Method and data processing apparatus for generating fingerprint map for unexplored region, and positioning apparatus
The use of a GAN-based generative model automates the creation of fingerprint maps for indoor positioning, addressing the inefficiencies of traditional methods by generating maps without signal collection, thus enhancing the versatility and efficiency of indoor positioning services.
Patent Information
- Application Number
- PCT/KR2023/010816
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-05-02
- Filing Date
- 2023-07-26
- Publication Date
- 2024-11-07
AI Technical Summary
Fingerprint-based indoor positioning methods require extensive time and resources for building fingerprint maps, making them inefficient for universalizing indoor positioning services.
A generative model, specifically a GAN-based approach, is used to automatically generate fingerprint maps by processing feature vectors from access points and reference points, allowing for the creation of fingerprint maps without a wireless signal collection process.
This method enables the rapid and cost-effective generation of fingerprint maps, enhancing the versatility of fingerprint-based positioning techniques and improving indoor positioning services by reducing the need for extensive data collection.
Smart Images

Figure KR2023010816_07112024_PF_FP_ABST
Abstract
Description
Method for generating a fingerprint map for an unexplored area, data processing device and positioning device
[0001] The technology described below is a technique for automatically generating a fingerprint map.
[0002] This application is the result of research conducted with the support of the National Research Foundation of Korea (No. NRF-2021R1A2B5B01002906) and the National IT Industry Promotion Agency (No. RS-2022-00155966, Artificial Intelligence Convergence Innovation Talent Development (Ewha Womans University)) with funding from the government (Ministry of Science and ICT) in 2023.
[0003] Indoor positioning refers to positioning in areas where it is difficult to receive satellite navigation system signals, such as indoors. A typical method for indoor positioning is using wireless signals such as Wi-Fi or Bluetooth. Furthermore, a representative example of wireless signal-based indoor positioning is the fingerprinting technique.
[0004] Fingerprint-based indoor positioning requires collecting wireless signals in the service area in advance to build a fingerprint map. However, fingerprint-based indoor positioning has the disadvantage of requiring significant time and expense to build the fingerprint map.
[0005] The technology described below aims to automatically build a fingerprint map using a generative model.
[0006] A method for generating a fingerprint map for an unexplored area includes a step in which a data processing device receives, by an input, the locations of reference points for a service area and the locations of a plurality of APs (access points), a step in which the data processing device extracts a feature vector for each of the APs based on the locations of the reference points and the APs, a step in which the data processing device inputs, as a condition, the feature vector extracted for each of the APs into a pre-learned generative model and inputs the arbitrary noise to generate a fingerprint map for each of the APs, and a step in which the data processing device combines the fingerprint maps generated for the APs to generate a final fingerprint map for the service area.
[0007] A data processing device for generating a fingerprint map for an unexplored area includes an input device for receiving locations of reference points for a service area and locations of APs, a storage device for receiving a fingerprint map of a specific area and a feature vector extracted based on APs at the reference points of the specific area as conditions and storing a generation model trained to generate a fingerprint map for the specific area, and a calculation device for extracting a distance from each of the reference points for the service area to the AP and an angle formed with the AP and inputting them as conditions for the generation model, and inputting arbitrary noise into the generation model to generate a fingerprint map for the service area.
[0008] A positioning device using a virtual fingerprint map includes a communication device that receives wireless signals of APs at a current location, a storage device that stores the virtual fingerprint map, and a calculation device that estimates the current location based on the strength of the wireless signals of the APs by referring to the virtual fingerprint map.
[0009] The technology described below can automatically build a fingerprint map using only the locations of access points (APs) deployed within a service area, without requiring wireless signal collection. Therefore, the technology described below contributes to the widespread adoption of indoor positioning services by providing versatility to fingerprint-based positioning techniques.
[0010] Figure 1 is an example of a fingerprint-based positioning system.
[0011] Figure 2 is an example of a schematic process for building a generative model.
[0012] Figure 3 is an example of a specific learning process of a generative model.
[0013] Figure 4 is an example of a process in which a data processing device creates a fingerprint map using a generative model.
[0014] Figure 5 is an example of a data processing device that generates a fingerprint map for an undetected area.
[0015] Figure 6 is an example of a positioning device using a virtual fingerprint map.
[0016] The technology described below is susceptible to various modifications and embodiments. Specific embodiments are illustrated and described in detail in the drawings. However, this does not limit the technology described below to specific embodiments, and it should be understood that all modifications, equivalents, and alternatives fall within the spirit and scope of the technology described below.
[0017] Terms such as first, second, A, and B may be used to describe various components, but these components are not limited by these terms and are used solely to distinguish one component from another. For example, without departing from the scope of the technology described below, a first component may be referred to as a second component, and similarly, a second component may also be referred to as a first component. The term "and / or" includes a combination of a plurality of related described items or any of a plurality of related described items.
[0018] As used herein, the singular expressions should be understood to include the plural expressions unless the context clearly dictates otherwise, and the term "comprises" and the like should be understood to mean the presence of a described feature, number, step, operation, component, part, or combination thereof, but not to exclude the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.
[0019] Before going into a detailed description of the drawings, it should be made clear that the division of components in this specification is merely a division based on the main function of each component. In other words, two or more components described below may be combined into a single component, or a single component may be further subdivided into two or more components with more detailed functions. In addition to its own main function, each component described below may additionally perform some or all of the functions of other components, and of course, some of the main functions of each component may be exclusively performed by other components.
[0020] Additionally, in performing a method or method of operation, each process constituting the method may occur in a different order than the stated order, unless the context clearly indicates a specific order. That is, each process may occur in the same order as the stated order, may be performed substantially simultaneously, or may be performed in the opposite order.
[0021] The technology described below is a technique for generating a fingerprint map for positioning in a service area.
[0022] Service area refers to an area where satellite navigation signals are difficult to receive. Therefore, the term "service area" is used hereinafter to include indoor areas.
[0023] The technique described below creates a fingerprint map using a generative model.
[0024] Various types of generative models are being studied. For example, generative models include autoencoders, Generative Adversarial Networks (GANs), and diffusion models. As described below, the researcher used a GAN-based model. Therefore, the following description focuses on the GAN model. However, any of these various generative models can be used to generate fingerprint maps.
[0025] Below, we explain how a learning device builds a generative model using learning data. A learning device is a device capable of data processing and model learning. It can also build learning data. For example, a learning device can be implemented as a PC, a networked server, a smart device, or a chipset embedded with a dedicated program.
[0026] The following describes how a data processing device generates a fingerprint map using a generative model. For example, the data processing device can be implemented as a PC, a networked server, a smart device, or a chipset embedded with a dedicated program.
[0027] In the description below, the wireless signal can be any of several types of signals, such as Wi-Fi or Bluetooth. A terminal receives the wireless signal in the service area. The terminal can be any of several types of devices, including smart devices, mobile devices, wearable devices, and robots.
[0028] The following describes how a positioning device uses a generated fingerprint map to determine positioning within a service area. The positioning device may be a device such as the terminal or server described above.
[0029] Figure 1 is an example of a fingerprint-based positioning system (100).
[0030] While the service area S is actually a three-dimensional space, fingerprint maps typically consist of information on a two-dimensional plane. The service area S can be defined by a specific coordinate system, including the location of reference points and the locations of access points (APs). APs are devices that transmit wireless signals, such as Wi-Fi, within the service area.
[0031] In Fig. 1, (A) illustrates the process of a data processing device (50) generating a fingerprint map. The data processing device (50) extracts certain vector information from AP location information. The data processing device (50) inputs arbitrary noise and vector information into a pre-built generation model to generate a fingerprint map M for a service area S. The fingerprint map is generally in the form of a grid map and contains information on the strength of a wireless signal for each rectangular section. The process of constructing the generation model and the specific process of generating a fingerprint map will be described later.
[0032] In Fig. 1, (B) and (C) illustrate a process in which a positioning device performs positioning in a service area using a fingerprint map.
[0033] A positioning device (110) such as a smart device can estimate its current location based on the strength of a wireless signal received from APs at the current location in the service area S by referring to a fingerprint map.
[0034] The terminal (80) collects the strength of the wireless signal received from APs at the current location in the service area S. The terminal (80) transmits the information on the strength of the wireless signal to a separate positioning device (120). The positioning device (120), such as a computer device or a server, can estimate the current location of the terminal (80) based on the strength of the wireless signal received from APs at the current location in the service area S by referring to a fingerprint map. The positioning device (120) can transmit the positioning result to the terminal (80).
[0035] Figure 2 illustrates a schematic example of the process (200) for building a generative model. It explains that a learning device builds the generative model. The learning device corresponds to a computer device that controls the fingerprint data processing and generative model learning process.
[0036] The learning device collects pre-generated fingerprint maps (210). The fingerprint maps used in the model building process are called "learning fingerprint maps." These learning fingerprint maps can be generated manually based on wireless signal strength collected from a specific area, using conventional methods. These learning fingerprint maps can also be extracted from public databases (DBs).
[0037] The learning device extracts wireless signal feature vectors from the learning fingerprint map (220). The wireless signal feature vectors are used as additional information (conditions) during the learning process of the generative model.
[0038] This section describes the process by which a learning device extracts wireless signal feature vectors. The learning device extracts wireless signal feature vectors for each AP from a learning fingerprint map.
[0039] A learning fingerprint map stores the strength of the wireless signal received at each reference point (or data point) in the service area. Typically, a learning fingerprint map is a grid map, storing wireless signal strength information for each rectangular section. Equation 1 below represents the fingerprint X(i) at the ith reference point in the service area.
[0040]
[0041] X(i) is the fingerprint at the i-th reference point s j is the strength of the wireless signal received from the jth AP (j is the index of the AP) at the ith reference point. N ap is the number of APs located in the service area. N data is the number of reference points in the service area.
[0042] The learning device can extract wireless signal feature vectors for each AP. The learning device can define the feature vectors for each AP in the fingerprint map as shown in Equation 2 below. X j (i) represents the wireless signal strength received from AP j at the ith reference point.
[0043]
[0044] The researcher extracted (i) the distance to the AP and (ii) the angle from the AP as feature values based on the reference point in the service area. In other words, a wireless signal feature vector can be defined as (distance value, angle) for each reference point. The learning device can extract wireless signal feature vectors based on the previously known location of the AP in the service area based on a specific coordinate system. The researcher used the azimuth (angle in a plane) between the reference point and the AP as the feature value.
[0045] Alternatively, the learning device can estimate the distance between a specific point and the AP based on the wireless signal strength at the specific point using a widely known wireless signal propagation model, such as the log-distance path loss model.
[0046] The learning device can extract the wireless signal feature vector expressed by the mathematical formula 3 below from the fingerprint map. In other words, the learning device can generate a new fingerprint map (feature vector map) expressed by the mathematical formula 3 below. X j represents the wireless signal feature vector for AP j.
[0047]
[0048] d i is the distance from the ith reference point to AP j, and a i is the angle between the ith reference point and AP j.
[0049] The learning device uses a learning fingerprint map and a wireless signal feature vector to learn a generative model (230).
[0050] Figure 3 is an example of a specific learning process of a generative model.
[0051] The researchers used a conditional generative adversarial network (cGAN) model. The cGAN (300) consists of a generator network (G, 310) and a discriminator network (D, 320) that are adversarial.
[0052] The generator network G(310) is the real fingerprint map x real Learn the distribution of a fake fingerprint map x from random noise z. fake= G(z|c) is generated. At this time, the generator network G(310) receives the aforementioned wireless signal feature vector as input as a condition. The wireless signal feature vector includes the distance and angle from the reference points to the AP. Figure 3 illustrates condition c, which is a wireless signal feature vector for n reference points.
[0053] The discriminator network D (320) receives the fingerprint map G (z|c) generated by the generator network G (310) and the actual fingerprint map as input and determines whether G (z|c) is real or fake. At this time, the discriminator network D (320) also receives a condition, which is a wireless signal feature vector, as input.
[0054] As the distribution of the generated fingerprint map G(z|c) becomes more similar to the distribution of the actual fingerprint map, the discriminant value D(G(z|c)) output by the discriminator network D(320) approaches 1. The objective function of cGAN(300) can be defined as in mathematical expression 4 below.
[0055]
[0056] p data (x real ) is the distribution of the actual fingerprint map, and z~p z (z) represents the distribution of the generated fingerprint map.
[0057] To prevent overfitting of the generator network G (310), the researcher utilized AC-GAN (Auxiliary Classifier GAN). AC-GAN includes a classifier (class loss) that predicts conditions (distance and angle) in addition to a discriminator (sample loss) that classifies whether the generated fingerprint map G (z|c) is real data. cGAN (300) is trained in the direction of minimizing both sample loss and class loss in the discriminator network D (320). The use of AC-GAN may be an optional configuration for fingerprint map generation.
[0058] Figure 4 is an example of a process (400) in which a data processing device generates a fingerprint map using a generation model. Figure 4 is an example of generating a fingerprint map for an unknown service area where wireless signal information has not been searched.
[0059] The data processing device receives information about the service area for which a fingerprint map is to be generated (410). The data processing device can receive location information about the service area (locations of reference areas within the service area) and the locations of AP(s) deployed in the service area. It is assumed that there are n APs deployed in the service area.
[0060] First, we will describe the construction of a fingerprint map for AP 1, focusing on its generation. The data processing device can generate the aforementioned wireless signal feature vector based on the locations of reference points in the service area and the location of AP 1 (420). The wireless signal feature vector consists of the distance and angle from each reference point to AP 1. The wireless signal feature vector is information used as a condition in the cGAN.
[0061] The data processing device inputs random noise z into the generator network G (430). The generator network G is a network that has completed training through the process of FIG. 3. The generator network G generates a fingerprint map for AP 1 (440). The left side of FIG. 4 illustrates an example of generating a fingerprint map for one AP.
[0062] The data processing device repeats the same process for each AP in the service area to generate n fingerprint maps (450).
[0063] The data processing device merges n fingerprint maps to generate a single fingerprint map containing intensity vectors of multiple wireless signals (460).
[0064] Thereafter, the positioning device can determine the location of the terminal using a single fingerprint map generated for the service area.
[0065] The researchers verified the performance of the generative model that produces fingerprint maps.
[0066] The researcher briefly describes the model and fingerprint map creation process.
[0067] The researchers used TensorFlow 1.15.0. Both the generator network G and the discriminator network D they built contain a single hidden layer with a ReLu activation function. The discriminator network D, on the other hand, has a sigmoid activation function in its output layer.
[0068] The researcher sampled 100 uniformly distributed variables from U[-1, 1] and used them as input noise. The researcher built a generative model using four real-world datasets. The four real-world domains are listed in Table 1 below.
[0069] Number of APs in the location Area size (number of reference points) Area description Hall 98.1×8.7m 2 (831) - Barrier-free Lounge (Lounge) 126.6×9.9m 2 (764)-Office space with few obstacles 1 (Ofc0Obs) 169.9×9.9m 2 (1067) - Medium Obstacle Level - 6 Office Spaces - Open Door Office Space 2 (Ofc-HighObs) 106.3 x 10.8 m 2 (756) - Many obstacles - 6 conference rooms with glass walls - Doors half open
[0070] The researchers evaluated (a) the distance (in meters) between the estimated location and the actual location using the generated fingerprint map, and (b) the precision of the estimated location using the generated fingerprint map. Accuracy was determined based on a tolerance of 1.5 meters from the correct answer.
[0071] The researchers evaluated the performance of a generative model for estimating fingerprints of unexplored points. The researchers constructed generative models for each of four domains (environments) and used them to estimate the fingerprints of unexplored points in each domain. In each of the four environments, the researchers deleted values of random points from the actual fingerprint map and then used the generative model to estimate the fingerprints of the deleted points. Table 2 below shows the results of estimating the fingerprints of the unexplored points. Table 2 also shows the average of the estimation results. Estimating the fingerprints of unexplored points can be considered fingerprint map augmentation.
[0072]
[0073] In Table 2, Unaug. is the result of positioning using the unaugmented fingerprint map of unexplored points. As mentioned above, the researcher built two generative models. In Table 2, cGAN is a typical conditional GAN, and AC-GAN is a GAN with added class loss. When positioning was performed using the augmented fingerprint map, performance (distance deviation) improved by up to 0.77 m. As shown in Table 2, the generative model showed higher performance in complex service areas with many obstacles.
[0074] The researchers evaluated the performance of a generative model that estimates fingerprint maps for areas where AP placement has changed. The researchers built generative models for four different areas (environments), modified the AP placement in each area, and then used the generative models to generate fingerprint maps. Table 3 below presents the results of evaluating the fingerprint map performance in areas where AP placement has changed.
[0075]
[0076] In Table 3, the upper row represents performance when using actual fingerprint maps. Even in complex environments with many obstacles (Ofc-HighObs), the generative model demonstrated significant performance for positioning services.
[0077] The researchers evaluated the performance of a generative model for estimating fingerprint maps in an entirely new domain. They built generative models for each of four domains (environments) and used them to generate fingerprint maps for other domains. Table 4 below presents the performance evaluation results for the fingerprint maps generated in the new environments.
[0078]
[0079] In Table 4, the upper row represents performance when using actual fingerprint maps. Examining Tables 3 and 4, creating a new fingerprint map using a generative model built in the same environment (structure, obstacle presence) showed somewhat higher performance. However, creating a fingerprint map for an entirely new environment using a generative model (Table 4) also showed significant results in terms of positioning services.
[0080] Figure 5 illustrates an example of a data processing device (500) that generates a fingerprint map for an undetected area. The data processing device (500) is a device that generates a fingerprint map for a new service area using the aforementioned generation model. The data processing device (500) can be physically implemented in various forms. For example, the data processing device (500) can take the form of a computer device such as a PC, a smart device, a network server, a data processing-dedicated chipset, etc.
[0081] The data processing device (500) may include a storage device (510), a memory (520), a computing device (530), an interface device (540), a communication device (550), and an output device (560).
[0082] The storage device (510) stores the aforementioned generation model. The generation model is a model that has completed learning using learning data through the aforementioned process.
[0083] The storage device (510) can store information about a new service area. Information about the service area can include the location of the area, the location of reference points, the location of each AP, etc.
[0084] The storage device (510) can store a program that controls the process of creating a fingerprint map using a generation model.
[0085] The storage device (510) can store the generated fingerprint map.
[0086] The memory (520) can store data and information generated during the process of the data processing device (500) generating a fingerprint map.
[0087] The interface device (540) is a device that receives certain commands and data from the outside.
[0088] The interface device (540) can receive a creation model from a physically connected input device or an external storage device. The interface device (540) can receive information about a service area from a physically connected input device or an external storage device.
[0089] The interface device (540) can also transmit the generated fingerprint map to an external object.
[0090] The interface device (540) is meant to include a configuration that transmits data or information received by the communication device (550) into the data processing device (500).
[0091] A communication device (550) refers to a configuration that receives and transmits certain information through a wired or wireless network.
[0092] The communication device (550) can receive a creation model from an external object.
[0093] The communication device (550) can receive information about the service area from an external object.
[0094] The communication device (550) may also transmit the generated fingerprint map to an external object such as a positioning device.
[0095] The output device (560) is a device that outputs certain information. The output device (560) can output interfaces, fingerprint maps, etc. required for the data processing process.
[0096] The computing device (530) uses information of the service area to compute the distance and angle between each reference point of the service area and the AP. The computing device (530) generates the feature vector described above. The computing device (530) generates the feature vector (d) for the reference point i. i , a i ) is generated. The computing device (530) generates a feature vector set for n reference points located in the entire service area. The computing device (530) calculates the feature vector set for each AP.
[0097] The computing device (530) generates a fingerprint map for each AP deployed in the service area. The computing device (530) inputs feature vectors extracted under arbitrary noise and conditions into the generation model to generate a fingerprint map. The computing device (530) generates a fingerprint map for each AP. The computing device (530) combines the fingerprint maps of all APs to generate a final fingerprint map containing strength information for multiple wireless signals. This process is as described in FIG. 4.
[0098] The computing device (530) may be a device such as a processor, AP, or chip embedded with a program that processes data and performs certain operations.
[0099] In some cases, the data processing device (500) may be a positioning device. In this case, the data processing device (500) can perform positioning based on wireless signal information received in the service area using the generated fingerprint map.
[0100] Fig. 6 is an example of a positioning device (600) using a virtual fingerprint map.
[0101] The positioning device (600) can collect information about the service area in which it is located. The positioning device (600) can transmit the service area information to the data processing device. The positioning device (600) can receive a fingerprint map for the service area generated by the data processing device using a generation model. The fingerprint map generated by the data processing device is not a map generated by collecting actual wireless signal strength, and thus can be referred to as a virtual fingerprint map. In the following description, it is assumed that the positioning device (600) possesses a virtual fingerprint map for the service area in which it is currently located.
[0102] The positioning device (600) may be any one of various types of devices. For example, the positioning device (600) may be any one of various types of devices, such as a smart device, a wearable device, a robot, or a vacuum cleaner.
[0103] The positioning device (600) may include a storage device (610), a memory (620), a computing device (630), an interface device (640), and a communication device.
[0104] The storage device (610) stores the virtual fingerprint map described above.
[0105] The storage device (610) can store a program that estimates the current location using a fingerprint map. For example, the program may be a program that estimates the location based on the Nearest Neighbor (NN) algorithm.
[0106] The memory (620) can store data and information generated during the process of positioning using the fingerprint map by the positioning device (600).
[0107] The interface device (640) is a device that receives certain commands and data from the outside.
[0108] The interface device (640) can receive a virtual fingerprint map from a physically connected input device or an external storage device.
[0109] The interface device (640) is meant to include a configuration that transmits data or information received by the communication device (650) into the positioning device (600).
[0110] A communication device (650) refers to a configuration that receives and transmits certain information through a wired or wireless network.
[0111] The communication device (650) can receive a fingerprint map from an external object.
[0112] The communication device (650) can receive wireless signals transmitted by APs in the service area.
[0113] The computing device (630) can determine the strength of the wireless signal transmitted by the APs at the current location.
[0114] The computing device (630) can estimate the current location by comparing the virtual fingerprint map with the wireless signal strength of the current location. The wireless signal strength may include strength information for each of a plurality of wireless signals.
[0115] The computing device (630) may be a device such as a processor, AP, or chip embedded with a program that processes data and performs certain operations.
[0116] Additionally, the above-described generative model construction method, generative model-based fingerprint map creation method, or positioning method may be implemented as a program (or application) including an executable algorithm that can be executed on a computer. The program may be stored and provided on a temporary or non-transitory computer-readable medium.
[0117] A non-transitory readable medium refers to a medium that stores data semi-permanently and can be read by a device, rather than a medium that stores data for a short period of time, such as a register, cache, or memory. Specifically, the various applications or programs described above may be stored and provided on a non-transitory readable medium, such as a CD, DVD, hard disk, Blu-ray disk, USB, memory card, ROM (read-only memory), PROM (programmable read only memory), EPROM (Erasable PROM, EPROM), EEPROM (Electrically EPROM), or flash memory.
[0118] Temporarily readable media refers to various types of RAM, such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDR SDRAM), Enhanced SDRAM (ESDRAM), Synchronous DRAM (Synclink DRAM, SLDRAM), and Direct Rambus RAM (DRRAM).
[0119] The present embodiment and the drawings attached to the present specification only clearly illustrate a part of the technical idea included in the above-described technology, and it is obvious that all modified examples and specific embodiments that can be easily inferred by a person skilled in the art within the scope of the technical idea included in the specification and drawings of the above-described technology are included in the scope of the rights of the above-described technology.
Claims
1. A step in which a data processing device receives input of the locations of reference points for a service area and the locations of multiple APs (access points); A step in which the data processing device extracts a feature vector based on the location of the reference point and the location of the AP for each of the APs; A step in which the data processing device inputs a feature vector extracted for each AP as a condition to a pre-learned generative model for each of the APs and inputs the random noise to generate a fingerprint map for each of the APs; and A method for generating a fingerprint map for an unexplored area, comprising the step of combining fingerprint maps generated by the data processing device for the APs to generate a final fingerprint map for the service area.
2. In paragraph 1, The above generative model is a method for generating a fingerprint map for an unexplored area, which is a conditional Generative Adversarial Network (GAN) or Auxiliary Classifier GAN (AC-GAN).
3. In paragraph 1, A method for generating a fingerprint map for an unexplored area, wherein the feature vector for any one of the above APs includes a distance between one reference point and the one AP and an angle formed between the one reference point and the one AP.
4. A step in which a data processing device receives input of the locations of reference points for the service area and the locations of APs (access points); A step in which the data processing device extracts the distance from each of the reference points to the AP and the angle formed with the AP; and The data processing device includes a step of inputting random noise into a pre-learned generative model and inputting the distance and angle between the reference points as conditions to generate a fingerprint map for the service area, A method for generating a fingerprint map for an unexplored area, wherein the above-mentioned generation model is a model trained to generate a fingerprint map for a specific area by receiving a fingerprint map of a specific area and a feature vector extracted based on APs at reference points of the specific area as input conditions.
5. Input device for receiving the location of reference points for the service area and the location of AP (access point); A storage device that stores a generative model trained to generate a fingerprint map for a specific area by receiving a fingerprint map of a specific area and a feature vector extracted based on APs at reference points of the specific area as input conditions; and A data processing device for generating a fingerprint map for an unexplored area, including a computational device for extracting the distance to the AP and the angle formed with the AP from each of the reference points for the service area and inputting the extracted distance and the angle formed with the AP as conditions for the generation model, and inputting arbitrary noise into the generation model to generate a fingerprint map for the service area.
6. In paragraph 5, A data processing device for generating a fingerprint map for an unexplored area, wherein the feature vector includes a distance between one reference point and the AP of the specific area for each of the reference points of the specific area and an angle formed between the one reference point and the AP of the specific area.
7. In paragraph 5, The above generative model is a data processing device that generates a fingerprint map for an unexplored area, which is a conditional Generative Adversarial Network (GAN) or Auxiliary Classifier GAN (AC-GAN).
8. A communication device that receives wireless signals from APs (access points) at the current location; A storage device for storing a virtual fingerprint map; and Including a calculation device that estimates the current location based on the strength of the wireless signals of the APs by referring to the above virtual fingerprint map, A positioning device using a virtual fingerprint map, wherein the virtual fingerprint map is generated by a generative model trained to generate a fingerprint map for a specific area by receiving a fingerprint map of a specific area and a feature vector extracted based on APs at reference points of the specific area as input conditions.
9. In paragraph 8, A positioning device using a virtual fingerprint map in which the above feature vector includes a distance between one reference point and the AP of the specific area for each of the reference points of the specific area and an angle formed by the one reference point and the AP of the specific area.
Citation Information
Patent Citations
Indoor positioning apparatus considering environmental parameters and method thereof
KR101600190B1
Localization system using fingerprint
KR1020110026229A
Method and apparatus for estimating location using access point in wireless communication system
KR1020170108376A
Lens driving member
KR1020230100581A
Method and Apparatus for Fingerprint technique based on Wi-Fi signal characteristics and indoor space structure
KR102064680B1
