Device and method

Abstract

A device according to one aspect is provided with an acquisition unit for acquiring an input data set including an outer shape diagram indicating the outer shape of a floor plan and information indicating an installation position of a wireless communication device within the outer shape diagram, and a generation unit for outputting a floor plan diagram having the same outer shape as the outer shape of the floor plan and having an excellent radio wave propagation characteristic, by inputting the input data set to an AI model trained through machine learning using a training data set including a training floor plan diagram having an excellent radio wave propagation characteristic and information indicating an installation position of the wireless communication device within the training floor plan diagram.

Description

Device and method

[0001] The present disclosure relates to a device and a method.

[0002] Patent Document 1 describes obtaining floor plan data representing the floor plan of a building, calculating the position coordinates of devices based on the floor plan data and the communication states of a wireless master unit and wireless devices, and generating screen data including the layout of the wireless devices in the floor plan data for display.

[0003] Japanese Patent Application Laid-Open No. 2020-95485

[0004] In recent years, many wireless devices have been introduced in houses or offices, and wireless networks have been constructed, increasing the importance of wireless communication. On the other hand, it is not common to design the floor plan of a building considering radio propagation characteristics. Therefore, depending on the floor plan of the building, radio waves may be attenuated by being shielded by the inner walls, fixtures, furniture, etc. of the building, and locally low radio wave intensity areas (radio wave dead zones) may be formed. When radio wave dead zones occur, stable wireless communication within the building cannot be achieved, and there is a risk of a floor plan with low convenience.

[0005] Therefore, an object of the present disclosure is to provide a device and a method capable of generating a floor plan excellent in radio propagation characteristics.

[0006] A device according to one aspect includes an acquisition unit that acquires an input data set including an outline drawing showing the outline of a floor plan and information indicating the installation position of a wireless communication device in the outline drawing, and an AI model learned by machine learning using a training data set including a training floor plan excellent in radio propagation characteristics and information indicating the installation position of a wireless communication device in the training floor plan, and inputs the input data set into the AI model, and a generation unit that outputs a floor plan having the same shape as the outline of the floor plan and excellent in radio propagation characteristics.

[0007] In the device according to the above aspect, based on an outline drawing showing the outline of a floor plan and the installation position of a wireless communication device in the outline drawing, a floor plan having the same shape as the outline of the floor plan and excellent in radio propagation characteristics is output. In this aspect, it is possible to provide a floor plan excellent in radio propagation characteristics according to the installation position of the wireless communication device while maintaining the outline of the floor plan that depends on the shape of the building.

[0008] According to this disclosure, it is possible to generate floor plans with excellent wireless propagation characteristics.

[0009] This is a block diagram showing the functional configuration of a floor plan generation device according to one embodiment. This is a diagram showing an example of a floor plan. This is a diagram showing multiple regions of a floor plan. This is a diagram showing an example of an outline drawing. This is a diagram showing an example of a floor plan generated by the floor plan generation unit. This is a block diagram showing the functional configuration of an AI model. This is a diagram showing the input and output of the AI ​​model in the generation phase. This is a flowchart showing the learning phase of the floor plan generation method. This is a flowchart showing the generation phase of the floor plan generation method. This is a block diagram showing the functional configuration of a floor plan generation program according to one embodiment. This is a block diagram showing the functional configuration of a floor plan generation device according to another embodiment. This is a block diagram showing the hardware configuration of a floor plan generation device.

[0010] Embodiments of the present disclosure will be described below with reference to the drawings. In the following description, the same or equivalent elements will be denoted by the same reference numerals, and redundant descriptions will not be repeated.

[0011] Figure 1 is a block diagram showing the functional configuration of a floor plan generation device according to one embodiment. The floor plan generation device 1 shown in Figure 1 provides information showing a floor plan that exhibits excellent wireless propagation characteristics. A floor plan is information that shows the arrangement of exterior walls, interior walls, doors, windows, stairs, etc. of a building. A floor plan is a diagram that visually shows the layout of a building. Typically, a floor plan is image data showing the arrangement of exterior walls, interior walls, doors, windows, stairs, etc. of a building.

[0012] A floor plan with excellent wireless propagation characteristics means a floor plan with few areas of low radio wave intensity (blind spots) and where stable wireless communication is possible at any location within the building. Note that "building" includes not only the entire building but also a section of a building (such as an apartment building).

[0013] As shown in Figure 1, the floor plan generation device 1 comprises a data set generation unit 10, a floor plan generation unit 20, and a storage unit 30. In the example shown in Figure 1, the data set generation unit 10, the floor plan generation unit 20, and the storage unit 30 are implemented by a single floor plan generation device 1, but these functional elements may be distributed and arranged across multiple devices.

[0014] The dataset generation unit 10 and the floor plan generation unit 20 can access the storage unit 30. The storage unit 30 is a storage device that stores the floor plan dataset DS1, the first training dataset DS2, the second training dataset DS3, and the AI ​​model 40. The storage unit 30 may be located inside the floor plan generation device 1, as shown in Figure 1, or it may be located outside the floor plan generation device 1 so that it can be accessed from the floor plan generation device 1.

[0015] The floor plan dataset DS1 includes multiple floor plans d11. These multiple floor plans d11 are collected, for example, by the administrator of the floor plan generation device 1 and stored in the storage unit 30 beforehand. The multiple floor plans d11 in the floor plan dataset DS1 show different floor plans from each other. That is, the shapes, dimensions, and arrangements of the exterior walls, interior walls, doors, windows, stairs, etc., in the multiple floor plans d11 are different from each other. The floor plan dataset DS1 also includes coordinate information d12 indicating the installation location of the wireless router WR in each floor plan d11.

[0016] Figure 2 shows an example of floor plan d11 from the floor plan dataset DS1. As shown in Figure 2, floor plan d11 shows the shape, dimensions, and arrangement of the exterior wall OW, interior wall IW, and door DR. The exterior wall OW defines the interior space of the building. That is, the shape of the exterior wall OW indicates the outer shape of the building's floor plan. Note that if floor plan d11 represents the floor plan of a section of a building such as an apartment, floor plan d11 may have boundary walls that define the boundary with adjacent sections instead of exterior walls. The interior wall IW and door DR are located inside the exterior wall OW. The shape, dimensions, and arrangement of the interior wall IW and door DR affect the propagation of radio waves radiated from the wireless router WR.

[0017] The coordinate information d12 of the wireless router WR indicates the installation location of the wireless router WR in the floor plan d11. The wireless router WR is a wireless communication device that provides internet connectivity and is connected to the internet via a wired or wireless line. The wireless router WR emits radio waves within the building to provide internet connectivity to terminals within the building via wireless communication (e.g., Wi-Fi). The installation location of the wireless router WR affects the electric field distribution within the building.

[0018] The floor plan dataset DS1 may also include coordinate information d13 indicating the position of furniture F within each floor plan d11. In the example shown in Figure 2, the position information of furniture F, such as kitchen equipment and a bed, is associated with the floor plan d11.

[0019] Terminals that communicate with the wireless communication device include, for example, personal computers, laptop computers, tablet devices, smartphones, home appliances, game consoles, PDAs (Personal Digital Assistants), etc., and are operated by the user. The number of terminals connected to the wireless router WR may be two or more. Details of the first training dataset DS2, the second training dataset DS3, and the AI ​​model 40 stored in the memory unit 30 will be described later.

[0020] The dataset generation unit 10 generates a first training dataset DS2 for training the AI ​​model 40. As shown in Figure 1, the dataset generation unit 10 includes an acquisition unit 11, a propagation estimation unit 12, an evaluation unit 13, and a storage unit 14. The acquisition unit 11 acquires the floor plan dataset DS1 from the storage unit 30.

[0021] The propagation estimation unit 12 generates a throughput map for each floor plan d11, showing the throughput of wireless communication, based on the floor plan dataset DS1. To generate the throughput map, the propagation estimation unit 12 divides each floor plan d11 into multiple regions r(x, y). For example, in Figure 3, the floor plan d11 is divided into multiple regions r(x, y) in a two-dimensional (mesh-like) structure.

[0022] The propagation estimation unit 12 calculates the electric field strength in each region r(x,y) based on the floor plan d11, the coordinate information d12 of the wireless router WR, and the coordinate information d13 of the furniture F. Specifically, the propagation estimation unit 12 calculates the electric field strength in each region r(x,y) by simulating the position of the wireless router WR, the antenna characteristics of the wireless router WR (radiation pattern, frequency, output power, etc.), the transmission loss, reflection, diffraction, etc. at the outer wall OW, inner wall IW, door DR, and furniture F. Alternatively, the propagation estimation unit 12 may obtain the electric field strength in each region r(x,y) of the floor plan d11 using a prediction model that has been machine-learned using multiple floor plans d11 and electric field strength distributions.

[0023] Furthermore, the propagation estimation unit 12 calculates the throughput in each region r(x, y) from the acquired electric field strength distribution using the Shannon-Hartley theorem. Throughput refers to the data transmission speed of wireless communication between the wireless router WR and the terminal in each region r(x, y). The propagation estimation unit 12 stores a throughput map showing the throughput in each region r(x, y) in the storage unit 30 for each floor plan d11.

[0024] The evaluation unit 13 evaluates whether each of the multiple floor plans d11 in the floor plan dataset DS1 has good radio wave propagation characteristics based on the throughput map generated by the propagation estimation unit 12. Specifically, the evaluation unit 13 obtains the minimum throughput value for each of the multiple regions r(x, y) of the throughput map for each floor plan d11 and determines whether the obtained minimum value is higher than the reference value. If the minimum throughput value is below the reference value, it can be said that there are radio wave blind spots in that floor plan d11, and therefore the radio wave propagation characteristics of that floor plan d11 are determined to be poor. On the other hand, if the minimum throughput value is higher than the reference value, it can be said that there are no radio wave blind spots in that floor plan d11, and therefore the electric field propagation characteristics of that floor plan d11 are determined to be good.

[0025] The storage unit 14 stores the floor plan d11, which the evaluation unit 13 has determined to have good electric field propagation characteristics, as a training floor plan d21 in the storage unit 30. That is, the training floor plan d21 is a floor plan with excellent electric field propagation characteristics. The storage unit 14 also stores the coordinate information d12 of the wireless router WR and the coordinate information d13 of the furniture F corresponding to the floor plan d11 that has been determined to have good electric field propagation characteristics as the coordinate information d22 of the wireless router WR and the coordinate information d23 of the furniture F, respectively, in the storage unit 30. The training floor plan d21, the coordinate information d22 of the wireless router WR, and the coordinate information d23 of the furniture F constitute the first training dataset DS2. Typically, the first training dataset DS2 includes multiple training floor plans d21, multiple coordinate information d22 of the wireless router WR, and multiple coordinate information d23 of the furniture F.

[0026] As shown in Figure 1, the floor plan generation unit 20 comprises an acquisition unit 21, a generation unit 22, and an output unit 23. The floor plan generation unit 20 generates floor plans with excellent radio wave propagation characteristics using a machine learning-trained AI model 40.

[0027] The floor plan generation unit 20 iteratively trains the AI ​​model 40 using the first training dataset DS2 and the second training dataset DS3. This is the training phase. The floor plan generation unit 20 also inputs an outline drawing showing the shape of the building's exterior to the trained AI model 40 and generates a floor plan that has the same exterior shape as the building and has excellent radio wave propagation characteristics. This is the generation phase (estimation phase).

[0028] The trained AI model 40 is portable between computer systems. Therefore, a trained model generated on one computer system can be used on another computer system. Of course, one computer system may perform both the generation and use of the trained AI model 40. That is, the floor plan generation unit 20 may perform both the training phase and the operation phase, or it may perform only one of the training phase or the generation phase.

[0029] First, let's explain the operation of the floor plan generation unit 20 during the learning phase. During the learning phase, the acquisition unit 21 of the floor plan generation unit 20 acquires the first training dataset DS2 and the second training dataset DS3 from the storage unit 30.

[0030] The second training dataset DS3 includes multiple outline drawings d31. An outline drawing d31 is a diagram that visually shows the outline of a building. Typically, an outline drawing d31 is image data showing the shape, dimensions, and arrangement of the building's exterior walls or boundary walls. In other words, the building's outline shows the shape, dimensions, and arrangement of the outer frame of the floor plan. Multiple outline drawings d31 are collected, for example, by the administrator of the floor plan generation device 1 and stored in the storage unit 30 in advance. Typically, multiple outline drawings d31 show different outlines from each other. That is, the shape, dimensions, or arrangement of the exterior walls in the multiple outline drawings d31 are different from each other. The second training dataset DS3 also includes coordinate information d32 indicating the installation position of the wireless router WR in each outline drawing d31.

[0031] Figure 4 shows an example of the outline drawing d31 of the second training dataset DS3. As shown in Figure 4, the outline drawing d31 includes information about the exterior wall OW that defines the outline of the building, but does not include information about the building's fixtures (interior wall IW and door DR) located inside the exterior wall OW.

[0032] As shown in Figure 4, the coordinate information d12 indicates the installation position of the wireless router WR in the outline drawing d31. The second training dataset DS3 may also include coordinate information d33 indicating the position of furniture F within the outline drawing d31. In the example shown in Figure 4, the position information of furniture F, such as kitchen equipment and a bed, is associated with the outline drawing d31.

[0033] The generation unit 22 inputs the outline drawing and the coordinate information of the wireless router WR into the machine-learned AI model 40 and outputs a floor plan that has the same shape as the building outline shown in the outline drawing and has excellent radio wave propagation characteristics. For example, when the outline drawing shown in Figure 4, the coordinate information of the wireless router WR, and the coordinate information of the furniture are input, the generation unit 22 generates a floor plan in which the interior wall IW and door DR are added inside the exterior wall OW, as shown in Figure 5. At this time, the generation unit 22 adds the interior wall IW and door DR while maintaining the positions of the wireless router WR and furniture F. The generation unit 22 also generates a floor plan that corresponds to the type of furniture F. That is, it generates a floor plan in which the area where the bed is placed becomes the bedroom, and the area where the kitchen equipment is placed becomes the kitchen. The floor plan generated by the generation unit 22 is a pseudo-floor plan that does not actually exist (it is not collected by the administrator). In the following explanation, the floor plan that is pseudo-generated by the AI ​​model 40 is called the "generated floor plan".

[0034] Figure 6 shows the functional configuration of AI model 40. AI model 40 is generated by machine learning a generative adversarial network (GAN). Specifically, as shown in Figure 6, AI model 40 includes a generative model 41 and a discriminative model 42. Each of the generative model 41 and the discriminative model 42 is a machine learning model that has a neural network.

[0035] The generation model 41 is configured to take the second training dataset DS3 as input and generate a generated floor plan having an exterior wall OW with the same shape as the exterior wall OW shown in the outline drawing of the second training dataset DS3. For example, when the outline drawing d31 shown in Figure 4 is input to the generation model 41, the generation model 41 generates a generated floor plan d50, as shown in Figure 5, in which an interior wall IW and a door DR are added inside the exterior wall OW. Coordinate information indicating the positions of the wireless router WR and furniture F is associated with the generated floor plan d50.

[0036] The discrimination model 42 receives the training floor plan d21 from the first training dataset DS2 acquired by the acquisition unit 21, and the generated floor plan d50 generated by the generation model 41 as input. The discrimination model 42 is configured to distinguish between a real floor plan and a pseudo-generated floor plan from the two input floor plans. A real floor plan is an actual existing floor plan collected by the administrator. In other words, the discrimination model 42 is trained to determine that the training floor plan d21 is a real floor plan and that the generated floor plan d50 is a pseudo-floor plan. In the learning phase, multiple training floor plans d21 and multiple generated floor plans d50 are input to the discrimination model 42 in sequence, and the parameters of the neural network of the discrimination model 42 are updated so that the loss function is minimized so that the loss value is small when the discrimination result of the discrimination model 42 matches the correct answer.

[0037] On the other hand, the generative model 41 is trained so that the discriminative model 42 determines that the generated floor plan d50 is a real floor plan. To this end, the generative model 41 is repeatedly input with the discriminative model 42's discrimination results, and the parameters of the neural network of the generative model 41 are updated so that the loss function is minimized so that the loss value is small when the discriminative model 42's discrimination result differs from the correct answer. As described above, the trained floor plan d21 is a floor plan with excellent radio wave propagation characteristics. By repeatedly training the generative model 41 to generate a generated floor plan d50 that the discriminative model 42 mistakenly identifies as a real floor plan, the generated floor plan d50 output from the generative model 41 becomes more and more superior in radio wave propagation characteristics.

[0038] Next, the operation of the floor plan generation unit 20 in the generation phase will be described. In the generation phase, the acquisition unit 21 acquires an input dataset DS4 from the user's terminal, which includes an outline drawing d41 showing the outline of the building's floor plan and coordinate information d42 of the wireless router WR. That is, the outline drawing d41 included in the input dataset DS4 is image data that includes information about the exterior wall OW that defines the outline of the building, as shown in the outline drawing d31 in Figure 4, but does not include information about the building's fixtures (interior wall IW and door DR) that are located inside the exterior wall OW. The input dataset DS4 is created, for example, by the user. In one embodiment, the input dataset DS4 may also include coordinate information d43 of furniture F.

[0039] As shown in Figure 7, the generation unit 22 inputs the input dataset DS4 to the generation model 41 of the trained AI model 40 and outputs a generated floor plan d50. As described above, the generation model 41 is trained to generate a generated floor plan d50 that the discrimination model 42 will mistakenly identify as a real floor plan. Therefore, the generated floor plan d50 output from the generation model 41 has an exterior wall OW with the same shape as the exterior wall OW of the building shown in the outline drawing d41, and is a floor plan with excellent radio wave propagation characteristics.

[0040] The output unit 23 outputs the generated floor plan d50 generated by the generation unit 22. The output unit 23 may display the generated floor plan d50 on the display device of the floor plan generation device 1, or it may transmit the generated floor plan d50 to the user's terminal via the network.

[0041] Next, a floor plan generation method according to one embodiment will be described with reference to Figures 8 and 9. Figure 8 is a flowchart showing the learning phase of the floor plan generation method.

[0042] As shown in FIG. 8, first, in the learning phase of the floor plan generation method, the acquisition unit 21 acquires the first training dataset DS2 and the second training dataset DS3 from the storage unit 30 (step ST1). The first training dataset DS2 includes a training floor plan d21 having good radio wave propagation characteristics and coordinate information d22 of the wireless router WR in the training floor plan d21. The first training dataset DS2 may include coordinate information d23 of the furniture F in the training floor plan d21. The second training dataset DS3 includes an outline drawing d31 showing the outline of the floor plan and coordinate information d32 of the wireless router WR in the outline drawing d31. The second training dataset DS3 may include coordinate information d33 of the furniture F in the outline drawing d31.

[0043] Next, the generation unit 22 inputs the second training dataset DS3 into the AI model 40 (step ST2). The generation model 41 of the AI model 40 generates a generated floor plan d50 having an outer wall OW having the same shape as the outer wall OW of the input outline drawing d31 of the second training dataset DS3 and including an inner wall IW, a door DR, and the like (step ST3).

[0044] Next, the generation unit 22 inputs the training floor plan d21 of the first training dataset DS2 and the generated floor plan d50 generated by the generation model 41 into the discrimination model 42 of the AI model 40 (step ST4). At this time, the coordinate information indicating the positions of the wireless router WR and the furniture F associated with the generated floor plan d50 is also input into the discrimination model 42.

[0045] Next, the discrimination model 42 outputs the probabilities that the input training floor plan d21 and generated floor plan d50 of the first training dataset DS2 are genuine floor plans, respectively (step ST5). A genuine floor plan means a training floor plan with excellent radio wave propagation characteristics, not a floor plan pseudo-generated by the generation model 41. That is, the discrimination model 42 discriminates whether the two input floor plans are the generated floor plan d50 generated by the generation model 41.

[0046] Next, the discrimination model 42 updates the parameters of the neural network of the discrimination model 42 so that the probability of determining that the training floor plan d21 of the first training data set DS2 is an actual floor plan becomes high and the probability of determining that the generated floor plan d50 is an actual floor plan becomes low (step ST6). For example, the parameters of the neural network of the discrimination model 42 are updated so as to minimize a loss function such that the loss value becomes small when the discrimination result of the discrimination model 42 matches the correct answer.

[0047] Also, the generation model 41 updates the parameters of the neural network of the generation model 41 so that the probability of determining that the generated floor plan d50 is an actual floor plan becomes high (step ST7). For example, the parameters of the neural network of the generation model 41 are updated so as to minimize a loss function such that the loss value becomes small when the discrimination result of the discrimination model 42 differs from the correct answer.

[0048] The above steps ST1 to ST7 are repeatedly executed until the end condition is satisfied (step ST8). By repeatedly executing the above steps ST1 to ST7, the generation model 41 is learned so that a generated floor plan d50 having an outer wall OW having the same shape as the outer wall OW of the external view d31 of the second training data set DS3 and having excellent radio wave propagation characteristics is generated.

[0049] Next, referring to FIG. 9, the generation phase of the floor plan generation method according to an embodiment will be described. FIG. 9 is a flowchart showing the generation phase of the floor plan generation method.

[0050] As shown in FIG. 9, in the generation phase of the floor plan generation method, the acquisition unit 21 acquires an input data set DS4 from, for example, a user's terminal (step ST11). The input data set DS4 includes an external view d41 indicating the external shape of the floor plan specified by the user and coordinate information d42 of the wireless router WR specified by the user. In one embodiment, the input data set DS4 may include coordinate information d43 of the furniture F specified by the user.

[0051] Next, the floor plan generation unit 20 inputs the input dataset DS4 into the trained AI model 40 (step ST12). Upon receiving the input dataset DS4, the generation model 41 of the AI ​​model 40 generates a generated floor plan d50 that has an exterior wall OW with the same shape as the exterior wall OW of the outline drawing d41 of the input dataset DS4, and includes interior walls IW and doors DR, etc. (step ST13). As described above, the generation model 41 of the AI ​​model 40 is trained to generate floor plans with excellent radio wave propagation characteristics, so the generated floor plan d50 is a floor plan with excellent radio wave propagation characteristics with few radio wave blind spots.

[0052] Next, the output unit 23 outputs the generated floor plan d50, which was created by the AI ​​model 40, to the user's terminal (step ST14).

[0053] Next, with reference to Figure 10, a floor plan generation program for causing a computer to function as a floor plan generation device 1 according to one embodiment will be described. Figure 9 is a diagram showing the functional configuration of the floor plan generation program. The floor plan generation program P comprises a main module M1, a data set generation module M10, and a floor plan generation module M20. The main module M1 comprehensively controls the processing of the floor plan generation device 1.

[0054] The dataset generation module M10 includes an acquisition module M11, a propagation estimation module M12, an evaluation module M13, and a storage module M14. The floor plan generation module M20 includes an acquisition module M21, a generation module M22, and an output module M23.

[0055] The functions realized by executing the acquisition module M11, propagation estimation module M12, evaluation module M13, storage module M14, acquisition module M21, generation module M22, and output module M23 are the same as the functions of the acquisition unit 11, propagation estimation unit 12, evaluation unit 13, storage unit 14, acquisition unit 21, generation unit 22, and output unit 23 described above.

[0056] The floor plan generation program P is provided, for example, in a form recorded on a recording medium M such as a CD-ROM or DVD-ROM, or on a semiconductor memory. Alternatively, the floor plan generation program P may be provided via a communication network as a computer data signal superimposed on a carrier wave.

[0057] As described above, in the floor plan generation device 1 according to one embodiment, a generated floor plan d50 is output based on an outline drawing d41 showing the outline of the floor plan and coordinate information d42 of the wireless router WR in the outline drawing d41. The generated floor plan d50 has exterior walls or boundary walls of the same shape as the exterior walls or boundary walls in the outline drawing d41 and has excellent radio wave propagation characteristics. The floor plan generation device 1 can provide a floor plan with excellent radio wave propagation characteristics according to the installation position of the wireless router WR while maintaining the shape of the exterior walls or boundary walls which depend on the shape of the building.

[0058] Furthermore, the floor plan generation device 1 is not limited to the configuration shown in Figure 1, and as shown in Figure 11, at least a portion of the floor plan dataset DS1, the first training dataset DS2, the second training dataset DS3, and the AI ​​model 40 may be stored in a storage unit 30 located outside the floor plan generation device 1 so that it can be accessed from the floor plan generation device 1. In addition, each component of the floor plan generation device 1 may be implemented on the user's terminal.

[0059] The apparatus and method relating to this disclosure may have the following configurations.

[0060] An apparatus according to one embodiment includes an acquisition unit that acquires an input dataset including an outline drawing showing the external shape of a floor plan and information indicating the installation location of a wireless communication device in the outline drawing, and a generation unit that inputs the input dataset to an AI model trained using a training dataset including a training floor plan with excellent radio wave propagation characteristics and information indicating the installation location of a wireless communication device in the training floor plan, and outputs a floor plan having the same shape as the external shape of the floor plan and having excellent radio wave propagation characteristics.

[0061] The device according to this embodiment can provide a floor plan with excellent radio wave propagation characteristics according to the installation location of the wireless communication device, while maintaining the external shape of the floor plan which depends on the shape of the building.

[0062] In one embodiment, the AI ​​model may be generated by machine learning a generative adversarial network (GAN). In this embodiment, by machine learning a GAN, it is possible to generate a floor plan with excellent radio wave propagation characteristics.

[0063] In one embodiment, the AI ​​model may include a generative model trained to generate floor plans having the same shape as the exterior of the floor plan, and a discriminative model trained to take the floor plan generated by the generative model and a training floor plan as input and determine whether or not the input floor plan was generated by the generation unit. In this embodiment, by making the two neural networks of the generative model and the discriminative model compete, the generative model can be trained to generate floor plans with excellent radio wave propagation characteristics.

[0064] An apparatus according to one embodiment may further include a dataset generation unit that generates a training dataset, the dataset generation unit including: an acquisition unit that acquires a floor plan dataset including a floor plan and information indicating the installation location of wireless communication devices in the floor plan; a propagation estimation unit that generates a throughput map showing the throughput of wireless communication for each of a plurality of areas that demarcate the floor plan of the floor plan dataset based on the floor plan dataset; an evaluation unit that evaluates whether the floor plan of the floor plan dataset has good radio wave propagation characteristics based on the throughput map; and a storage unit that, when the floor plan of the floor plan dataset is evaluated to have good radio wave propagation characteristics, stores the floor plan in a storage unit as a training floor plan. In this embodiment, a training floor plan with good radio wave propagation characteristics can be extracted from the floor plan dataset.

[0065] In one embodiment, the evaluation unit may evaluate that the floor plan in the floor plan dataset has excellent radio wave propagation characteristics when the minimum throughput value in multiple regions is greater than a reference value. In this embodiment, it is possible to generate a training floor plan with excellent radio wave propagation characteristics that does not include regions with locally low throughput.

[0066] In one embodiment, the training dataset and the input dataset may further include information indicating the location of furniture. In this embodiment, a floor plan can be generated taking into account radio wave loss, reflection, and diffraction caused by furniture.

[0067] A method according to one embodiment includes the steps of: acquiring an input dataset including an outline drawing showing the external shape of a floor plan and information indicating the installation location of a wireless communication device in the outline drawing; and inputting the input dataset into an AI model trained using a training dataset including a training floor plan with excellent radio wave propagation characteristics and information indicating the installation location of a wireless communication device in the training floor plan, and outputting a floor plan having the same external shape as the external shape of the floor plan and having excellent radio wave propagation characteristics.

[0068] The block diagram shown in Figure 1 represents functional units. These functional blocks (components) are realized by any combination of at least one of hardware and software. Furthermore, the method of realizing each functional block is not particularly limited. That is, each functional block may be realized using one device that is physically or logically coupled, or it may be realized using two or more physically or logically separated devices that are directly or indirectly connected (for example, using wired or wireless connections). A functional block may also be realized by combining software with the one or more devices described above.

[0069] Functions include, but are not limited to, judgment, decision, determination, calculation, calculation, processing, derivation, investigation, exploration, confirmation, reception, transmission, output, access, resolution, selection, selection, establishment, comparison, assumption, expectation, assumption, broadcasting, notifying, communicating, forwarding, configuring, reconfiguring, allocating (mapping), and assigning. For example, a functional block (configuration part) that enables transmission is called a transmitting unit or transmitter. In all cases, as mentioned above, the method of implementation is not particularly limited.

[0070] For example, the floor plan generation device 1 in one embodiment may function as a computer. Figure 12 shows an example of the hardware configuration of the floor plan generation device 1 according to this embodiment. Physically, the floor plan generation device 1 may be configured as a computer device including a processor 1001, memory 1002, storage 1003, communication device 1004, input device 1005, output device 1006, bus 1007, etc.

[0071] In the following explanation, the term "device" can be replaced with "circuit," "device," "unit," etc. The hardware configuration of the floor plan generation device 1 may include one or more of the devices shown in Figure 12, or it may be configured without some of the devices.

[0072] Each function in the floor plan generation device 1 is realized by loading predetermined software (programs) onto hardware such as the processor 1001 and memory 1002, causing the processor 1001 to perform calculations and control communication by the communication device 1004, as well as the reading and / or writing of data in the memory 1002 and storage 1003.

[0073] The processor 1001 controls the entire computer, for example, by running an operating system. The processor 1001 may consist of a central processing unit (CPU) that includes interfaces with peripheral devices, control units, arithmetic units, registers, etc. For example, each component shown in Figure 1 may be implemented in the processor 1001.

[0074] Furthermore, the processor 1001 reads programs (program code), software modules, and data from the storage 1003 and / or communication device 1004 into the memory 1002, and executes various processes accordingly. The program used is one that causes the computer to execute at least a part of the operations described in the above embodiment. For example, each component of the floor plan generation device 1 may be stored in the memory 1002 and implemented by a control program that runs on the processor 1001. Although the above-described processes have been explained as being executed by one processor 1001, they may be executed simultaneously or sequentially by two or more processors 1001. The processor 1001 may be implemented on one or more chips. The program may also be transmitted from a network via a telecommunications line.

[0075] The memory 1002 is a computer-readable recording medium and may consist of at least one of the following: ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), RAM (Random Access Memory), etc. The memory 1002 may also be called a register, cache, main memory, etc. The memory 1002 can store executable programs (program code), software modules, etc., for carrying out an information processing method according to one embodiment of the present invention.

[0076] The storage 1003 is a computer-readable recording medium and may consist of at least one of the following: an optical disc such as a CD-ROM (Compact Disc ROM), a hard disk drive, a flexible disk, a magneto-optical disk (e.g., a compact disc, a digital multipurpose disc, a Blu-ray® disc), a smart card, flash memory (e.g., a card, a stick, a key drive), a floppy® disk, a magnetic strip, etc. The storage 1003 may also be called an auxiliary storage device. The above-mentioned storage medium may be, for example, a database, server, or other suitable medium including memory 1002 and / or storage 1003.

[0077] The communication device 1004 is hardware (transceiver / receiver device) for communicating between computers via a wired and / or wireless network, and is also referred to as a network device, network controller, network card, communication module, etc.

[0078] The input device 1005 is an input device that accepts input from an external source (e.g., a keyboard, mouse, microphone, switch, button, sensor, etc.). The output device 1006 is an output device that outputs to an external source (e.g., a display, speaker, LED lamp, etc.). The input device 1005 and the output device 1006 may be configured as an integrated unit (e.g., a touch panel).

[0079] Furthermore, each device, such as the processor 1001 and the memory 1002, is connected by a bus 1007 for communicating information. The bus 1007 may consist of a single bus, or different buses may be used for communication between devices.

[0080] Furthermore, the floor plan generation device 1 may be configured to include hardware such as a microprocessor, a digital signal processor (DSP), an ASIC (Application Specific Integrated Circuit), a PLD (Programmable Logic Device), and an FPGA (Field Programmable Gate Array), and some or all of each functional block may be realized by such hardware. For example, the processor 1001 may be implemented using at least one of these hardware components.

[0081] The notification of information is not limited to the embodiments described herein and may be carried out by other means. For example, the notification of information may be carried out by physical layer signaling (e.g., DCI (Downlink Control Information), UCI (Uplink Control Information)), upper layer signaling (e.g., RRC (Radio Resource Control) signaling, MAC (Medium Access Control) signaling, broadcast information (MIB (Master Information Block), SIB (System Information Block))), other signals, or combinations thereof. RRC signaling may also be called RRC messages, and may be, for example, RRC Connection Setup messages, RRC Connection Reconfiguration messages, etc.

[0082] Each aspect / embodiment described in this disclosure may be applied to at least one of the following systems: LTE (Long Term Evolution), LTE-A (LTE-Advanced), SUPER 3G, IMT-Advanced, 4G (4th generation mobile communication system), 5G (5th generation mobile communication system), FRA (Future Radio Access), NR (new Radio), W-CDMA®, GSM®, CDMA2000, UMB (Ultra Mobile Broadband), IEEE 802.11 (Wi-Fi®), IEEE 802.16 (WiMAX®), IEEE 802.20, UWB (Ultra-WideBand), Bluetooth®, and other appropriate systems, as well as next-generation systems extended based thereon. Furthermore, multiple systems may be applied in combination (for example, a combination of at least one of LTE and LTE-A with 5G).

[0083] The processing procedures, sequences, flowcharts, etc., of each aspect / embodiment described in this disclosure may be reordered, provided they do not contradict each other. For example, the methods described in this disclosure present various step elements using exemplary order and are not limited to the specific order presented.

[0084] The specific operations described in this disclosure as being performed by a base station may, in some cases, be performed by its upper node. In a network consisting of one or more network nodes having a base station, it is clear that various operations performed for communication with a terminal can be performed by the base station and at least one other network node (for example, an MME or S-GW, but not limited to these). Although the above example illustrates the case where there is one other network node besides the base station, it may also be a combination of multiple other network nodes (for example, an MME and an S-GW).

[0085] Information can be output from a higher layer (or lower layer) to a lower layer (or higher layer). Input and output may also occur via multiple network nodes.

[0086] Input and output information may be stored in a specific location (e.g., memory) or managed in a management table. Input and output information may be overwritten, updated, or appended to. Output information may be deleted. Input information may be sent to other devices.

[0087] The determination may be made by a value represented by one bit (0 or 1), by a boolean value (true or false), or by a numerical comparison (for example, a comparison with a predetermined value).

[0088] Each aspect / embodiment described in this disclosure may be used individually, in combination, or switched between as needed during implementation. Furthermore, notification of specific information (e.g., notification that "X is") is not limited to explicit notification, but may also be implicit (e.g., by not providing such notification).

[0089] Although the present disclosure has been described in detail above, it will be clear to those skilled in the art that the present disclosure is not limited to the embodiments described herein. The present disclosure can be implemented in modified and altered forms without departing from the intent and scope of the present disclosure as defined by the claims. Therefore, the descriptions in the present disclosure are illustrative and not intended to be restrictive in any way.

[0090] Software should be broadly interpreted to mean instructions, instruction sets, code, code segments, program code, programs, subprograms, software modules, applications, software applications, software packages, routines, subroutines, objects, executable files, execution threads, procedures, functions, and so on, whether they are called software, firmware, middleware, microcode, hardware description languages, or by any other name.

[0091] Furthermore, software, instructions, etc., may be transmitted and received via a transmission medium. For example, if software is transmitted from a website, server, or other remote source using wired technologies such as coaxial cable, fiber optic cable, twisted pair, and digital subscriber lines (DSL) and / or wireless technologies such as infrared, radio, and microwave, these wired and / or wireless technologies are included in the definition of a transmission medium.

[0092] The information, signals, etc. described in this disclosure may be represented using any of the various different techniques. For example, the data, instructions, commands, information, signals, bits, symbols, chips, etc. that may be referred to throughout the above description may be represented by voltage, current, electromagnetic waves, magnetic fields or magnetic particles, optical fields or photons, or any combination thereof.

[0093] In addition, terms described in this disclosure and / or terms necessary for understanding this specification may be replaced with terms having the same or similar meaning.

[0094] The terms “system” and “network” as used in this disclosure are interchangeable.

[0095] Furthermore, the information, parameters, etc., described in this disclosure may be expressed as absolute values, relative values ​​from a given value, or by corresponding other information. For example, wireless resources may be indicated by an index.

[0096] The names used for the parameters described above are not restrictive in any way. Furthermore, the formulas and other expressions using these parameters may differ from those expressly disclosed in this disclosure. Various channels (e.g., PUCCH, PDCCH, etc.) and information elements can be identified by any suitable name, and therefore, the various names assigned to these various channels and information elements are not restrictive in any way.

[0097] As used in this disclosure, the terms “determining” and “determining” may encompass a wide variety of actions. “Determining” may include, for example, judging, calculating, computing, processing, deriving, investigating, looking up, searching, or inquiring (e.g., searching in a table, database, or other data structure), or ascertaining. “Determining” may also include receiving (e.g., receiving information), transmitting (e.g., sending information), inputting, outputting, or accessing (e.g., accessing data in memory). Furthermore, "judgment" and "decision" can include considering something as having been "judged" or "decided" after resolving, selecting, choosing, establishing, comparing, etc. In other words, "judgment" and "decision" can include considering something as having been "judged" or "decided" after some action. Also, "judgment (decision)" can be reinterpreted as "assuming," "expecting," or "considering."

[0098] As used in this disclosure, the phrase "based on" does not mean "based solely on" unless otherwise specified. In other words, the phrase "based on" means both "based solely on" and "based on at least."

[0099] Where the terms “first,” “second,” etc., are used in this disclosure, no reference to those elements shall generally limit the quantity or order of those elements. These terms may be used herein as a convenient way to distinguish between two or more elements. Accordingly, references to the first and second elements shall not imply that only two elements may be employed therein, or that the first element must precede the second element in any way.

[0100] In the configuration of each of the above devices, "means" may be replaced with "part," "circuit," "device," etc.

[0101] To the extent that “include,” “including,” and their variations are used herein or in the claims, these terms are intended to be inclusive, as is the term “comprising.” Furthermore, the term “or” as used herein or in the claims is not intended to be exclusive OR.

[0102] In this disclosure, if articles are added through translation, such as a, an, and the in English, this disclosure may include the fact that the noun following these articles is plural.

[0103] In this disclosure, the term "A and B are different" may mean "A and B are different from each other." The term may also mean "A and B are each different from C." Terms such as "separate" and "combine" may be interpreted similarly to "different."

[0104] The floor plan generation device 1 and information processing method disclosed herein may have the following configurations.

[0105] [1] An apparatus comprising: an acquisition unit that acquires an input dataset including an outline drawing showing the outline of a floor plan and information indicating the installation location of a wireless communication device in the outline drawing; and a generation unit that inputs the input dataset to an AI model that has been trained using a training dataset including a training floor plan with excellent radio wave propagation characteristics and information indicating the installation location of the wireless communication device in the training floor plan, and outputs a floor plan having the same shape as the outline of the floor plan and having excellent radio wave propagation characteristics. [2] The apparatus according to [1], wherein the AI ​​model is generated by training a generative adversarial network (GAN). [3] The apparatus according to [1] or [2], wherein the AI ​​model comprises: a generation model that has been trained to generate a floor plan having the same shape as the outline of the floor plan; and a discrimination model that has been trained to input a floor plan generated by the generation model and a training floor plan, and to determine whether the input floor plan is a floor plan generated by the generation unit. [4] The apparatus according to any one of [1] to [3], further comprising a dataset generation unit for generating the training dataset, wherein the dataset generation unit includes: an acquisition unit for acquiring a floor plan dataset including a floor plan and information indicating the installation location of the wireless communication device in the floor plan; a propagation estimation unit for generating a throughput map showing the throughput of wireless communication for each of a plurality of areas that demarcate the floor plan of the floor plan dataset based on the floor plan dataset; an evaluation unit for evaluating whether the floor plan of the floor plan dataset has good radio wave propagation characteristics based on the throughput map; and a storage unit for storing the floor plan as the training floor plan in a storage unit when the floor plan of the floor plan dataset is evaluated to have good radio wave propagation characteristics. [5] The apparatus according to [4], wherein the evaluation unit evaluates that the floor plan of the floor plan dataset has good radio wave propagation characteristics when the minimum value of the throughput of the plurality of areas is greater than a reference value. [6] The apparatus according to any one of [1] to [5], wherein the training dataset and the input dataset further include information indicating the location of furniture.[7] A method comprising the steps of: acquiring an input dataset including an outline drawing showing the external shape of a floor plan and information indicating the installation location of a wireless communication device in the outline drawing; and inputting the input dataset into an AI model trained using a training dataset including a training floor plan with excellent radio wave propagation characteristics and information indicating the installation location of the wireless communication device in the training floor plan, and outputting a floor plan having the same external shape as the external shape of the floor plan and having excellent radio wave propagation characteristics.

[0106] 1...Floor plan generation device, 10...Data set generation unit, 11...Acquisition unit, 12...Propagation estimation unit, 13...Evaluation unit, 14...Storage unit, 20...Floor plan generation unit, 21...Acquisition unit, 22...Generation unit, 30...Storage unit, 40...AI model, 41...Generation model, 42...Discrimination model, d11...Floor plan, d21...Training floor plan, d31, d41...Outline drawing, DS1...Floor plan dataset, DS2...First training dataset, DS3...Second training dataset, DS4...Input dataset, F...Furniture, WR...Wireless router (wireless communication device).

Claims

1. An apparatus comprising: an acquisition unit that acquires an input dataset including an outline drawing showing the external shape of a floor plan and information indicating the installation location of a wireless communication device in the outline drawing; and a generation unit that inputs the input dataset to an AI model trained using a training dataset including a training floor plan with excellent radio wave propagation characteristics and information indicating the installation location of the wireless communication device in the training floor plan, and outputs a floor plan having the same shape as the external shape of the floor plan and having excellent radio wave propagation characteristics.

2. The apparatus according to claim 1, wherein the AI ​​model is generated by machine learning a generative adversarial network (GAN).

3. The apparatus according to claim 1, wherein the AI ​​model includes: a generation model trained by machine learning to generate a floor plan having the same shape as the exterior of the floor plan; and a discrimination model trained by machine learning to input a floor plan generated by the generation model and a training floor plan, and to determine whether or not the input floor plan is a floor plan generated by the generation unit.

4. The apparatus according to claim 1, further comprising a dataset generation unit for generating the training dataset, the dataset generation unit comprising: an acquisition unit for acquiring a floor plan dataset including a floor plan and information indicating the installation location of the wireless communication device in the floor plan; a propagation estimation unit for generating a throughput map showing the throughput of wireless communication for each of a plurality of areas that demarcate the floor plan in the floor plan dataset based on the floor plan dataset; an evaluation unit for evaluating whether the floor plan in the floor plan dataset has good radio wave propagation characteristics based on the throughput map; and a storage unit for storing the floor plan in the storage unit as the training floor plan when the floor plan in the floor plan dataset is evaluated to have good radio wave propagation characteristics.

5. The apparatus according to claim 4, wherein the evaluation unit evaluates that the floor plan of the floor plan dataset has excellent radio wave propagation characteristics when the minimum value of the throughput of the plurality of regions is greater than a reference value.

6. The apparatus according to claim 1, wherein the training dataset and the input dataset further include information indicating the location of furniture.

7. A method comprising the steps of: acquiring an input dataset including an outline drawing showing the external shape of a floor plan and information indicating the installation location of a wireless communication device in the outline drawing; and inputting the input dataset into an AI model trained using a training dataset including a training floor plan with excellent radio wave propagation characteristics and information indicating the installation location of the wireless communication device in the training floor plan, and outputting a floor plan having the same external shape as the external shape of the floor plan and having excellent radio wave propagation characteristics.

Images