WO2023201549A1

WO2023201549A1 - Positioning method, model generation method, and device

Info

Publication number: WO2023201549A1
Application number: PCT/CN2022/087775
Authority: WO
Inventors: 沈渊; 李玉箫; 尤心; 卢前溪; 刘洋
Original assignee: Oppo广东移动通信有限公司; 清华大学
Priority date: 2022-04-19
Filing date: 2022-04-19
Publication date: 2023-10-26

Abstract

The present application relates to a positioning method, a model generation method, a device, a computer-readable storage medium, a computer program product, and a computer program. The method comprises: a first device receiving a wireless signal; the first device processing the wireless signal on the basis of a target model, so as to obtain a positioning parameter; and the first device determining positioning information of same on the basis of the positioning parameter.

Description

Positioning method, model generation method and equipment

Technical field

The present application relates to the field of communications, and more specifically, to a positioning method, a model generation method, a device, a computer-readable storage medium, a computer program product, and a computer program.

Background technique

In communication scenarios, electronic devices usually need to be positioned. In the related art, electronic devices are usually positioned through wireless signals from signal sources. However, how to position electronic devices more accurately is a problem that needs to be solved.

Contents of the invention

Embodiments of the present application provide a positioning method, a model generation method, a device, a computer-readable storage medium, a computer program product, and a computer program.

The embodiment of this application provides a positioning method, including:

The first device receives wireless signals;

The first device processes the wireless signal based on the target model to obtain positioning parameters;

The first device determines positioning information of the first device based on the positioning parameter.

The embodiment of this application provides a model generation method, including:

Use training samples to train the preset model to obtain the trained target model;

Wherein, the target model is used to process wireless signals to obtain positioning parameters, and the positioning parameters are used to determine positioning information.

The embodiment of the present application provides a first device, including:

communication unit for receiving wireless signals;

A processing unit, configured to process the wireless signal based on a target model to obtain positioning parameters; and determine positioning information of the first device based on the positioning parameters.

An embodiment of the present application provides an electronic device, including:

A training unit is used to train a preset model using training samples to obtain a trained target model; wherein the target model is used to process wireless signals to obtain positioning parameters, and the positioning parameters are used to determine positioning information.

An embodiment of the present application provides a first device, including a processor and a memory. The memory is used to store computer programs, and the processor is used to call and run the computer program stored in the memory, so that the first device performs the above method.

An embodiment of the present application provides an electronic device, including a processor and a memory. The memory is used to store computer programs, and the processor is used to call and run the computer programs stored in the memory, so that the electronic device performs the above method.

An embodiment of the present application provides a chip for implementing the above method.

Specifically, the chip includes: a processor, configured to call and run a computer program from a memory, so that the device installed with the chip executes the above method.

Embodiments of the present application provide a computer-readable storage medium for storing a computer program, which when the computer program is run by a device, causes the device to perform the above method.

An embodiment of the present application provides a computer program product, which includes computer program instructions, and the computer program instructions cause a computer to execute the above method.

An embodiment of the present application provides a computer program that, when run on a computer, causes the computer to perform the above method.

In the embodiment of the present application, by adopting the solution provided by this embodiment, when the first device receives the wireless signal, the wireless signal is processed through the target model to obtain positioning parameters, and then the positioning of the first device can be determined based on the positioning parameters. information. In this way, the parameters used for positioning in the wireless signal can be more accurately identified through the target model, thereby ensuring the accuracy of the final positioning information and ensuring processing efficiency.

Description of the drawings

Figure 1 is a schematic diagram of an application scenario according to an embodiment of the present application.

Figure 2 is a schematic flow chart of a positioning method according to an embodiment of the present application.

Figure 3 is a CIR schematic diagram of a wireless signal according to an embodiment of the present application.

Figure 4 is a schematic diagram of a distance-based positioning scenario according to an embodiment of the present application.

Figure 5 is a schematic diagram of the processing structure of the target model according to an embodiment of the present application.

Figure 6 is a schematic flow chart of a model generation method according to an embodiment of the present application.

Figure 7 is a schematic diagram of the processing structure of a preset model according to an embodiment of the present application.

FIG. 8 is a schematic diagram of wireless signals received under different environmental characteristics according to an embodiment of the present application.

Figure 9 is a schematic block diagram of a first device according to an embodiment of the present application.

FIG. 10 is a schematic block diagram of an electronic device according to an embodiment of the present application.

Figure 11 is a schematic block diagram of a communication device according to an embodiment of the present application.

Figure 12 is a schematic block diagram of a chip according to an embodiment of the present application.

Figure 13 is a schematic block diagram of a communication system according to an embodiment of the present application.

Detailed ways

The technical solutions in the embodiments of the present application will be described below with reference to the drawings in the embodiments of the present application.

The technical solutions of the embodiments of the present application can be applied to various communication systems, such as: Global System of Mobile communication (GSM) system, Code Division Multiple Access (Code Division Multiple Access, CDMA) system, broadband code division multiple access (Wideband Code Division Multiple Access, WCDMA) system, General Packet Radio Service (GPRS), Long Term Evolution (LTE) system, Advanced long term evolution (LTE-A) system , New Radio (NR) system, evolution system of NR system, LTE (LTE-based access to unlicensed spectrum, LTE-U) system on unlicensed spectrum, NR (NR-based access to unlicensed spectrum) unlicensed spectrum (NR-U) system, Non-Terrestrial Networks (NTN) system, Universal Mobile Telecommunication System (UMTS), Wireless Local Area Networks (WLAN), wireless fidelity (Wireless Fidelity, WiFi), fifth-generation communication (5th-Generation, 5G) system or other communication systems, etc.

Generally speaking, traditional communication systems support a limited number of connections and are easy to implement. However, with the development of communication technology, mobile communication systems will not only support traditional communication, but also support, for example, Device to Device, D2D) communication, Machine to Machine (M2M) communication, Machine Type Communication (MTC), Vehicle to Vehicle (V2V) communication, or Vehicle to everything (V2X) communication, etc. , the embodiments of the present application can also be applied to these communication systems.

In a possible implementation manner, the communication system in the embodiment of the present application can be applied to a carrier aggregation (Carrier Aggregation, CA) scenario, a dual connectivity (Dual Connectivity, DC) scenario, or an independent ( Standalone, SA) network deployment scenario.

In a possible implementation, the communication system in the embodiment of the present application can be applied to unlicensed spectrum, where the unlicensed spectrum can also be considered as shared spectrum; or, the communication system in the embodiment of the present application can also be applied to Licensed spectrum, where licensed spectrum can also be considered as unshared spectrum.

The embodiments of this application describe various embodiments in combination with network equipment and terminal equipment. The terminal equipment may also be called user equipment (User Equipment, UE), access terminal, user unit, user station, mobile station, mobile station, remote station, remote terminal, mobile device, user terminal, terminal, wireless communication equipment, user agent or user device, etc.

The terminal device can be a station (STAION, STA) in the WLAN, a cellular phone, a cordless phone, a Session Initiation Protocol (Session Initiation Protocol, SIP) phone, a wireless local loop (Wireless Local Loop, WLL) station, or a personal digital processing unit. (Personal Digital Assistant, PDA) devices, handheld devices with wireless communication capabilities, computing devices or other processing devices connected to wireless modems, vehicle-mounted devices, wearable devices, next-generation communication systems such as terminal devices in NR networks, or in the future Terminal equipment in the evolved Public Land Mobile Network (PLMN) network, etc.

In the embodiment of this application, the terminal device can be deployed on land, including indoor or outdoor, handheld, wearable or vehicle-mounted; it can also be deployed on water (such as ships, etc.); it can also be deployed in the air (such as aircraft, balloons and satellites). superior).

In the embodiment of this application, the terminal device may be a mobile phone (Mobile Phone), a tablet computer (Pad), a computer with a wireless transceiver function, a virtual reality (Virtual Reality, VR) terminal device, or an augmented reality (Augmented Reality, AR) terminal. Equipment, wireless terminal equipment in industrial control, wireless terminal equipment in self-driving, wireless terminal equipment in remote medical, wireless terminal equipment in smart grid , wireless terminal equipment in transportation safety, wireless terminal equipment in smart city, or wireless terminal equipment in smart home, etc.

As an example and not a limitation, in this embodiment of the present application, the terminal device may also be a wearable device. Wearable devices can also be called wearable smart devices. It is a general term for applying wearable technology to intelligently design daily wear and develop wearable devices, such as glasses, gloves, watches, clothing and shoes, etc. A wearable device is a portable device that is worn directly on the body or integrated into the user's clothing or accessories. Wearable devices are not just hardware devices, but also achieve powerful functions through software support, data interaction, and cloud interaction. Broadly defined wearable smart devices include full-featured, large-sized devices that can achieve complete or partial functions without relying on smartphones, such as smart watches or smart glasses, and those that only focus on a certain type of application function and need to cooperate with other devices such as smartphones. Use, such as various types of smart bracelets, smart jewelry, etc. for physical sign monitoring.

In the embodiment of this application, the network device may be a device used to communicate with mobile devices. The network device may be an access point (Access Point, AP) in WLAN, or a base station (Base Transceiver Station, BTS) in GSM or CDMA. , it can also be a base station (NodeB, NB) in WCDMA, or an evolved base station in LTE

(Evolutional Node B, eNB or eNodeB), or a relay station or access point, or a vehicle-mounted device, a wearable device, and a network device in an NR network (gNB) or a network device in a future evolved PLMN network or a network in an NTN network Equipment etc.

As an example and not a limitation, in the embodiment of the present application, the network device may have mobile characteristics, for example, the network device may be a mobile device. Optionally, the network device can be a satellite or balloon station. For example, the satellite can be a low earth orbit (LEO) satellite, a medium earth orbit (MEO) satellite, a geosynchronous orbit (geostationary earth orbit, GEO) satellite, a high elliptical orbit (High Elliptical Orbit, HEO) satellite ) satellite, etc. Optionally, the network device may also be a base station installed on land, water, etc.

In this embodiment of the present application, network equipment can provide services for a cell, and terminal equipment communicates with the network equipment through transmission resources (for example, frequency domain resources, or spectrum resources) used by the cell. The cell can be a network equipment ( For example, the cell corresponding to the base station), the cell can belong to the macro base station, or it can belong to the base station corresponding to the small cell (Small cell). The small cell here can include: urban cell (Metro cell), micro cell (Micro cell), pico cell ( Pico cell), femto cell (Femto cell), etc. These small cells have the characteristics of small coverage and low transmission power, and are suitable for providing high-rate data transmission services.

Figure 1 illustrates a communication system 100. The communication system includes a network device 110 and two terminal devices 120. In a possible implementation, the communication system 100 may include multiple network devices 110 , and the coverage of each network device 110 may include other numbers of terminal devices 120 , which is not limited in this embodiment of the present application.

In a possible implementation, the communication system 100 may also include other network entities such as a Mobility Management Entity (MME), an Access and Mobility Management Function (AMF), etc. The application examples do not limit this.

Among them, network equipment may include access network equipment and core network equipment. That is, the wireless communication system also includes multiple core networks used to communicate with access network equipment. The access network equipment can be a long-term evolution (long-term evolution, LTE) system, a next-generation (mobile communication system) (next radio, NR) system or authorized auxiliary access long-term evolution (LAA- Evolutionary base station (evolutional node B, abbreviated as eNB or e-NodeB) macro base station, micro base station (also known as "small base station"), pico base station, access point (access point, AP), Transmission point (TP) or new generation base station (new generation Node B, gNodeB), etc.

It should be understood that in the embodiments of this application, devices with communication functions in the network/system may be called communication devices. Taking the communication system shown in Figure 1 as an example, the communication equipment may include network equipment and terminal equipment with communication functions. The network equipment and terminal equipment may be specific equipment in the embodiments of the present application, which will not be described again here; the communication equipment also It may include other devices in the communication system, such as network controllers, mobility management entities and other network entities, which are not limited in the embodiments of this application.

It should be understood that the terms "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. In addition, the character "/" in this article generally indicates that the related objects are an "or" relationship.

It should be understood that the "instruction" mentioned in the embodiments of this application may be a direct instruction, an indirect instruction, or an association relationship. For example, A indicates B, which can mean that A directly indicates B, for example, B can be obtained through A; it can also mean that A indirectly indicates B, for example, A indicates C, and B can be obtained through C; it can also mean that there is an association between A and B. relation.

In the description of the embodiments of this application, the term "correspondence" can mean that there is a direct correspondence or indirect correspondence between the two, it can also mean that there is an associated relationship between the two, or it can mean indicating and being instructed, configuration and being. Configuration and other relationships.

In order to facilitate understanding of the technical solutions of the embodiments of the present application, the relevant technologies of the embodiments of the present application are described below. The following related technologies can be optionally combined with the technical solutions of the embodiments of the present application, and they all belong to the embodiments of the present application. protected range.

Figure 2 is a schematic flow chart of a positioning method according to an embodiment of the present application. The method includes at least part of the following.

S210. The first device receives wireless signals;

S220. The first device processes the wireless signal based on the target model to obtain positioning parameters;

S230. The first device determines the positioning information of the first device based on the positioning parameter.

In S210, the first device receives wireless signals, which may be: the first device receives N wireless signals; wherein the N wireless signals are sent by N second devices. N may be an integer greater than or equal to 1; preferably, N is an integer greater than or equal to 2.

The N wireless signals are sent by N second devices, which specifically means that there is a one-to-one correspondence between the wireless signals and the second devices. For example, the first wireless signal is sent by the first second device, The 2nd wireless signal is sent by the 2nd second device and so on.

Among the N wireless signals, the signal type of each wireless signal may be the same. For example, assuming that any wireless signal is the i-th wireless signal, the signal type of the i-th wireless signal may be one of the following: Bluetooth signal , WIFI signal, cellular network signal, Ultra Wide Band (UWB) signal, etc. Where, i is an integer greater than or equal to 1 and less than or equal to N. It should be understood that the foregoing signal types are only illustrative and do not limit all possible signal types of wireless signals. In fact, any signal type of wireless signals that can be used for positioning can be used in this embodiment. Within the scope of protection, just not exhaustive.

In this embodiment, the first device may be any electronic device, such as a terminal device. In this embodiment, the N second devices may be of the same device type. For example, the N second devices may be access network devices in a cellular network, such as base stations, eNBs, gNBs, etc.; or, the N second devices may be access points (APs) ( Or it can also be called an AP station (STA, Station)); or, the N second devices can be other devices that can send Bluetooth signals, such as devices set at a certain location, etc. It should be understood that any device that can send wireless signals of any of the aforementioned signal types can be the second device in this embodiment, and this embodiment does not exhaustively list all possible device types of the second device.

The target model includes a first sub-model and a second sub-model; wherein the first sub-model is used to process wireless signals to obtain location feature information; the second sub-model is used to process the location feature information. Process and obtain positioning parameters. Wherein, the target model may be a pre-trained model. The training process of the target model may be performed in the first device, or may be performed in other devices than the first device. The other devices may be other electronic devices, such as personal computers, laptops, servers, etc., which are not exhaustive here.

In S220, the first device processes the wireless signal based on the target model to obtain positioning parameters, which may include: the first device inputs the wireless signal into the first sub-model of the target model to obtain the positioning parameters. The location feature information output by the first sub-model is input into the second sub-model of the target model to obtain the positioning parameters output by the second sub-model.

As mentioned above, the wireless signal may specifically be N wireless signals. Correspondingly, the first device processes the wireless signal based on the target model to obtain positioning parameters, which specifically includes: the first device processes all the wireless signals. The j-th wireless signal among the N wireless signals is input to the first sub-model to obtain the j-th location feature information output by the first sub-model; where j is an integer greater than or equal to 1 and less than or equal to N; Input the j-th position feature information into the second sub-model to obtain the j-th positioning parameter output by the second sub-model.

The aforementioned j-th wireless signal may be any one of the N wireless signals. In this embodiment, all the N wireless signals are processed in the following manner, but will not be described in detail one by one. Correspondingly, the j-th positioning parameter is also It can be any one of all the positioning parameters finally obtained. Since the number of wireless signals received by the first device is N, N positioning parameters can be obtained through the foregoing processing.

In this embodiment, inputting the j-th wireless signal among the N wireless signals into the first sub-model of the target model may specifically refer to inputting the CIR (Channel Impulse Response) of the j-th wireless signal. Channel impulse response) is input into the first sub-model of the target model. Here, the CIR of the j-th wireless signal may be as shown in Figure 3, and may include the attenuation (or attenuation value) of the j-th wireless signal received by the first device after channel attenuation at different times in the time domain. Assume that the first device is represented as k, the j-th wireless signal is the wireless signal sent by the j-th second device, and the CIR of the j-th wireless signal received by the first device is r _kj (t). The r _kj (t) can be expressed by the following formula:

Among them, s _j (t) is the signal representation of the j-th second device transmitting the wireless signal, Y is the number of multipaths, y={1,2,...,Y} is the indication of different transmission links of the wireless signal,

and

are respectively the amplitude and delay corresponding to the signal of the transmission link from the y-th j-th second device to the first device k, and n(t) is Gaussian white noise.

Further, as the specific functions of the target model are different, the specific content of the aforementioned positioning parameters will also be different; accordingly, the first device determines the processing method of the positioning information of the first device based on the positioning parameters. will be different, let’s talk about them separately:

In one implementation, the target model is specifically used to process wireless signals to obtain the distance between the first device and the second device. Correspondingly, the N positioning parameters include: the distance between the first device and each of the N second devices.

As mentioned above, the first device can receive N wireless signals, specifically, it can receive N wireless signals respectively sent by N second devices. In processing the j-th wireless signal, the target model is specifically used to process the CIR of the j-th wireless signal to obtain the j-th distance between the first device and the j-th second device.

The positioning information of the first device may be determined based on the N distances and the position information of the N second devices. The above N may specifically be an integer greater than or equal to 3. Specifically, in the aforementioned S230, the first device determines the positioning information of the first device based on the positioning parameter, which may include: the first device determines the position of the first device based on the N distances and the N second devices. information to determine the positioning information of the first device. Taking the geographical coordinate system as an example, the positioning information of the first device may at least include: the longitude and latitude of the first device. The location information of the N second devices may at least include: the longitude and latitude of each of the N second devices. It should be understood that the above-mentioned positioning information and position information are described here only by taking the geographical coordinate system as an example. In actual processing, other coordinate systems can also be used, but this is not exhaustive.

At least some of the N second devices have longitudes and/or latitudes different from other second devices; that is, the N second devices are not on the same straight line.

This implementation may determine N possible location ranges of the first device based on the location information of the N second devices and the distances between the N second devices (ie, N source nodes) and the first device. , there is a unique intersection position among N possible position ranges, and this intersection position is used as the positioning information of the first device. For example, assuming that the first device is represented as k in the standard coordinate system, the distance d _kj from the first device to the j-th second device among the N second devices can be expressed as: d _kj =||p _k -p _j | |; Among them, p _k is the positioning information of the first device, and p _j is the position information of the j-th second device. When the aforementioned p _j is known, the possible position range of p _k can be determined by obtaining d _kj (which can be a circular range); and then based on the relationship between the other second devices among the N second devices and the first The distance between a device can determine other possible value ranges, and a unique intersection position can be determined based on N possible position ranges, and the intersection position can be used as the positioning information of the first device.

For example, N equals 3, the first distance refers to the distance between the first device and the first second device; the second distance refers to the distance between the first device and the second second device. The distance between devices; the third distance refers to the distance between the first device and the third second device. The first device determines the positioning information of the first device based on the three distances and the location information of the three second devices, as shown in Figure 4, specifically: the first device uses the first second device The first position range 401 is determined with the position information of the second device as the center o1 and the first distance as the radius r1; similarly, the position information of the second second device is used as the center o2 and the second distance is the radius r2 to determine the first position range 401. Two position ranges 402; determine the third position range 403 with the position information of the third second device as the center o3 and the third distance as the radius r3; combine the first position range 401 and the second position range 402 and the longitude and latitude of the unique intersection position of the third location range 403 as the positioning information of the first device.

In one implementation, the target model is specifically used to process wireless signals to obtain distance deviations. Correspondingly, the N positioning parameters are N distance deviations.

As mentioned above, the first device can receive N wireless signals, specifically, it can receive N wireless signals respectively sent by N second devices. In the processing of the jth wireless signal, the target model is specifically used to process the CIR of the jth wireless signal to obtain the jth distance deviation between the first device and the jth second device. .

The positioning information of the first device may be determined based on the N distance deviations, N initial distances, and the position information of the N second devices. The above N may specifically be an integer greater than or equal to 3.

Before performing the aforementioned S230, it may also include: the first device determining the N initial distances between the first device and the N second devices based on the N wireless signals. Wherein, the first device may process each of the N wireless signals separately. Specifically, the first device determines the jth wireless signal among the N wireless signals. The j-th initial distance between the first device and the j-th second device among the N second devices. That is to say, the j-th initial distance may be estimated by the first device based on the CIR of the j-th wireless signal. Here, the definition and expression of the CIR of any wireless signal are the same as in the previous embodiments and will not be described again; the method of estimating the j-th initial distance based on the CIR of the j-th wireless signal can be set according to the actual situation. For example, a first reach path estimation algorithm may be used, which is not limited in this embodiment.

The aforementioned S230, the first device determines the positioning information of the first device based on the positioning parameters, may include: the first device corrects the N initial distances based on the N distance deviations to obtain N adjustments. based on the N adjusted distances and the position information of the N second devices, determine the positioning information of the first device.

Any distance deviation among the N distance deviations may be a positive number, a negative number, or zero.

The first device determines N adjusted distances based on the N distance deviations and N initial distances. Specifically, the first device determines N adjusted distances based on the jth distance deviation among the N distance deviations and The j-th initial distance among the N initial distances is calculated, and the j-th adjusted distance among the N adjusted distances is obtained.

For example, assuming that the j-th initial distance is the initial distance between the j-th second device and the first device k, it is expressed as

The jth distance deviation is expressed as b _kj , then the jth adjusted distance d _kj can be calculated using the following formula:

In this embodiment, the positioning information of the first device and the position information of the N second devices are the same as those in the previous embodiment, and will not be repeated.

For example, N is equal to 3. The first initial distance among the three initial distances refers to the initial distance between the first device and the first second device. Subtract the first initial distance from the first initial distance. distance deviation, the first adjusted distance is obtained; the second initial distance refers to the initial distance between the first device and the second second device, and the second initial distance is subtracted from the second distance deviation, the second adjusted distance is obtained; the third initial distance refers to the initial distance between the first device and the third second device, and the third initial distance is subtracted from the third distance deviation to obtain the third adjusted distance.

Regarding the specific processing method for the first device to determine the positioning information of the first device based on the three adjusted distances and the position information of the three second devices, it is different from the above-mentioned method based on the three distances and the three second devices. The specific processing methods for determining the location information of the second device and the location information of the first device are similar, so repeated descriptions will not be made.

In one implementation, the target model is specifically used to process wireless signals to obtain the RSS value of the wireless signals. Correspondingly, the N positioning parameters are N signal received strength RSS (Received Signal Strength) values.

As mentioned above, the first device can receive N wireless signals, specifically, it can receive N wireless signals respectively sent by N second devices. In the processing of the j-th wireless signal, the target model is specifically used to process the CIR of the j-th wireless signal to obtain the RSS value of the j-th wireless signal received by the first device.

The positioning information of the first device may be determined based on the N RSS values and the signal feature library. Specifically, in the aforementioned S230, the first device determines the positioning information of the first device based on the positioning parameters, which may include: the first device determines the positioning information of the first device based on N positioning parameters and a signal feature library. The positioning information; wherein, the signal feature library contains the feature quantity of each reference position in a plurality of reference locations; the feature quantity of each reference position includes the reference signal of each second device in the N second devices. RSS reference value.

It should be pointed out that the above-mentioned signal feature database can also be called a fingerprint database, or a fingerprint database, or a signal fingerprint information database, etc., and all possible names are not exhaustive here.

The multiple reference locations may refer to multiple locations within the area where the first device is currently located; where the area where the first device is currently located may refer to indoors or outdoors, which is not performed in this embodiment. limited. Each reference position among the multiple reference positions can be represented by a geographical coordinate system. For example, the signal feature library contains 10 reference positions, and the first reference position is represented as (longitude 1, latitude 1), The second reference position is expressed as (longitude 2, latitude 2), and so on without being exhaustive; it should be understood that each reference position can also be represented by other coordinate systems, but this is not exhaustive here.

The above-mentioned multiple reference positions may be multiple grid reference positions. Each grid reference position can be represented by the center point of each grid reference position; or, each grid reference position can be represented by at least two coordinate points (such as the upper left coordinate point of the grid, the lower right coordinate point of the grid Coordinate points). Taking each coordinate point as an example, it can be represented by a geographical coordinate system. Assume that the first grid reference position in the signal feature library is expressed as {(longitude 11, latitude 11), (longitude 12, latitude 12)}, and the second The grid reference position is expressed as {(longitude 21, latitude 21), (longitude 22, latitude 22)}, and so on. This is not exhaustive.

It should also be pointed out that the RSS reference value of the reference signal of each second device mentioned above may be the RSS value of the wireless signal of each second device measured at each reference position when constructing the signal feature library.

Specifically, the N positioning parameters are N RSS values; correspondingly, the first device determines the positioning information of the first device based on the N positioning parameters and the signal feature library, which refers to: the first device The device determines the positioning information of the first device based on the N RSS values and the signal feature library. The above N may specifically be an integer greater than or equal to 2.

The first device determines the positioning information of the first device based on the N RSS values and the signal feature database. Specifically, the first device may: combine the N RSS values with the characteristics of each reference position in the signal feature database. The reference position where the feature quantity matching the N RSS values is located is used as the target position, and the target position is used as the positioning information of the first device.

Among them, whether the N RSS values match the feature quantities at each reference position can be determined by using any one of the NN (Nearest Neighbor, nearest neighbor) algorithm, NNN (N-Nearest Neighbor, N nearest neighbor) algorithm, Bass algorithm, etc. To achieve this, we will not exhaustively list all possible algorithms here.

Taking the NN algorithm as an example, whether the N RSS values match the feature quantities at each reference position can be determined using Euclidean distance. For example, if N is equal to 3, similarity matching is performed between the three RSS values and the feature quantity of each reference position in the signal feature library (for example, including three RSS reference values); the reference position with the smallest Euclidean distance between the two is regarded as The positioning information of the first device. For example, the three RSS values obtained by the first device are expressed as (RSS 1, RSS 2, RSS 3). There is a reference position A (coordinates = (longitude 11, latitude 11)) in the signal feature database, and the corresponding RSS reference value = (RSS 1, RSS 2, RSS 4)); reference position B (coordinates = (longitude 21, latitude 21), corresponding RSS reference value = (RSS 3, RSS 5, RSS 4)). Through comparison, it can be obtained that the three RSS values are most similar to the feature quantities corresponding to the reference location A (that is, the RSS reference value). Therefore, it can be determined that the positioning information of the first device is equal to the coordinates of the reference location A (longitude 11, latitude 11).

In one implementation, the target model is specifically used to process wireless signals to obtain RSS deviations. Correspondingly, the N positioning parameters are N RSS deviations.

The positioning information of the first device may be determined based on the N RSS values and the signal feature library. The aforementioned S230, the first device determines the positioning information of the first device based on the positioning parameters, may include: the first device determines the positioning information of the first device based on N positioning parameters and a signal feature library. ; Wherein, the signal feature database contains the feature quantity of each reference position among the plurality of reference positions; the feature quantity of each reference position includes the RSS reference value of the reference signal of each of the N second devices. .

Specifically, the first device determining the positioning information of the first device based on the N positioning parameters and the signal feature library may include: the first device based on the N RSS deviations and N initial intensity estimation values, Determine N adjusted RSS values; determine the positioning information of the first device based on the N adjusted RSS values and the signal feature library.

Any one of the N initial strength estimate values, such as the j-th initial strength estimate value, may be the first device's received strength of the j-th wireless signal among the N wireless signals. Measured. The aforementioned j-th initial intensity estimate value may specifically be the j-th initial RSS value.

Any RSS deviation among the N RSS deviations may be a positive number, a negative number, or zero.

The first device determines N adjusted RSS values based on the N RSS deviations and N initial intensity estimation values. Specifically, the first device determines N adjusted RSS values based on the jth of the N RSS deviations. The RSS deviation is calculated with the j-th initial intensity estimate value among the N initial intensity estimate values, and the j-th adjusted RSS value among the N adjusted RSS values is obtained.

The content that may be included in the foregoing signal feature database is the same as that of the foregoing implementation, and will not be repeated here.

Determining the positioning information of the first device based on the N adjusted RSS values and the signal feature library may be based on the N adjusted RSS values and each reference position in the signal feature library. The feature quantities are compared, the reference location where the feature quantity matching the N adjusted RSS values is located is used as the target location, and the positioning information of the first device is used as the target location. Here, the specific comparison method is similar to the previous embodiment, and repeated description will not be repeated.

In one implementation, the target model is specifically used to process wireless signals to obtain the reception angle of the wireless signals by the first device. The N positioning parameters are N receiving angles of the first device for each of the N wireless signals.

Wherein, the reception angle at which the first device receives any one of the N wireless signals may be the angle between any one of the wireless signals received by the first device and the equivalent center point of its own antenna array relative to the reference direction. angle. Wherein, the antenna array of the first device may be an antenna array composed of two or more antennas. The reference direction may be preset. Specifically, it may be the same reference direction preset in each device in the system. For example, it may be true north, true south, etc., and no exhaustive list will be made here.

As mentioned above, the first device can receive N wireless signals, specifically, it can receive N wireless signals respectively sent by N second devices. In the processing of the j-th wireless signal, the target model is specifically used to process the CIR of the j-th wireless signal to obtain the j-th reception angle between the first device and the j-th second device. .

The positioning information of the first device may be determined based on the N receiving angles and the position information of the N second devices. The above N may specifically be an integer greater than or equal to 3.

Specifically, in the aforementioned S230, the first device determines the positioning information of the first device based on the positioning parameters, which may include: the first device determines the positioning information of the first device based on the N receiving angles and the N second devices. Location information determines the positioning information of the first device.

For example, N is equal to 3, and the first receiving angle among the three receiving angles refers to the receiving angle of the first wireless signal sent by the first device to the first second device; the second receiving angle refers to is the receiving angle of the second wireless signal sent by the first device to the second second device; the third receiving angle refers to the third receiving angle of the first device sent by the third second device The reception angle of wireless signals. The first device determines the positioning information of the first device based on the three receiving angles and the position information of the three second devices, specifically:

The first device uses the first receiving angle and the second receiving angle to determine the angle 1 between the first second device, the second second device and the first device; similarly, the first device The device uses the second receiving angle and the third receiving angle to determine the angle 2 between the second second device, the third second device and the first device; the first device uses the third receiving angle , the first receiving angle is determined by the angle 3 between the first second device, the third second device and the first device;

Based on the included angle 1 and the position of the first second device and the position of the second second device, determine the first circular range; based on the included angle 2 and the position of the second second device and the second second device Determine the second circular range based on the position of the device; determine the third circular range based on the included angle 3 and the position of the first second device and the third second device; convert the first circular range The position of the intersection point of the range, the second circular range and the third circular range is used as the positioning information of the first device.

In one implementation, the target model is specifically used to process wireless signals to obtain the angle deviation at which the first device receives the wireless signal; correspondingly, the N positioning parameters are N angle deviations.

The positioning information of the first device may be determined based on the N angular deviations, N initial receiving angles, and the position information of the N second devices. The above N may specifically be an integer greater than or equal to 3. Specifically, in the aforementioned S230, the first device determines the positioning information of the first device based on the positioning parameters, which may include: the first device determines the positioning information of the first device based on N positioning parameters and the position information of the N second devices. , determine the positioning information of the first device. Specifically: the first device corrects the N initial receiving angles based on the N angular deviations to obtain N adjusted angles; based on the N adjusted angles and the N second devices Location information determines the positioning information of the first device.

Any one of the N initial reception angles, such as the j-th initial reception angle, can be estimated by the first device based on the j-th wireless signal among the N wireless signals; here, it can is the incident angle of the j-th wireless signal determined by the first device based on its own antenna array. Wherein, the antenna array of the first device may be an antenna array composed of two or more antennas.

Any one of the N angle deviations may be a positive number, a negative number, or zero.

The first device determines N adjusted angles based on the N angular deviations and N initial receiving angles. Specifically, the first device determines N adjusted angles based on the jth angular deviation among the N angular deviations. Calculate with the j-th initial receiving angle among the N initial receiving angles to obtain the j-th adjusted angle among the N adjusted angles. Here, the calculation can be set to addition or subtraction according to the actual situation.

Exemplarily, N is equal to 3. The first angular deviation among the 3 angular deviations refers to the angular deviation between the first wireless signal sent by the first device and the first second device. 3 initial receptions The first angle estimate in the angle refers to the angle estimate between the first device and the first wireless signal sent by the first second device; the second angle deviation refers to the angle between the first device and the first wireless signal sent by the second device. The angle deviation between the first device and the second wireless signal sent by the second second device. The second angle estimate refers to the angle between the first device and the second wireless signal sent by the second second device. The angle estimate value; the third angle deviation refers to the angle deviation between the third wireless signal sent by the first device and the third second device, and the third angle estimate value refers to the angle deviation of the third wireless signal sent by the third second device. An estimated value of the angle between a device and a third wireless signal sent by a third second device.

The specific processing method by which the first device determines the positioning information of the first device based on the three adjusted angles and the position information of the three second devices is similar to the previous embodiment and will not be described again here.

It should be understood that in the foregoing embodiments, N equals 3 is only an example. In actual processing, the number of N can be adjusted according to the accuracy requirements, for example, it can be 4 or 5, or more or less, which is not done here. Repeat.

It should be noted that in the various aforementioned embodiments, the processing instructions based on the N positioning parameters each include one of the following: the distance between the first device and each of the N second devices; The receiving angle of each of the N wireless signals by the first device; N signal reception strength RSS values; N distance deviations; N angle deviations; and N RSS deviations. However, in actual processing, each of the N positioning parameters may contain two or more of the above contents. More possible implementations are described below:

In a possible implementation, the j-th positioning parameter among the N positioning parameters may include the distance between the first device and the j-th second device, the response of the first device to the j-th wireless signal. Reception angle; that is to say, the target model processes one wireless signal at a time and can obtain the above distance and the above reception angle at the same time.

In this embodiment, N may be 1. That is to say, the first device determines the positioning information of the first device based on a positioning parameter and a position information of the second device. Specifically, the first device determines the transmission angle at which the second device transmits the wireless signal based on the receiving angle of the wireless signal by the first device and the reference direction included in the positioning parameter; based on the location information of the second device , the emission angle and the distance between the first device and the second device, a target position can be determined, and the target position can be used as the positioning information of the first device. The reference direction can be the same reference direction set by all devices in the system, such as true north. Of course, it can also be set as true south. As long as the reference directions set by each device in the system are the same, the reference direction can be set to the same reference direction. within the protection scope of this embodiment.

It should be understood that in this embodiment, the distance between the first device and the second device can also be replaced by a distance deviation, and/or the reception angle of the wireless signal by the first device can be replaced by an angle deviation. For example, when the distance between the first device and the second device is replaced by a distance deviation, the specific processing may be: the first device based on the receiving angle of the wireless signal by the first device included in the positioning parameter, and The reference angle determines the transmission angle at which the second device transmits the wireless signal; based on calculation based on the initial distance between the first device and the second device and the aforementioned distance deviation, the distance between the first device and the second device can be obtained The adjusted distance; based on the position information of the second device, the launch angle and the adjusted distance, a target position can be determined, and the target position can be used as the positioning information of the first device.

Regarding the reception angle of the wireless signal by the first device, the processing of the angle deviation is similar to the above and will not be repeated.

It should be pointed out that as long as the positioning parameters output by the target model include at least one of the foregoing contents, they are all within the protection scope of this embodiment, but the various possible implementation methods are not exhaustive here.

In one implementation, the target model further includes a third sub-model. The first sub-model of the target model is also used to process the wireless signal to obtain environmental feature information; the target model also includes a third sub-model, and the third sub-model is used to process the environmental feature information. Perform processing to obtain scene type information.

For example, see Figure 5. Figure 5 illustrates that the target model 500 includes a first sub-model 501, a second sub-model 502, and a third sub-model 503; among which, the first sub-model 501 is used to process the wireless signal 510. , obtain the location feature information 511 and environment feature information 512 of the wireless signal 510; the second sub-model 502 is used to process the location feature information 511 to obtain the positioning parameters 521; the third sub-model 503 is used to process the environment feature information 512 Process to obtain scene type information 522.

Specifically, the scene type information may include probability values of multiple scene types. The number of the multiple scene types can be determined according to the actual situation, for example, it can be 3, 6, 7 or more or less, which is not limited in this embodiment. For example, the scene type information includes: {scene type 1: probability value 1, scene type 2: probability value 2, scene type 3: probability value 3}.

Among them, the specific meanings represented by various scene types can be set according to the actual situation. For example, they can include: indoor, outdoor, and NLOS (non line of sight, non-line of sight transmission) obstacle types (such as metal obstacles, plastic obstacles objects, etc.), and LOS (line of sight, line of sight transmission), which is the type of scene without obstacles. Different scene types can be represented by preset values. Assume that the scene type of the environment where the first device k is located is represented by l _k . The values of different scene types can represent different obstacle information, or represent different locations. Environmental information; for example, l _k =0 indicates that the scene type is indoor, l _k =1 indicates that the scene type is outdoor; for another example, it may also include: l _k =00 indicates that the scene type is indoor without obstructions, l _k =01 indicates that The scene type is indoor with wooden shields, l _k =10 means the scene type is outdoor without shields, l _k =11 means the scene type is outdoor with metal shields, etc. It should be understood that this is only an illustrative explanation, and it does not mean that there are only the above scene types. More scene types can be set according to the actual situation, such as plastic shields indoors, glass shields outdoors, etc. We are no longer exhausted here. Lift.

Correspondingly, the method further includes: the first device determining the environment in which the first device is located based on the scene type information output by the third sub-model.

It has been pointed out in the foregoing embodiments that the first device may receive N wireless signals, and N may be greater than or equal to 1.

In one example, N is equal to 1, that is, when the first device only receives one wireless signal, the location of the first device can be determined directly based on the scene type information output by the third sub-model. environment. It is assumed that the scene type information obtained at one time can include probability values of multiple scene types, specifically {l _k =00: probability value 0.1, l _k =01: probability value 0.7, l _k =10: probability value 0.2, l _k = 11: Probability value 0.3}, then the scene type l _k =01 with the largest probability value can be used as the environment where the first device is located. For example, l _k =01 means there is a wooden obstruction indoors, then the location of the first device can be determined. The environment is indoors and there are wooden shelters.

In one example, N is greater than 1, that is, when the first device receives multiple wireless signals, the first sub-model of the target model is also used to calculate the i-th wireless signal among the N wireless signals. The signal is processed to obtain the i-th environmental feature information corresponding to the i-th wireless signal; the target model also includes a third sub-model, and the third sub-model is used to perform processing on the i-th environmental feature information. Process to obtain the i-th scene type information. The i-th scene type information specifically includes probability values corresponding to multiple scene types.

Correspondingly, the first device determines that the environment in which the first device is located based on the scene type information output by the third sub-model includes: the N information output by the first device based on the third sub-model. Scene type information determines the environment in which the first device is located.

Specifically, in the N scene type information, if the scene type with the largest probability value is the same, the first device uses the scene type with the largest probability value as the environment in which the first device is located;

Or, in the case where the scene types with the largest probability values in the N scene type information are partially different, the scene type with the largest number of the same scene type with the largest probability value in the N scene type information is used as the first device. environment.

Or, when the scene types with the highest probability values corresponding to the N scene type information are all different, the scene type with the highest probability value among them is used as the environment in which the first device is located.

Of course, in the above case where there is multiple scene type information, the description of determining the environment where the first device is located is only an exemplary description, and other determination methods may be used in actual processing. This embodiment is not exhaustive, as long as The environment in which the first device is located can be determined based on multiple scene type information, which is within the protection scope of this embodiment.

It can be seen that by adopting the above solution, when the first device receives the wireless signal, the wireless signal can be processed through the target model to obtain positioning parameters, and then the positioning information of the first device can be determined based on the positioning parameters. In this way, the characteristic information used for positioning in the wireless signal can be more accurately identified through the target model, thereby ensuring the accuracy of the final positioning information and ensuring processing efficiency.

Figure 6 is a schematic flow chart of a model generation method according to an embodiment of the present application. The method includes at least part of the following.

S610. Use the training samples to train the preset model to obtain the trained target model;

The model generation method provided in this embodiment can be applied to an electronic device, which may be the same as or different from the first device that performs the positioning method in the foregoing embodiment. In the case where the electronic device that executes the model generation method is the same as the first device that executes the positioning method, the target model may be directly used to execute the positioning method after the first device completes the training of the preset model to obtain the target model. . In the case where the electronic device that executes the model generation method is different from the first device that executes the positioning method, it may be that after the electronic device that executes the model generation method completes the training of the preset model to obtain the target model, the target model (specifically, (parameters of the first sub-model, the second sub-model and the third sub-model in the target model) are sent to the first device that executes the positioning method, so that the first device that executes the positioning method executes the aforementioned positioning method based on the received target model. .

The above-mentioned training samples may include: a wireless training signal, a first preset label corresponding to the wireless training signal, and a second preset label corresponding to the wireless training signal. Here, the training sample may be any one of multiple training samples included in the training data set.

The first preset tag corresponding to the wireless training signal includes at least one of the following: a distance tag corresponding to the wireless training signal; an angle tag corresponding to the wireless training signal; an RSS tag corresponding to the wireless training signal; The distance deviation tag corresponding to the wireless training signal; the angle deviation tag corresponding to the wireless training signal; the RSS deviation tag corresponding to the wireless training signal.

In actual use, only one of the above-mentioned first preset tags can be used according to the training target selection. For example, the training target is to obtain a target model that can obtain the distance positioning parameter based on the received wireless signal processing. Then the distance label corresponding to the wireless training signal can be used as the first preset label. Of course, you can also use multiple of them according to actual needs. For example, the training goal is to obtain a target model that can obtain distance and angle based on the received wireless signal processing, and use the distance and angle as positioning parameters, then wireless training can be used The distance label and angle label corresponding to the signal are used as the first preset label. The above is only an illustrative description, and does not mean that only the above methods can be used in actual use. The specific content of the first preset label can be flexibly selected and set according to actual needs.

The second preset tag corresponding to the wireless training signal includes: a scene type tag corresponding to the environment in which the wireless training signal is located.

The scene type tag contained in the above-mentioned second preset tag may specifically be a preset value corresponding to the scene type. For example, 0 indicates that the scene type is indoor, 1 indicates that the scene type is outdoor, etc. This is only an exemplary explanation. The preset values that do not represent the scene type are only the above two possibilities. As the scene types are divided into more detailed , there may be more preset value possibilities, and this embodiment does not list them exhaustively.

The above-mentioned wireless training signal acquisition method can be: placing the signal source node and the target node at any two locations, and recording the received signal of the target node. Wherein, placing the signal source node and the target node at any two locations may mean placing the signal source node and the target node at any two locations within any area. Here, the signal source node may be a device capable of sending wireless signals of any one signal type among Bluetooth signals, WIFI signals, cellular network signals, and ultra-wideband (Ultra Wide Band, UWB) signals; accordingly, the target node It can be a device capable of receiving wireless signals of any of the above signal types.

While acquiring the wireless training signal, the first preset label and the second preset label may be set based on the relevant information of the signal source node and the target node. For example, if there is an obstruction between the signal source node and the target node, the scene type can be determined according to the material of the obstruction, and then the scene type label (ie, the aforementioned second preset label) can be determined; the materials of the aforementioned obstruction include: Not limited to glass, wood, plastic, metal, etc. For another example, the distance label in the first preset label is determined based on the distance between the signal source node and the target node; or the aforementioned first preset label is determined based on the angle between the signal source node and the target node. or, based on the RSS value measured by the target node, determine the RSS label in the first preset label. For example, the distance deviation can also be determined based on the distance between the signal source node and the target node and the initial distance estimated by the current target node, that is, the distance deviation label in the aforementioned first preset label; or, based on the signal source The angle between the node and the target node, and the initial angle estimated by the current target node, determine the angle deviation, that is, the angle deviation label in the first preset label; or, based on the RSS value measured by the target node and the actual RSS value to determine the RSS deviation label in the aforementioned first preset label.

The training samples can be obtained through the above processing. The following is a detailed description of how to train the preset model based on the training samples:

The use of training samples to train the preset model includes:

Input the wireless training signal in the training sample into the first preset sub-model of the preset model to obtain positioning feature prediction information and scene feature prediction information output by the first preset sub-model;

Input the positioning feature prediction information and the scene feature prediction information into the fourth preset sub-model in the preset model to obtain the reconstructed signal of the wireless training signal output by the fourth preset sub-model. ;

Input the positioning feature prediction information into the second preset sub-model in the preset model, obtain the positioning estimation parameters output by the second preset sub-model, and input the scene feature prediction information into the preset model. The third preset sub-model is used to obtain the scene type estimate output by the third preset sub-model;

Determine a loss function based on at least one of the reconstructed signal, the positioning estimation parameter, the scene type estimate, the first preset label, and the second preset label;

Based on the loss function, reverse conduction updates at least one of the first preset sub-model, the second preset sub-model, the third preset sub-model and the fourth preset sub-model.

The preset model may include a first preset sub-model 701, a second preset sub-model 702, a third preset sub-model 703, and a fourth preset sub-model 704, as shown in FIG. 7 . It can be seen from Figure 7 that the wireless training signals in the training samples can be input into the first preset sub-model 701 in the preset model, and the positioning feature prediction information and scene feature prediction information can be obtained through the first preset sub-model 701; The inputs of the four preset sub-models 704 are positioning feature prediction information and scene feature prediction information, and the output of the fourth preset sub-model is the reconstructed signal; the input of the second preset sub-model 702 is the positioning feature prediction information, and the output information is the predicted positioning estimation parameter; the input of the third preset sub-model 703 is the scene feature prediction information, and the output is the predicted scene type estimate.

The above positioning estimation parameters may include at least one of the following: distance prediction value, angle prediction value, RSS prediction value, distance deviation prediction value, angle deviation prediction value, and RSS deviation prediction value.

The above-mentioned positioning estimation parameters may vary with different goals of this training, and the positioning estimation parameters correspond to the specific content of the aforementioned first preset label. For example, the goal of this training is to obtain a target model that can obtain the positioning parameter of distance based on the received wireless signal processing. Then the distance label corresponding to the wireless training signal can be used as the first preset label. Similarly, the positioning estimation parameter is distance. Predictive value. For example, the goal of this training is to obtain a target model that can obtain the positioning parameter of the RSS value based on the received wireless signal processing, then the RSS value label corresponding to the wireless training signal can be used as the first preset label. Similarly, the positioning estimation parameter is the RSS predicted value. The goal of this training is to obtain a target model that can obtain the distance and angle based on the received wireless signal processing, and use the distance and angle as positioning parameters. Then the distance label and angle label corresponding to the wireless training signal can be used as the first preset Labels, similarly, positioning estimation parameters can include distance and angle. The above is only an illustrative explanation. It does not mean that only the above methods can be used in actual use. You can set the parameters of each sub-model in the preset model according to actual needs, so that the positioning estimation parameters output by the preset model are the ones required for this training. required type or content.

The training of the preset model can be processed in multiple loops. The following is a description of the implementation methods that can be included in any loop processing:

In one implementation, the loss is determined based on at least one of the reconstructed signal, the positioning estimation parameter, the scene type estimate, the first preset label and the second preset label. A function includes: determining a first loss function based on the reconstructed signal, the wireless training signal, the positioning estimation parameter, and the scene type estimation value.

For example, the transformation process of the first preset sub-network is expressed as f _θ , and the change process of the fourth preset sub-network is expressed as f _φ ; where θ and φ represent the first preset sub-network and the fourth preset sub-network. Parameters of the subnetwork (or network parameters). Denote the input wireless training signal as r, the location feature prediction information z _p and the scene feature prediction information z _l output by the first preset sub-network. The fourth preset sub-network outputs the reconstructed received signal

Specifically expressed as

Then the first loss function is expressed as:

L _AE (θ,φ; r)=||rf _φ (f _θ (r))|| ² +||z _p || ² +‖z _l ‖ ² ;

Among them, L _AE (θ,φ; r) is the first loss function; rf _φ (f _θ (r)) is

‖.‖ means calculating the absolute value.

The step of updating at least one of the first preset sub-model, the second preset sub-model, the third preset sub-model and the fourth preset sub-model based on the loss function includes: based on the first preset sub-model. A loss function is used to conduct reverse conduction to update the parameters of the first preset sub-model and the parameters of the fourth preset sub-model. The above example is also used to illustrate, specifically, the aforementioned θ and φ are updated based on the first loss function reverse conduction.

Determining the loss function based on at least one of the reconstructed signal, the positioning estimation parameter, the scene type estimation value, the first preset label and the second preset label further includes: based on The positioning estimation parameter and the first preset label determine a second loss function; and based on the scene type estimate and the second preset label, a third loss function is determined.

For example, assume that the first preset label is distance deviation; accordingly, the positioning estimation parameter includes a distance deviation estimate. Assume that the transformation relationship of the second preset sub-model is expressed as

in

are the parameters of the second preset sub-model (or network parameters); the input position feature prediction information of the second preset sub-model is expressed as z _p , and the distance deviation estimate is output after passing through the second preset sub-model.

Right now

Compare the distance deviation estimate with the distance deviation b contained in the first preset label to calculate the second loss function (for example, expressed as

The second loss function term is specifically:

Assume that the transformation relationship of the third preset sub-model is expressed as

in

are the parameters of the third preset sub-model (or network parameters); the input of the third preset sub-model is scene feature prediction information z _l , and the scene type estimate is output after passing through the third preset sub-model.

Right now

Compare the scene type estimate with the scene type label l contained in the second preset label, and calculate the third loss function

Specifically, the third loss function can be:

Based on the loss function, reverse conduction updates at least one of the first preset sub-model, the second preset sub-model, the third preset sub-model and the fourth preset sub-model, including: based on the The second loss function is used to conduct backward conduction to update the parameters of the second preset sub-model, and based on the third loss function, the reverse conduction is used to update the parameters of the third preset sub-model. The same is explained with the aforementioned example. Specifically, based on the second loss function and the third loss function, reverse conduction updates the aforementioned

It should be understood that since the training of the preset model is performed in a loop, in any loop, such as the t-th loop processing (t is a positive integer), the above-mentioned first loss function can be used. Based on the above-mentioned first loss The function performs reverse conduction to update the parameters of the first preset sub-model and the fourth preset sub-model; in the t+1 cycle processing, the above-mentioned second loss function and third loss function can be used to conduct reverse conduction, and the parameters of the first preset sub-model and the fourth preset sub-model are updated. The parameters of the second preset model and the parameters of the third preset sub-model; then in the t+2 loop processing, reverse conduction is performed again to update the first preset sub-model and the fourth preset sub-model based on the above-mentioned first loss function. Parameters of the model; in the t+3 loop processing, the above-mentioned second loss function and third loss function are used for reverse conduction, and the parameters of the second preset model and the parameters of the third preset sub-model are updated. By analogy, until the preset model converges as a whole, the training of the preset model is determined to be completed.

It should be pointed out that the conditions for determining whether the preset model has converged may include at least one of the following: the number of training cycles reaches the preset threshold; the first loss function, the second loss function, and the third loss function no longer change or respectively less than the specified value, etc. The preset threshold value can be set according to the actual situation, for example, it can be 100 times, 50 times, etc. This embodiment does not exhaust all possible convergence conditions of the preset model.

The derivation process of the aforementioned first loss function is explained as follows: the received wireless signal is the wireless signal after the wireless signal transmitted by the signal source node passes through the obstacle, and different environmental characteristics will make the received wireless signal different, such as in Figure 8 As shown, the received wireless signal obtained when the wireless signal passes through the environmental features including glass obstacles (for example, the CIR of the received wireless signal is shown in the upper part of Figure 8), and the received wireless signal obtained when the wireless signal passes through the environmental features including plastic obstacles. The received wireless signal (for example, the CIR of the received wireless signal shown in the lower part of Figure 8) is different between the two.

The received wireless signal r (that is, the wireless training signal in the aforementioned training sample) can be decomposed into two conditionally independent features of the latent space, namely the aforementioned location feature prediction information z _p and the environment feature prediction information z _l . The distribution of latent variables can be expressed as follows: p(z _p ,z _l |r)=p(z _p |r)p(z _l |r). Among them, p() represents the calculation of hidden variable distribution. Among them, z _p contains position-related information only related to the wireless training signal in the training sample and is not affected by environmental factors, so the positioning estimation parameters can be completely determined by z _p . z _l contains information only related to environmental factors and is not affected by location, so the relevant information of the scene can be completely determined by the environmental characteristics z _l . Then using the location feature prediction information z _p and the environment feature prediction information z _l , the conditional distribution p(b|z _p ), p(l|z _l ) can be obtained; where b represents the positioning estimation parameter and l represents the scene type estimate) . Then use the variational inference method to construct the variational distribution q(z _p ,z _l |r) (where q() is the variational distribution function) to approximate the unknown joint characteristic distribution p(z _p ,z _l |r) . Assume that the variational distribution also has the following decomposition relationship: q(z _p ,z _l |r)=q(z _p |r)q(z _l |r). According to variational theory, it can be deduced that the evidence variational lower bound L for the variational distribution q(z _p ,z _l |r) and observation r is:

-D _KL (q(z _l |r)||p(z _l )); where, D _KL (·||·) is the Kullback-Leibler (KL) divergence between distributions; E is the empirical distribution function.

Assume that the prior distribution p(z _p |r)p(z _l |r) is a Gaussian distribution with a mean of 0 and a variance of 1 to ensure a certain degree of randomness in the encoding; assume that the variational distribution q(z _p |r)q (z _l |r) is a random distribution with mean and variance of 1, and its mean is obtained by the first preset sub-model (encoder). Substituting the above assumptions into D _KL (q(z _p |r)||p(z _p )) and D _KL (q(z _l |r)||p(z _l )), we can get D _KL (q(z _p |r)||p(z _p )) is ||z _p || ² , and D _KL (q(z _l |r)||p(z _l )) is ‖z _l ‖ ² . of the aforementioned functions

The second preset sub-model (decoder) inputs the results of any sampling in the distribution of the first preset sub-model z _p , z _l ~ q (z _p , z _l |r), and outputs the reconstructed signal r ~ p(r|z _p =z _p ,z _l =z _l ), the empirical distribution function of the reconstruction result can be used to express

We hope to maximize L, that is, to maximize this term, which is equivalent to making the reconstructed r close to the r of the initial input first subnetwork, which can be expressed by ||rf _φ (f _θ (r))|| ² . In this way, through the above derivation process, the first loss function can be expressed as: L _AE (θ,φ; r)=||rf _φ (f _θ (r))|| ² +||z _p || ² +‖ z _l ‖ ² .

In another implementation, based on at least one of the reconstructed signal, the positioning estimation parameter, the scene type estimation value, the first preset label and the second preset label, Determining a loss function includes: determining a first loss function based on the reconstructed signal, the wireless training signal, the positioning estimation parameter, and the scene type estimation value; and determining a first loss function based on the positioning estimation parameter and the first prediction value. Assume labels, determine the second loss function; determine the third loss function based on the scene type estimate and the second preset label; determine the third loss function based on the first loss function, the second loss function, and the third loss function The weighted calculation yields the total loss function.

Among them, the methods of constructing the first loss function, the second loss function, and the third loss function have been described in the foregoing embodiments and will not be described again here.

Based on the examples of the aforementioned embodiments, the total loss function is further described. The total loss function can be expressed as:

Among them, L _AE (θ, φ; r) is the aforementioned first loss function. The specific calculation method has been explained above and will not be repeated. α _AE is the weight of the first loss function;

That is, the aforementioned second loss function, the specific calculation method is the same as the aforementioned embodiment, α _p is the weight of the second loss function;

That is, the third loss function, the specific calculation method is the same as the previous embodiment, α _l is the weight of the third loss function. The specific values of α _AE , α _p , and α _l can be determined according to the actual situation. For example, α _AE =α _p =α _l =1 can be set, or it can also be set to other values, which are not limited here.

Based on the loss function, reverse conduction updates at least one of the first preset sub-model, the second preset sub-model, the third preset sub-model and the fourth preset sub-model, including: based on the The total loss function conducts backwards to update the parameters of the first preset sub-model, the parameters of the second preset sub-model, the parameters of the third preset sub-model and the parameters of the fourth preset sub-model.

It should be understood that since the training for the preset model can be performed in a loop, in this embodiment, in any cycle, the parameters of the first preset model, the second preset model, and the second preset model can be updated by reverse conduction using the above total loss function in any cycle. The parameters of the preset model, the parameters of the third preset model and the parameters of the fourth preset model; and so on, until the entire preset model converges, the training of the preset model is determined to be completed. It should be noted that the conditions for determining whether the preset model has converged may include at least one of the following: the number of training cycles reaches a preset threshold; the total loss function no longer changes or is less than the preset value, etc. The preset threshold value can be set according to the actual situation, for example, it can be 100 times, 50 times, etc. This embodiment does not exhaust all possible convergence conditions of the preset model.

After it is determined that the training of the preset model is completed, the first preset sub-model in the preset model is used as the first sub-model in the target model, and the second preset sub-model in the preset model is used as the first sub-model in the target model. The second sub-model of , and the third preset sub-model in the preset model is used as the third sub-model in the target model. It can be seen that the fourth preset sub-model is only used when training the preset model, and when using the trained preset model (i.e., the target model) for actual prediction, as shown in Figure 5, only Just use the first sub-model, the second sub-model and the third sub-model.

It can be seen that by adopting the above solution, the preset model can be trained to obtain a target model. The target model can be used to process wireless signals to obtain positioning parameters, and the positioning parameters are used to determine positioning information. In this way, wireless signals can be identified through the target model to obtain more accurate parameters for positioning, ensuring positioning accuracy and processing efficiency.

In addition, in the aforementioned scheme, the signal can be decomposed into a latent space. On the one hand, advanced semantic features can be adaptively extracted from high-dimensional data, making full use of signal information compared with traditional triangulation and other estimation-based positioning schemes. The positioning accuracy can be improved through the trained target model; on the other hand, the environmental characteristics of the signal and the positioning-related features can be decoupled through the aforementioned solution, and the environmental influence can be separated from the positioning features, which is further compared to the traditional positioning solution. Explicitly eliminate environmental interference to improve positioning robustness. Moreover, the algorithm of the aforementioned solution has low complexity and can meet the high real-time requirements of actual application scenarios. Since the above algorithm is designed based on deep learning technology, after completing offline training to obtain the target model, it only needs to be performed on the first device (i.e., the node to be tested). Side) Storing the parameters of the model can achieve real-time location and environmental feature estimation, which is easy to deploy and process for hardware computing.

Figure 9 is a schematic structural diagram of a first device according to an embodiment of the present application, including:

Communication unit 901, used to receive wireless signals;

The processing unit 902 is configured to process the wireless signal based on a target model to obtain positioning parameters; and determine positioning information of the first device based on the positioning parameters.

The communication unit 901 is configured to receive N wireless signals; the N wireless signals are sent by N second devices; N is an integer greater than or equal to 1.

The target model includes a first sub-model and a second sub-model;

The processing unit 902 is configured to input the j-th wireless signal among the N wireless signals into the first sub-model to obtain the j-th location feature information output by the first sub-model; where j is An integer greater than or equal to 1 and less than or equal to N; input the j-th position feature information into the second sub-model to obtain the j-th positioning parameter output by the second sub-model.

The processing unit 902 is configured to determine the positioning information of the first device based on the N positioning parameters and the position information of the N second devices.

The N positioning parameters include at least one of the following:

The distance between the first device and each of the N second devices;

The receiving angle of each of the N wireless signals by the first device.

The N positioning parameters include: N distance deviations;

The processing unit 902 is configured to correct N initial distances based on the N distance deviations to obtain N adjusted distances; based on the N adjusted distances and the positions of the N second devices information to determine the positioning information of the first device.

The processing unit 902 is configured to determine the N initial distances between the first device and the N second devices based on the N wireless signals.

The N positioning parameters include: N angular deviations; the processing unit is used to correct the N initial receiving angles based on the N angular deviations to obtain N adjusted angles; based on the N The adjusted angle and the position information of the N second devices determine the positioning information of the first device.

The processing unit 902 is configured to determine the N initial reception angles for receiving the N wireless signals.

The processing unit 902 is configured to determine the positioning information of the first device based on N positioning parameters and a signal feature library; wherein the signal feature library contains the characteristic amount of each reference position in a plurality of reference positions; The feature quantity of each reference position includes the RSS reference value of the reference signal of each of the N second devices.

The N positioning parameters include N signal reception strength RSS values.

The N positioning parameters include N RSS deviations; the processing unit is configured to determine N adjusted RSS values based on the N RSS deviations and N initial intensity estimates; based on the N adjusted The RSS value and the signal feature library are used to determine the positioning information of the first device.

The first sub-model of the target model is also used to process wireless signals to obtain environmental feature information corresponding to the wireless signals;

The target model also includes a third sub-model, which is used to process the environmental feature information to obtain scene type information.

The processing unit 902 is configured to determine the environment in which the first device is located based on the scene type information output by the third sub-model.

The signal type of the i-th wireless signal among the N wireless signals is one of the following: Bluetooth wireless signal, WIFI wireless signal, cellular network wireless signal, or ultra-wideband UWB wireless signal.

Figure 10 is a schematic structural diagram of an electronic device according to an embodiment of the present application, including:

The training unit 1001 is used to train a preset model using training samples to obtain a trained target model; wherein the target model is used to process wireless signals to obtain positioning parameters, and the positioning parameters are used to determine positioning information.

The training samples include: a wireless training signal, a first preset label corresponding to the wireless training signal, and a second preset label corresponding to the wireless training signal.

The training unit 1001 is configured to input the wireless training signal in the training sample into the first preset sub-model of the preset model, and obtain the positioning feature prediction information and scene features output by the first preset sub-model. Prediction information; input the positioning feature prediction information and the scene feature prediction information into the fourth preset sub-model in the preset model to obtain the wireless training signal output by the fourth preset sub-model. Reconstruct the signal; input the positioning feature prediction information into the second preset sub-model in the preset model, obtain the positioning estimation parameters output by the second preset sub-model, and input the scene feature prediction information into the The third preset sub-model in the preset model is used to obtain the scene type estimate output by the third preset sub-model; based on the reconstructed signal, the positioning estimation parameter, the scene type estimate, and the At least one of the first preset label and the second preset label determines a loss function; based on the loss function, reverse conduction updates the first preset sub-model, the second preset sub-model, and the third preset sub-model. At least one of the preset sub-model and the fourth preset sub-model.

The training unit 1001 is configured to determine a first loss function based on the reconstructed signal, the wireless training signal, the positioning estimation parameter, and the scene type estimation value.

The training unit 1001 is configured to conduct reverse conduction to update the parameters of the first preset sub-model and the parameters of the fourth preset sub-model based on the first loss function.

The training unit 1001 is configured to determine a second loss function based on the positioning estimation parameter and the first preset label; and determine a third loss function based on the scene type estimate and the second preset label.

The training unit 1001 is configured to conduct backward conduction to update the parameters of the second preset sub-model based on the second loss function, and conduct reverse conduction to update the third preset based on the third loss function. Parameters of the submodel.

The training unit 1001 is configured to determine a first loss function based on the reconstructed signal, the wireless training signal, the positioning estimation parameter, and the scene type estimation value; based on the positioning estimation parameter and the third A preset label, determine the second loss function; based on the scene type estimate and the second preset label, determine the third loss function; based on the first loss function, the second loss function, the third loss function The loss function is weighted to obtain the total loss function.

The training unit 1001 is configured to update the parameters of the first preset sub-model, the parameters of the second preset sub-model, the parameters of the third preset sub-model and the fourth preset sub-model based on the total loss function reverse conduction. Set the parameters of the submodel.

The electronic device in the embodiment of the present application can implement the corresponding functions in the model generation method in the foregoing method embodiment. For the corresponding processes, functions, implementation methods and beneficial effects of each module (sub-module, unit or component, etc.) in the electronic device, please refer to the corresponding description in the above method embodiment, and will not be described again here. It should be understood that in addition to the above-mentioned training unit, the above-mentioned electronic device may also include a receiving unit, a sending unit, a storage unit, etc., for example, after obtaining the trained target model, it can be sent to the aforementioned first device through the sending unit; and For example, the aforementioned training samples can be stored in a storage unit and so on. It should be noted that the functions described for each module (sub-module, unit or component, etc.) in the electronic device of the embodiment of the application may be implemented by different modules (sub-module, unit or component, etc.), or may be implemented by the same module. (Submodule, unit or component, etc.) implementation.

Figure 11 is a schematic structural diagram of a communication device 1100 according to an embodiment of the present application. The communication device 1100 includes a processor 1110, and the processor 1110 can call and run a computer program from the memory, so that the communication device 1100 implements the method in the embodiment of the present application.

In a possible implementation, the communication device 1100 may further include a memory 1120. The processor 1110 can call and run the computer program from the memory 1120, so that the communication device 1100 implements the method in the embodiment of the present application.

The memory 1120 may be a separate device independent of the processor 1110, or may be integrated into the processor 1110.

In a possible implementation, the communication device 1100 may also include a transceiver 1130, and the processor 1110 may control the transceiver 1130 to communicate with other devices. Specifically, the communication device 1100 may send information or data to, or receive data from, other devices. Information or data sent.

Among them, the transceiver 1130 may include a transmitter and a receiver. The transceiver 1130 may further include an antenna, and the number of antennas may be one or more.

In a possible implementation, the communication device 1100 may be the first device in the embodiment of the present application, and the communication device 1100 may implement the corresponding processes implemented by the first device in the various methods of the embodiment of the present application. For the sake of simplicity , which will not be described in detail here.

In a possible implementation manner, the communication device 1100 may be an electronic device according to the embodiment of the present application, and the communication device 1100 may implement the corresponding processes implemented by the electronic device in each method of the embodiment of the present application. For simplicity, in This will not be described again.

Figure 12 is a schematic structural diagram of a chip 1200 according to an embodiment of the present application. The chip 1200 includes a processor 1210, and the processor 1210 can call and run a computer program from the memory to implement the method in the embodiment of the present application.

In a possible implementation, the chip 1200 may also include a memory 1220. The processor 1210 can call and run the computer program from the memory 1220 to implement the method executed by the aforementioned first device or electronic device in the embodiment of the present application.

The memory 1220 may be a separate device independent of the processor 1210, or may be integrated into the processor 1210.

In a possible implementation, the chip 1200 may also include an input interface 1230. The processor 1210 can control the input interface 1230 to communicate with other devices or chips, and specifically, can obtain information or data sent by other devices or chips.

In a possible implementation, the chip 1200 may also include an output interface 1240. The processor 1210 can control the output interface 1240 to communicate with other devices or chips. Specifically, it can output information or data to other devices or chips.

In a possible implementation, the chip can be applied to the first device in the embodiment of the present application, and the chip can implement the corresponding processes implemented by the first device in the various methods of the embodiment of the present application. For simplicity, in This will not be described again.

In a possible implementation manner, the chip can be applied to the electronic device in the embodiment of the present application, and the chip can implement the corresponding processes implemented by the electronic device in the various methods of the embodiment of the present application. For the sake of brevity, this is not mentioned here. Again.

The chips applied to the first device and the electronic device may be the same chip or different chips.

It should be understood that the chips mentioned in the embodiments of this application may also be called system-on-chip, system-on-a-chip, system-on-chip or system-on-chip, etc.

The processor mentioned above can be a general-purpose processor, a digital signal processor (DSP), an off-the-shelf programmable gate array (FPGA), an application specific integrated circuit (ASIC), or Other programmable logic devices, transistor logic devices, discrete hardware components, etc. The above-mentioned general processor may be a microprocessor or any conventional processor.

The memory mentioned above may be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. Among them, 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).

It should be understood that the above memory is an exemplary but not restrictive description. For example, the memory in the embodiment of the present application can also be a static random access memory (static RAM, SRAM), a dynamic random access memory (dynamic RAM, DRAM), Synchronous dynamic random access memory (synchronous DRAM, SDRAM), double data rate synchronous dynamic random access memory (double data rate SDRAM, DDR SDRAM), enhanced synchronous dynamic random access memory (enhanced SDRAM, ESDRAM), synchronous connection Dynamic random access memory (synch linN DRAM, SLDRAM) and direct memory bus random access memory (Direct Rambus RAM, DR RAM) and so on. That is, memories in embodiments of the present application are intended to include, but are not limited to, these and any other suitable types of memories.

Figure 13 is a schematic block diagram of a communication system 1300 according to an embodiment of the present application. The communication system 1300 includes a first device 1310 and a second device 1320.

The first device 1310 can be used to implement the corresponding functions implemented by the first device in the above method, and the second device 1320 can be used to implement the corresponding functions implemented by the second device in the above method. For the sake of brevity, no further details will be given here.

In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. 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 instructions. When the computer program instructions are loaded and executed on a 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 coaxial cable, optical fiber, Digital Subscriber Line (DSL)) or wireless (such as infrared, wireless, microwave, etc.) means to transmit to another website, computer, server or data center. 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 or data center integrated with one or more available media. The available media may be magnetic media (eg, floppy disk, hard disk, tape), optical media (eg, DVD), or semiconductor media (eg, Solid State Disk (Solid State Disk, SSD)), etc.

It should be understood that in the various embodiments of the present application, the size of the sequence numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its functions and internal logic, and should not be used in the embodiments of the present application. The implementation process constitutes any limitation.

Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, devices and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be described again here.

The above are only specific embodiments of the present application, but the protection scope of the present application is not limited thereto. Any person familiar with the technical field can easily think of changes or replacements within the technical scope disclosed in the present application. are covered by the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

Claims

A positioning method that includes:

The first device receives wireless signals;

The first device processes the wireless signal based on the target model to obtain positioning parameters;

The first device determines positioning information of the first device based on the positioning parameter.
The method of claim 1, wherein the first device receiving a wireless signal includes:

The first device receives N wireless signals; the N wireless signals are sent by N second devices; N is an integer greater than or equal to 1.
The method of claim 2, wherein the target model includes a first sub-model and a second sub-model;

The first device processes the wireless signal based on the target model to obtain positioning parameters, including:

The first device inputs the j-th wireless signal among the N wireless signals into the first sub-model to obtain the j-th location feature information output by the first sub-model; where j is greater than or equal to 1 and an integer less than or equal to N;

Input the j-th position feature information into the second sub-model to obtain the j-th positioning parameter output by the second sub-model.
The method of claim 3, wherein the first device determines the positioning information of the first device based on the positioning parameter, including:

The first device determines the positioning information of the first device based on the N positioning parameters and the position information of the N second devices.
The method according to claim 4, wherein the N positioning parameters include at least one of the following:

The distance between the first device and each of the N second devices;

The receiving angle of each of the N wireless signals by the first device.
The method according to claim 4, wherein the N positioning parameters include: N distance deviations;

The first device determines the positioning information of the first device based on the N positioning parameters and the position information of the N second devices, including:

The first device corrects the N initial distances based on the N distance deviations to obtain N adjusted distances; based on the N adjusted distances and the position information of the N second devices, determine The positioning information of the first device.
The method of claim 6, further comprising:

The first device determines the N initial distances between the first device and the N second devices based on the N wireless signals.
The method according to claim 4, 6 or 7, wherein the N positioning parameters include: N angular deviations;

The first device determines the positioning information of the first device based on the N positioning parameters and the position information of the N second devices, including:

The first device corrects the N initial receiving angles based on the N angular deviations to obtain N adjusted angles; based on the N adjusted angles and the position information of the N second devices, Determine the positioning information of the first device.
The method of claim 8, further comprising:

The first device determines the N initial reception angles for receiving the N wireless signals.
The method of claim 3, wherein the first device determines the positioning information of the first device based on the positioning parameter, including:

The first device determines the positioning information of the first device based on N positioning parameters and a signal feature library; wherein the signal feature library contains a characteristic amount of each reference position in a plurality of reference positions; each of the The feature quantities of the reference positions include the RSS reference values of the reference signals of each of the N second devices.
The method of claim 10, wherein the N positioning parameters include N signal reception strength RSS values.
The method of claim 10, wherein the N positioning parameters include N RSS deviations;

The first device determines the positioning information of the first device based on the N positioning parameters and the signal feature library, including:

The first device determines N adjusted RSS values based on the N RSS deviations and N initial intensity estimation values; determines the first adjusted RSS value based on the N adjusted RSS values and the signal feature library. A device’s location information.
The method according to any one of claims 1-12, wherein,

The first sub-model of the target model is also used to process wireless signals to obtain environmental feature information corresponding to the wireless signals;

The target model also includes a third sub-model, which is used to process the environmental feature information to obtain scene type information.
The method of claim 13, wherein the method further includes:

The first device determines the environment in which the first device is located based on the scene type information output by the third sub-model.
The method according to any one of claims 2 to 14, wherein the signal type of the i-th wireless signal among the N wireless signals is one of the following: Bluetooth wireless signal, WIFI wireless signal, cellular network wireless signal, Ultra-wideband UWB wireless signal.
A model generation method including:

Use training samples to train the preset model to obtain the trained target model;

Wherein, the target model is used to process wireless signals to obtain positioning parameters, and the positioning parameters are used to determine positioning information.
The method of claim 16, wherein the training sample includes: a wireless training signal, a first preset label corresponding to the wireless training signal, and a second preset label corresponding to the wireless training signal.
The method according to claim 17, wherein the first preset tag corresponding to the wireless training signal includes at least one of the following:

The distance tag corresponding to the wireless training signal;

The angle label corresponding to the wireless training signal;

The RSS tag corresponding to the wireless training signal;

The distance deviation label corresponding to the wireless training signal;

The angle deviation label corresponding to the wireless training signal;

The RSS deviation label corresponding to the wireless training signal.
The method according to claim 17 or 18, wherein the second preset tag corresponding to the wireless training signal includes: a scene type tag corresponding to the environment in which the wireless training signal is located.
The method according to any one of claims 17-19, wherein said using training samples to train a preset model includes:

Input the wireless training signal in the training sample into the first preset sub-model of the preset model to obtain positioning feature prediction information and scene feature prediction information output by the first preset sub-model;

Input the positioning feature prediction information and the scene feature prediction information into the fourth preset sub-model in the preset model to obtain the reconstructed signal of the wireless training signal output by the fourth preset sub-model. ;

Input the positioning feature prediction information into the second preset sub-model in the preset model, obtain the positioning estimation parameters output by the second preset sub-model, and input the scene feature prediction information into the preset model. The third preset sub-model is used to obtain the scene type estimate output by the third preset sub-model;

Determine a loss function based on at least one of the reconstructed signal, the positioning estimation parameter, the scene type estimate, the first preset label, and the second preset label;

Based on the loss function, reverse conduction updates at least one of the first preset sub-model, the second preset sub-model, the third preset sub-model and the fourth preset sub-model.
The method according to claim 20, wherein the method is based on at least one of the reconstructed signal, the positioning estimation parameter, the scene type estimation value, the first preset label and the second preset label. One, determine the loss function, including:

A first loss function is determined based on the reconstructed signal, the wireless training signal, the positioning estimation parameter, and the scene type estimation value.
The method according to claim 21, wherein, based on the loss function, reverse conduction updates the first preset sub-model, the second preset sub-model, the third preset sub-model and the fourth preset sub-model. At least one of the submodels, including:

Based on the first loss function, reverse conduction updates the parameters of the first preset sub-model and the parameters of the fourth preset sub-model.
The method according to any one of claims 20-22, wherein the method is based on the reconstructed signal, the positioning estimation parameter, the scene type estimation value, the first preset label and the second At least one of the preset labels determines the loss function, including:

A second loss function is determined based on the positioning estimation parameter and the first preset label; and a third loss function is determined based on the scene type estimate and the second preset label.
The method according to claim 23, wherein, based on the loss function, reverse conduction updates the first preset sub-model, the second preset sub-model, the third preset sub-model and the fourth preset sub-model. At least one of the submodels, including:

Based on the second loss function, reverse conduction updates the parameters of the second preset sub-model, and based on the third loss function, reverse conduction updates the parameters of the third preset sub-model.
The method of claim 20, wherein the loss is determined based on at least one of the reconstructed signal, the first estimate, the scene type estimate, the first preset label, and the second preset label. Functions, including:

Based on the reconstructed signal, the wireless training signal, the positioning estimation parameter, and the scene type estimation value, a first loss function is determined; based on the positioning estimation parameter and the first preset label, a second loss function is determined. Loss function; determine a third loss function based on the scene type estimate and the second preset label;

A total loss function is obtained by weighted calculation based on the first loss function, the second loss function, and the third loss function.
The method of claim 25, wherein, based on the loss function, backward conduction updates the first preset sub-model, the second preset sub-model, the third preset sub-model and the fourth preset sub-model. At least one of the submodels, including:

The parameters of the first preset sub-model, the parameters of the second preset sub-model, the parameters of the third preset sub-model and the parameters of the fourth preset sub-model are updated based on the total loss function in reverse conduction.
A first device comprising:

communication unit for receiving wireless signals;

A processing unit, configured to process the wireless signal based on a target model to obtain positioning parameters; and determine positioning information of the first device based on the positioning parameters.
The first device according to claim 27, wherein the communication unit is configured to receive N wireless signals; the N wireless signals are sent by N second devices; N is an integer greater than or equal to 1.
The first device of claim 28, wherein the target model includes a first sub-model and a second sub-model;

The processing unit is configured to input the j-th wireless signal among the N wireless signals into the first sub-model to obtain the j-th location feature information output by the first sub-model; where j is greater than An integer equal to 1 and less than or equal to N; input the j-th position feature information into the second sub-model to obtain the j-th positioning parameter output by the second sub-model.
The first device according to claim 29, wherein the processing unit is configured to determine the positioning information of the first device based on N positioning parameters and the position information of the N second devices.
The first device according to claim 30, wherein the N positioning parameters include at least one of the following:

The distance between the first device and each of the N second devices;

The receiving angle of each of the N wireless signals by the first device.
The first device according to claim 29, wherein the N positioning parameters include: N distance deviations;

The processing unit is configured to correct N initial distances based on the N distance deviations to obtain N adjusted distances; based on the N adjusted distances and the position information of the N second devices , determine the positioning information of the first device.
The first device according to claim 32, wherein the processing unit is configured to determine the N initial distance between the first device and the N second devices based on the N wireless signals. distance.
The first device according to claim 30, 32 or 33, wherein the N positioning parameters include: N angular deviations;

The processing unit is configured to correct N initial receiving angles based on the N angular deviations to obtain N adjusted angles; based on the N adjusted angles and the positions of the N second devices information to determine the positioning information of the first device.
The first device according to claim 34, wherein the processing unit is configured to determine the N initial reception angles for receiving the N wireless signals.
The first device according to claim 29, wherein the processing unit is configured to determine the positioning information of the first device based on N positioning parameters and a signal feature library; wherein the signal feature library contains a plurality of The characteristic quantity of each reference position among the reference positions; the characteristic quantity of each reference position includes the RSS reference value of the reference signal of each of the N second devices.
The first device of claim 36, wherein the N positioning parameters include N signal reception strength RSS values.
The first device of claim 36, wherein the N positioning parameters include N RSS deviations;

The processing unit is configured to determine N adjusted RSS values based on the N RSS deviations and N initial intensity estimation values; determine the N adjusted RSS values based on the N adjusted RSS values and the signal feature library. The positioning information of the first device.
The first device according to any one of claims 27-38, wherein,

The first sub-model of the target model is also used to process wireless signals to obtain environmental feature information corresponding to the wireless signals;

The target model also includes a third sub-model, which is used to process the environmental feature information to obtain scene type information.
The first device according to claim 39, wherein the processing unit is configured to determine the environment in which the first device is located based on the scene type information output by the third sub-model.
The first device according to any one of claims 28 to 40, wherein the signal type of the i-th wireless signal among the N wireless signals is one of the following: Bluetooth wireless signal, WIFI wireless signal, cellular network wireless signal signal, ultra-wideband UWB wireless signal.
An electronic device including:

A training unit is used to train a preset model using training samples to obtain a trained target model; wherein the target model is used to process wireless signals to obtain positioning parameters, and the positioning parameters are used to determine positioning information.
The electronic device according to claim 42, wherein the training sample includes: a wireless training signal, a first preset tag corresponding to the wireless training signal, and a second preset tag corresponding to the wireless training signal.
The electronic device according to claim 43, wherein the first preset tag corresponding to the wireless training signal includes at least one of the following:

The distance tag corresponding to the wireless training signal;

The angle label corresponding to the wireless training signal;

The RSS tag corresponding to the wireless training signal;

The distance deviation label corresponding to the wireless training signal;

The angle deviation label corresponding to the wireless training signal;

The RSS deviation label corresponding to the wireless training signal.
The electronic device according to claim 43 or 44, wherein the second preset tag corresponding to the wireless training signal includes: a scene type tag corresponding to the environment in which the wireless training signal is located.
The electronic device according to any one of claims 43 to 45, wherein the training unit is configured to input the wireless training signal in the training sample into the first preset sub-model of the preset model to obtain the The positioning feature prediction information and scene feature prediction information output by the first preset sub-model; input the positioning feature prediction information and the scene feature prediction information into the fourth preset sub-model in the preset model, and obtain The reconstructed signal of the wireless training signal output by the fourth preset sub-model; input the positioning feature prediction information into the second preset sub-model in the preset model to obtain the second preset sub-model output positioning estimation parameters, and input the scene feature prediction information into the third preset sub-model in the preset model to obtain the scene type estimate output by the third preset sub-model; based on the reconstructed signal , at least one of the positioning estimation parameters, the scene type estimation value, the first preset label and the second preset label, determine a loss function; based on the loss function, reverse conduction updates the At least one of the first preset sub-model, the second preset sub-model, the third preset sub-model and the fourth preset sub-model.
The electronic device according to claim 46, wherein the training unit is configured to determine a first loss function based on the reconstructed signal, the wireless training signal, the positioning estimation parameter, and the scene type estimation value. .
The electronic device according to claim 47, wherein the training unit is configured to conduct backward conduction to update the parameters of the first preset sub-model and the parameters of the fourth preset sub-model based on the first loss function.
The electronic device according to any one of claims 46-48, wherein the training unit is used to determine a second loss function based on the positioning estimation parameter and the first preset label; and based on the scene The type estimate and the second preset label determine the third loss function.
The electronic device according to claim 49, wherein the training unit is configured to conduct backward conduction to update parameters of the second preset sub-model based on the second loss function, and to update parameters of the second preset sub-model based on the third loss function. , reverse conduction updates the parameters of the third preset sub-model.
The electronic device according to claim 46, wherein the training unit is configured to determine a first loss function based on the reconstructed signal, the wireless training signal, the positioning estimation parameter, and the scene type estimation value. ; Based on the positioning estimation parameter and the first preset label, determine a second loss function; Based on the scene type estimate and the second preset label, determine a third loss function; Based on the first loss function, The second loss function and the third loss function are weighted and calculated to obtain a total loss function.
The electronic device according to claim 51, wherein the training unit is configured to update parameters of the first preset sub-model, parameters of the second preset sub-model, and the third preset sub-model based on the total loss function through reverse conduction. The parameters of the third preset sub-model and the parameters of the fourth preset sub-model.
A first device, including: a processor and a memory, the memory is used to store a computer program, the processor is used to call and run the computer program stored in the memory, so that the terminal device executes the instructions of claims 1 to The method described in any one of 15.
An electronic device, including: a processor and a memory, the memory is used to store a computer program, the processor is used to call and run the computer program stored in the memory, so that the terminal device executes claims 16 to 26 any one of the methods.
A chip includes: a processor, configured to call and run a computer program from a memory, so that a device installed with the chip executes the method according to any one of claims 1 to 15 or 16 to 26.
A computer-readable storage medium for storing a computer program that, when executed by a device, causes the device to perform the method according to any one of claims 1 to 15 or 16 to 26.
A computer program product comprising computer program instructions that cause a computer to perform the method according to any one of claims 1 to 15 or 16 to 26.
A computer program that causes a computer to perform the method according to any one of claims 1 to 15 or 16 to 26.