different spatial locations, and then select the spatial location corresponding to the maximum probability, The spatial location is determined as the actual location of the second device, so that the actual location information of the second device can be obtained.
  • the first device can update the probabilities of the second device at different spatial locations based on the first location information and the second location information, so that the updated probabilities are closer to each other.
  • the actual location of the second device can be updated.
  • Step 308 Optimize the neural network based on the first position information and the second position information, combined with the loss function of the neural network.
  • the first device can optimize the neural network based on the first location information and the second location information, so that the neural network can output more accurate second location information, thereby Improved accuracy in locating second devices.
  • the first device may input the first location information and the second location information into the localization loss function of the neural network, use the first location information as label data, and use the second location information as output data, By comparing the first position information and the second position information, the parameters in the position loss function are adjusted to achieve optimization of the neural network.
  • one second device can be positioned by multiple first devices.
  • one second device can be positioned by two first devices.
  • FIG 4 is a schematic diagram of a framework in which two first devices (first device A and first device B) position a second device.
  • both the first device A and the first device B can obtain the CSI of the communication channel with the second device.
  • the first device B may associate the multiple extracted sparse points in a manner similar to steps 302 to 304, and send the associated sparse point information to the first device A.
  • the first device B can input the information of the residual points into the neural network in a similar manner to step 306, process the information of the residual points through the encoder of the neural network, obtain intermediate data, and then provide the information to the first device A sends the processed intermediate data.
  • the first device A can also associate the multiple extracted sparse points in a manner similar to steps 302 to 304, and receive the information of the multiple associated sparse points sent by the first device B. After that, the first device A can calculate in a similar way to step 305 through the preset SLAM algorithm based on the information of the associated sparse points and the received information of the associated sparse points of the first device B, and obtain The first position information is generated by combining the sparse point information respectively extracted by the two first devices.
  • the first device A can also use a method similar to step 306 to input the information of the residual points into the neural network, process the information of the residual points through the encoder of the neural network, obtain intermediate data, and then combine it with the first device B to send The intermediate data is processed through the decoder of the neural network to obtain the second position information.
  • the first device A can obtain the actual location information of the second device based on the first location information and the second location information in a similar manner to step 307, and complete the positioning of the second device.
  • the embodiment of the present application provides a positioning method that collects the CSI of the communication channel between the first device and the second device and extracts the CSI to obtain the sparse one-dimensional or multi-dimensional spectrum of the CSI.
  • the information corresponding to the points and residual points respectively is calculated through the SLAM algorithm to calculate the information of the associated sparse points to obtain the first position information.
  • the information corresponding to the residual points is input into the pre-trained neural network and output through the neural network.
  • the second location information is then fused and calculated based on the first location information and the second location information to obtain the actual location information of the second device.
  • the residual point information in the one-dimensional or multi-dimensional spectrum of the CSI is obtained, and the second device is positioned again through the trained neural network combined with the residual point information to obtain the second position information, so that the second device can be positioned through the third
  • the second position information corrects the first position information, thereby improving the accuracy of positioning the second device.
  • the above-mentioned neural network is based on the spatial consistency characteristics of each residual point, combined with the mutual information included in the information of each residual point, and generates a message format field that is convenient for describing multiple feature combinations through disentangled representation, thereby The accuracy of the second position information can be improved and the feature set learned by the neural network can be expanded.
  • the second device is positioned through multiple first devices, and during the positioning process, each first device can interact with the location information of the second device, so that the second device can be positioned through more location information. Improved accuracy in locating second devices.
  • sequence number of each step in the above embodiment does not mean the order of execution.
  • the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present application.
  • FIG. 5 is a structural block diagram of a positioning device provided by the embodiment of the present application. For convenience of explanation, only the parts related to the embodiment of the present application are shown.
  • the device includes:
  • the extraction module 501 is used to extract the channel state information of the communication channel between the first device and the second device, and determine the sparse points and residual points of the second device;
  • the calculation module 502 is used to calculate the information corresponding to the sparse points according to a preset algorithm to obtain the first position information, and to calculate the information corresponding to the residual points through a pre-trained neural network to obtain the second position information,
  • the first location information and the second location information are both used to represent the location of the second device;
  • the first determining module 503 is configured to determine the actual location information of the second device based on the first location information and the second location information.
  • the device also includes:
  • the first receiving module 504 is used to receive the reference signal sent by the second device
  • the second determination module 505 is used to determine the channel state information according to the reference signal.
  • the calculation module 502 is specifically used to input the information corresponding to the residual point into the neural network; calculate the information corresponding to the residual point according to the pre-trained algorithm and pre-trained parameters in the neural network, and obtain the second location information.
  • the calculation module 502 is also specifically configured to associate the sparse point with a historical feature point based on the information corresponding to the sparse point.
  • the historical feature point is the sparse point of the second device that has been extracted; by The preset algorithm combines the historical feature points associated with the sparse point to calculate the information corresponding to the sparse point to obtain the first location information.
  • the device also includes:
  • the second receiving module 506 is used to receive the sparse points and residual points of the second device sent by other second devices;
  • the calculation module 502 is also specifically configured to calculate the information corresponding to the sparse point and other information corresponding to the sparse point sent by the second device according to a preset algorithm to obtain the first location information;
  • the calculation module 502 is also specifically configured to calculate the information corresponding to the residual point and other information corresponding to the residual point sent by the second device through a pre-trained neural network to obtain the second position information.
  • the extraction module 501 is specifically used to extract the channel state information to obtain a one-dimensional or multi-dimensional spectrum; detect each spectrum and determine sparse points and residual points in each spectrum.
  • the device also includes:
  • the optimization module 507 is configured to optimize the neural network based on the second position information, through the loss function of the neural network, and in combination with the first position information.
  • the positioning device collects the CSI of the communication channel between the first device and the second device and extracts the CSI to obtain the sparse one-dimensional or multi-dimensional spectrum of the CSI.
  • the information corresponding to the points and residual points respectively is calculated through the SLAM algorithm to calculate the information of the associated sparse points to obtain the first position information.
  • the information corresponding to the residual points is input into the pre-trained neural network and output through the neural network.
  • the second location information is then fused and calculated based on the first location information and the second location information to obtain the actual location information of the second device.
  • the residual point information in the one-dimensional or multi-dimensional spectrum of the CSI is obtained, and the second device is positioned again through the trained neural network combined with the residual point information to obtain the second position information, so that the second device can be positioned through the third
  • the second position information corrects the first position information, thereby improving the accuracy of positioning the second device.
  • Figure 6 is a schematic structural diagram of a terminal device provided by this application.
  • the above-mentioned first device or second device may be configured in the terminal device 600.
  • the first device or the second device itself may be the terminal device 600.
  • the terminal device 600 can perform the actions performed by the terminal device in the methods corresponding to FIG. 3A, FIG. 3B, and FIG. 4.
  • FIG. 6 only shows the main components of the terminal device.
  • the terminal device 600 includes a processor, a memory, a control circuit, an antenna, and an input and output device.
  • the processor is mainly used to process communication protocols and communication data, and to control the entire terminal device, execute software programs, and process data of the software programs. For example, it is used to support the terminal device to execute the above instruction method of the transmission precoding matrix in the embodiment. the action described.
  • the memory is mainly used to store software programs and data, such as the codebook described in the above embodiment.
  • the control circuit is mainly used for conversion of baseband signals and radio frequency signals and processing of radio frequency signals.
  • the control circuit and the antenna together can also be called a transceiver, which is mainly used to send and receive radio frequency signals in the form of electromagnetic waves.
  • Input and output devices such as touch screens, display screens, keyboards, etc., are mainly used to receive data input by users and output data to users.
  • the processor can read the software program in the storage unit, interpret and execute the instructions of the software program, and process the data of the software program.
  • the processor performs baseband processing on the data to be sent, and then outputs the baseband signal to the radio frequency circuit.
  • the radio frequency circuit performs radio frequency processing on the baseband signal and then sends the radio frequency signal out in the form of electromagnetic waves through the antenna.
  • the radio frequency circuit receives the radio frequency signal through the antenna, converts the radio frequency signal into a baseband signal, and outputs the baseband signal to the processor.
  • the processor converts the baseband signal into data and processes the data.
  • FIG. 6 only shows one memory and processor. In an actual terminal device, there may be multiple processors and memories.
  • the memory may also be called a storage medium or a storage device, which is not limited in the embodiments of the present application.
  • the processor may include a baseband processor and a central processing unit.
  • the baseband processor is mainly used to process communication protocols and communication data.
  • the central processing unit is mainly used to control the entire terminal device, execute software programs, and process software programs. data.
  • the processor in Figure 6 integrates the functions of a baseband processor and a central processor.
  • the baseband processor and the central processor can also be independent processors, interconnected through technologies such as buses.
  • the terminal device may include multiple baseband processors to adapt to different network standards, the terminal device may include multiple central processors to enhance its processing capabilities, and various components of the terminal device may be connected through various buses.
  • the baseband processor can also be expressed as a baseband processing circuit or a baseband processing chip.
  • the central processing unit can also be expressed as a central processing circuit or a central processing chip.
  • the function of processing communication protocols and communication data can be built into the processor, or can be stored in the storage unit in the form of a software program, and the processor executes the software program to implement the baseband processing function.
  • the antenna and the control circuit with the transceiver function can be regarded as the transceiver unit 601 of the terminal device 600
  • the processor with the processing function can be regarded as the processing unit 602 of the terminal device 600
  • the terminal device 600 includes a transceiver unit 601 and a processing unit 602.
  • the transceiver unit may also be called a transceiver, a transceiver, a transceiver device, etc.
  • the devices in the transceiver unit 601 used to implement the receiving function can be regarded as receiving units
  • the devices used in the transceiver unit 601 used to implement the transmitting function can be regarded as the transmitting unit.
  • the transceiver unit 601 includes a receiving unit and a transmitting unit.
  • the receiving unit may also be called a receiver, a receiver, a receiving circuit, etc.
  • the sending unit may be called a transmitter, a transmitter, a transmitting circuit, etc.
  • FIG. 7 is a schematic structural diagram of a network device provided by an embodiment of the present application, which can be used to implement the functions of the network device in the above method.
  • the network device 700 includes one or more radio frequency units, such as a remote radio unit (RRU) 701 and one or more baseband units (BBU) (also called a digital unit, DU) 702.
  • the RRU 701 may be called a transceiver unit, a transceiver, a transceiver circuit, a transceiver, etc., and may include at least one antenna and a radio frequency unit.
  • the RRU 701 part is mainly used for transmitting and receiving radio frequency signals and converting radio frequency signals and baseband signals, for example, for sending signaling messages in the above embodiment to terminal equipment.
  • the BBU 702 part is mainly used for baseband processing, base station control, etc.
  • the RRU 701 and the BBU 702 can be physically set together or physically separated, that is, a distributed base station.
  • the BBU 702 is the control center of the base station, which can also be called a processing unit. It is mainly used to complete baseband processing functions, such as channel coding, multiplexing, modulation, spread spectrum, etc.
  • the BBU (processing unit) 702 can be used to control the base station to execute the operation process related to the network equipment in the above method embodiment.
  • the BBU 702 can be composed of one or more single boards. Multiple single boards can jointly support a single access standard wireless access network (such as an LTE system or a 5G system), or can support different access networks respectively. standard wireless access network.
  • the BBU 702 also includes memory and a processor. This memory is used to store necessary instructions and data. For example, the memory stores the codebook in the above embodiment, etc.
  • the processor is used to control the base station to perform necessary actions, for example, to control the base station to perform the operation process of the network equipment in the above method embodiment.
  • the memory and processor can serve one or more single boards. In other words, the memory and processor can be set independently on each board. It is also possible for multiple boards to share the same memory and processor. In addition, necessary circuits can also be installed on each board.
  • SoC system-on-chip
  • all or part of the functions of part 702 and part 701 can be implemented by SoC technology, for example, by a base station function chip Implementation, the base station function chip integrates processor, memory, antenna interface and other devices.
  • the program of the base station related functions is stored in the memory, and the processor executes the program to realize the related functions of the base station.
  • the base station function chip can also read the memory external to the chip to implement related functions of the base station.
  • the processor may be a central processing unit (CPU).
  • the processor may also be other general-purpose processors, digital signal processors (DSP), or dedicated integrated processors.
  • Circuit application specific integrated circuit, ASIC
  • off-the-shelf programmable gate array field programmable gate array, FPGA
  • a general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.
  • non-volatile memory may be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory.
  • non-volatile memory can be read-only memory (ROM), programmable ROM (PROM), erasable programmable read-only memory (erasable PROM, EPROM), electrically removable memory. Erase electrically programmable read-only memory (EPROM, EEPROM) or flash memory.
  • Volatile memory can be random access memory (RAM), which is used as an external cache.
  • RAM random access memory
  • static random access memory static random access memory
  • DRAM dynamic random access memory
  • RAM synchronous dynamic random access memory
  • SDRAM synchronous dynamic random access memory
  • double data rate SDRAM double data rate SDRAM
  • DDR SDRAM double data rate SDRAM
  • enhanced SDRAM enhanced synchronous dynamic random access memory
  • SLDRAM synchronous connection dynamic random access memory access memory
  • direct rambus RAM direct rambus RAM, DR RAM
  • the above embodiments may be implemented in whole or in part by software, hardware, firmware, or any other combination.
  • the above-described embodiments may be implemented in whole or in part in the form of a computer program product.
  • the computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on the computer, the processes or functions according to the embodiments of the present application are generated in whole or in part.
  • the computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device.
  • the computer instructions may be stored in or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted over a wired connection from a website, computer, server, or data center (such as infrared, wireless, microwave, etc.) to another website, computer, server or data center.
  • the computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that contains one or more sets of available media.
  • the available media may be magnetic media (eg, floppy disk, hard disk, magnetic tape), optical media (eg, DVD), or semiconductor media.
  • the semiconductor medium may be a solid state drive.
  • An embodiment of the present application also provides a communication system, which includes: the above-mentioned terminal device and the above-mentioned network device.
  • Embodiments of the present application also provide a computer-readable medium for storing computer program code.
  • the computer program includes instructions for executing the positioning method of the embodiment of the present application in the methods corresponding to FIG. 3A, FIG. 3B, and FIG. 4. .
  • the readable medium may be a read-only memory (ROM) or a random access memory (RAM), which is not limited in the embodiments of this application.
  • This application also provides a computer program product, which includes instructions that, when executed, cause the terminal device to perform terminal device operations corresponding to the above method, or to cause the network device to perform the terminal device operation corresponding to the above method. operation of network equipment.
  • An embodiment of the present application also provides a system chip.
  • the system chip includes: a processing unit and a communication unit.
  • the processing unit may be, for example, a processor.
  • the communication unit may be, for example, an input/output interface, a pin, or a circuit.
  • the processing unit can execute computer instructions to cause the chip in the communication device to execute any positioning method provided by the above embodiments of the present application.
  • any communication device provided in the above embodiments of the present application may include the system chip.
  • the computer instructions are stored in a storage unit.
  • the storage unit is a storage unit within the chip, such as a register, cache, etc.
  • the storage unit may also be a storage unit located outside the chip within the terminal, such as a ROM or other storage unit that can store static information and instructions. Type of static storage device, RAM, etc.
  • the processor mentioned in any of the above places may be a CPU, a microprocessor, an ASIC, or one or more integrated circuits for program execution that control the above-mentioned main system information transmission method.
  • the processing unit and the storage unit can be decoupled, respectively installed on different physical devices, and connected through wired or wireless methods to realize the respective functions of the processing unit and the storage unit to support the system chip to implement the above embodiments. various functions in .
  • the processing unit and the memory may be coupled on the same device.
  • non-volatile memory can be read-only memory (ROM), programmable ROM (PROM), erasable programmable read-only memory (erasable PROM, EPROM), electrically removable memory. Erase electrically programmable read-only memory (EPROM, EEPROM) or flash memory. Volatile memory can be random access memory (RAM), which is used as an external cache.
  • RAM random access memory
  • RAM random access memory
  • static random access memory static random access memory
  • DRAM dynamic random access memory
  • RAM synchronous dynamic random access memory
  • SDRAM synchronous dynamic random access memory
  • double data rate SDRAM double data rate SDRAM
  • DDR SDRAM double data rate SDRAM
  • enhanced SDRAM enhanced synchronous dynamic random access memory
  • SLDRAM synchronous connection dynamic random access memory access memory
  • direct rambus RAM direct rambus RAM, DR RAM
  • system and “network” are often used interchangeably herein.
  • the term “and/or” in this article is just an association relationship that describes related objects, indicating that three relationships can exist. For example, A and/or B can mean: A exists alone, A and B exist simultaneously, and they exist alone. B these three situations.
  • the character "/" in this article generally indicates that the related objects are an "or” relationship.
  • uplink and downlink appearing in this application are used to describe the direction of data/information transmission in specific scenarios.
  • the "uplink” direction generally refers to the direction of data/information transmission from the terminal to the network side, or distribution The direction of transmission from the centralized unit to the centralized unit.
  • the “downstream” direction generally refers to the direction of data/information transmission from the network side to the terminal, or the direction of transmission from the centralized unit to the distributed unit. It can be understood that “uplink” and “downlink” " is only used to describe the transmission direction of data/information, and the specific starting and ending equipment of the data/information transmission is not limited.
  • the methods in the embodiments of the present application can be implemented in whole or in part by software, hardware, firmware, or any combination thereof.
  • software When implemented using software, it may be implemented in whole or in part in the form of a computer program product.
  • the computer program product includes one or more computer programs or instructions. When the computer program or instructions are loaded and executed on the computer, the processes or functions described in the embodiments of the present application are executed in whole or in part.
  • the computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device.
  • the computer program or instructions may be stored in or transmitted over a computer-readable storage medium.
  • the computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server that integrates one or more available media.
  • the disclosed systems, devices and methods can be implemented in other ways.
  • the device embodiments described above are only illustrative.
  • the division of the units is only a logical function division. In actual implementation, there may be other division methods.
  • multiple units or components may be combined or may be Integrated into another system, or some features can be ignored, or not implemented.
  • the coupling or direct coupling or communication connection between each other shown or discussed may be through some interfaces, and the indirect coupling or communication connection of the devices or units may be in electrical, mechanical or other forms.
  • 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 they may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.
  • each functional unit in each embodiment of the present application can be integrated into one processing unit, each unit can exist physically alone, or two or more units can be integrated into one unit.
  • the functions 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.
  • the technical solution of the present application is essentially or the part that contributes to the existing technology or the part of the technical solution can be embodied in the form of a software product.
  • the computer software product is stored in a storage medium, including Several instructions are used to cause 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 this application.
  • the aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM), and random access.

Abstract

本申请适用于无线定位技术领域，提供了一种定位方法、电子设备、计算机可读存储介质及芯片系统，所述方法包括：对信道状态信息进行提取，确定第二设备的稀疏点和残差点；根据预先设置的算法对稀疏点对应的信息进行计算，得到第一位置信息，并通过预先训练的神经网络对残差点对应的信息进行计算，得到第二位置信息；根据第一位置信息和第二位置信息，确定第二设备的实际位置信息。本申请实施例提供的方法，通过获取信道状态信息的一维或多维频谱中的残差点的信息，通过训练的神经网络结合残差点的信息，再次对第二设备进行定位，得到第二位置信息，从而可以通过第二位置信息修正第一位置信息，进而可以提高对第二设备进行定位的准确性。

Description

定位方法、电子设备、计算机可读存储介质及芯片系统 技术领域
本申请涉及无线定位技术领域，尤其涉及一种定位方法、电子设备、计算机可读存储介质及芯片系统。
背景技术
随着无线通信技术的不断发展，无线信号在用于数据通信的基础上，还可以应用在无线定位系统中，对无线定位系统的发送端或接收端进行定位。
例如，无线定位系统可以采集发送端与接收端之间的信道所对应的信道状态信息(channel state information，CSI)，再采用即时定位与地图构建(simultaneous localization and mapping，SLAM)技术，对CSI进行处理，得到发送端或接收端所在的位置。
但是，在实际应用中，部分物体(如建筑和家具等)具有不规则的几何形状和电磁特性，SLAM基于CSI信息获取上述部分物体的特征比较困难，从而影响对发送端或接收端的定位。
发明内容
本申请提供一种定位方法、电子设备、计算机可读存储介质及芯片系统，解决了现有技术中对发送端或接收端定位不准确的问题。
为达到上述目的，本申请采用如下技术方案：
第一方面，提供一种定位方法，应用于无线通信系统的任一第一设备，所述无线通信系统包括：多个所述第一设备和第二设备，所述第二设备与每个所述第一设备均无线连接，任意两个所述第一设备之间均无线连接，所述方法包括：
对所述第一设备与所述第二设备之间的通信信道的信道状态信息进行提取，确定所述第二设备的稀疏点和残差点；
根据预先设置的算法对所述稀疏点对应的信息进行计算，得到第一位置信息，并通过预先训练的神经网络对所述残差点对应的信息进行计算，得到第二位置信息，所述第一位置信息和所述第二位置信息均用于表示所述第二设备的位置；
根据所述第一位置信息和所述第二位置信息，确定所述第二设备的实际位置信息。
在第一方面的第一种可能的实现方式中，在所述对所述第一设备与所述第二设备之间的通信信道的信道状态信息进行提取，确定所述第二设备的稀疏点和残差点之前，所述方法还包括：
接收所述第二设备发送的参考信号；
根据所述参考信号，确定所述信道状态信息。
在第一方面的第二种可能的实现方式中，所述通过预先训练的神经网络对所述残差点对应的信息进行计算，得到第二位置信息，包括：
将所述残差点对应的信息输入所述神经网络；
根据所述神经网络中预先训练的算法以及预先训练的参数，对所述残差点对应的信息进行计算，得到所述第二位置信息。
在第一方面的第三种可能的实现方式中，所述根据预先设置的算法对所述稀疏点对应的信息进行计算，得到第一位置信息，包括：
根据所述稀疏点对应的信息，将所述稀疏点与历史特征点进行关联，所述历史特征点为已经提取得到的所述第二设备的稀疏点；
通过预先设置的算法，结合所述稀疏点关联的历史特征点，对所述稀疏点对应的信息进行计算，得到所述第一位置信息。
基于第一方面的上述任意一种可能的实现方式，在第一方面的第四种可能的实现方式中，所述方法还包括：
接收其他第二设备发送的所述第二设备的稀疏点和残差点；
所述根据预先设置的算法对所述稀疏点对应的信息进行计算，得到第一位置信息，包括：
根据预先设置的算法，对所述稀疏点对应的信息、以及其他所述第二设备发送的所述稀疏点对应的信息进行计算，得到所述第一位置信息；
所述通过预先训练的神经网络对所述残差点对应的信息进行计算，得到第二位置信息，包括：
通过预先训练的神经网络，对所述残差点对应的信息、以及其他所述第二设备发送的所述残差点对应的信息进行计算，得到所述第二位置信息。
基于第一方面的上述任意一种可能的实现方式，在第一方面的第五种可能的实现方式中，所述对所述第一设备与所述第二设备之间的通信信道的信道状态信息进行提取，得到稀疏点和残差点，包括：
对所述信道状态信息进行提取，得到一维或多维频谱；
对每个所述频谱进行检测，确定每个所述频谱中的稀疏点和残差点。
基于第一方面的上述任意一种可能的实现方式，在第一方面的第六种可能的实现方式中，所述方法还包括：
基于所述第二位置信息，通过所述神经网络的损失函数，结合所述第一位置信息，对所述神经网络进行优化。
综上所述，本申请实施例提供的一种定位方法，通过采集第一设备与第二设备之间的通信信道的CSI，并对CSI进行提取，得到CSI的一维或多维频谱中的稀疏点和残差点分别对应的信息，再通过SLAM算法对关联后的稀疏点的信息进行计算，得到第一位置信息，并将残差点对应的信息输入预先训练完毕的神经网络，通过神经网络输出得到第二位置信息，再根据第一位置信息和第二位置信息，融合计算得到第二设备的实际位置信息。通过对CSI进行提取，得到CSI的一维或多维频谱中的残差点的信息，通过训练的神经网络结合残差点的信息，再次对第二设备进行定位，得到第二位置信息，从而可以通过第二位置信息修正第一位置信息，进而可以提高对第二设备进行定位的准确性。
而且，上述神经网络基于各个残差点所具备的空间一致性的特征，并结合各个残差点的信息中所包括的互信息，通过解纠缠的表示生成便于描述多种特征组合的消息 格式字段，从而可以提高第二位置信息的准确性，并可以扩展神经网络所学习到的特征集合。
另外，通过多个第一设备对第二设备进行定位，并在定位过程中各个第一设备可以对第二设备的位置信息进行交互，从而可以通过更多的位置信息对第二设备进行定位，提高对第二设备进行定位的准确性。
第二方面，提供一种定位装置，应用于无线通信系统的任一第一设备，所述无线通信系统包括：多个所述第一设备和第二设备，所述第二设备与每个所述第一设备均无线连接，任意两个所述第一设备之间均无线连接，所述装置包括：
提取模块，用于对所述第一设备与所述第二设备之间的通信信道的信道状态信息进行提取，确定所述第二设备的稀疏点和残差点；
计算模块，用于根据预先设置的算法对所述稀疏点对应的信息进行计算，得到第一位置信息，并通过预先训练的神经网络对所述残差点对应的信息进行计算，得到第二位置信息，所述第一位置信息和所述第二位置信息均用于表示所述第二设备的位置；
第一确定模块，用于根据所述第一位置信息和所述第二位置信息，确定所述第二设备的实际位置信息。
在第二方面的第一种可能的实现方式中，所述装置还包括：
第一接收模块，用于接收所述第二设备发送的参考信号；
第二确定模块，用于根据所述参考信号，确定所述信道状态信息。
在第二方面的第二种可能的实现方式中，所述计算模块，具体用于将所述残差点对应的信息输入所述神经网络；根据所述神经网络中预先训练的算法以及预先训练的参数，对所述残差点对应的信息进行计算，得到所述第二位置信息。
在第二方面的第三种可能的实现方式中，所述计算模块，还具体用于根据所述稀疏点对应的信息，将所述稀疏点与历史特征点进行关联，所述历史特征点为已经提取得到的所述第二设备的稀疏点；通过预先设置的算法，结合所述稀疏点关联的历史特征点，对所述稀疏点对应的信息进行计算，得到所述第一位置信息。
基于第二方面的上述任意一种可能的实现方式，在第一方面的第四种可能的实现方式中，所述装置还包括：
第二接收模块，用于接收其他第二设备发送的所述第二设备的稀疏点和残差点；
所述计算模块，还具体用于根据预先设置的算法，对所述稀疏点对应的信息、以及其他所述第二设备发送的所述稀疏点对应的信息进行计算，得到所述第一位置信息；
所述计算模块，还具体用于通过预先训练的神经网络，对所述残差点对应的信息、以及其他所述第二设备发送的所述残差点对应的信息进行计算，得到所述第二位置信息。
基于第二方面的上述任意一种可能的实现方式，在第一方面的第五种可能的实现方式中，所述提取模块，具体用于对所述信道状态信息进行提取，得到一维或多维频谱；对每个所述频谱进行检测，确定每个所述频谱中的稀疏点和残差点。
基于第二方面的上述任意一种可能的实现方式，在第一方面的第六种可能的实现方式中，所述装置还包括：
优化模块，用于基于所述第二位置信息，通过所述神经网络的损失函数，结合所 述第一位置信息，对所述神经网络进行优化。
第三方面，提供一种电子设备，包括：处理器，所述处理器用于运行存储器中存储的计算机程序，以使得所述电子设备实现如第一方面中任一项所述的定位方法。
第四方面，提供一种计算机可读存储介质，所述计算机可读存储介质存储有计算机程序，所述计算机程序被处理器执行时实现如第一方面中任一项所述的定位方法。
第五方面，提供一种芯片系统，所述芯片系统包括存储器和处理器，所述处理器执行所述存储器中存储的计算机程序，以实现如第一方面中任一项所述的定位方法。
可以理解的是，上述第二方面至第五方面的有益效果可以参见上述第一方面中的相关描述，在此不再赘述。
附图说明
图1为本申请实施例提供的一种信道脉冲响应的波形示意图；
图2A为本申请实施例提供的一种无线通信系统的系统示意图；
图2B为本申请实施例提供的另一种无线通信系统的系统示意图；
图3A为本申请实施例提供的一种定位方法的示意性流程图；
图3B为本申请实施例提供的一种第一设备对第二设备进行定位的框架示意图；
图4为本申请实施例提供的另一种第一设备对第二设备进行定位的框架示意图；
图5为本申请实施例提供的一种定位装置的结构框图；
图6为本申请提供的一种终端设备的结构示意图；
图7为本申请实施例提供的一种网络设备的结构示意图。
具体实施方式
以下描述中，为了说明而不是为了限定，提出了诸如特定系统结构、技术之类的具体细节，以便透彻理解本申请实施例。然而，本领域的技术人员应当清楚，在没有这些具体细节的其它实施例中也可以实现本申请。在其它情况中，省略对众所周知的SLAM算法、对CSI进行提取得到稀疏点和残差点的方法、贝叶斯融合算法、终端设备和基站的详细说明，以免不必要的细节妨碍本申请的描述。
以下实施例中所使用的术语只是为了描述特定实施例的目的，而并非旨在作为对本申请的限制。如在本申请的说明书和所附权利要求书中所使用的那样，单数表达形式“一个”、“所述”、“上述”和“该”旨在也包括例如“一个或多个”这种表达形式，除非其上下文中明确地有相反指示。
随着无线通信技术的不断发展，在通过无线信号进行数据通信的基础上，还可以基于无线信号进行定位，从而对发射无线信号的发送端，或者，对接收无线信号的接收端实现定位，确定发送端或接收端所在的位置。
相关技术中，可以采用超宽带(ultra wide band，UWB)系统进行定位。由于UWB系统中显著增加了信道的带宽，从而可以提高UWB系统的分辨率，进而可以实现更准确地定位。而且，UWB系统还可以结合过采样技术，从而提高接收端的有效信噪比。
具体地，UWB系统中可以包括多个UWB基站。针对每个UWB基站，该UWB基站可以针对待定位设备获取相对应的信道脉冲响应(channel impulse response，CIR)，并基于获取的CIR，通过首径上升沿检测(leading edge detection，LED)算法确定CIR中首峰的上升沿，确定首径延迟，乘以光速即可计算得到首径传播距离。再根据各个 UWB基站所在的位置，以及每个UWB基站与待定位设备之间的首径传播距离，即可确定待定位设备所在的位置。
但是，参见图1，图1为本申请实施例提供的一种信道脉冲响应的波形示意图，图中示出了根据LED算法确定的噪低强度，以及预先设置的首径上升坡信号强度门限。结合信道脉冲响应对应的波形，可以确定该信道脉冲响应中波峰A的峰值大于首径上升坡信号强度门限，而波峰B的峰值虽然大于噪低强度，但是小于首径上升坡信号强度门限。
基于噪低强度和首径上升坡信号强度门限，可以认为波峰A所在的横轴0值附近所对应的距离为视距径(line of sight path，LoS path)，在波峰A的幅值大于首径上升坡信号强度门限时，可以将此时横轴位置对应的延迟确定为首径延迟，从而可以根据首径延迟确定LoS径距离。
但是，波峰B与波峰A相比，波峰B的峰值小于首径上升坡信号强度门限，则说明波峰B属于NLoS径的情形，此时首径的能量被严重衰减，无法根据首径延迟进行准确定位。
例如，在500兆赫兹(MHz)带宽的UWB系统中，时域采样间隔0.2纳秒(ns)，典型多径分辨率可以达到亚米级。但是，在实际环境中的物体尺寸，很多物体的尺寸可以达到分米级甚至更小，则UWB系统无法分辨这些物体的特征，从而导致多径无法成功分辨，或者，受制于多径混叠效应而无法准确估参。
因此，基于在UWB系统中出现的问题，可以基于无线多径即时定位与地图构建(simultaneous localization and mapping，SLAM)的定位技术对发送端或接收端进行定位。
具体地，发送端或接收端可以先获取CSI，并根据CSI提取得到一维或多维频谱(如延迟频谱、角谱和多普勒频谱中的至少一种)，再对一维或多维频谱进行特征点检测，确定一维或多维频谱中的稀疏点。之后，可以将各个特征点中的稀疏点与已标记的各个特征点进行关联，再通过SLAM算法进行计算，得到发送端或接收端的位置信息，从而在视距径被遮挡时，将观测到的各个特征点与已标记的各个特征点进行关联匹配，实现对发送端或接收端的定位。
进一步地，可以通过多个发送端或多个接收端采集的CSI进行定位。由于多个发送端或多个接收端分别位于空间中的不同位置，则每个发送端或每个接收端采集的CSI具有空间分集，从而可以大幅降低在特定位置和特定方向出现系统性的位置精度恶化的概率。而且，由于多个发送端或多个接收端采集的CSI具有一定的重复性，从而可以提升定位的精度。
但是，在实际应用中，受到建筑、家具及其他物体具有不规则的几何形状和电磁特性的影响，模型中简单的点源相对于实际的扩展目标过于抽象，会导致模型不匹配造成定位精度不够理想的问题。而且，无线系统的分辨率又不足以直接获取扩展目标的几何形状，因此获取扩展目标的特征比较困难。
因此，本申请实施例提出一种定位方法，第一设备通过获取第二设备对应的CSI，并通过预先设置的神经网络对CSI进行处理，得到用于描述第二设备的特征描述字，再根据特征描述字，结合基于无线多径SLAM的定位技术，确定第二设备的位置。通 过对第二设备对应的CSI进行提取，得到第二设备的特征描述字，可以根据该特征描述字提高第二设备的定位精度，可以提高对第二设备进行定位的准确性。
下述对本申请实施例提供的一种定位方法所涉及的无线通信系统进行介绍，参见图2A，图2A为本申请实施例提供的一种无线通信系统的系统示意图，如图2A所示，该无线通信系统中可以包括：第一设备210和第二设备220。
上述无线通信系统可以包括：窄带物联网系统(narrow band-internet of things，NB-IoT)、全球移动通信系统(global system for mobile communications，GSM)、增强型数据速率GSM演进系统(enhanced data rate for GSM evolution，EDGE)、宽带码分多址系统(wideband code division multiple access，WCDMA)、码分多址2000系统(code division multiple access，CDMA2000)、时分同步码分多址系统(time division-synchronization code division multiple Access，TD-SCDMA)、长期演进系统(long term evolution，LTE)以及下一代5G移动通信系统的三大应用场景：增强移动宽带(enhanced mobile broadband，eMBB)，超高可靠与低时延通信(ultra-reliable and low latency communications,URLLC)和大规模物联网(massive machine type of communication，mMTC)，本申请实施例对无线通信系统不做限定。
其中，无线通信系统通常由小区组成，每个小区包含一个基站(base station，BS)，基站可以向多个移动台(mobile station，MS)提供通信服务。例如，上述无线通信系统中的第一设备210可以为基站，第二设备220可以为移动台；或者，上述无线通信系统中的第一设备210可以为移动台，第二设备220可以为基站，本申请实施例对第一设备210和第二设备220不做限定。
而且，基站可以包括：基带单元(baseband unit，BBU)和远端射频单元(remote radio unit，RRU)。其中BBU和RRU可以分别放置在不同的地方，例如：RRU可以放置于高话务量的区域，BBU放置于中心机房。当然，BBU和RRU也可以放置在同一区域，例如，BBU和RRU可以放置在同一机房，BBU和RRU也可以分别作为同一机架下的不同部件，本申请实施例对BBU和RRU的位置均不做限定。
另外，移动台可以为终端设备或其他无线通信节点(或还包含其他辅助通信、计算节点)，本申请实施例对移动台不做限定。此外，本申请实施例提供的定位方法可以由基站或移动台中的无线通信芯片计算实现，也可由基站或移动台中的其他处理芯片计算实现，本申请实施例对基站或移动台中实现定位方法的芯片不做限定。
下述以第一设备210是基站、第二设备220是移动台为例，对本申请实施例提供的定位方法以及无线通信系统进行介绍说明。
参见图2B，图2B为本申请实施例提供的另一种无线通信系统的系统示意图，如图2B所示，无线通信系统可以包括：第一设备A、第一设备B和第二设备C，第二设备C与第一设备A和第一设备B之间均存在通信信道。
相应的，第二设备C可以分别向第一设备A和第一设备B发送参考信号(如参考符号(reference symbol)或长训练字段(long training field，LTF)等)，参考信号经过多种路径到达第一设备A和第一设备B，则第一设备A和第一设备B均可以根据该参考信号，对第二设备C与第一设备A和第一设备B之间的信道状态进行检测，得到第一CSI和第二CSI。其中，第一CSI用于表示第一设备A与第二设备C之间的信道状 态，第二CSI用于表示第一设备A与第二设备C之间的信道状态。
第一设备A可以先对第一CSI进行预处理，再将预处理后的CSI输入预先设置的神经网络中，得到第二设备C对应的第一特征描述字。类似的，第一设备B也可以先对第二CSI进行特征提取，得到第二特征描述字。
之后，第一设备A将第一特征描述字作为先验信息与第一设备B共享，则第一设备B可以根据第二特征描述字，结合第一特征描述字，对第二设备C进行定位，确定第二设备C的位置。
其中，参考信号可以从第二设备C直线传播至第一设备A和第一设备B，也可以经由规则点和/或非规则点的反射，传播至第一设备A和第一设备B，还可以采用其他路径从第二设备C传播至第一设备A和第一设备B，本申请实施例对传播参考信号的多种路径不做限定。
而且，规则点和非规则点均为能够反射参考信号的物体，规则点可以为具有规则形状的物体，非规则点可以为具有非规则形状的物体，也可以为多个规则点所组成的具有非规则形状的物体，本申请实施例对规则点和非规则点不做限定。
需要说明的是，先验信息可以包括：待测设备的标识信息、待测设备的估计位置信息、规则点的目标特征、以及非规则点的目标特征和属性。
其中，待测设备为需要通过检测设备对待测设备进行定位的设备；待测设备的标识信息用于指示待测设备；待测设备的估计位置信息为检测设备估计的待测设备的位置。
例如，待测设备可以为第二设备C、检测设备可以为第一设备A和第一设备B，待测设备的标识信息可以为第二设备C的标识信息，待测设备的估计位置信息可以为第一设备A或第一设备B预估的第二设备C的位置信息，规则点的目标特征可以为规则点的位置信息，例如，规则点的位置信息可以为坐标信息(如二维坐标信息、三维坐标信息、极坐标信息或某一特定坐标系中的一个或多个维度的参数)，而且，可以采用分布的方式进行描述，如均值、均值方差或其他分布表示。
而且，非规则点的目标特征和属性可以为多段信道脉冲响应(channel impulse response，CIR)片段(如延迟、角度和幅值各片段)，例如，非规则点的目标特征和属性可以包括该CIR片段(如图1中所示的波峰A和波峰B)的相对延迟、该CIR片段的延迟个数、以及该CIR片段内各抽头的幅值和相位等信息。
下述先以1个第一设备对1个第二设备进行定位为例，对本申请实施例提供的一种定位方法进行说明。
图3A为本申请实施例提供的一种定位方法的示意性流程图，图3B为本申请实施例提供的一种1个第一设备对1个第二设备进行定位的框架示意图，作为示例而非限定，该方法可以应用于上述无线通信系统的第一设备中，参见图3A和图3B，该方法包括：
步骤301、根据第二设备发送的参考信号，获取第一设备与第二设备之间的通信信道的CSI。
其中，参考信号可以为参考符号或LTF等，本申请实施例对参考信号不做限定。
在无线通信系统中，第一设备可以通过通信信道向第二设备发送无线信号，也可 以通过通信信道接收第二设备发送的无线信号，从而实现第一设备与第二设备之间的通信。而在无线通信系统中，第一设备不但可以与第二设备进行通信，还可以对第二设备进行定位，确定第二设备在空间中的位置。
具体地，第二设备可以通过通信信道向第一设备发送参考信号，则第一设备可以根据通信信道接收该参考信号，并对接收的参考信号进行检测，确定第一设备与第二设备之间的通信信道的CSI，以便在后续步骤中，第一设备可以根据获取的CSI对第二设备进行定位。
步骤302、对CSI进行提取，得到一维或多维频谱。
其中，一维或多维频谱可以包括：延迟谱、角谱和多普勒频谱等或它们的组合，例如延迟-角谱，本申请实施例对提取得到的一维或多维频谱不做限定。
第一设备在得到CSI后，可以对CSI进行预处理，获取CSI中的CIR片段，以便可以根据CIR片段中的一维或多维频谱，结合预先训练的神经网络，对第二设备进行定位，确定第二设备在空间中的位置。
具体地，第一设备可以先采用低通滤波的方式对CSI进行去噪，之后再对CSI中的相位信号和幅度信号进行分离，并对相位信号进行线性变换，消除同步误差，同时，还可以对幅度信号进行归一化处理。第一设备可以将归一化的幅度信号与线性变换的相位信号进行重组，再采用快速傅里叶逆变换的方式对重组的信号进行处理，得到多个CIR片段，从而可以根据多个CIR片段提取得到包括延迟频谱、角谱和多普勒频谱等一维或多维频谱。
步骤303、对一维或多维频谱进行检测，确定一维或多维频谱中的稀疏点和残差点。
其中，稀疏点为频谱中能量密度大于第一阈值的点，残差点为频谱中能量密度小于第二阈值的点，且第二阈值小于第一阈值。也即是，稀疏点为频谱中排布较为稀疏的点，残差点为频谱中不清晰的点。
在第一设备对第二设备进行定位的过程中，第二设备可能会不断移动，引起第一设备与第二设备之间的无线传播路径发生变化，则第二设备与第一设备之间的通信信道的CSI也随之发生变化。相应的，根据CSI提取得到的一维或多维频谱中的稀疏点和残差点也会发生变化。
因此，第一设备可以根据一维或多维频谱，对一维或多维频谱中的稀疏点和残差点进行识别，以便在后续步骤中，第一设备可以根据各个稀疏点和各个残差点，确定第二设备的位置。
具体地，对于每个频谱，第一设备可以根据预先设置的计算公式，根据频谱计算得到频谱中各个点对应的参数(如频率、能量密度、方差和期望值等)，再根据预先设置的第一阈值和第二阈值等多个阈值，结合每个点对应参数进行比较，确定每个频谱中所包括的稀疏点和残差点。
例如，第一设备可以采用压缩感知的检测方式，对一维或多维频谱进行检测，得到一维或多维频谱中的稀疏点和残差点。
步骤304、对于每个稀疏点，将稀疏点与历史特征点进行关联。
其中，历史特征点为第一设备在本次提取得到稀疏点之前，根据频谱提取得到的 稀疏点。例如，在第一设备对第二设备进行定位的过程中，第二设备可以多次向第一设备发送参考信号，则第一设备在第一次接收到参考信号后，可以根据参考信号提取得到稀疏点；之后，第一设备可以第二次接收参考信号，并再次提取得到稀疏点。此时，根据第一次接收的参考信号所提取的稀疏点即为历史特征点。相应的，第一设备可以将第二次提取的稀疏点与历史特征点进行关联。
与步骤303相对应的，第一设备在确定各个稀疏点后，可以基于各个稀疏点进行识别，确定第二设备所在的位置。而在步骤303确定多个稀疏点之前，第一设备已经得到多个历史特征点，则第一设备可以利用第二设备的位置状态和运动状态估计信息等数据，以及第二设备的历史状态等信息，进行关联判决或关联的似然度计算，从而将步骤303确定的多个稀疏点，与已经获取的多个历史特征点进行关联。
具体地，第一设备在提取得到各个稀疏点后，针对每个稀疏点，第一设备可以将该稀疏点对应的参数(如距离、角度和速度等参数)，与每个历史特征点对应的参数进行比较，确定与稀疏点对应的参数相同或相似的历史特征点对应的参数，从而可以确定该参数所对应的历史特征点与稀疏点为同一特征点，进而可以将该稀疏点与确定的历史特征点进行关联。
类似的，第一设备可以按照上述方式，确定每个稀疏点是否对应有历史特征点，并将对应有历史特征点的稀疏点进行关联，得到多个关联后的稀疏点。
例如，某个历史特征点的历史信息包括历史速度和历史位置，第二设备相对于第一设备的历史速度为(V x＝0米每秒(m/s)，V y＝+1m/s)，且历史位置的坐标为(X＝0米(m)，Y＝5m)，而某个稀疏点对应的坐标为(X＝0m，Y＝6m)，则可以确定该稀疏点与该历史特征点大概率关联。
步骤305、基于SLAM算法，根据一维或多维频谱中的每个稀疏点进行计算，得到第二设备的第一位置信息。
第一设备在对各个稀疏点关联完毕后，即可通过预先设置的SLAM算法，基于各个稀疏点进行计算，得到第二设备所在的第一位置信息，以便在后续步骤中，第一设备可以基于该第一位置信息，确定第二设备的实际位置信息。
具体地，第一设备可以基于提取得到的各个稀疏点的信息，通过SLAM算法中的状态转移矩阵进行计算，再结合预先设置的输入控制模型、控制输入信息和噪音，计算得到第二设备的第一位置信息。
步骤306、将残差点输入神经网络，得到第二设备的第二位置信息。
与步骤304相对应的，第一设备不但可以对提取得到稀疏点进行关联、得到第一位置信息，还可以将提取得到的残差点的信息输入预先设置的神经网络，通过神经网络对残差点的信息进行运算，得到第二位置信息。
具体地，第一设备可以将各个残差点的信息所组成的消息格式字段输入预先设置的神经网络，通过神经网络确定消息格式字段中每个信息所包括的特征属性。针对每个信息，第一设备可以根据每个特征属性针对所属残差点所包括的互信息，选取包括互信息最大的两个特征属性，结合预先设置的参数，计算得到第二设备的第二位置信息。
例如，第j个残差点对应的信息为S j，神经网络在对S j进行识别后，得到与第j 个残差点具有最大互信息的特征属性可以为
Figure PCTCN2022117371-appb-000001
与第j个残差点具有第二大互信息的特征属性可以为
Figure PCTCN2022117371-appb-000002
其中，
Figure PCTCN2022117371-appb-000003
Figure PCTCN2022117371-appb-000004
之后，神经网络可以根据预先设置的公式，结合预先设置的参数进行计算，得到第二设备的第二位置信息
Figure PCTCN2022117371-appb-000005
其中，d pos用于表示第二位置信息，C length用于表示多个残差点的信息所组成的消息格式字段的长度，H(S j)为预先设置的参数，用于表示与第j个残差点对应的熵。
需要说明的是，在实际应用中，可以通过不同的消息格式字段表示多个残差点的信息，也即是，可以通过多种描述方式表示第二设备。例如，可以采用限定描述的方式，通过在消息格式字段中的第1个消息中描述本消息格式字段所具备的消息数目；也可以采用持续描述的方式，不断增加消息格式字段中的消息。
例如，一个托盘中放置有多个金属杯，则可以通过4个消息组成的消息格式字段进行表示，第1个消息可以描述这是一个椭圆分布散射物集合(而非长方形分布的)，第2个消息可以描述该椭圆分布的散射物集合的长轴长度，第3个消息可以描述该椭圆分布的散射物集合的短轴长度，第4个消息可以描述该椭圆分布的散射物密度，或空间泊松到达率。
另外需要说明的是，在实际应用中，受到噪声和干扰等因素的影响，获取的残差点中可能包括虚假残差点，该虚假残差点在第二设备所处的环境中，并未对应有实际特征点。因此，第一设备在通过神经网络对第二设备进行定位的过程中，神经网络还可以对多个残差点中的虚假残差点进行识别泛化，以提高第二位置信息的准确性。
而且，在实际应用中，受到环境特征点无线路径强度衰弱的影响，第一设备在根据信号强度对残差点的信息进行编码的过程中，会出现乱序的情况(如在不同时刻的第二设备所处的环境中多个环境特征点的信息整体发生交换，或者，上述举例中的4个消息的顺序被打乱)。因此，神经网络还可以对各个残差点的信息的排列顺序进行泛化，从而提高第二位置信息的准确性。
相应的，在对神经网络进行训练、以及在步骤308对神经网络进行优化的过程中，第一设备均可以针对第二位置信息，也即是上述举例中的d pos，结合针对上述提及的虚假残差点和排列顺序进行泛化的方式，对神经网络进行训练，使得训练得到的神经网络可以对虚假残差点和排列顺序进行泛化，从而输出得到更为准确的第二位置信息。
步骤307、根据第一位置信息和第二位置信息，确定第二设备的实际位置信息。
第一设备在确定第一位置信息和第二位置信息后，可以进一步对第一位置信息和第二位置信息进行分析，确定第二设备实际所在的位置，也即是第二设备的实际位置信息，完成对第二设备的定位。
具体地，第一设备可以通过预先设置的贝叶斯函数，根据第一位置信息和第二位置信息进行计算，得到第二设备在不同空间位置的概率，再选取最大概率所对应的空间位置，并将该空间位置确定为第二设备实际所在的位置，从而可以得到第二设备的 实际位置信息。
例如，第一设备在得到第一位置信息和第二位置信息后，可以基于第一位置信息和第二位置信息，对第二设备在不同空间位置的概率进行更新，使得更新后的概率更接近第二设备实际所在的位置。
步骤308、根据第一位置信息和第二位置信息，结合神经网络的损失函数，对神经网络进行优化。
与步骤305相对应的，第一设备在得到第一位置信息后，可以根据第一位置信息与第二位置信息，对神经网络进行优化，使得神经网络能够输出更加准确的第二位置信息，从而提高对第二设备进行定位的准确性。
具体地，第一设备可以将第一位置信息和第二位置信息输入神经网络的位置损失函数(localization loss function)，将第一位置信息作为标签数据，并将第二位置信息作为输出的数据，通过比较第一位置信息和第二位置信息，对位置损失函数中的参数进行调整，实现对神经网络的优化。
需要说明的是，上述以1个第一设备对1个第二设备进行定位为例进行说明，而在实际应用中，可以通过多个第一设备对1个第二设备进行定位。例如，与图2B相对应的，可以通过2个第一设备对1个第二设备进行定位。
参见图4，图4为2个第一设备(第一设备A和第一设备B)对1个第二设备进行定位的框架示意图。具体地，第一设备A和第一设备B均可以获取与第二设备之间的通信信道的CSI。第一设备B可以采用与步骤302至步骤304类似的方式，对提取得到的多个稀疏点进行关联，并向第一设备A发送关联后的稀疏点的信息。而且，第一设备B可以采用与步骤306类似的方式，将残差点的信息输入神经网络，通过神经网络的编码器(encoder)对残差点的信息进行处理，得到中间数据，再向第一设备A发送处理后的中间数据。
类似的，第一设备A也可以采用与步骤302至步骤304类似的方式，对提取得到的多个稀疏点进行关联，并接收第一设备B发送的多个关联后的稀疏点的信息。之后，第一设备A可以采用与步骤305类似的方式，通过预先设置的SLAM算法，根据关联后的稀疏点的信息、以及接收的第一设备B关联后的稀疏点的信息进行计算，得到由两个第一设备分别提取的稀疏点的信息结合后所生成的第一位置信息。
而且，第一设备A也可以采用与步骤306类似的方式，将残差点的信息输入神经网络，通过神经网络的编码器对残差点的信息进行处理，得到中间数据，再结合第一设备B发送的中间数据，通过神经网络的译码器(decoder)对两个中间数据进行处理，得到第二位置信息。
之后，第一设备A可以采用与步骤307类似的方式，根据第一位置信息和第二位置信息得到第二设备的实际位置信息，完成对第二设备的定位。
需要说明的是，上述第一设备A和第一设备B对第二设备进行定位的过程，与步骤301至步骤307的定位过程类似，在此不再赘述。
综上所述，本申请实施例提供的一种定位方法，通过采集第一设备与第二设备之间的通信信道的CSI，并对CSI进行提取，得到CSI的一维或多维频谱中的稀疏点和残差点分别对应的信息，再通过SLAM算法对关联后的稀疏点的信息进行计算，得到 第一位置信息，并将残差点对应的信息输入预先训练完毕的神经网络，通过神经网络输出得到第二位置信息，再根据第一位置信息和第二位置信息，融合计算得到第二设备的实际位置信息。通过对CSI进行提取，得到CSI的一维或多维频谱中的残差点的信息，通过训练的神经网络结合残差点的信息，再次对第二设备进行定位，得到第二位置信息，从而可以通过第二位置信息修正第一位置信息，进而可以提高对第二设备进行定位的准确性。
而且，上述神经网络基于各个残差点所具备的空间一致性的特征，并结合各个残差点的信息中所包括的互信息，通过解纠缠的表示生成便于描述多种特征组合的消息格式字段，从而可以提高第二位置信息的准确性，并可以扩展神经网络所学习到的特征集合。
另外，通过多个第一设备对第二设备进行定位，并在定位过程中各个第一设备可以对第二设备的位置信息进行交互，从而可以通过更多的位置信息对第二设备进行定位，提高对第二设备进行定位的准确性。
应理解，上述实施例中各步骤的序号的大小并不意味着执行顺序的先后，各过程的执行顺序应以其功能和内在逻辑确定，而不应对本申请实施例的实施过程构成任何限定。
对应于上文实施例所述的定位方法，图5为本申请实施例提供的一种定位装置的结构框图，为了便于说明，仅示出了与本申请实施例相关的部分。
参照图5，该装置包括：
提取模块501，用于对该第一设备与该第二设备之间的通信信道的信道状态信息进行提取，确定该第二设备的稀疏点和残差点；
计算模块502，用于根据预先设置的算法对该稀疏点对应的信息进行计算，得到第一位置信息，并通过预先训练的神经网络对该残差点对应的信息进行计算，得到第二位置信息，该第一位置信息和该第二位置信息均用于表示该第二设备的位置；
第一确定模块503，用于根据该第一位置信息和该第二位置信息，确定该第二设备的实际位置信息。
可选的，该装置还包括：
第一接收模块504，用于接收该第二设备发送的参考信号；
第二确定模块505，用于根据该参考信号，确定该信道状态信息。
可选的，该计算模块502，具体用于将该残差点对应的信息输入该神经网络；根据该神经网络中预先训练的算法以及预先训练的参数，对该残差点对应的信息进行计算，得到该第二位置信息。
可选的，该计算模块502，还具体用于根据该稀疏点对应的信息，将该稀疏点与历史特征点进行关联，该历史特征点为已经提取得到的该第二设备的稀疏点；通过预先设置的算法，结合该稀疏点关联的历史特征点，对该稀疏点对应的信息进行计算，得到该第一位置信息。
可选的，该装置还包括：
第二接收模块506，用于接收其他第二设备发送的该第二设备的稀疏点和残差点；
该计算模块502，还具体用于根据预先设置的算法，对该稀疏点对应的信息、以 及其他该第二设备发送的该稀疏点对应的信息进行计算，得到该第一位置信息；
该计算模块502，还具体用于通过预先训练的神经网络，对该残差点对应的信息、以及其他该第二设备发送的该残差点对应的信息进行计算，得到该第二位置信息。
可选的，该提取模块501，具体用于对该信道状态信息进行提取，得到一维或多维频谱；对每个该频谱进行检测，确定每个该频谱中的稀疏点和残差点。
可选的，该装置还包括：
优化模块507，用于基于该第二位置信息，通过该神经网络的损失函数，结合该第一位置信息，对该神经网络进行优化。
综上所述，本申请实施例提供的一种定位装置，通过采集第一设备与第二设备之间的通信信道的CSI，并对CSI进行提取，得到CSI的一维或多维频谱中的稀疏点和残差点分别对应的信息，再通过SLAM算法对关联后的稀疏点的信息进行计算，得到第一位置信息，并将残差点对应的信息输入预先训练完毕的神经网络，通过神经网络输出得到第二位置信息，再根据第一位置信息和第二位置信息，融合计算得到第二设备的实际位置信息。通过对CSI进行提取，得到CSI的一维或多维频谱中的残差点的信息，通过训练的神经网络结合残差点的信息，再次对第二设备进行定位，得到第二位置信息，从而可以通过第二位置信息修正第一位置信息，进而可以提高对第二设备进行定位的准确性。
图6为本申请提供的一种终端设备的结构示意图。上述第一设备或者第二设备可以配置在该终端设备600中。或者，该第一设备或者第二设备本身可以即为该终端设备600。或者说，该终端设备600可以执行上述图3A、图3B和图4对应的方法中终端设备执行的动作。可选的，为了便于说明，图6仅示出了终端设备的主要部件。如图6所示，终端设备600包括处理器、存储器、控制电路、天线以及输入输出装置。
处理器主要用于对通信协议以及通信数据进行处理，以及对整个终端设备进行控制，执行软件程序，处理软件程序的数据，例如用于支持终端设备执行上述传输预编码矩阵的指示方法实施例中所描述的动作。存储器主要用于存储软件程序和数据，例如存储上述实施例中所描述的码本。控制电路主要用于基带信号与射频信号的转换以及对射频信号的处理。控制电路和天线一起也可以叫做收发器，主要用于收发电磁波形式的射频信号。输入输出装置，例如触摸屏、显示屏，键盘等主要用于接收用户输入的数据以及对用户输出数据。
当终端设备开机后，处理器可以读取存储单元中的软件程序，解释并执行软件程序的指令，处理软件程序的数据。当需要通过无线发送数据时，处理器对待发送的数据进行基带处理后，输出基带信号至射频电路，射频电路将基带信号进行射频处理后将射频信号通过天线以电磁波的形式向外发送。当有数据发送到终端设备时，射频电路通过天线接收到射频信号，将射频信号转换为基带信号，并将基带信号输出至处理器，处理器将基带信号转换为数据并对该数据进行处理。
本领域技术人员可以理解，为了便于说明，图6仅示出了一个存储器和处理器。在实际的终端设备中，可以存在多个处理器和存储器。存储器也可以称为存储介质或者存储设备等，本申请实施例对此不做限制。
例如，处理器可以包括基带处理器和中央处理器，基带处理器主要用于对通信协 议以及通信数据进行处理，中央处理器主要用于对整个终端设备进行控制，执行软件程序，处理软件程序的数据。图6中的处理器集成了基带处理器和中央处理器的功能，本领域技术人员可以理解，基带处理器和中央处理器也可以是各自独立的处理器，通过总线等技术互联。本领域技术人员可以理解，终端设备可以包括多个基带处理器以适应不同的网络制式，终端设备可以包括多个中央处理器以增强其处理能力，终端设备的各个部件可以通过各种总线连接。该基带处理器也可以表述为基带处理电路或者基带处理芯片。该中央处理器也可以表述为中央处理电路或者中央处理芯片。对通信协议以及通信数据进行处理的功能可以内置在处理器中，也可以以软件程序的形式存储在存储单元中，由处理器执行软件程序以实现基带处理功能。
示例性的，在本申请实施例中，可以将具有收发功能的天线和控制电路视为终端设备600的收发单元601，将具有处理功能的处理器视为终端设备600的处理单元602。如图6所示，终端设备600包括收发单元601和处理单元602。收发单元也可以称为收发器、收发机、收发装置等。可选的，可以将收发单元601中用于实现接收功能的器件视为接收单元，将收发单元601中用于实现发送功能的器件视为发送单元，即收发单元601包括接收单元和发送单元。示例性的，接收单元也可以称为接收机、接收器、接收电路等，发送单元可以称为发射机、发射器或者发射电路等。
图7为本申请实施例提供的一种网络设备的结构示意图，可以用于实现上述方法中的网络设备的功能。网络设备700包括一个或多个射频单元，如远端射频单元(remote radio unit，RRU)701和一个或多个基带单元(baseband unit，BBU)(也可称为数字单元，digital unit，DU)702。该RRU 701可以称为收发单元、收发机、收发电路、或者收发器等等，其可以包括至少一个天线和射频单元。该RRU 701部分主要用于射频信号的收发以及射频信号与基带信号的转换，例如用于向终端设备发送上述实施例中的信令消息。该BBU 702部分主要用于进行基带处理，对基站进行控制等。该RRU 701与BBU 702可以是物理上设置在一起，也可以物理上分离设置的，即分布式基站。
该BBU 702为基站的控制中心，也可以称为处理单元，主要用于完成基带处理功能，如信道编码，复用，调制，扩频等等。例如该BBU(处理单元)702可以用于控制基站执行上述方法实施例中关于网络设备的操作流程。
在一个示例中，该BBU 702可以由一个或多个单板构成，多个单板可以共同支持单一接入制式的无线接入网(如LTE系统，或5G系统)，也可以分别支持不同接入制式的无线接入网。该BBU 702还包括存储器和处理器。该存储器用以存储必要的指令和数据。例如存储器存储上述实施例中的码本等。该处理器用于控制基站进行必要的动作，例如用于控制基站执行上述方法实施例中关于网络设备的操作流程。该存储器和处理器可以服务于一个或多个单板。也就是说，可以每个单板上单独设置存储器和处理器。也可以是多个单板共用相同的存储器和处理器。此外每个单板上还可以设置有必要的电路。
在一种可能的实施方式中，随着片上系统(system-on-chip，SoC)技术的发展，可以将702部分和701部分的全部或者部分功能由SoC技术实现，例如由一颗基站功能芯片实现，该基站功能芯片集成了处理器、存储器、天线接口等器件，基站相关功能的程序存储在存储器中，由处理器执行程序以实现基站的相关功能。可选的，该基 站功能芯片也能够读取该芯片外部的存储器以实现基站的相关功能。
应理解，图7示例的网络设备的结构仅为一种可能的形态，而不应对本申请实施例构成任何限定。本申请并不排除未来可能出现的其他形态的基站结构的可能。
应理解，本申请实施例中，该处理器可以为中央处理单元(central processing unit，CPU)，该处理器还可以是其他通用处理器、数字信号处理器(digital signal processor，DSP)、专用集成电路(application specific integrated circuit，ASIC)、现成可编程门阵列(field programmable gate array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。
还应理解，本申请实施例中的存储器可以是易失性存储器或非易失性存储器，或可包括易失性和非易失性存储器两者。其中，非易失性存储器可以是只读存储器(read-only memory，ROM)、可编程只读存储器(programmable ROM，PROM)、可擦除可编程只读存储器(erasable PROM，EPROM)、电可擦除可编程只读存储器(electrically EPROM，EEPROM)或闪存。易失性存储器可以是随机存取存储器(random access memory，RAM)，其用作外部高速缓存。通过示例性但不是限制性说明，许多形式的随机存取存储器(random access memory，RAM)可用，例如静态随机存取存储器(static RAM，SRAM)、动态随机存取存储器(DRAM)、同步动态随机存取存储器(synchronous DRAM，SDRAM)、双倍数据速率同步动态随机存取存储器(double data rate SDRAM，DDR SDRAM)、增强型同步动态随机存取存储器(enhanced SDRAM，ESDRAM)、同步连接动态随机存取存储器(synchlink DRAM，SLDRAM)和直接内存总线随机存取存储器(direct rambus RAM，DR RAM)。
上述实施例，可以全部或部分地通过软件、硬件、固件或其他任意组合来实现。当使用软件实现时，上述实施例可以全部或部分地以计算机程序产品的形式实现。该计算机程序产品包括一个或多个计算机指令或计算机程序。在计算机上加载或执行该计算机指令或计算机程序时，全部或部分地产生按照本申请实施例的流程或功能。该计算机可以为通用计算机、专用计算机、计算机网络、或者其他可编程装置。该计算机指令可以存储在计算机可读存储介质中，或者从一个计算机可读存储介质向另一个计算机可读存储介质传输，例如，该计算机指令可以从一个网站站点、计算机、服务器或数据中心通过有线(例如红外、无线、微波等)方式向另一个网站站点、计算机、服务器或数据中心进行传输。该计算机可读存储介质可以是计算机能够存取的任何可用介质或者是包含一个或多个可用介质集合的服务器、数据中心等数据存储设备。该可用介质可以是磁性介质(例如，软盘、硬盘、磁带)、光介质(例如，DVD)、或者半导体介质。半导体介质可以是固态硬盘。
本申请实施例还提供了一种通信系统，该通信系统包括：上述的终端设备和上述的网络设备。
本申请实施例还提供了一种计算机可读介质，用于存储计算机程序代码，该计算机程序包括用于执行上述图3A、图3B和图4对应的方法中本申请实施例的定位方法的指令。该可读介质可以是只读存储器(read-only memory，ROM)或随机存取存储器(random access memory，RAM)，本申请实施例对此不做限制。
本申请还提供了一种计算机程序产品，该计算机程序产品包括指令，当该指令被执行时，以使得终端设备执行对应于上述方法的终端设备操作，或者，以使得网络设备执行对应于上述方法的网络设备的操作。
本申请实施例还提供了一种系统芯片，该系统芯片包括：处理单元和通信单元，该处理单元，例如可以是处理器，该通信单元例如可以是输入/输出接口、管脚或电路等。该处理单元可执行计算机指令，以使该通信装置内的芯片执行上述本申请实施例提供的任一种定位方法。
可选地，上述本申请实施例中提供的任意一种通信装置可以包括该系统芯片。
可选地，该计算机指令被存储在存储单元中。
可选地，该存储单元为该芯片内的存储单元，如寄存器、缓存等，该存储单元还可以是该终端内的位于该芯片外部的存储单元，如ROM或可存储静态信息和指令的其他类型的静态存储设备，RAM等。其中，上述任一处提到的处理器，可以是一个CPU，微处理器，ASIC，或一个或多个用于控制上述的主系统信息传输的方法的程序执行的集成电路。该处理单元和该存储单元可以解耦，分别设置在不同的物理设备上，通过有线或者无线的方式连接来实现该处理单元和该存储单元的各自的功能，以支持该系统芯片实现上述实施例中的各种功能。或者，该处理单元和该存储器也可以耦合在同一个设备上。
可以理解，本申请实施例中的存储器可以是易失性存储器或非易失性存储器，或可包括易失性和非易失性存储器两者。其中，非易失性存储器可以是只读存储器(read-only memory，ROM)、可编程只读存储器(programmable ROM，PROM)、可擦除可编程只读存储器(erasable PROM，EPROM)、电可擦除可编程只读存储器(electrically EPROM，EEPROM)或闪存。易失性存储器可以是随机存取存储器(random access memory，RAM)，其用作外部高速缓存。通过示例性但不是限制性说明，许多形式的随机存取存储器(random access memory，RAM)可用，例如静态随机存取存储器(static RAM，SRAM)、动态随机存取存储器(DRAM)、同步动态随机存取存储器(synchronous DRAM，SDRAM)、双倍数据速率同步动态随机存取存储器(double data rate SDRAM，DDR SDRAM)、增强型同步动态随机存取存储器(enhanced SDRAM，ESDRAM)、同步连接动态随机存取存储器(synchlink DRAM，SLDRAM)和直接内存总线随机存取存储器(direct rambus RAM，DR RAM)。
本文中术语“系统”和“网络”在本文中常被可互换使用。本文中术语“和/或”，仅仅是一种描述关联对象的关联关系，表示可以存在三种关系，例如，A和/或B，可以表示：单独存在A，同时存在A和B，单独存在B这三种情况。另外，本文中字符“/”，一般表示前后关联对象是一种“或”的关系。
本申请中出现的术语“上行”和“下行”，用于在特定场景描述数据/信息传输的方向，比如，“上行”方向一般是指数据/信息从终端向网络侧传输的方向，或者分布式单元向集中式单元传输的方向，“下行”方向一般是指数据/信息从网络侧向终端传输的方向，或者集中式单元向分布式单元传输的方向，可以理解，“上行”和“下行”仅用于描述数据/信息的传输方向，该数据/信息传输的具体起止的设备都不作限定。
在本申请中可能出现的对各种消息/信息/设备/网元/系统/装置/动作/操作/流程/概 念等各类客体进行了赋名，可以理解的是，这些具体的名称并不构成对相关客体的限定，所赋名称可随着场景，语境或者使用习惯等因素而变更，对本申请中技术术语的技术含义的理解，应主要从其在技术方案中所体现/执行的功能和技术效果来确定。
本领域普通技术人员可以意识到，本申请的实施例中的方法可以全部或部分地通过软件、硬件、固件或者其任意组合来实现。当使用软件实现时，可以全部或部分地以计算机程序产品的形式实现。所述计算机程序产品包括一个或多个计算机程序或指令。在计算机上加载和执行所述计算机程序或指令时，全部或部分地执行本申请实施例所述的流程或功能。所述计算机可以是通用计算机、专用计算机、计算机网络、或者其他可编程装置。所述计算机程序或指令可以存储在计算机可读存储介质中，或者通过所述计算机可读存储介质进行传输。所述计算机可读存储介质可以是计算机能够存取的任何可用介质或者是集成一个或多个可用介质的服务器等数据存储设备。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的系统、装置和单元的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
在本申请所提供的几个实施例中，应该理解到，所揭露的系统、装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，该单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。
所述功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(read-only memory，ROM)、随机存取。
以上所述，仅为本申请的具体实施方式，但本申请的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本申请揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本申请的保护范围之内。因此，本申请的保护范围应以所述权利要求的保护范围为准。

Claims (10)

  1. 一种定位方法，其特征在于，应用于无线通信系统的任一第一设备，所述无线通信系统包括：多个所述第一设备和第二设备，所述第二设备与每个所述第一设备均无线连接，任意两个所述第一设备之间均无线连接，所述方法包括：
    对所述第一设备与所述第二设备之间的通信信道的信道状态信息进行提取，确定所述第二设备的稀疏点和残差点；
    根据预先设置的算法对所述稀疏点对应的信息进行计算，得到第一位置信息，并通过预先训练的神经网络对所述残差点对应的信息进行计算，得到第二位置信息，所述第一位置信息和所述第二位置信息均用于表示所述第二设备的位置；
    根据所述第一位置信息和所述第二位置信息，确定所述第二设备的实际位置信息。
  2. 根据权利要求1所述的方法，其特征在于，在所述对所述第一设备与所述第二设备之间的通信信道的信道状态信息进行提取，确定所述第二设备的稀疏点和残差点之前，所述方法还包括：
    接收所述第二设备发送的参考信号；
    根据所述参考信号，确定所述信道状态信息。
  3. 根据权利要求1所述的方法，其特征在于，所述通过预先训练的神经网络对所述残差点对应的信息进行计算，得到第二位置信息，包括：
    将所述残差点对应的信息输入所述神经网络；
    根据所述神经网络中预先训练的算法以及预先训练的参数，对所述残差点对应的信息进行计算，得到所述第二位置信息。
  4. 根据权利要求1所述的方法，其特征在于，所述根据预先设置的算法对所述稀疏点对应的信息进行计算，得到第一位置信息，包括：
    根据所述稀疏点对应的信息，将所述稀疏点与历史特征点进行关联，所述历史特征点为已经提取得到的所述第二设备的稀疏点；
    通过预先设置的算法，结合所述稀疏点关联的历史特征点，对所述稀疏点对应的信息进行计算，得到所述第一位置信息。
  5. 根据权利要求1至4任一所述的方法，其特征在于，所述方法还包括：
    接收其他第二设备发送的所述第二设备的稀疏点和残差点；
    所述根据预先设置的算法对所述稀疏点对应的信息进行计算，得到第一位置信息，包括：
    根据预先设置的算法，对所述稀疏点对应的信息、以及其他所述第二设备发送的所述稀疏点对应的信息进行计算，得到所述第一位置信息；
    所述通过预先训练的神经网络对所述残差点对应的信息进行计算，得到第二位置信息，包括：
    通过预先训练的神经网络，对所述残差点对应的信息、以及其他所述第二设备发送的所述残差点对应的信息进行计算，得到所述第二位置信息。
  6. 根据权利要求1至5任一所述的方法，其特征在于，所述对所述第一设备与所述第二设备之间的通信信道的信道状态信息进行提取，得到稀疏点和残差点，包括：
    对所述信道状态信息进行提取，得到一维或多维频谱；
    对每个所述频谱进行检测，确定每个所述频谱中的稀疏点和残差点。
  7. 根据权利要求1至6任一所述的方法，其特征在于，所述方法还包括：
    基于所述第二位置信息，通过所述神经网络的损失函数，结合所述第一位置信息，对所述神经网络进行优化。
  8. 一种电子设备，其特征在于，包括：处理器，所述处理器用于运行存储器中存储的计算机程序，以使得所述电子设备实现如权利要求1至7中任一项所述的定位方法。
  9. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质存储有计算机程序，所述计算机程序被处理器执行时实现如权利要求1至7中任一项所述的定位方法。
  10. 一种芯片系统，其特征在于，所述芯片系统包括存储器和处理器，所述处理器执行所述存储器中存储的计算机程序，以实现如权利要求1至7中任一项所述的定位方法。
PCT/CN2022/117371 2022-09-06 2022-09-06 定位方法、电子设备、计算机可读存储介质及芯片系统 WO2024050701A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2022/117371 WO2024050701A1 (zh) 2022-09-06 2022-09-06 定位方法、电子设备、计算机可读存储介质及芯片系统

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2022/117371 WO2024050701A1 (zh) 2022-09-06 2022-09-06 定位方法、电子设备、计算机可读存储介质及芯片系统

Publications (1)

Publication Number Publication Date
WO2024050701A1 true WO2024050701A1 (zh) 2024-03-14

Family

ID=90192671

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/117371 WO2024050701A1 (zh) 2022-09-06 2022-09-06 定位方法、电子设备、计算机可读存储介质及芯片系统

Country Status (1)

Country Link
WO (1) WO2024050701A1 (zh)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111428819A (zh) * 2020-04-27 2020-07-17 北京工业大学 一种基于堆叠自编码网络和svm的csi室内定位方法
CN112469119A (zh) * 2021-02-03 2021-03-09 网络通信与安全紫金山实验室 定位方法、装置、计算机设备和存储介质
CN113015093A (zh) * 2021-01-29 2021-06-22 辽宁大学 一种基于三维深度残差神经网络的室内无线定位方法
CN113260044A (zh) * 2021-04-12 2021-08-13 北京邮电大学 基于双层字典学习的csi指纹定位方法、装置及设备
US20220196787A1 (en) * 2019-05-01 2022-06-23 The Regents Of The University Of California Wireless device localization

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20220196787A1 (en) * 2019-05-01 2022-06-23 The Regents Of The University Of California Wireless device localization
CN111428819A (zh) * 2020-04-27 2020-07-17 北京工业大学 一种基于堆叠自编码网络和svm的csi室内定位方法
CN113015093A (zh) * 2021-01-29 2021-06-22 辽宁大学 一种基于三维深度残差神经网络的室内无线定位方法
CN112469119A (zh) * 2021-02-03 2021-03-09 网络通信与安全紫金山实验室 定位方法、装置、计算机设备和存储介质
CN113260044A (zh) * 2021-04-12 2021-08-13 北京邮电大学 基于双层字典学习的csi指纹定位方法、装置及设备

Similar Documents

Publication Publication Date Title
Liu et al. Survey on CSI-based indoor positioning systems and recent advances
Xu et al. Pyramid: Real-time lora collision decoding with peak tracking
WO2020258106A1 (zh) 手势识别的方法和设备、定位追踪的方法和设备
US9810766B2 (en) Method and apparatus for modeling time relationships between clocks
CN109490826B (zh) 一种基于无线电波场强rssi的测距与位置定位方法
Xiong et al. Synchronicity: Pushing the envelope of fine-grained localization with distributed mimo
CA2903456A1 (en) Methods and systems for improving time of arrival determination
WO2016101656A1 (zh) 无线定位的方法
Liu et al. A research on CSI-based human motion detection in complex scenarios
EP3550890B1 (en) Method for recognizing line-of-sight path, and wireless device
CN114729982A (zh) 用于定位的方法及装置
Rahman et al. Lochunt: Angle of arrival based location estimation in harsh multipath environments
Zhang et al. A training-free multipath enhancement (TFME-RTI) method for Device-Free Multi-target localization
Li et al. Passive motion detection via mmWave communication system
US20230362039A1 (en) Neural network-based channel estimation method and communication apparatus
WO2024050701A1 (zh) 定位方法、电子设备、计算机可读存储介质及芯片系统
CN105589063B (zh) 基于偏度的脉冲无线电60GHz测距方法
Li et al. DeFe: Indoor localization based on channel state information feature using deep learning
WO2017049914A1 (zh) 一种终端定位方法、装置及系统
Yang et al. Mobi-Track: Distilling direct path in time domain for high accuracy WiFi tracking
CN109061562B (zh) 基于定位芯片内部参数的对象定位方法、装置和电子设备
Machaj et al. Impact of radio map simulation on positioning in indoor environtment using finger printing algorithms
Jiang et al. CS-DICT: Accurate indoor localization with CSI selective amplitude and phase based regularized dictionary learning
Wang et al. Multipath-Assisted Single-Anchor Localization via Deep Variational Learning
Polo et al. Machine Learning-based Inversion of Wireless Signals for Real-Time Gesture Recognition

Legal Events

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

Ref document number: 22957674

Country of ref document: EP

Kind code of ref document: A1