Network planning method and device, computing device cluster and storage medium

Abstract

The invention discloses a network planning method and device, a computing device cluster and a storage medium, and belongs to the technical field of communication. According to the method, the room areas for indicating the rooms in the building structure chart are identified by identifying the building structure chart, so that the deployment mode of the wireless local network equipment in the building is determined based on the plurality of room areas in the building structure chart, and the network planning of the building indicated by the building structure chart is completed.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a network planning method, apparatus, computing device cluster, and storage medium. Background Art

[0002] Wireless local area networks (WLANs) transmit data via wireless signals. Signal strength weakens over increasing transmission distance, and overlapping interference between adjacent wireless signals can degrade signal quality and even render the network unusable. To improve wireless network quality and meet customer network construction standards, a network planning method is urgently needed. This method involves planning and designing WLAN networks and the deployment of wireless LAN devices, such as access points (APs), within buildings to ensure network coverage and enhance the user experience. Summary of the Invention

[0003] This application provides a network planning method, apparatus, computing device cluster, and storage medium that can plan and design a WLAN network to be constructed in a building. The technical solution is as follows:

[0004] In a first aspect, a network planning method is provided, which is executed by a computing device. The method includes: first obtaining a first building structure diagram of a building, then identifying the first building structure diagram to obtain a second building structure diagram, and determining a deployment method of wireless local area network equipment in the building based on multiple room areas in the second building structure diagram, wherein the room area indicates a room in the building.

[0005] The method identifies a building structure diagram to identify room areas in the building structure diagram used to indicate rooms, so as to determine the deployment method of wireless local area network equipment in the building based on multiple room areas in the building structure diagram to complete the network planning of the building indicated by the building structure diagram.

[0006] In one possible implementation, the first category labels corresponding to the identified multiple room areas indicate the category of the room indicated by the corresponding room area. Based on this, another way for the method to determine the deployment method can also be: based on the multiple room areas in the second building structure diagram and the first category labels corresponding to the multiple room areas, determine the deployment method of the wireless LAN equipment in the building.

[0007] The above possible implementation method determines the deployment method based on the category label of each room area, which can meet the WLAN network requirements of different categories of rooms, making the determined deployment method more reasonable and the planned deployment location of WLAN equipment more reasonable, thereby further improving the coverage of the WLAN network in the building and the network speed when the terminal connects to the WLAN network.

[0008] In one possible implementation, the application scenario of the building corresponds to multiple second category labels, and the second category labels are used to indicate the category of the rooms in the building under the application scenario. Based on this, the first category labels corresponding to the multiple room areas are obtained as follows: for any room area among the multiple room areas, the text information in the room area is identified, and the room category information is extracted from the identified text information; the second category label with the highest semantic similarity to the room category information among the multiple second category labels is determined as the first category label corresponding to the room area.

[0009] In one possible implementation, the application scenario of the building corresponds to multiple second category labels, and the second category label is used to indicate the category of the room in the building under the application scenario. Based on this, the method also includes: for any room area among the multiple room areas, if the first category label corresponding to the room area is not the second category label under the application scenario, based on the first category label corresponding to the room area, determine the target category label from the multiple second category labels, and replace the first category label corresponding to the room area with the target category label, wherein the function of the room under the target category label is similar to the function of the room under the first category label corresponding to the room area.

[0010] The above possible implementation method can calibrate the category labels corresponding to multiple room areas to improve the accuracy of the category labels of the room areas. Subsequently, when determining the deployment method of WLAN equipment in the building based on the category labels of the room areas, the determined deployment method can be made more reasonable, thereby improving the coverage effect of the WLAN network and the network speed when the terminal connects to the WLAN network.

[0011] In one possible implementation, the second building structure diagram also includes an obstacle area, which indicates obstacles in the building, and the obstacles include walls or load-bearing columns. Based on this, another way of determining the deployment method of the method may be: based on multiple room areas and obstacle areas in the second building structure diagram, determine the deployment method of the wireless LAN device in the building.

[0012] The above possible implementation method can avoid deploying WLAN devices on obstacles such as walls and / or load-bearing columns through the deployment method determined by the obstacle area, and can also avoid WLAN devices being close to obstacles to reduce the impact of obstacles on network signals, thereby making the planned deployment location of WLAN devices more reasonable, thereby further improving the coverage effect of the WLAN network in the building and the network rate when the terminal connects to the WLAN network.

[0013] In one possible implementation, the second building structure diagram also includes an open area, which refers to an area in the second building structure diagram that is outside the multiple room areas and within the outline of the building; based on this, another way of determining the deployment method of the method can also be: based on the multiple room areas and the open area in the second building structure diagram, determine the deployment method of the wireless LAN equipment in the building.

[0014] The above possible implementation method can enable WLAN devices to be deployed in open spaces in buildings through the deployment method determined by the open area to meet the needs of open spaces for WLAN networks, making the determined deployment method more reasonable and the planned deployment location of WLAN devices more reasonable, thereby further improving the coverage effect of the WLAN network in the building and the network speed when the terminal connects to the WLAN network.

[0015] In one possible implementation, the above-mentioned open area also includes a seating area, and the seating area indication refers to the area where seats are deployed in the open area; based on this, another way of determining the deployment method of the method can also be: based on multiple room areas, open areas and at least one seating area in the second building structure diagram, determine the deployment method of the wireless LAN equipment in the building.

[0016] The above-mentioned possible implementation method can meet the different capacity requirements of WLAN equipment between seating space and non-seating space in the open space through the deployment method determined by the seating area in the open area, thereby making the deployment method determined by the deployment position of the WLAN equipment in the corresponding open space more reasonable, and the planned deployment position of the WLAN equipment is also more reasonable, thereby further improving the coverage effect of the WLAN network in the building and the network speed when the terminal connects to the WLAN network.

[0017] In one possible implementation, the room area is a first room area or a second room area, the first room area indicates an independent room in a building, and the second room area indicates a room with sub-rooms in the building; based on this, the above-mentioned identification of the first building structure diagram includes: performing room area identification on the first building structure diagram to obtain multiple first room areas; based on the overlap of contours between the multiple first room areas and adjacent room areas, performing area fusion on the multiple first room areas in the first building structure diagram to obtain at least one second room area, where the second room area includes two adjacent first room areas.

[0018] The above possible implementation method merges at least two first room areas into one room area through area fusion to identify the room area of ​​the combined room. When the deployment method of the WLAN device in the building is subsequently determined based on the fused room area, it is possible to avoid deploying WLAN devices in small sub-rooms in the combined room, avoid redundant deployment of WLAN devices, and make the determined deployment method more reasonable.

[0019] In a possible implementation, the above-mentioned regional fusion of multiple first room areas in the first building structure diagram based on the outline overlap between the multiple first room areas and the adjacent room areas to obtain at least one second room area includes: for any room area among the multiple first room areas, obtaining the outline overlap between the room area and the adjacent room area; if the outline overlap is greater than or equal to the overlap threshold, regional fusion of the room area and the adjacent area on the first building structure diagram to obtain a second room area, wherein the adjacent room area refers to a first room area adjacent to the room area, and the outline overlap indicates the overlap between the outline of the room area and the outline of the adjacent room area.

[0020] In one possible implementation, the method further includes: predicting the signal coverage in the building under the deployment mode to obtain a first prediction result; predicting the network rate of the terminal in the building under the deployment mode to obtain a second prediction result; and determining a network planning report for the building based on the first prediction result, the second prediction result and the deployment mode, the network planning report including the deployment location of the wireless LAN device in the building.

[0021] The above possible implementation method predicts the signal coverage in the building and the network rate of the terminal in the building under this deployment method, and generates a network planning report for the building through the prediction results and the deployment method, so that the prediction results can provide evidence for the deployment location of the WLAN device in the network planning report, so as to increase the credibility of the deployment location of the WALIN device, so that the WALIN device can be deployed in the building in the future according to the deployment location of the WALIN device in the network planning report.

[0022] In one possible implementation, the method also includes: if the first prediction result does not reach the first expected result and / or the second prediction result does not reach the second expected result, adjusting the deployment method based on the first prediction result and the second prediction result, the first expected result refers to the expected signal coverage in the building, and the second expected result refers to the expected network rate of the terminal in the building.

[0023] The above possible implementation method adjusts the deployment method so that the adjusted deployment method can ensure that the network deployment in the building and the network rate of the terminal in the building achieve the expected effect, making the adjusted deployment method more reasonable and the planned deployment location of the WLAN equipment more reasonable, thereby further improving the coverage effect of the WLAN network in the building and the network rate when the terminal connects to the WLAN network.

[0024] In a second aspect, a network planning device is provided for executing the method provided in the first aspect or any optional manner of the first aspect.

[0025] In a third aspect, a computing device cluster is provided, which includes at least one computing device, each computing device including a processor, and the processor of the at least one computing device is used to execute instructions stored in the memory of the at least one computing device, so that the computer device cluster executes to implement the method provided in the first aspect or any optional manner of the first aspect.

[0026] In a fourth aspect, a computer-readable storage medium is provided, in which computer program instructions are stored. When the computer program instructions are executed by a computing device cluster, the computing device cluster executes the method provided in the first aspect or any optional manner of the first aspect.

[0027] In a fifth aspect, a computer program product or a computer program is provided, which includes instructions (such as computer program instructions). When the instructions are executed by a computing device cluster, the computing device cluster executes the method provided in the first aspect or any optional method of the first aspect.

[0028] Based on the implementation methods provided in the above aspects, this application can also be further combined to provide more implementation methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 is a schematic diagram of a network planning system provided in an embodiment of the present application;

[0030] Figure 2 This is a flow chart of a network planning method provided by an embodiment of the present application;

[0031] Figure 3 This is a schematic diagram of a region identification process provided in an embodiment of the present application;

[0032] Figure 4 This is a schematic diagram of a room area recognition process provided by an embodiment of the present application;

[0033] Figure 5 This is a schematic diagram of a category label calibration process provided by an embodiment of the present application;

[0034] Figure 6 is a schematic diagram of another category label calibration process provided in an embodiment of the present application;

[0035] Figure 7 This is a partial building structure diagram provided in an embodiment of the present application;

[0036] Figure 8 is a schematic diagram of a method for determining contour coincidence provided in an embodiment of the present application;

[0037] Figure 9 is a schematic diagram of a region recognition result provided by an embodiment of the present application;

[0038] Figure 10 This is a flowchart of another network planning method provided by an embodiment of the present application;

[0039] Figure 11 is a schematic diagram of an open area identification process provided in an embodiment of the present application;

[0040] Figure 12 is a schematic diagram of another region recognition result provided by an embodiment of the present application;

[0041] Figure 13 This is a flowchart of another network planning method provided by an embodiment of the present application;

[0042] Figure 14 is a schematic diagram of an obstacle recognition result provided by an embodiment of the present application;

[0043] Figure 15 This is a flowchart of another network planning method provided by an embodiment of the present application;

[0044] Figure 16 This is a schematic diagram of a network planning process provided by an embodiment of the present application;

[0045] Figure 17 This is a schematic diagram of a deployment method of a WLAN device in a building provided by an embodiment of the present application;

[0046] Figure 18 This is a coverage signal simulation diagram under a deployment method provided in an embodiment of the present application;

[0047] Figure 19 This is a rate simulation diagram under a deployment method provided in an embodiment of the present application;

[0048] Figure 20 is a schematic diagram of an adjusted deployment method provided in an embodiment of the present application;

[0049] Figure 21 This is a coverage signal simulation diagram under an adjusted deployment mode provided in an embodiment of the present application;

[0050] Figure 22 This is a rate simulation diagram under an adjusted deployment mode provided in an embodiment of the present application;

[0051] Figure 23 This is a schematic diagram of the structure of a network planning device provided in an embodiment of the present application;

[0052] Figure 24 is a structural diagram of a computing device provided in an embodiment of the present application;

[0053] Figure 25 This is an architectural diagram of a computing device cluster provided in an embodiment of the present application;

[0054] Figure 26 This is an architecture diagram of another computer cluster provided in an embodiment of the present application. DETAILED DESCRIPTION

[0055] For a building where a WLAN network is to be deployed, the network planning process can be as follows: the customer manually marks the wall boundaries of the WLAN network coverage area on the building's floor plan, and uploads the marked floor plan and other information such as the scale of the floor plan to the cloud. The cloud then defines the overall coverage range of the building based on the wall boundaries marked on the floor plan, deploys AP points within the coverage range based on the scale of the floor plan, and returns to the customer the deployment method of the wireless LAN equipment (i.e., WLAN equipment) in the building based on the AP points.

[0056] Because the aforementioned network planning process uses the building floor plan as the scale, with the entire building as the global area, to determine the locations of WLAN devices such as APs, this ignores the building's internal environmental information. This can lead to inappropriate placement of WLAN devices within the building (for example, some WLAN devices may be located close to obstacles such as walls within the building), resulting in poor WLAN network coverage. Furthermore, the aforementioned network planning process relies on manual annotation, making automated network planning ineffective.

[0057] Based on this, the present application provides a network planning method that can obtain internal building environmental information by parsing a building structure diagram without relying on manual annotation, and then perform network planning for a WLAN network to be constructed in the building based on the internal building environmental information. The building structure diagram indicates the internal and / or external structure of the building, and the building structure diagram can be a floor plan, a three-dimensional structure diagram, or a structure diagram of another dimension of the building. The embodiments of the present application do not limit the dimensions of the building structure diagram.

[0058] Next, the system to which this method is applied is introduced as follows.

[0059] Figure 1 1 is a schematic diagram of a network planning system provided in an embodiment of the present application. The network planning system 100 is used to perform network planning for a WLAN network to be constructed in a building. The WLAN network includes multiple WLAN devices. Network planning for the WLAN network means planning the deployment method of the WLAN devices in the building. WLAN devices refer to wireless network devices required to construct a WLAN network, such as APs, Power over Ethernet (PoE) switches, and wireless access controllers. Here, the embodiment of the present application does not limit the device type of WLAN devices.

[0060] like Figure 1 As shown, the network planning system 100 includes a drawing parser 101, a device deployer 102 and a deployment verifier 103. For any building where a WLAN network is to be constructed, a user can upload a building structure diagram of the building to the network planning system 100. The uploading method is, for example, uploading through a terminal or directly uploading the building structure diagram to the drawing parser 101 without going through a terminal. The terminal can be a user device such as a mobile phone or a personal computer.

[0061] The drawing parser 101 is configured to parse the uploaded building structure drawing to obtain the building's internal environment information and send the internal environment information to the device deployer 102. The internal environment information includes information about each room in the building and information about each obstacle. The room information includes at least one of the room's location, area, outline, and category. The categories of rooms in a building vary depending on the building's application scenario. For example, rooms in buildings used in the education industry are categorized as dormitories, classrooms, conference rooms, and libraries, while rooms in buildings used in the hotel industry are categorized as single rooms, double rooms, and suites. The categories of rooms in a building can also be the same in different application scenarios. For example, categories such as bathrooms and stairwells are applicable to rooms in buildings in a variety of application scenarios. This embodiment of the present application does not limit the categories of rooms in the building. The internal environment information can be marked on the building structure drawing. The internal environment information on the building structure drawing can be displayed as text, graphics, specific colors, etc. This embodiment of the present application does not limit the form of display of the internal environment information on the building structure drawing. Of course, the internal environment information may not be marked on the building structure drawing. If the internal environment information is marked on the construction structure drawing, the drawing parser 101 sends the construction structure drawing marked with the internal environment information to the device deployer 102.

[0062] The device deployer 102 is configured to determine the deployment method of the WLAN devices in the building based on the internal environment information sent by the drawing parser 101, and to send the internal environment information and the deployment method to the deployment verifier 103. The deployment method may or may not be marked on the building structure drawing along with the internal environment information. If marked on the building structure drawing, the device deployer 102 sends the building structure drawing with the internal environment information and the deployment method marked to the deployment verifier 103.

[0063] The deployment verifier 103 is configured to receive the internal environment information and the deployment method sent by the device deployer 102, and based on the internal environment information, verify whether the network status of the WLAN network in the building meets the standards under the deployment method, obtain a verification result, and return the verification result to the device deployer 102. The verification result indicates whether the network status of the WLAN network in the building meets the standards under the deployment method. The network status of the WLAN network includes the signal coverage in the building and / or the network speed of the terminal in the building. The terminal can be a smartphone, tablet computer, music player, wearable smart device, laptop computer, desktop computer, etc. Here, the embodiment of the present application does not limit the type of terminal.

[0064] The device deployer 102 is further configured to receive a verification result returned by the deployment verifier 103. If the verification result indicates that the network status of the WLAN network in the building meets the standard under the deployment mode, the deployment mode is used as the final deployment mode of the WLAN devices in the building. If the verification result indicates that the network status of the WLAN network in the building does not meet the standard under the deployment mode, the deployment mode is adjusted, and the adjusted deployment mode is sent to the deployment verifier 103 for verification until a deployment mode that can make the network status of the WLAN network meet the standard is obtained, and the deployment mode is used as the final deployment mode of the WLAN devices in the building. After the final deployment mode of the WLAN devices in the building is obtained, the deployment mode can be returned to the user. For example, the return mode can be returned to the terminal that uploaded the building structure diagram, or the deployment mode can be printed by a technician and the printed deployment mode is returned to the user so that the user can deploy the WLAN devices in the building according to the deployment mode and build a WLAN network.

[0065] The deployment verifier 103 is an optional device in the network planning system 100. In another possible implementation, the network planning system 100 does not include the deployment verifier 103, and the device deployer 102 can provide the determined deployment mode to the user.

[0066] The network planning system 100 can be deployed in the cloud. The blueprint parser 101, device deployer 102, and deployment verifier 103 can all be implemented in software or hardware. For example, the implementation of blueprint parser 101 will be described below using blueprint parser 101 as an example. Similarly, the implementation of device deployer 102 and deployment verifier 103 can refer to the implementation of blueprint parser 101.

[0067] As an example of a software functional unit, the drawing parser 101 may include code running on a computing instance. The computing instance may include at least one of a physical host (computing device), a virtual machine, and a container. Furthermore, the computing instance may be one or more. For example, the drawing parser 101 may include code running on multiple hosts / virtual machines / containers. It should be noted that the multiple hosts / virtual machines / containers used to run the code may be distributed in the same region or in different regions. Furthermore, the multiple hosts / virtual machines / containers used to run the code may be distributed in the same availability zone (AZ) or in different AZs, each AZ including one data center or multiple geographically close data centers. Typically, a region may include multiple AZs.

[0068] Similarly, multiple hosts / virtual machines / containers running the code can be distributed within the same virtual private cloud (VPC) or across multiple VPCs. Typically, a VPC is set up within a region. Cross-region communication between two VPCs within the same region, or between VPCs in different regions, requires a communication gateway within each VPC to interconnect the VPCs.

[0069] As an example of a hardware functional unit, the drawing parser 101 may include at least one computing device, such as a server. Alternatively, the drawing parser 101 may be implemented using an application-specific integrated circuit (ASIC) or a programmable logic device (PLD). The PLD may be a complex programmable logical device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof.

[0070] The multiple computing devices included in the drawing parser 101 can be distributed in the same region or in different regions. The multiple computing devices included in the drawing parser 101 can be distributed in the same AZ or in different AZs. Similarly, the multiple computing devices included in the drawing parser 101 can be distributed in the same VPC or in multiple VPCs. The multiple computing devices can be any combination of computing devices such as servers, ASICs, PLDs, CPLDs, FPGAs, and GALs.

[0071] It should be noted that, in other embodiments, the drawing parser 101 can be used to execute any step in the following network planning method, the device deployer 102 can be used to execute any step in the following network planning method, and the deployment verifier 103 can be used to execute any step in the following network planning method. The steps that the drawing parser 101, the device deployer 102 and the deployment verifier 103 are responsible for implementing can be specified as needed. The full functions of the network planning system 100 are realized by respectively implementing different steps in the following network planning method through the drawing parser 101, the device deployer 102 and the deployment verifier 103.

[0072] The network planning system 100 may not be deployed in the cloud. In this case, the blueprint parser 101, device deployer 102, and deployment verifier 103 may all be implemented in software or hardware. Regardless of whether the network planning system is deployed in the cloud, if implemented in hardware, at least two of the blueprint parser 101, device deployer 102, and deployment verifier 103 may be integrated into the same computing device or different computing devices. The computing device may also be a network device with computing capabilities.

[0073] Next, based on the network planning system 100 introduced above, the network planning method provided by this application is introduced in detail.

[0074] Figure 2 This is a flowchart of a network planning method provided by an embodiment of the present application. The method is executed by a computing device, which may include at least one of the drawing parser 101, device deployer 102 and deployment verifier 103 in the above-mentioned network planning system 100. The method includes the following steps.

[0075] 201. A computing device obtains a first architectural structure drawing of a building.

[0076] The building is any building where a WLAN network is to be constructed, such as a certain building. The first building structure diagram is the building structure diagram of the building, which may be a building structure diagram of a certain floor in the building or a building structure diagram of multiple floors in the building.

[0077] A user can use a terminal to send a first building structure diagram to a computing device, so that the computing device obtains the first building structure diagram. For example, based on the first building structure diagram uploaded by the user, the terminal sends a network planning request to the computing device. The network planning request instructs the computing device to perform network planning for a WLAN network to be constructed in the building based on the first building structure diagram. The computing device receives the network planning request, obtains the first building structure diagram from the network planning request, and executes steps 202 and 203 below based on the first building structure diagram to perform network planning for the WLAN network to be constructed.

[0078] A user can also send a first building structure diagram to a computing device without using a terminal. For example, the user uploads the first building structure diagram to the computing device, and then performs a network planning operation on the computing device. The network planning operation instructs the computing device to plan a WLAN network to be constructed in the corresponding building based on the building structure diagram. In response to the network planning operation, the computing device obtains a network planning request and, based on the network planning request, executes steps 202 and 203 below to plan the WLAN network.

[0079] In a possible implementation, the computing device includes the drawing parser 101 in the network planning system 100 , and the drawing parser 101 performs this step 201 .

[0080] 202. The computing device identifies the first building structure diagram to obtain a second building structure diagram, where the second building structure diagram includes a plurality of room areas, where the room areas indicate rooms in the building.

[0081] Among them, any room area refers to the area where a room is located in the building structure diagram, and the room area is a closed area in the building structure diagram. The multiple room areas are room areas identified from the first building structure diagram. Optionally, the second building structure diagram also includes multiple room area identifiers, and the multiple room area identifiers correspond to the multiple room areas respectively. Any room area identifier is used to identify the corresponding room area. The room area identifier can be a mask of the corresponding room area, covering the corresponding room area. The room area identifier can also be a room outline of the room area. The room outline is the boundary line of the room area, which is used to indicate the outline of the room area.

[0082] The building includes multiple rooms, any of which may or may not have sub-rooms. For ease of description, rooms without sub-rooms are referred to as independent rooms, and rooms with sub-rooms are referred to as combined rooms. Examples of independent rooms include independent bathrooms, classrooms, and dormitories without bathrooms. Combined rooms are suites, which include at least two sub-rooms, such as dormitories with bathrooms, family suites in hotels, and double rooms with bathrooms. Accordingly, the area in the building structure diagram where independent rooms are located is referred to as a first room area, and the area in the building structure diagram where combined rooms are located is referred to as a second room area. Therefore, any room area in the multiple room areas is referred to as a first room area or a second room area. The first room area refers to an independent room in the building, and the second room area refers to a room with sub-rooms.

[0083] In one possible implementation, identifying the first building structure diagram includes identifying a room area in the first building structure diagram, such as Figure 3 As shown, the room area recognition process includes independent room area recognition, room category recognition, category calibration, and area fusion. Next, the recognition process is introduced as follows through the following steps A1 to A4.

[0084] Step A1: The computing device identifies room areas on the first building structure diagram to obtain a plurality of first room areas.

[0085] The room area recognition is used to identify the first room area in the first building structure diagram, that is, to identify the independent room area.

[0086] like Figure 4 In the illustrated room area recognition process, a computing device performs room area recognition on a first building structure drawing based on a room recognition model. For example, the first building structure drawing is input into the room recognition model, which performs room area recognition on the input first building structure drawing, labels each identified first room area with a room area identifier, and outputs a first building structure drawing including the room area identifier. The computing device obtains the output first building structure drawing as a second building structure drawing, and segments the first room area in the first building structure drawing by labeling the room area identifier of the first room area on the first building structure drawing. In another possible implementation, the computing device inputs the first building structure drawing and the scale of the first building structure drawing into the room recognition model. The room recognition model segments the first building structure drawing into multiple partial building structure drawings based on the input scale, performs room area recognition on each of the multiple partial building structure drawings, labels each identified first room area with a room area identifier, and outputs the first building structure drawing including the room area identifier.

[0087] The room recognition model refers to an artificial intelligence (AI) model that can recognize and mark the first room area in the building structure diagram. The room recognition model supports recognition of closed room areas of various shapes.

[0088] For example, the room recognition model can be an interactive large model, such as the segmentation anything model (SAM), the grounding DINO-segment anything model (Grounded-SAM), etc. The interactive large model segments the elements in the input first building structure diagram, determines the room area from the segmented elements, and outputs the building structure diagram including the room area. The element refers to the items presented on the first building structure diagram, such as a bathroom, and the infrastructure in the bathroom (such as a toilet, a cubicle, a washbasin, etc.). The way to determine the room area from the segmented elements is, for example, to determine the element whose area exceeds an area threshold as the room area, or to determine the N elements with the largest area among the segmented elements as the room area, or to determine the element whose object image element information accounts for a proportion of the outer contour that exceeds a preset proportion as room information, or to determine the element including sub-elements as the room area, etc., where N is an integer greater than 0.

[0089] For another example, the room recognition model can also be a semantic segmentation model, such as a mask region-based convolutional neural network (Mask R-CNN), a general segmentation model (segment everything in context, SegGPT), etc. If the room recognition model is a semantic segmentation model, the room recognition model can be trained based on multiple first sample building structure drawings, wherein the first sample building structure drawings are building structure drawings used to train the room recognition model, the first sample building structure drawings include multiple room area labels, and the room area label is used to indicate a room area (such as a first room area) in the first sample building structure drawings. The training device uses the room area indicated by the room area label as the recognition target, and trains the room recognition model based on the first sample building structure drawings and the room area label in the first sample building structure drawings, so that the trained room recognition model can learn the ability to recognize and label room areas in the building structure drawings, so that the computing device can use the trained room recognition model to recognize and label each first room area on the first building structure drawing. The training device and the computing device can be the same computing device or different computing devices.

[0090] Step A2: The computing device performs room category recognition on the multiple first room areas to obtain first category labels corresponding to the multiple first room areas.

[0091] Among them, the first category label is the category label corresponding to the first room area, and the first category label indicates the category of the room indicated by the corresponding room area (i.e., the room category). For example, the first category label includes the room category, or the first category label also includes the confidence of the room category, which is used to indicate the degree of trustworthiness of the room category. For the introduction to the room category, please refer to the above and will not be repeated here.

[0092] For example, in the process of identifying room areas in the first building structure diagram, the computing device identifies room categories for the multiple first room areas. Figure 4As shown, the above-mentioned room recognition model also has the function of room category recognition. Taking the room recognition model as a semantic recognition model as an example, the room recognition model is also trained based on multiple target category labels corresponding to the first sample building structure diagram. The target category label indicates the room category of a first room area in the first sample building structure diagram. The training device uses the target category label as the category label expected to be output by the room recognition model, and trains the room recognition model based on multiple first sample building structure diagrams, multiple target category labels and room area labels in the first building structure diagram, so that the trained room recognition model can learn the function of recognizing the room category of the first room area, so that subsequent computing devices can use the room recognition model to obtain each first room area and the category label of each room area in various building structure diagrams.

[0093] Exemplarily, a computing device inputs a first building structure diagram into a room recognition model. The room recognition model identifies room areas in the first building structure diagram, identifies room categories for each identified first room area, labels each first room area in the first building structure diagram with a room area identifier, and outputs a category label corresponding to each first room area and the first building structure diagram including the room area identifier. The room recognition model can determine the category label corresponding to the room area by identifying infrastructure within the room area. For example, if a bed is identified in the first room area, "dormitory" is used as the category of the first room area. If multiple rows of seats are identified in the first room area, "conference room" is used as the category of the first room area. Given that the shapes of some infrastructure in the room area may be similar to the shapes of the room area, for example, the bed and the room may both be rectangles in the building structure diagram, the computing device can also input the first building structure diagram and the scale of the first building structure diagram into the first room recognition model, so that the room recognition model can determine whether a certain area in the building structure diagram is the area where the infrastructure is located or the room area through the scale. For example, if the actual area of ​​the object indicated by a rectangular area after conversion through the scale is larger than the area of ​​a normal bed and is similar to the area of ​​the room, then the area can be identified as the room area.

[0094] Alternatively, the room area in the first building structure diagram includes text information, which may be the name of the room corresponding to the room area, such as "double room" or "stairwell." After obtaining a second building structure diagram including multiple first room areas, the computing device identifies the text information in any of the multiple first room areas and extracts room category information from the identified text information. The room category information is used to indicate the category of the room area. For example, the computing device identifies the text information in the room area based on an optical character recognition (OCR) model and extracts the room category information from the identified text information. After obtaining the room category information, multiple second category labels corresponding to the building's application scenario are obtained. The second category labels are used to indicate the category of the room in the building under the application scenario. For an introduction to the building's application scenario and room category, please refer to the above and will not be repeated here. For example, the computing device stores a room category library in the computing device. The room category library includes category labels corresponding to multiple application scenarios, each application scenario corresponding to at least one category label. The application scenario of the building corresponding to the first building structure diagram is used as the target application scenario, and the category label corresponding to the target application scenario in the room category library is obtained as the second category label. After obtaining multiple second category tags, the computing device obtains the semantic similarities between the multiple second category tags and the room category information, and determines the second category tag with the highest semantic similarity to the room category information among the multiple second category tags as the first category tag corresponding to the room area, wherein the semantic similarity between the second category tag and the room category information indicates the semantic similarity between the second category tag and the room category information.

[0095] Step A3: The computing device calibrates the first category tags corresponding to the multiple first room areas.

[0096] The first category of tags is calibrated using either calibration method 1 or calibration method 2 described below.

[0097] Calibration method 1: Calibrate the first category tags corresponding to multiple first room areas using the second category tags corresponding to the application scenario of the building.

[0098] For any room area among multiple first room areas, if the first category label corresponding to the room area is not the second category label under the application scenario, the computing device determines the target category label from multiple second category labels based on the first category label corresponding to the room area, and replaces the first category label corresponding to the room area with the target category label.

[0099] Among them, the function of the room under the target category label is similar to the function of the room under the category label corresponding to the room area, and the target category label is the first category label of the room area after correction. Exemplarily, the room category library also includes the correspondence between category labels of rooms with similar functions (or the same functions) in multiple application scenarios. For example, wards in the medical industry, dormitories in the education industry, and single rooms in the hotel industry are all rooms used for accommodation, then the category label "ward" in the medical industry corresponds to the category label "dormitory" in the education industry and the category label "single room" in the hotel industry. For any room area among the multiple first room areas, if the first category label corresponding to the room area is not the second category label in the application scenario, the computing device determines the second category label corresponding to the first category label in the correspondence as the target category label.

[0100] by Figure 5 Taking the calibration process shown as an example, assuming that the application scenario of the building corresponding to the first building structure diagram is the education industry, each first category label is matched with each category label in the education industry. Assuming that a first category label is "ward", the category label "ward" and the category labels in the education industry are different (that is, they do not match), and the category label "dormitory" in the education industry and the room indicated by the category label "ward" have similar functions. Therefore, the category label "dormitory" is used as the target category label, and the category label of the first room area is corrected from "ward" to the category label "dormitory".

[0101] Calibration method 2: calibrating first category labels corresponding to a plurality of first room areas using text information in the first building structure drawing.

[0102] Assume that the first building structure diagram is marked with text information of each room area, and the text information may be the name of the room corresponding to the room area. If the first category labels corresponding to the multiple first room areas are not identified through the text information in the first building structure diagram, for example, they are identified through a room recognition model, the computing device can calibrate the first category labels corresponding to the multiple first room areas through calibration method 2.

[0103] The calibration method is, for example, for any room area among multiple first room areas, the computing device identifies the text information in the room area, extracts the room category information from the identified text information, and the computing device determines the second category label with the highest semantic similarity to the room category information among multiple second category labels as the target category label, and replaces the first category label corresponding to the room area with the target category label, wherein the process has been introduced above and will not be repeated here.

[0104] For example, assuming that the first category labels corresponding to multiple first room areas are identified by the room recognition model, the room recognition model relies on the infrastructure in the first room area when identifying the room category of the first room area. If there is no infrastructure in a room area, the category label of the identified first room area is "other". At this time, the category label can be calibrated through calibration method 2.

[0105] by Figure 6 Taking the calibration process shown above as an example, let's assume the building's application scenario is the education industry. Category labels in the education industry include "office / conference room" and "classroom." "Office / conference room" has a high semantic similarity with multiple category labels, such as "office" and "classroom / office." "Classroom" also has a high semantic similarity with multiple category labels, such as "classroom" and "study room." Assume a room area in the building includes the text "classroom / office." The category label (i.e., the original category) identified by room type is "other," and the confidence level for the original category is 0.859. The computing device uses an OCR model to identify the text in the room area, extracts the room category information "classroom / office" from the identified text information, and performs semantic similarity matching between "classroom / office" and various category labels in the education industry. For example, the semantic similarity between "classroom / office" and various category labels in the education industry is calculated. Assuming that the semantic similarity (0.778) between the category labels "office / conference room" in the education industry and "classroom / office" is the highest, the category label of the room area is changed to the matching category label "office / conference room", and the relative semantic similarity of 0.778 is used as the confidence level of the category label. In other words, if the original category label is "other", it is replaced by the matching category label by default.

[0106] By calibrating the category labels corresponding to multiple first room areas, the accuracy of the category labels of the first room areas can be improved. Subsequently, when determining the deployment method of WLAN devices in the building based on the category labels of the first room areas, the determined deployment method can be made more reasonable, thereby improving the coverage effect of the WLAN network and the network speed when the terminal connects to the WLAN network.

[0107] Step A4: The computing device performs region fusion on the multiple first room areas in the first building structure diagram based on the overlap of the outlines of the multiple first room areas and the adjacent room areas to obtain at least one second room area, where the second room area includes at least two adjacent first room areas.

[0108] The adjacent room area of ​​any first room area refers to the first room area adjacent to the first room area, and any first room area has at least one adjacent room area. The second room area is obtained by merging at least two adjacent first room areas.

[0109] For any room area among the multiple first room areas, the computing device obtains the outline overlap between the room area and the adjacent room area. If the outline overlap is greater than or equal to the overlap threshold, the room area and the adjacent area are regionally fused on the first building structure diagram to obtain a second room area, so that the second room area includes the room area and the adjacent room area.

[0110] The outline overlap indicates the overlap between the outline of the room area and the outline of the adjacent room area (i.e., the outline overlap). The outline overlap can be represented by the Intersection over Union (IoU) between the room area and the adjacent room area. Figure 7 As an example, let's assume that Figure 7 The area where the double room is located is identified as two first room areas, one room area is the bathroom area, and the other room area is the double room area. Figure 8 In the local building structure diagram, the computing device uses the bounding box 1 to frame the identified bathroom area, and uses the bounding box 2 to frame the identified double room area. The bounding box 1 and the bounding box 2 are used as the outlines of the bathroom area and the double room area respectively, and the IoU between the bounding box 1 and the bounding box 2 is calculated. Since the bounding box 1 is entirely within the bounding box 2, the calculated IoU is relatively large and exceeds the overlap threshold. Figure 9 As shown, the same mask is added to the identified bathroom area and double room area to merge the bathroom area and the double room area into one room area (ie, the second room area).

[0111] For any room area among the multiple first room areas, if the outline overlap between the room area and the adjacent room area is less than the overlap threshold, regional fusion of the room area and the adjacent area is not performed. Therefore, the first room area in the first building structure diagram may or may not be regionally fused. The first building structure diagram after regional fusion is used as the second building structure diagram. In this case, the second building structure diagram includes at least one first room area and / or at least one second room area.

[0112] A combined room includes multiple sub-rooms, among which there may be some relatively small sub-rooms (such as bathrooms). Normally, WLAN equipment is not deployed in rooms with relatively small areas. However, when deploying WLAN equipment such as APs in a building as a global area, it is very likely that WLAN equipment will be deployed in relatively small sub-rooms, resulting in an unreasonable deployment method. However, this application uses the overlap of contours to merge the first room area with a smaller area and the first room area with a relatively larger area into one room area to identify the room area of ​​the combined room. Subsequently, when determining the deployment method of WLAN equipment in the building based on the merged room area, it is possible to avoid deploying WLAN equipment in small sub-rooms in the combined room, avoid redundant WLAN equipment deployment, and make the determined deployment method more reasonable.

[0113] In the case where multiple first room areas have corresponding first category labels, after area fusion, for any second room area, the first category label corresponding to the first room area with the largest area in the second room area is used as the category label of the second room area, and the first category label is still used as the category label of the second room area. Figure 8 For example, assuming that the category labels of the room areas framed by external frame 1 and external frame 2 are "bathroom" and "double room" respectively, after the room areas framed by external frame 1 and external frame 2 are merged into a second room area, "double room" is used as the category label of the second room area.

[0114] The above description is based on the example of first calibrating the category label of the first room area and then performing regional fusion. In another possible implementation, Figure 4 As shown, the computing device may first perform region fusion on the identified first room region, and then calibrate the category labels of each room region (including the first room region and / or the second room region). The calibration method may be any of the calibration methods 1 and 2 described above. Through region identification and region fusion, each room region can be marked on the second building structure diagram, indicating the boundaries of each room region; through room category identification and category label calibration, the category of the room indicated by each room region on the second building structure diagram can be determined.

[0115] The above is explained by taking the example of first identifying the room category of the first room area and then performing regional fusion. In another possible implementation method, after identifying the first room area, regional fusion can also be performed first, and then room identification is performed on each room area in the first building structure diagram after regional fusion. After room identification, the category label corresponding to each identified room area is used as the first category label, and the first category label can be calibrated or not.

[0116] In one possible implementation, different first category labels correspond to different room area identifiers. After determining each room area (first room area and / or second room area) in the first building structure diagram and the first category label corresponding to each room area, the room area identifier corresponding to the corresponding category label is added to each room area in the first building structure diagram, so that the category label of the corresponding room area can also be reflected through the room area identifier. Taking the room area identifier as an example, different category labels correspond to different masks, such as Figure 9 The area recognition results shown in the figure indicate that the areas where the masks are located are room areas, and the masks of room areas with different category labels are different.

[0117] Steps A2 through A4 are all optional. For example, after executing step A1 to obtain a first building structure diagram including multiple first room areas, the computing device does not execute steps A3 through A4, but instead executes step 203 using the first room areas on the first building structure diagram as the ultimately identified room areas and the first building structure diagram including the multiple first room areas as the second building structure diagram. For another example, after executing step A1, steps A2 and A3 are not executed, and the computing device skips to step A4 to perform region fusion on the identified first room areas, using the fused first building structure diagram as the second building structure diagram, and executing step 203. For another example, after executing steps A1 and A2, step A3 is not executed, and step A4 is executed to obtain the second building structure diagram.

[0118] In a possible implementation, when the computing device includes a drawing parser 101, step 202 may be performed by the drawing parser 101. Optionally, the drawing parser 101 includes a room identification unit, which is used to complete the room identification process for the first building structure drawing.

[0119] 203. The computing device determines a deployment method of WLAN devices in the building based on multiple room areas in the second building structure diagram.

[0120] The multiple room areas include at least one first room area and / or at least one second room area. The deployment method includes the deployment locations of multiple WLAN devices in the building. The multiple WLAN devices are used to establish a WLAN network in the building, and the deployment locations are located in the rooms indicated by the room areas. Optionally, the deployment method also includes the model and installation method of each WLAN device. The WLAN device can be an AP or other type of WLAN device that can release a network signal.

[0121] Each room type can correspond to a room deployment rule, which refers to the rules for deploying WLAN devices in the room. Different room types can correspond to different room deployment rules or the same room deployment rules.

[0122] Table 1

[0123]

[0124] As shown in Table 1, the room deployment rules include coverage identifiers, the device model, capacity, spacing, and installation method of the WLAN device. The coverage identifier is used to indicate whether the room in the corresponding room category needs to be covered by the network signal of the WLAN network. For example, a coverage identifier of "yes" indicates that the room in the corresponding room category needs to be covered by the network signal, and a coverage identifier of "no" indicates that the room in the corresponding room category does not need to be covered by the network signal. Taking Table 1 as an example, the coverage identifier corresponding to the room category "office" is "yes", and the coverage identifier corresponding to the room category "bathroom" is "no". The determined deployment method needs to ensure that: all rooms in the building with the room category of "office" are covered by the network signal of the WLAN device, and all rooms in the building with the room category of "bathroom" may not be covered by the network signal of the WLAN device. The network signal covered by any room includes the network signal of the WLAN device deployed in the room and / or the network signal of the WLAN device deployed in the area outside the room in the building. The network signal covered by the room depends on whether a WLAN device is deployed in the room and whether the network signal of the WLAN device deployed outside the room can be transmitted to the room. The coverage identifier is optional, and the room deployment rule does not have to include the coverage identifier. By default, the rooms under the corresponding room category need to cover the network signal. The device model refers to the model of WLAN devices that can be deployed in the rooms under the corresponding room type, such as panel type, mounted type, etc. Taking Table 1 as an example, the device model corresponding to the room category "Office" is model 1. When deploying WLAN devices in each room with the room category of "Office", WLAN devices of model 1 need to be deployed in each room. Capacity refers to the number of people that can be used by each WLAN device. The installation method of the WLAN device depends on the device model of the WLAN device. For example, some device models are installed ceiling-mounted (such as installed on the ceiling of the room), some device models are installed wall-mounted (such as installed on the wall), such as panel-type APs, and some device models are installed away from the wall, such as mounted APs.

[0125] In the absence of identifying the room type of the room area, for each of the multiple room areas, the deployment position of the WLAN device in the corresponding room is determined according to the same room deployment rule. In this case, the room deployment rule may or may not include a coverage identifier. If the coverage identifier is not included, it is assumed that the rooms in the corresponding room category need to be covered by the network signal, and the step of determining the deployment position of the WLAN device in the corresponding room according to the room deployment rule is executed. If the coverage identifier is included, the coverage identifier indicates that the rooms in the corresponding room category need to be covered by the network signal of the WLAN network, and based on the indication of the coverage identifier, the step of determining the deployment position of the WLAN device in the corresponding room according to the room deployment rule is executed.

[0126] Among them, the process of determining the deployment position is, for example: the computing device determines that the room area corresponds to the installation area in the room based on the installation method of the WLAN device in the room deployment rule. The installation area refers to the area where the WLAN device can be installed. For example, if the installation method is ceiling-mounted, the installation area is the ceiling of the room; if the installation method is wall-mounted, the installation area is the wall of the room; if the installation method is far from the wall, the installation area is the area far from the wall in the room; after determining the installation area, based on the scale of the first building structure drawing and the area of ​​the installation area in the second building structure drawing, the actual area of ​​the installation area in the room is determined; based on the actual area of ​​the installation area and the spacing in the room deployment rule, the deployment position of each WLAN device in the installation area is determined, the spacing between the deployment positions of each WLAN device meets the spacing requirements in the room deployment rule, and the deployment positions of each WLAN device are all located in the installation area.

[0127] In a similar manner, the computing device may determine the deployment locations of the WLAN devices in the rooms corresponding to the respective room areas, and determine the deployment method of the WLAN devices in the building based on the deployment locations of the WLAN devices in the rooms corresponding to the respective room areas.

[0128] When the room type of the room area has been identified, multiple room areas in the second building structure correspond to first category labels, and the first category label corresponding to any room area is an uncalibrated first category label or a calibrated first category label (such as a target category label). The computing device determines the deployment method of the wireless LAN device in the building based on the multiple room areas in the second building structure diagram and the first category labels corresponding to the multiple room areas.

[0129] Exemplarily, for any room area among the multiple room areas, the computing device determines the deployment position of the WLAN device in the corresponding room according to the room deployment rule corresponding to the room category indicated by the first category label corresponding to the room area. The determination method is the same as the above-mentioned method of determining the deployment position of the WLAN device in the corresponding room according to the same room deployment rule. The difference is that if the coverage identifier in the room deployment rule corresponding to the room category indicates that the network signal does not need to cover the rooms under the room category, the computing device determines the deployment method of the WLAN device in the corresponding room as the first deployment method. The first deployment method indicates that the WLAN device does not need to be deployed in the room. If the coverage indication in the room deployment rule corresponding to the room category indicates that the network signal needs to cover the rooms under the room category, the network signal coverage requirements of rooms of different categories are met. According to a similar determination method, the computing device can determine the deployment position of the WLAN device in the rooms corresponding to each room area, and determine the deployment method of the WLAN device in the building based on the deployment position of the WLAN device in the rooms corresponding to each room area.

[0130] Based on the category labels of each room area, the deployment method of wireless LAN equipment in the building is determined, so that the determined deployment method can meet the WLAN network requirements of different types of rooms, making the determined deployment method more reasonable and the planned deployment location of WLAN equipment more reasonable, thereby further improving the coverage of the WLAN network in the building and the network speed when the terminal connects to the WLAN network.

[0131] In one possible implementation, the computing device includes Figure 1 In the case of the device deployer 102 in , this step 203 can be performed by the device deployer 102.

[0132] Figure 2 In the illustrated embodiment, a building diagram is identified to identify room areas used to indicate rooms in the diagram. Based on the multiple room areas in the diagram, a deployment pattern for wireless local area network (WLAN) devices in the building is determined, thereby completing network planning for the building indicated by the diagram. This deployment pattern enables WLAN devices to be deployed within the rooms of the building, making their deployment locations more reasonable. This improves WLAN network coverage in the building and network speeds for terminals connecting to the WLAN network.

[0133] In addition to enclosed spaces such as rooms, buildings also contain open spaces, such as corridors, balconies, open office areas, and open rest areas. Accordingly, building diagrams also contain areas where open spaces are located. For ease of description, the areas where open spaces are located in building diagrams are referred to as open areas. In other words, open areas indicate open spaces in a building. In another possible implementation, the computing device can also determine the deployment method of WLAN network devices in the building based on the open areas in the building diagram.

[0134] For example, Figure 10 The flowchart shown is another network planning method provided by an embodiment of the present application. The method is executed by a computing device, which may include the above-mentioned network planning system 100. The method includes the following steps.

[0135] 1001. A computing device obtains a first architectural structure drawing of a building.

[0136] Among them, this step 1001 is similar to the above-mentioned step 201. Here, the embodiment of the present application will not repeat this step 1001.

[0137] 1002. The computing device identifies the first building structure drawing to obtain a second building structure drawing, where the second building structure drawing includes a plurality of room areas and an open area. The open area refers to an area outside the plurality of room areas and within a building outline in the second building structure drawing.

[0138] The building outline is the boundary of a building area identified from the first building structure drawing. The building area is the area where the building is located on the building structure drawing. The multiple room areas include at least one first room area and / or at least one second room area. The multiple room areas and the open area are all areas identified on the first building structure drawing. Accordingly, identifying the first building structure drawing includes the following steps B1 and B2.

[0139] Step B1: The computing device identifies the room areas in the first building structure diagram to obtain a second building structure diagram, where the second building structure diagram includes a plurality of room areas.

[0140] Here, step B1 is the same as step 202 and will not be described again here.

[0141] In another possible implementation, in the process of identifying the room area in the first building structure diagram, the computing device may further identify the room category of the room area to obtain a first category label, or may further calibrate the identified first category label. This process is described in the above step 202 and will not be repeated here.

[0142] by Figure 3The embodiment of the present application shown provides a process for regional recognition of a building structure diagram as an example. Assume that the computing device includes Figure 1 The drawing parser 101 in the drawing parser 101 is used to execute step 1002. The drawing parser 101 includes a room recognition unit. The room recognition unit recognizes the room area of ​​the first building structure drawing and performs area fusion and category calibration on the recognized first room area to complete this step B1. The workflow of the room recognition unit is as follows: Figure 4 As shown, no further details are given here.

[0143] Step B2: The computing device identifies the open area in the second building structure diagram to obtain the identified open area.

[0144] like Figure 3 As shown, the open area recognition process includes boundary detection and open area recognition. Next, the recognition process is introduced as follows through the following steps B21 to B22.

[0145] Step B21: The computing device performs boundary detection on the first building structure drawing to obtain a third building structure drawing, where the third building structure drawing includes a building outline.

[0146] The computing device may perform boundary detection on the first building structure diagram based on the boundary detection model. Figure 11 In the open area identification process shown, a computing device inputs a first building structure drawing into a boundary detection model. The boundary detection model performs boundary detection on the input first building structure drawing, annotates the identified boundaries on the first building structure drawing with building outlines, and outputs the first building structure drawing including the building outlines. The computing device uses the output first building structure drawing as the third building structure drawing. In another possible implementation, the computing device inputs the first building structure drawing and its scale into the boundary detection model. The boundary detection model, based on the input scale, segments the first building structure drawing into multiple partial building structure drawings, performs boundary detection on each of the multiple partial building structure drawings, annotates the identified boundaries on each of the multiple partial building structure drawings with building outlines, and outputs the first building structure drawing including the building outlines.

[0147] Among them, the boundary detection model refers to an AI model that can identify and mark the boundaries of building areas in building structure drawings, and supports the recognition of building areas with known shapes. Figure 11As shown, the boundary detection model includes a wall recognition sub-model and a contour extraction sub-model. The wall recognition sub-model is an AI model that can recognize walls in building structure drawings and supports the recognition of walls of various shapes. Exemplarily, the wall recognition sub-model can be a semantic segmentation model, such as an interlaced sparse self-attention for semantic segmentation network (ISANet) for semantic segmentation. The wall recognition sub-model can also be an AI model other than a semantic segmentation model, which can recognize walls in building structure drawings. The contour extraction sub-model is an AI model that can extract the contours of building areas in building structure drawings, such as a contour extraction algorithm model.

[0148] like Figure 11 As shown, the computing device inputs the first building structure drawing into the boundary detection model. The wall recognition sub-model in the boundary detection model performs wall recognition on the first building structure drawing, labels the recognized wall area with wall identification lines, and outputs the first building structure drawing including the wall identification lines to the contour extraction sub-model. The wall area refers to the area where the wall identified in the building structure drawing is located, the wall area indicates the wall in the building, and the wall identification lines represent the recognized wall area. The contour extraction sub-model performs threshold segmentation and edge detection on the mask of the wall area, labels the building contour lines at the boundary of the building area detected in the first building structure drawing, and outputs a third building structure drawing.

[0149] Step B22: The computing device identifies open areas on the second building structure drawing based on the building outlines on the third building structure drawing.

[0150] The computing device determines the area within the building outline and outside the plurality of room areas in the second building structure diagram as an open area, and outputs the second building structure diagram including the open area and the room area. Figure 11 Taking the open area recognition shown in the figure as an example, for the second building structure diagram including room area boundaries (such as room area identifiers) and the third building structure diagram output by the boundary detection model, the building outline in the third building structure diagram is fused into the second building structure diagram, so that the second building structure diagram includes room area boundaries and building outlines, the building outline in the second building structure diagram is used as the outer boundary, and the boundaries of each room area are used as the inner boundary. The area between the inner boundary and the outer boundary in the second building structure diagram is determined as the open area, and the open area is marked with the open area identifier in the second building structure diagram to achieve Figure 11 The open area determination process shown in , enables the second building structure to include room area identifiers and open area identifiers.

[0151] The open area identifier is used to identify the open area. The open area identifier can be a mask of the open area, covering the open area. The open area identifier can also be a contour line (boundary line) of the open area. The open area identifier is different from the room area identifier, so that the room area and the open area can be distinguished from each other on the second building structure diagram. For example, if both the room area identifier and the open area identifier are masks, Figure 7 The room area recognition and open area recognition of the partial building structure diagram shown in the figure can be obtained. Figure 12 The local building structure diagram after region recognition is shown. The mask of the open area is different from that of the room area. The masks of the room areas corresponding to different category labels are different. Figure 12 and Figure 9 The difference is that Figure 12 The mask of the open area is added. Figure 12 Other content in Figure 9 Similar. Among them, Figure 9 and Figure 12 The difference between the masks is reflected by the different patterns on the masks. In another possible implementation, the difference between the masks can also be reflected in the different colors of the masks.

[0152] In another possible implementation, Figure 11 As shown, in the open area recognition process, after determining the open area, the computing device can also perform seat area recognition on the open area to obtain the seat area in the open area. The seat area refers to the area where seats are deployed in the open area, and the seat area is obtained through seat recognition. Exemplarily, the computing device inputs the second building structure drawing including the open area identifier into the seat recognition model, and the seat recognition model performs seat area recognition on the open area in the input second building structure drawing based on the open area identifier in the second building structure drawing, marks the identified seat area as the seat area identifier, and outputs the second building structure drawing including the open area identifier and the seat area identifier. In the case where the open area identifier is a mask, the seat recognition model can first remove the open area identifier, and then perform seat area recognition on the open area in the second building structure drawing. The seat area identifier is used to identify the identified seat area. The identification method can refer to the open area identifier and will not be repeated here.

[0153] In another possible implementation, after determining the open area, an outline of the open area is added to the second architectural drawing to indicate the boundary of the open area. The second architectural drawing including the outline is then input into a seating recognition model. The seating recognition model then identifies seating areas for the open area in the second architectural drawing based on the outline in the second architectural drawing. The identified seating areas are labeled as seating area identifiers, and the second architectural drawing including the outline and seating area identifiers is output. The outline may or may not be an open area identifier. If it is not an open area identifier, after obtaining the second architectural drawing including the outline and seating area identifiers, the computing device adds the open area identifier to the area within the outline on the second architectural drawing.

[0154] The above-mentioned seat recognition model is an AI model that can identify the seating area in the building structure diagram and supports the recognition of seating areas of various shapes. The seat recognition model can be a target detection model, such as a fast region-based convolutional network (Fast R-CNN). Of course, the seat recognition model can also be other AI models other than the target detection model, as long as it can detect the seating area. The seat recognition model can be trained based on multiple second sample building structure diagrams, wherein the second sample building structure diagram is the building structure diagram used to train the seat recognition model, and the second sample building structure diagram includes multiple seat area labels, which are used as area labels to indicate a seat area in the second sample building structure diagram. The training device uses the area indicated by the area label as the recognition target, and trains the seat recognition model based on the second sample building structure diagram and the seat area label in the second sample building structure diagram, so that the trained room recognition model can learn the ability to recognize and mark the area in the building structure diagram, so that the computing device can use the trained room recognition model to recognize and mark each seating area on the second building structure diagram. Through the open area recognition process, the boundaries of the open area in the first building structure diagram and whether there is a seating area in the open area can be identified.

[0155] like Figure 3 As shown, the computing device performs room area recognition and open area recognition on the first building room structure diagram to obtain area recognition results, which include the location and category label of each room area on the first building structure diagram, the location of each open area, and the location of the seating area in each open area. The area recognition results can be reflected by the second building structure diagram, for example, as shown in FIG. Figure 12 In the identified building structure diagram shown, different masks can be used to indicate open areas and room areas respectively.

[0156] The above description uses the example of performing boundary detection on the first building structure drawing. In another possible implementation, the computing device may also perform boundary detection on the second building structure drawing, and then perform step B2 on the second building structure drawing, so that the second building structure drawing can include room areas and open areas.

[0157] In one possible implementation, when the computing device includes a drawing parser 101, this step 1002 can be performed by the drawing parser 101. Optionally, the drawing parser 101 includes an open area recognition unit, which is used to complete the open area recognition process of the first building structure drawing.

[0158] 1003. The computing device determines a deployment method of wireless local area network devices in the building based on multiple room areas and open areas in the second building structure diagram.

[0159] Wherein, relative to step 203, in step 1003, the deployment position of the WLAN device in the building in the deployment mode is also located in the open space indicated by the open area in the building.

[0160] The computing device determines the deployment location of the WLAN device in the corresponding room based on the multiple room areas in the second building structure diagram. This process has been described in the above 203 and will not be repeated here.

[0161] The computing device determines, based on the open areas in the second building structure diagram, deployment locations of the WLAN devices corresponding to the open spaces in the building. For example, the deployment locations of the WLAN devices in each open space in the building are determined according to the open space deployment rules. The open space deployment rules refer to rules for deploying WLAN devices in open spaces. As shown in Table 2, different types of open spaces may correspond to different open space deployment rules or may correspond to the same open space deployment rules. The open space deployment rules corresponding to the types of open spaces are not limited herein.

[0162] Table 2

[0163]

[0164] For any open area in the second building structure diagram, the deployment position of the WLAN device in the corresponding open space is determined according to the open space deployment rules corresponding to the type of the open area. This process is similar to the process in step 203 of "determining the deployment position of the WLAN device in the corresponding room according to the room deployment rules corresponding to the room category indicated by the first category label corresponding to the room area", and will not be repeated here.

[0165] In a similar manner, the computing device can determine the deployment position of each open area of ​​the WLAN device corresponding to the open space, and determine the deployment method of the WLAN device in the building based on the deployment position of the WLAN device in each room area corresponding to the room and the deployment position of each open area corresponding to the open space.

[0166] In another possible implementation, if a seating area exists in any open area in the second building structure diagram, the deployment location of the WLAN device in the corresponding open space is determined based on the open space deployment rules corresponding to the seating area and the category of the open area, wherein the number of WLAN devices in the seating space in the open space is greater than the number of WLAN devices in the non-seating space, the seating space refers to the space corresponding to the seating area in the open space (i.e., the seating area), and the non-seating space refers to the space in the open space other than the seating space. For example, taking the open space as a corridor, if one WLAN device needs to be deployed in the corridor and there is a seating space in the corridor, then one WLAN device can be deployed in the seating space in the corridor or near the seating space.

[0167] In an open space, the population density of the seating space is greater than that of the non-seating space. The seating space and the non-seating space have different capacity requirements for WLAN devices. Through seating area identification, the open area in the building structure diagram can be divided into seating areas and non-seating areas, that is, the open space in the building is divided into seating areas and non-seating spaces. Based on the seating area, the deployment position of the WLAN device in the corresponding open space is determined, so that the number of WLAN devices in the seating space in the open space is greater than the number of WLAN devices in the non-seating space, thereby meeting the different capacity requirements for WLAN devices between the seating space and the non-seating space in the open space, so that the deployment method determined by the deployment position of the WLAN device in the corresponding open space is more reasonable, and the planned deployment position of the WLAN device is also more reasonable, thereby further improving the coverage effect of the WLAN network in the building and the network rate when the terminal connects to the WLAN network.

[0168] After determining the deployment position of the WLAN device in the corresponding open space, the computing device determines the deployment method of the WLAN device in the building based on the deployment position of the WLAN device in the corresponding open space, the deployment position of the WLAN device in the corresponding room of each room area, and the deployment position of each open area corresponding to the open space.

[0169] In one possible implementation, the computing device includes Figure 1 In the case of the device deployer 102 in , this step 1003 can be performed by the device deployer 102.

[0170] Figure 10The embodiment shown can achieve Figure 2 The beneficial effects of the embodiment are that by identifying the open areas in the building structure diagram and determining the deployment method based on the open areas, WLAN devices can also be deployed in the open spaces in the building to meet the needs of the open spaces for the WLAN network, making the determined deployment method more reasonable and the planned deployment location of the WLAN devices more reasonable, thereby further improving the coverage effect of the WLAN network in the building and the network rate when the terminal connects to the WLAN network.

[0171] Buildings also contain obstacles that affect network signal propagation, such as walls and load-bearing columns. Accordingly, the building structure diagram also contains areas where these obstacles are located. For ease of description, the areas where these obstacles are located in the building structure diagram are referred to as obstacle areas. That is, obstacle areas indicate obstacles in the building. In another possible implementation, the computing device can also determine the deployment method of wireless local area network devices in the building based on the obstacle areas in the building structure diagram.

[0172] For example, Figure 13 The flowchart shown is another network planning method provided by an embodiment of the present application. The method is executed by a computing device, which may include the above-mentioned network planning system 100. The method includes the following steps.

[0173] 1301. A computing device obtains a first architectural structure drawing of a building.

[0174] Among them, this step 1301 is similar to the above-mentioned step 201. Here, the embodiment of the present application will not repeat this step 301.

[0175] 1302. The computing device identifies the first building structure diagram to obtain a second building structure diagram and a fourth building structure diagram, wherein the second building structure diagram includes multiple room areas, and the fourth building structure diagram includes an obstacle area.

[0176] Optionally, the second building structure diagram further includes an open area. The fourth building structure diagram includes at least one obstacle area, the obstacle area being the obstacle area identified from the first building structure diagram. Optionally, the fourth building structure diagram includes an obstacle area identifier, which is marked on the obstacle area to indicate the identified obstacle area. Figure 14 As shown, the obstacle area identifier is divided into a first obstacle area identifier and a second obstacle area identifier, wherein the first obstacle area identifier is used to identify the wall area in the building structure drawing, and the second obstacle area identifier is used to identify the load-bearing column area in the building structure drawing. Both the wall area and the load-bearing column area are obstacle areas, and the load-bearing column area is the area where the load-bearing columns of the building in the building structure drawing are located.

[0177] The room area and the obstacle area are both obtained by identifying the first building structure diagram. The identifying of the first building structure diagram includes the following steps C1 and C2.

[0178] Step C1: The computing device identifies an obstacle area in the first building structure diagram to obtain a fourth building structure diagram, where the fourth building structure diagram includes the obstacle area.

[0179] The computing device can identify the obstacle area in the first building structure diagram based on the obstacle recognition model. For example, the first building structure diagram is input to the obstacle recognition model, and the obstacle recognition model identifies the obstacle area of ​​the input first building structure diagram, marks the corresponding obstacle area identifier in the identified obstacle area, and outputs the first building structure diagram including the obstacle area identifier. The computing device uses the output first building structure diagram as the fourth building structure diagram, so that the fourth building structure diagram includes various identified obstacle areas. Exemplarily, by identifying Figure 7 The obstacle area in the local building structure diagram shown can be obtained Figure 14 The figure shows the local building structure after obstacle area identification.

[0180] Among them, the obstacle recognition model refers to an AI model that can identify obstacle areas in the building structure diagram, and supports identifying obstacle areas of various shapes in the building structure diagram. The obstacle recognition model can be a semantic segmentation model, such as ISANet, Mask R-CNN, etc. The obstacle recognition model can also be other AI models other than the semantic segmentation model, as long as it can identify obstacle areas in the building structure diagram.

[0181] The obstacle model can be obtained by training based on multiple third sample building structure drawings, wherein the third sample building structure drawings are building structure drawings used to train the obstacle recognition model, and the third sample building structure drawings include multiple obstacle area labels, and the obstacle area label is used to indicate an obstacle area in the third sample building structure drawing. The training device uses the obstacle area indicated by the obstacle area label as the recognition target, and trains the obstacle recognition model based on the third sample building structure drawings and the obstacle area labels in the third building structure drawings, so that the trained obstacle recognition model can learn the ability to recognize and mark obstacle areas in the building structure drawings, so that the computing device can use the trained obstacle recognition model to recognize and mark each obstacle area on the first building structure drawing.

[0182] Step C2: The computing device identifies the room areas in the first building structure diagram to obtain a second building structure diagram, where the second building structure diagram includes a plurality of room areas.

[0183] Among them, this step C2 is the same as step 202 or step 1002. Here, the embodiment of the present application will not repeat this step C2.

[0184] In another possible implementation, the computing device may also execute step C2 first and then execute step C1. Here, the embodiment of the present application does not limit the execution order of step C1 and step C2.

[0185] In one possible implementation, when the computing device includes a drawing parser 101, this step 1302 can be performed by the drawing parser 101. Optionally, the drawing parser 101 includes an obstacle recognition module and a room recognition module. The obstacle recognition module is used to indicate the above step C1, and the room recognition module is used to execute the above step C2. Optionally, the obstacle recognition module includes a room recognition unit and an open area recognition unit.

[0186] 1303. If the second building structure diagram includes multiple room areas, the computing device determines a deployment method of the WLAN device in the building based on the multiple room areas in the second building structure diagram and the obstacle area in the fourth building structure diagram.

[0187] Among them, relative to step 203, in this step 1303, in this deployment method, the distance between the deployment position of the WLAN device in the building and the load-bearing column in the building is greater than or equal to the first distance threshold, so as to reduce the impact of the load-bearing column on the network signal sent by the WLAN device. If the installation method of the deployed WLAN device is away from the wall, the distance between the deployment position of the WLAN device in the building and the wall is greater than or equal to the second distance threshold, so as to reduce the impact of the wall on the network signal sent by the WLAN device. Among them, the first distance threshold and the second distance threshold can be set according to the actual implementation scenario. Here, the embodiment of the present application does not limit the first distance threshold and the second distance threshold.

[0188] For any room area among the multiple room areas, a deployment position of the WLAN device in the corresponding room is determined based on the room deployment rule corresponding to the room area and the obstacle area within the room area. For example, after determining the actual area of ​​the installation area in the corresponding room in the building, the location of the obstacle in the room is determined based on the obstacle area in the room area and the scale of the first building structure drawing. The deployment position of each WLAN device in the installation area is determined based on the actual area of ​​the installation area, the spacing specified in the room deployment rule, and the location of the obstacle in the room. When the installation method of each WLAN device meets the installation method and spacing requirements specified in the room deployment rule, the distance between the deployment of each WLAN device and the location of the obstacle in the room is greater than or equal to a target distance threshold. If the obstacle is a load-bearing column, the target distance threshold is a first distance threshold; if the obstacle is a wall, the target distance threshold is a second distance threshold. The process of determining the actual area of ​​the installation area is described in step 203 and will not be repeated here.

[0189] In a similar manner, the computing device may determine the deployment locations of the WLAN devices in the rooms corresponding to the respective room areas, and determine the deployment method of the WLAN devices in the building based on the deployment locations of the WLAN devices in the rooms corresponding to the respective room areas.

[0190] The deployment method determined in step 1303 can reduce the impact of obstacles in the rooms of the building on the network signals of the WLAN devices deployed in the rooms, making the planned deployment locations of the WLAN devices more reasonable, thereby further improving the coverage of the WLAN network in the building and the network speed when the terminal connects to the WLAN network.

[0191] 1304. If the second building structure diagram includes multiple room areas and open areas, determine a deployment method of WLAN devices in the building based on the room areas and open areas in the second building structure diagram and the obstacle areas in the fourth building structure diagram.

[0192] For any area among multiple room areas and open areas, the deployment position of the WLAN device in the corresponding space of the building is determined based on the deployment rules corresponding to the area and the obstacle area within the room area, where if the area is a room area, the deployment rule corresponding to the area is the room deployment rule, and the space corresponding to the building is the room; if the area is an open area, the deployment rule corresponding to the area is the open space deployment rule, and the space corresponding to the building is the open space. The determination process is the same as the process of "determining the deployment position of the WLAN device in the corresponding room based on the room deployment rules corresponding to the room area and the obstacle area within the room area" in step 1303, and will not be repeated here.

[0193] In a similar manner, the computing device can determine the deployment locations of WLAN devices in the spaces corresponding to each of the multiple room areas and open areas, and determine a deployment method for the WLAN devices in the building based on the deployment locations of the WLAN devices in the spaces corresponding to the respective areas. This deployment method allows the WLAN devices to be deployed in the rooms and open spaces of the building, while also preventing the WLAN devices from being deployed on or near obstacles in the building.

[0194] The deployment method determined in step 1304 can reduce the impact of various obstacles in the building on the network signals of the WLAN devices deployed in the room, making the planned deployment location of the WLAN devices more reasonable, thereby further improving the coverage of the WLAN network in the building and the network speed when the terminal connects to the WLAN network.

[0195] In another possible implementation, the computing device may also identify obstacle areas in the second building structure diagram, such that the second building structure diagram also includes obstacle areas. The computing device then determines the deployment method of the WLAN devices in the building based on the multiple room areas and the obstacle areas in the second building structure diagram. This process is similar to step 1303. Alternatively, the computing device determines the deployment method of the WLAN devices in the building based on the multiple room areas, open areas, and obstacle areas in the second building structure diagram. This process is similar to step 1304.

[0196] In one possible implementation, the computing device includes Figure 1 In the case of the device deployer 102 in , step 1303 and step 1304 can be performed by the device deployer 102.

[0197] Figure 13 The embodiment shown can achieve Figure 10 The beneficial effect of the embodiment is that by identifying obstacle areas in the building structure diagram, the deployment method of WLAN devices in the building is determined based on the obstacle areas. Under this deployment method, WLAN devices can be avoided from being deployed on obstacles such as walls and / or load-bearing columns, and WLAN devices can be avoided from being close to obstacles, thereby reducing the impact of obstacles on network signals. As a result, the planned deployment location of WLAN devices is also more reasonable, thereby further improving the coverage effect of the WLAN network in the building and the network rate when the terminal connects to the WLAN network.

[0198] In another possible implementation, after determining the deployment mode, the computing device generates a network planning report based on the deployment mode. Figure 15The flowchart of another network planning method provided by an embodiment of the present application is shown. The method is executed by a computing device, which may include the above-mentioned network planning system 100. The method includes the following steps.

[0199] 1501. A computing device obtains a first architectural structure drawing of a building.

[0200] Here, step 1501 is the same as step 201 above and will not be described again here.

[0201] 1502. The computing device identifies the first building structure drawing to obtain a second building structure drawing.

[0202] Among them, this step 1501 is the same as the above-mentioned step 202, step 1002 or step 1302, and will not be repeated here.

[0203] by Figure 16 Taking the schematic diagram of a network planning process provided by an embodiment of the present application as an example, for a certain building structure diagram obtained, the computing device performs intelligent drawing parsing on the building structure diagram to mark the obstacles in the building and the scenes of each area in the building (i.e., area type) in the building structure diagram, wherein the marking of obstacles can be achieved by identifying the obstacle area in the building structure diagram, such as step C1 in step 1302, and the marking of domain scenes can be achieved by identifying the room area and / or open area in the building structure diagram, such as step 202 for identification of the room area, and step B2 for identification of the open area in step 1002.

[0204] As introduced above, the annotation of regional scenes and obstacles can be completed by AI models (such as room recognition models, boundary detection models, and obstacle recognition models). In another possible implementation method, for the AI ​​model that has been trained, the computing device or any of the training devices can also obtain a building structure diagram manually annotated by the user. The building structure diagram can be annotated with the outline of at least one of the room area, building area, and obstacle area. The building structure diagram is used as a new sample building structure diagram, and the outline of an area manually annotated by the user is used as a new area label (such as a room area label, a building area label, and an obstacle area label) to update the label set of the AI ​​model. The label set includes the area label used to train the AI ​​model. The new area label is used as the recognition target to train the trained AI model again to fine-tune the AI ​​model. The device can periodically obtain new area labels and fine-tune the AI ​​model based on the new area labels to achieve online training of the AI ​​model. By fine-tuning the AI ​​model, the AI ​​model can recognize room areas, building areas or obstacle areas of more shapes in the building structure diagram, thereby further improving the recognition accuracy of the AI ​​model, so that subsequent computing devices can accurately identify room areas, building areas or obstacle areas in the first building structure diagram based on the AI ​​model, so that the deployment method determined based on the accurate recognition results of the AI ​​model is more reasonable.

[0205] like Figure 16 As shown, after obtaining the drawing parsing result of the building structure drawing (such as the second building structure drawing), the computing device automatically deploys the WLAN device on the building structure drawing based on the drawing parsing result, as shown in step 1503 below.

[0206] 1503. The computing device determines a deployment method of WLAN devices in the building based on multiple room areas in the second building structure diagram.

[0207] Among them, this step 1503 is the same as the above-mentioned step 203, step 1003, step 1303 or step 1304, and will not be repeated here.

[0208] The deployment method includes the deployment locations of the multiple WLAN devices in the building, or the deployment method also includes the device models, number, installation method, and configuration information of the multiple WLAN network devices. After determining the deployment method of the WLAN devices in the building, the deployment locations of the multiple WLAN devices in the building are obtained. The computing device adds device identifiers for the deployment locations of the multiple WLAN devices on the second building structure diagram to obtain a fifth building structure diagram, thereby automatically deploying the WLAN devices on the building structure diagram. The fifth building structure diagram includes multiple device identifiers, each device identifier indicating the deployment location of a WLAN device in the building.

[0209] Assumptions Figure 12 The local building structure diagram of the second building structure diagram is used. Taking the WLAN device as an AP as an example, the computing device adds the device identifier of the AP to the deployment position of each WLAN device in the local building structure diagram, removes the area identifier of each area in the building structure diagram (such as the room area identifier, the obstacle area identifier, the open area identifier, etc.), and obtains Figure 17 The partial building structure diagram shown is from Figure 17 It can be seen that APs are deployed in rooms and open spaces in buildings, and the APs are kept away from obstacles such as walls and load-bearing columns in the buildings.

[0210] The above is explained using the example of adding device identification to the second building structure diagram to obtain the fifth building structure diagram. In another possible implementation method, the computing device can also add device identifications for the deployment locations of multiple WLAN devices in the building on the first building structure diagram to obtain the fifth building structure diagram, or add device identifications for the deployment locations of multiple WLAN devices in the building on the fourth building structure diagram, remove the obstacle area identification on the fourth building structure diagram, and obtain the fifth building structure diagram.

[0211] 1504. The computing device predicts the signal coverage in the building under the deployment mode and obtains a first prediction result.

[0212] Among them, the first prediction result includes the location coordinates of multiple points in the building and the signal strength of multiple points. The multiple points can be individual locations in the building or partial locations in the building. Here, the embodiment of the present application does not limit the number and location of the points.

[0213] The computing device predicts signal coverage in the building under the deployment method based on the deployment locations of the multiple WLAN devices and the identified obstacle areas. For example, based on the deployment locations of the multiple WLAN devices, the distance between each of the multiple points and the multiple WLAN devices is obtained. Based on the deployment locations of the multiple WLAN devices and the obstacle areas in the second building structure diagram (or the fourth building structure diagram), the obstacle distribution between each point and the multiple WLAN devices is obtained. Based on the distance between each point and the multiple WLAN devices and the obstacle distribution between each point and the multiple WLAN devices, a classic signal attenuation model or ray tracing technology is used to predict the signal strength of each point to obtain the predicted signal strength of each point.

[0214] In a possible implementation, the computing device generates a coverage signal simulation diagram based on the predicted signal strength of each point and the fifth building structure diagram, and the coverage signal simulation diagram is used to display the signal strength of each point. Figure 17 For example, the deployed APs Figure 17 The predicted signal strength at each point under the deployment mode shown is as follows Figure 18 As shown, signal strength can be represented by signal identifiers. Different signal identifiers represent different signal strengths. For example, signal identifiers are divided into triangles, squares, four-pointed stars, small circles, etc. The signal strengths represented by triangles, squares, four-pointed stars, and small circles decrease in sequence. The signal strengths corresponding to the points in the areas where the triangles and squares are located meet the standards, and the signal strengths corresponding to the points in the areas where the four-pointed stars and small circles are located do not meet the standards. For example, the signal strength corresponding to the area where the four-pointed star in the large circle is located is relatively weak, the signal strength corresponding to this area does not meet the standards, and the signal coverage effect in this area is poor.

[0215] 1505. The computing device predicts the network rate of the terminal in the building under the deployment mode to obtain a second prediction result.

[0216] The second prediction result includes the network speeds at the multiple points in the building. The network speed at any point is the network speed of the terminal at that point in the building in this deployment configuration. The terminal can be a smartphone, tablet, music player, wearable smart device, laptop, or desktop computer, etc. The present embodiment does not limit the terminal type.

[0217] The computing device predicts the signal coverage in the building under the deployment mode based on the deployment locations of multiple WLAN devices, the device information of multiple WLAN devices and the first prediction result, wherein the device information of any WLAN device includes the device model, bandwidth, frequency band, concurrency number and at least one of the business distribution of the WLAN device, the concurrency number refers to the number of terminals mounted on the WLAN device, the business distribution refers to the distribution of running applications in the terminals mounted on the WLAN device, the running of applications affects the network rate of the terminals mounted on the WLAN device, the business distribution includes the number of terminals running the first application and the number of terminals running the second application, wherein the first application and the second application are different applications, and the application can be a game application, a video application, an instant messaging application, etc. Here, the embodiment of the present application does not limit the category of the application.

[0218] For example, the computing device takes the deployment locations of multiple WLAN devices, the device information of multiple WLAN devices, and the point information of multiple points as input data, and inputs them into the network rate prediction model. The network rate prediction model predicts the network rates of the multiple points based on the input data, and outputs a second prediction result, wherein the point information of any point is obtained based on the first prediction result, and the point information includes the location coordinates of the point and the signal strength of the point. If the point is located in a room in a building, the point also includes the room type of the room.

[0219] The network rate prediction model is an AI model used to predict the network rate of a terminal in a building, such as a deep neural network (DNN), a support vector machine (SVM), and a decision tree. The network rate prediction model is trained based on the deployment locations of multiple sample devices in a sample building, the device information of the multiple sample devices, the actual signal strengths of multiple points in the sample building, and the actual network rates of the multiple points. The sample building refers to any building, the multiple sample devices refer to the WLAN devices deployed in the sample building, the actual signal strength of any point is the signal strength of the point under the action of the multiple sample devices in the building, and the actual network rate of any point is the network rate of the terminal at the point under the action of the multiple sample devices in the building. The actual network rates of the multiple points are labels for training the network rate prediction model. The training device uses the actual network rates of the multiple points as the output targets of the network rate prediction model, and trains the network rate prediction model based on the deployment positions of the multiple sample devices in the sample building, the device information of the multiple sample devices, and the actual signal strengths of the multiple points in the sample building. When the difference between the network rates of the multiple points that can be output by the trained network rate prediction model and the actual network rates of the multiple points is less than the difference threshold, the training of the network rate prediction model is completed. After the training is completed, the network rate prediction model learns the following capabilities: based on the deployment positions of the multiple WLAN devices, the configuration information of the multiple WLAN devices, and the signal strengths of the multiple points in the building, the network rates of the multiple points are predicted, so that the subsequent computing device executes this step 1505 through the trained network rate prediction model.

[0220] In a possible implementation, the computing device generates a rate simulation diagram based on the predicted network rate of each point and the fifth building structure diagram, and the rate simulation diagram is used to display the network rate of each point. Figure 17 For example, the deployed APs Figure 17 The predicted network speeds at each point under the deployment method shown are as follows: Figure 19 As shown, the network rate can be represented by a rate identifier. Different rate identifiers represent different network rates. For example, the rate identifiers are divided into triangles, squares, four-pointed stars, and small circles. The network rates represented by triangles, squares, four-pointed stars, and small circles decrease in sequence. Among them, the network rates corresponding to the points in the areas where the triangles and squares are located meet the standards, while the network rates corresponding to the points in the areas where the four-pointed stars and small circles are located do not meet the standards.

[0221] In one possible implementation, the computing device includes Figure 1In the case of the deployment verifier 103 in , steps 1504 and 1505 can be performed by the deployment verifier 103.

[0222] 1506. The computing device determines a network planning report for the building based on the first prediction result, the second prediction result, and the deployment method. The network planning report includes a deployment location of the WLAN network device in the building.

[0223] The network planning report may further include a first prediction result and a second prediction result.

[0224] After obtaining the first prediction result and the second prediction result, the computing device generates the network planning report based on the first prediction result, the second prediction result and the deployment method.

[0225] In another possible implementation, Figure 16 As shown, by executing step 1504, the coverage simulation of the WLAN network in the building is realized and the first prediction result is obtained. By executing the above step 1505, the experienced network rate in the building is simulated and the second prediction result is obtained. After obtaining the first prediction result and the second prediction result, the computing device verifies whether the first prediction result and the second prediction result meet the standards.

[0226] A method for verifying whether the first prediction result meets the standard is, for example, the computing device compares the signal strength of each point in the first prediction result with the first expected result. If there is a point in the first prediction result whose signal strength is lower than the first expected result, then the first prediction result does not meet the first expected result; if the signal strength of each point in the first prediction result is greater than or equal to the first expected result, then the first prediction result meets the first expected result and the first prediction result meets the standard, wherein the first expected result refers to the expected signal coverage in the building, and the first prediction result can be the expected minimum signal strength in the building.

[0227] A method for verifying whether the second prediction result meets the standard is, for example, that the computing device compares the network rate of each point in the second prediction result with the second expected result respectively. If there is a point in the second prediction result whose network rate is lower than the second expected result, then the second prediction result does not meet the second expected result. If the network rate of each point in the second prediction result is greater than or equal to the second expected result, then the second prediction result meets the second expected result and the second prediction result meets the standard, wherein the second expected result refers to the expected network rate of the terminal in the building. For example, the second prediction result can be the minimum network rate of the expected terminal in the building.

[0228] Continue to refer Figure 16When both the first prediction result and the second prediction result meet the standards, the computing device generates a network planning report. For example, if the first prediction result reaches the first expected result and the second prediction result reaches the second expected result, a network planning report is generated based on the first prediction result, the second prediction result and the deployment method.

[0229] If at least one of the first and second prediction results fails to meet the requirements, the computing device adjusts the deployment method determined by the automatic deployment. For example, if the first prediction result fails to meet the first expected result and / or the second prediction result fails to meet the second expected result, the computing device adjusts the deployment method based on the first and second prediction results to obtain an adjusted deployment method. The following describes the deployment method adjustment process for different scenarios where the requirements are not met, combining Scenarios 1 to 3.

[0230] Case 1: If the first prediction result does not reach the first expected result and the second prediction result reaches the second expected result, the deployment mode is adjusted to obtain an adjusted deployment mode.

[0231] If the first prediction result does not reach the first expected result and the second prediction result reaches the second expected result, the computing device displays a signal simulation diagram. For the area where the signal strength does not reach the first expected result, the computing device can deploy a new WLAN device in the area on the signal simulation diagram, or adjust the WLAN device adjacent to the area to a position close to the area on the signal simulation diagram to obtain an adjusted signal simulation diagram, and determine a new deployment method based on the position of each device identifier in the adjusted signal simulation diagram.

[0232] Among them, the method of deploying new WLAN devices in the area is, for example, a user performs a device deployment operation on at least one point in the area, and the computing device responds to the device deployment operation by adding a device identifier to at least one point to indicate that a WLAN device is deployed at the point. The device deployment operation indicates that a WLAN device is deployed at the selected point, and by adding a new WLAN device in the area, the signal strength of each point in the area is increased. Figure 18 Take the signal simulation diagram shown as an example, Figure 18 The signal strength of the points in the area where the four-pointed star in the middle circle is located does not meet the first expected result. A new WLAN device, AP13, can be added to the area or near the area to obtain Figure 20 The new deployment method is shown.

[0233] A method of adjusting the WLAN devices adjacent to the area, for example, a user performs a position adjustment operation on a device identifier near the area, and the computing device adjusts the device identifier from a first position in the signal simulation diagram to a second position in response to the position adjustment operation, wherein the position adjustment operation is used to indicate that the deployment position of the WLAN device indicated by the device identifier is adjusted from the first position to the second position, and the distance between the first position and the center of the area is greater than the distance between the second position and the center, and the WLAN devices near the area are adjusted to a position close to the area to increase the signal strength of each point in the area.

[0234] In another possible implementation, the user may also select an area where the coverage does not meet the standard. For example, for any area where the signal strength does not reach the first expected result, if the user performs a first confirmation operation on the area, the computing device responds to the confirmation operation by not deploying a new WLAN device in the area, or by adjusting the WLAN device adjacent to the area to a position close to the area on the signal simulation diagram, wherein the first confirmation operation is used to indicate confirmation that the signal strength in the area is allowed to be substandard, thereby reducing the number of areas where the signal strength needs to be adjusted so that the deployment method can be adjusted quickly.

[0235] like Figure 16 As shown, after obtaining the adjusted deployment method, the computing device simulates the coverage of the WLAN network in the building again for the adjusted deployment method to obtain a new first prediction result, and verifies whether the new first prediction result meets the standard. If the new first prediction result does not meet the standard, the deployment method is adjusted again according to the processing method of Case 1, and so on, until a first prediction result that meets the standard is obtained. After obtaining the first prediction result that meets the standard, the computing device simulates the experienced network rate in the building again for the deployment method corresponding to the first prediction result that meets the standard to obtain a new second prediction result. Figure 20 The deployment mode shown is a new deployment mode, and the new first prediction result can be as follows: Figure 21 As shown, the new second prediction result can be Figure 22 As shown, relative to Figure 18 , Figure 21 AP13 is newly added, and the signal strength around AP13 is increased. Figure 19 , Figure 22 AP13 is added, and the network rate around AP13 increases. The computing device uses the deployment method corresponding to the first prediction result that meets the standard as the final deployment method, and generates a network planning report based on the first prediction result that meets the standard, the deployment method corresponding to the first prediction result that meets the standard, and the new second prediction result.

[0236] It should be noted that Figure 18 、 Figure 19 、 Figure 21 and Figure 22 There is also text information in the figure, such as "AP1", "double room" and "stairwell". In order to distinguish it from the signal identifier / rate identifier in the figure, this text information is shown with a black background in the figure. In another possible implementation, the background color of this text information may also be the signal identifier / rate identifier corresponding to the area, rather than black or other colors, to avoid the background color of the text information obscuring the signal identifier / rate identifier corresponding to the area. The text information such as "double room" and "stairwell" in the figure used to describe the room type may be the category label obtained through room type identification, or it may be the text information provided by the building structure diagram. Here, there is no limitation on the source of the text information used to describe the room type. Figure 18 、 Figure 19 、 Figure 21 and Figure 22 In the figure, the signal identifiers / rate identifiers are represented by patterns. In another possible implementation, the signal identifiers / rate identifiers can also be represented by colors, and different colors or different depths of the same color represent different signal identifiers / rate identifiers. Taking the use of colors to represent signal identifiers as an example, different colors represent different signal strength ranges. For example, the signal strength ranges corresponding to the purple points, green points, gray points, and white points in the signal simulation diagram decrease in sequence. The signal strengths corresponding to the purple points and green points meet the standards, while the signal strengths corresponding to the gray points and white points do not meet the standards. Taking the use of colors to represent rate identifiers as an example, different colors represent the speed of the network. For example, the network speed ranges corresponding to the green points, yellow points, red points, and white points decrease in sequence. Among them, the network speeds corresponding to the green points and yellow points meet the standards, while the network speeds corresponding to the red points and white points do not meet the standards.

[0237] Case 2: If the first prediction result reaches the first expected result and the second prediction result does not reach the second expected result, the deployment mode is adjusted to obtain an adjusted deployment mode.

[0238] If the first prediction result reaches the first expected result and the second prediction result does not reach the second expected result, the computing device displays the rate simulation graph. For the area where the network rate does not reach the second expected result, the computing device adjusts the rate simulation graph in accordance with the method of adjusting the signal simulation graph in Case 1 to obtain an adjusted rate simulation graph. Based on the location of each device identifier in the adjusted rate simulation graph, a new deployment method is determined.

[0239] like Figure 16As shown, after obtaining the adjusted deployment method, the computing device simulates the experienced network rate in the building again for the adjusted deployment method to obtain a new second prediction result, and verifies whether the new second prediction result meets the standard. If the new second prediction result does not meet the standard, the deployment method is adjusted again according to the processing method of Case 2, and so on, until a second prediction result that meets the standard is obtained. After obtaining the second prediction result that meets the standard, the computing device simulates the coverage of the WLAN network in the building again for the deployment method corresponding to the second prediction result that meets the standard to obtain a new first prediction result. The deployment method corresponding to the second prediction result that meets the standard is used as the final deployment method, and a network planning report is generated based on the second prediction result that meets the standard, the deployment method corresponding to the second prediction result that meets the standard, and the new first prediction result.

[0240] Case 3: If the first prediction result does not reach the first expected result and the second prediction result does not reach the second expected result, the deployment mode is adjusted to obtain an adjusted deployment mode.

[0241] If the first prediction result reaches the first expected result and the second prediction result does not reach the second expected result, the computing device displays a rate simulation graph. The computing device first adjusts the deployment method according to situation 1 to obtain a deployment method (referred to as the first deployment method) that can make the first prediction result meet the standard. Then, according to situation 2, the first deployment method is adjusted to obtain a deployment method (referred to as the second deployment method) that can make the second prediction result meet the standard. For the second deployment method, the computing device again simulates the coverage of the WLAN network in the building to obtain the first prediction result under the second deployment method. Based on the first prediction result under the second deployment method, the second prediction result under the second deployment method, and the second deployment method, a network planning report is generated.

[0242] The above-mentioned method determines whether the deployment method is reasonable by verifying whether the first prediction result and the second prediction result meet the standards. If the first prediction result and / or the second prediction result do not meet the standards, the deployment method is adjusted so that the adjusted deployment method can ensure that the network deployment in the building and the network rate of the terminal in the building achieve the expected effect, making the adjusted deployment method more reasonable and the planned deployment location of the WLAN equipment more reasonable, thereby further improving the coverage effect of the WLAN network in the building and the network rate when the terminal connects to the WLAN network.

[0243] In another possible embodiment, the user confirms the final deployment method based on the first prediction result and the second prediction result. For example, if the first prediction result does not reach the first expected result and / or the second prediction result does not reach the second expected result, the user can perform a second confirmation operation on the first prediction result and / or the second prediction result, and the computing device generates a network planning report based on the first prediction result, the second prediction result and the deployment method in response to the second confirmation operation. Alternatively, if the first prediction result reaches the first expected result and the second prediction result reaches the second expected result, the user can also perform a third confirmation operation on the first prediction result and / or the second prediction result, and the computing device generates a network planning report based on the first prediction result, the second prediction result and the deployment method in response to the second confirmation operation, wherein the second confirmation operation is used to indicate that the deployment method corresponding to the first prediction result and the second prediction result is the final deployment method.

[0244] The above is explained by first determining an initial deployment method and then predicting whether the network conditions (network coverage and network speed) under the deployment method meet the standards. In another possible implementation method, when determining the deployment method, the first expected result and the second expected result are used as constraints, and the deployment method is determined based on multiple room areas in the second building structure diagram. The deployment method refers to a deployment method that makes the network conditions in the building meet the standards. Then, for this deployment method, deployment 1504 and step 1505 are executed to obtain the first prediction result and the second prediction result, and a network planning report is generated based on the deployment method, the first prediction result and the second prediction result.

[0245] If a user uploads a first building structure diagram to a computing device via a terminal, the computing device can send a network planning report to the terminal after determining the network planning report. Alternatively, the computing device may not send the network planning report to the terminal, but instead a technician may print the network planning report from the computing device and provide the printed network planning report to the user. This allows the user to deploy WLAN devices in the building according to the deployment method specified in the network planning report.

[0246] In one possible implementation, the computing device includes Figure 1 In the case of the device deployer 102 and the deployment verifier 103, the process of determining the deployment mode in this step can be performed by the device deployer 102, and the process of verifying whether the first prediction result and the second prediction result meet the standards can be performed by the deployment verifier 103.

[0247] Figure 15 The embodiment shown can achieve Figure 2 、 Figure 10 or Figure 13The beneficial effect of the embodiment is that the signal coverage in the building and the network rate of the terminal in the building under the deployment method are predicted, and a network planning report for the building is generated based on the prediction results and the deployment method, so that the prediction results can provide evidence for the deployment location of the WLAN device in the network planning report, thereby increasing the credibility of the deployment location of the WLAN device, so that the WLAN device can be deployed in the building in the future according to the deployment location of the WLAN device in the building in the network planning report.

[0248] The above is explained by taking the example of the annotation of regional scenes and the annotation of obstacles completed by AI models (such as room recognition models, boundary detection models, and obstacle recognition models). In another possible way, the identification of regional scenes and the identification of obstacles are completed by AI models, and the annotation of regional scenes and the annotation of obstacles are not implemented by AI models. Taking obstacle area recognition as an example, the computing device identifies the obstacle area in the first building structure diagram based on the obstacle recognition model and obtains obstacle area information. The obstacle area information indicates the position of the obstacle area in the first building structure diagram. Afterwards, the computing device adds an obstacle area identifier in the first building structure diagram based on the obstacle area information to realize the annotation of the obstacle area. The annotation process of the regional scene is similar to the annotation process of the obstacle area and will not be repeated here. In another possible implementation, after the computing device obtains room area information, obstruction area information, open area information, or seating area information based on AI model identification, it may not mark the corresponding area identifiers on the first building structure diagram to obtain the second building structure diagram. Instead, the computing device determines the deployment mode of the WLAN devices in the building based on at least one of the room area information, obstruction area information, open area information, and seating area information. The determination process may refer to steps 203, 1003, 1303, or 1304 above. The room area information indicates the location of the room area in the first building structure diagram, the open area information indicates the location of the open area in the first building structure diagram, and the seating area information indicates the location of the seating area within the open area in the first building structure diagram.

[0249] The above describes the method of the embodiment of the present application, and the following describes the device of the embodiment of the present application. It should be understood that the device described below has any function of the computing device in the above method or the above network planning system. Figures 2 to 22 The network planning method according to the embodiment of the present application is described in detail. Figures 23 to 26 The apparatus of the network planning method according to the embodiment of the present application is described. It should be understood that the technical features described in the method embodiment are also applicable to the following apparatus embodiment.

[0250] Figure 23This is a schematic diagram of the structure of a network planning device provided in an embodiment of the present application. Figure 23 The device 2300 shown is used to execute any network planning method provided in the embodiments of the present application, such as Figure 23 As shown, the apparatus 2300 includes:

[0251] An acquisition module 2301 is configured to acquire a first architectural structure drawing of a building;

[0252] an identification module 2302 for identifying the first building structure diagram to obtain a second building structure diagram, wherein the second building structure diagram includes a plurality of room areas, where the room areas indicate rooms in the building;

[0253] The determination module 2303 is configured to determine a deployment mode of wireless local area network devices in the building based on multiple room areas in the second building structure diagram.

[0254] In a possible implementation, the determining module 2303 is further configured to:

[0255] Deployment of wireless local area network devices in the building is determined based on multiple room areas in the second building structure diagram and first category labels corresponding to the multiple room areas, where the first category labels indicate categories of rooms indicated by the corresponding room areas.

[0256] In a possible implementation, the application scenario of the building corresponds to a plurality of second category labels, and the second category labels are used to indicate the category of the rooms in the building in the application scenario; the determination module 2303 is further configured to:

[0257] For any room area among the multiple room areas, identifying text information in the room area, and extracting room category information from the identified text information;

[0258] The second category label having the highest semantic similarity with the room category information among the multiple second category labels is determined as the first category label corresponding to the room area.

[0259] In a possible implementation, the application scenario of the building corresponds to a plurality of second category labels, and the second category labels are used to indicate the category of the rooms in the building in the application scenario; the determination module 2303 is further configured to:

[0260] For any room area among the multiple room areas, if the first category label corresponding to the room area is not the second category label in the application scenario, a target category label is determined from the multiple second category labels based on the first category label corresponding to the room area, and the function of the room under the target category label is similar to the function of the room under the first category label corresponding to the room area;

[0261] Replace the first category label corresponding to the room area with the target category label.

[0262] In a possible implementation, the second building structure diagram further includes an obstacle area, where the obstacle area indicates obstacles in the building, and the obstacles include walls or load-bearing columns. The determining module 2303 is further configured to:

[0263] Based on the multiple room areas and obstacle areas in the second building structure diagram, a deployment mode of wireless local area network equipment in the building is determined.

[0264] In a possible implementation, the second building structure diagram further includes an open area, which refers to an area outside the multiple room areas and within the outline of the building in the second building structure diagram; the determining module 2303 is further configured to:

[0265] Based on the multiple room areas and open areas in the second building structure diagram, a deployment method of wireless local area network equipment in the building is determined.

[0266] In a possible implementation, the open area also includes a seating area, and the seating area indication refers to an area in the open area where seats are arranged; the determining module 2303 is further configured to:

[0267] Based on the multiple room areas, the open area, and the at least one seating area in the second building structure diagram, a deployment mode of the wireless local area network devices in the building is determined.

[0268] In a possible implementation, the room area is a first room area or a second room area, the first room area indicates an independent room in a building, and the second room area indicates a room having sub-rooms in the building; the identifying module 2302 includes:

[0269] an identification unit, configured to identify room areas on the first building structure diagram to obtain a plurality of first room areas;

[0270] The fusion unit is used to perform regional fusion on multiple first room areas in the first building structure diagram based on the overlap of the outlines of the multiple first room areas and the adjacent room areas, so as to obtain at least one second room area, where the second room area includes two adjacent first room areas.

[0271] In a possible implementation, the fusion unit is configured to:

[0272] For any room area among the multiple first room areas, obtaining a degree of outline overlap between the room area and an adjacent room area, where the adjacent room area refers to a first room area adjacent to the room area, and the degree of outline overlap indicates a degree of overlap between a outline of the room area and a outline of the adjacent room area;

[0273] If the outline overlap is greater than or equal to the overlap threshold, the room area and the adjacent area are fused on the first building structure diagram to obtain a second room area.

[0274] In a possible implementation, the apparatus 2300 further includes:

[0275] A first prediction module is used to predict the signal coverage in the building under the deployment mode and obtain a first prediction result;

[0276] A second prediction module is used to predict the network rate of the terminal in the building under the deployment mode to obtain a second prediction result;

[0277] The determination module 2303 is further configured to determine a network planning report of the building based on the first prediction result, the second prediction result and the deployment method, where the network planning report includes the deployment location of the wireless local area network device in the building.

[0278] In a possible implementation, the determining module 2303 is further configured to:

[0279] If the first prediction result does not achieve the first expected result and / or the second prediction result does not achieve the second expected result, the deployment method is adjusted based on the first prediction result and the second prediction result. The first expected result refers to the expected signal coverage in the building, and the second expected result refers to the expected network rate of the terminal in the building.

[0280] It should be understood that device 2300 corresponds to the computing device in the above-mentioned method embodiment, and the various modules in device 2300 and the above-mentioned other operations and / or functions are respectively various steps and methods implemented by the computing device in order to realize the method embodiment. For specific details, please refer to the above-mentioned method embodiment. For the sake of brevity, they will not be repeated here.

[0281] It should be understood that the division of the above-mentioned functional modules is only used as an example to illustrate the network planning of device 2300. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of device 2300 can be divided into different functional modules to complete all or part of the functions described above. In addition, the device 2300 provided in the above embodiment and the above-mentioned method embodiment are based on the same concept. The specific implementation process is detailed in the above-mentioned method embodiment and will not be repeated here.

[0282] For any module among the above-mentioned acquisition module 2301, identification module 2302, determination module 2303, first prediction module or second prediction module, the implementation method of the module is the same as the implementation method of the above-mentioned drawing parser 101. Here, the implementation method of the module in the embodiment of the present application will not be repeated.

[0283] The present application also provides a computing device 2400. Figure 24 As shown, computing device 2400 includes a bus 2402, a processor 2404, a memory 2406, and a communication interface 2408. Processor 2404, memory 2406, and communication interface 2408 communicate with each other via bus 2402. Computing device 2400 can be a server or a terminal device. It should be understood that this application does not limit the number of processors and memories in computing device 2400.

[0284] The bus 2402 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 The bus 2404 may include a path for transmitting information between various components of the computing device 2400 (eg, the memory 2406, the processor 2404, and the communication interface 2408).

[0285] The processor 2404 may include any one or more processors such as a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP), or a digital signal processor (DSP).

[0286] The memory 2406 may include volatile memory, such as random access memory (RAM). The processor 2404 may also include non-volatile memory, such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid state drive (SSD).

[0287] The memory 2406 stores executable program code, and the processor 2404 executes the executable program code to respectively implement the functions of the aforementioned acquisition module 2031, identification module 2302, and determination module 2303, thereby implementing the network planning method. In other words, the memory 2406 stores instructions for executing the network planning method.

[0288] Alternatively, the memory 2406 stores executable codes, and the processor 2404 executes the executable codes to respectively implement the functions of the aforementioned network planning apparatus 2300, thereby implementing the network planning method. That is, the memory 2406 stores instructions for executing the network planning method.

[0289] The communication interface 2408 uses a transceiver module such as, but not limited to, a network interface card or a transceiver to implement communication between the computing device 2400 and other devices or a communication network.

[0290] Embodiments of the present application also provide a computing device cluster. The computing device cluster includes at least one computing device. The computing device can be a server, such as a central server, an edge server, or a local server in a local data center. In some embodiments, the computing device can also be a terminal device such as a desktop computer, a laptop computer, or a smartphone.

[0291] like Figure 25 As shown, the computing device cluster includes at least one computing device 2400. The memory 2406 in one or more computing devices 2400 in the computing device cluster may store the same instructions for executing the network planning method.

[0292] In some possible implementations, the memory 2406 of one or more computing devices 2400 in the computing device cluster may also store some instructions for executing the network planning method. In other words, the combination of one or more computing devices 2400 can jointly execute the instructions for executing the network planning method.

[0293] It should be noted that the memory 2406 in different computing devices 2400 in the computing device cluster can store different instructions, each for executing a portion of the functions of the network planning apparatus. In other words, the instructions stored in the memory 2406 in different computing devices 2400 can implement the functions of one or more of the acquisition module 2031, the identification module 2302, and the determination module 2303.

[0294] In some possible implementations, one or more computing devices in a computing device cluster may be connected via a network, which may be a wide area network or a local area network. Figure 26 A possible implementation is shown. Figure 26As shown, two computing devices 2400A and 2400B are connected via a network. Specifically, the connection to the network is achieved through a communication interface in each computing device. In this possible implementation, the memory 2406 in computing device 2400A stores instructions for executing the functions of identification module 2302. Simultaneously, the memory 2406 in computing device 2400B stores instructions for executing the functions of acquisition module 2301 and determination module 2303.

[0295] Figure 26 The connection method between the computing device clusters shown may be that considering that a large amount of calculation is required to identify the building structure diagram in the network planning method provided in this application, it is considered to entrust the functions of the acquisition module 2301 and the determination module 2303 to the computing device 2400B for execution.

[0296] It should be understood that Figure 26 The functionality of the computing device 2400A shown in FIG2 may also be implemented by multiple computing devices 2400. Similarly, the functionality of the computing device 2400B may also be implemented by multiple computing devices 2400.

[0297] The present application embodiment also provides another computing device cluster. The connection relationship between the computing devices in the computing device cluster can be similarly referred to as Figure 25 and Figure 26 The connection mode of the computing device cluster is different in that the memory 2406 of one or more computing devices 2400 in the computing device cluster may store the same instructions for executing the network planning method.

[0298] In some possible implementations, the memory 2406 of one or more computing devices 2400 in the computing device cluster may also store some instructions for executing the network planning method. In other words, the combination of one or more computing devices 2400 can jointly execute the instructions for executing the network planning method.

[0299] It should be noted that the memory 2406 in different computing devices 2400 in the computing device cluster can store different instructions for executing partial functions of the network planning system. In other words, the instructions stored in the memory 2406 in different computing devices 2400 can implement the functions of one or more of the graph parser 101, the device deployer 102, and the deployment verifier 103.

[0300] Embodiments of the present application also provide a computer program product including instructions. The computer program product may be software or a program product including instructions that can be run on a computing device or stored in any available medium. When the computer program product is run on at least one computing device, the at least one computing device executes a network planning method.

[0301] The embodiment of the present application also provides a computer-readable storage medium. The computer-readable storage medium can be any available medium that can be stored by a computing device or a data storage device such as a data center that contains one or more available media. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a magnetic tape), an optical medium (e.g., a high-density digital video disc (DVD)), or a semiconductor medium (e.g., a solid-state drive). The computer-readable storage medium includes instructions that instruct the computing device to execute the network planning method.

[0302] In addition, an embodiment of the present application also provides a device, which can specifically be a chip, component or module, and the device may include a connected processor and memory; wherein the memory is used to store computer execution instructions, and when the device is running, the processor can execute the computer execution instructions stored in the memory to enable the chip to execute the network planning method in the above-mentioned method embodiments.

[0303] Among them, the devices, equipment, computer-readable storage media, computer program products or chips provided in this embodiment are all used to execute the corresponding methods provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding methods provided above, and will not be repeated here.

[0304] In the description of this application, unless otherwise specified, " / " means "or", for example, A / B can mean A or B. "And / or" in this article is merely a description of the association relationship of associated objects, indicating that three relationships can exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, "at least one" means one or more, and "a plurality" means two or more. Words such as "first" and "second" do not limit the quantity and execution order, and words such as "first" and "second" do not limit them to be necessarily different.

[0305] In this application, words such as "exemplary" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described in this application as "exemplary" or "for example" should not be construed as preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0306] It should be noted that the information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data used for analysis, storage, and display, etc.), and signals involved in this application are all authorized by the user or fully authorized by all parties, and the collection, use, and processing of relevant data must comply with the relevant laws, regulations, and standards of the relevant countries and regions. For example, the building structure drawings involved in this application were obtained with full authorization.

[0307] All the above optional technical solutions can be arbitrarily combined to form optional embodiments of the present disclosure, and will not be described in detail here.

[0308] The above description is merely an optional embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A network planning method, characterized in that: The method is executed by a computing device, and includes: Obtaining a first architectural structural drawing of the building; identifying the first building structure diagram to obtain a second building structure diagram, wherein the second building structure diagram includes a plurality of room areas, and the room areas indicate rooms in the building; Based on the multiple room areas in the second building structure diagram, a deployment mode of wireless local area network devices in the building is determined.

2. The method according to claim 1, characterized in that The method further comprises: Determine a deployment mode of wireless local area network devices in the building based on the multiple room areas in the second building structure diagram and first category labels corresponding to the multiple room areas, where the first category labels indicate categories of rooms indicated by corresponding room areas.

3. The method according to claim 2, characterized in that The application scenario of the building corresponds to a plurality of second category labels, and the second category labels are used to indicate the category of the rooms in the building in the application scenario; the method further includes: For any room area among the multiple room areas, identifying text information in the room area, and extracting room category information from the identified text information; The second category label having the highest semantic similarity with the room category information among the multiple second category labels is determined as the first category label corresponding to the room area.

4. The method according to claim 2 or 3, characterized in that The application scenario of the building corresponds to a plurality of second category labels, and the second category labels are used to indicate the category of the rooms in the building in the application scenario; the method further includes: For any room area among the multiple room areas, if the first category label corresponding to the room area is not the second category label in the application scenario, determining a target category label from the multiple second category labels based on the first category label corresponding to the room area, where the function of the room under the target category label is similar to the function of the room under the first category label corresponding to the room area; The first category label corresponding to the room area is replaced with the target category label.

5. The method according to any one of claims 1 to 4, characterized in that The second building structure diagram further includes an obstacle area, where the obstacle area indicates obstacles in the building, and the obstacles include walls or load-bearing columns. The method further includes: Based on the multiple room areas and the obstacle areas in the second building structure diagram, a deployment mode of wireless local area network devices in the building is determined.

6. The method according to any one of claims 1 to 5, characterized in that The second building structure diagram further includes an open area, and the open area refers to an area outside the plurality of room areas and within the outline of the building in the second building structure diagram; the method further includes: Based on the multiple room areas and the open area in the second building structure diagram, a deployment mode of wireless local area network devices in the building is determined.

7. The method according to claim 6, characterized in that The open area further includes a seating area, where seats are arranged in the open area. The method further includes: Based on the multiple room areas, the open area, and the at least one seating area in the second building structure diagram, a deployment mode of wireless local area network devices in the building is determined.

8. The method according to any one of claims 1 to 7, characterized in that The room area is a first room area or a second room area, the first room area indicates an independent room in the building, and the second room area indicates a room having sub-rooms in the building; and identifying the first building structure diagram includes: Performing room area recognition on the first building structure diagram to obtain a plurality of first room areas; Based on the overlap of outlines between the multiple first room areas and adjacent room areas, the multiple first room areas in the first building structure diagram are regionally merged to obtain at least one second room area, where the second room area includes two adjacent first room areas.

9. The method according to claim 8, characterized in that The performing region fusion on the multiple first room areas in the first building structure diagram based on the overlap of outlines between the multiple first room areas and adjacent room areas to obtain at least one second room area includes: For any room area among the plurality of first room areas, obtaining a degree of outline overlap between the room area and an adjacent room area, where the adjacent room area refers to a first room area adjacent to the room area, and the degree of outline overlap indicates a degree of overlap between an outline of the room area and an outline of the adjacent room area; If the outline overlap is greater than or equal to an overlap threshold, the room area and the adjacent area are fused on the first building structure diagram to obtain a second room area.

10. The method according to any one of claims 1 to 9, characterized in that The method further comprises: Predicting signal coverage in the building under the deployment mode to obtain a first prediction result; Predicting a network rate of the terminal in the building under the deployment mode to obtain a second prediction result; Based on the first prediction result, the second prediction result and the deployment method, a network planning report for the building is determined, where the network planning report includes a deployment location of the wireless local area network device in the building.

11. The method according to claim 10, characterized in that The method further comprises: If the first prediction result does not achieve the first expected result and / or the second prediction result does not achieve the second expected result, the deployment method is adjusted based on the first prediction result and the second prediction result. The first expected result refers to the expected signal coverage in the building, and the second expected result refers to the expected network rate of the terminal in the building.

12. A network planning device, characterized in that: The device comprises: An acquisition module, configured to acquire a first building structure drawing of a building; an identification module, configured to identify the first building structure diagram to obtain a second building structure diagram, wherein the second building structure diagram includes a plurality of room areas, and the room areas indicate rooms in the building; A determination module is used to determine a deployment mode of wireless local area network devices in the building based on the multiple room areas in the second building structure diagram.

13. The device according to claim 12, characterized in that The determining module is further configured to: Determine a deployment mode of wireless local area network devices in the building based on the multiple room areas in the second building structure diagram and first category labels corresponding to the multiple room areas, where the first category labels indicate categories of rooms indicated by corresponding room areas.

14. The device according to claim 13, characterized in that The application scenario of the building corresponds to a plurality of second category labels, where the second category labels are used to indicate the category of the rooms in the building under the application scenario; and the determining module is further configured to: For any room area among the multiple room areas, identifying text information in the room area, and extracting room category information from the identified text information; The second category label having the highest semantic similarity with the room category information among the multiple second category labels is determined as the first category label corresponding to the room area.

15. The device according to claim 13 or 14, characterized in that The application scenario of the building corresponds to a plurality of second category labels, where the second category labels are used to indicate the category of the rooms in the building under the application scenario; and the determining module is further configured to: For any room area among the multiple room areas, if the first category label corresponding to the room area is not the second category label in the application scenario, determining a target category label from the multiple second category labels based on the first category label corresponding to the room area, where the function of the room under the target category label is similar to the function of the room under the first category label corresponding to the room area; The first category label corresponding to the room area is replaced with the target category label.

16. The device according to any one of claims 12 to 15, characterized in that The second building structure diagram further includes an obstacle area, where the obstacle area indicates obstacles in the building, and the obstacles include walls or load-bearing columns; the determining module is further configured to: Based on the multiple room areas and the obstacle areas in the second building structure diagram, a deployment mode of wireless local area network devices in the building is determined.

17. The device according to any one of claims 12 to 16, characterized in that The second building structure diagram further includes an open area, which refers to an area outside the multiple room areas and within the outline of the building in the second building structure diagram; the determining module is further configured to: Based on the multiple room areas and the open area in the second building structure diagram, a deployment mode of wireless local area network devices in the building is determined.

18. The device according to claim 17, characterized in that The open area also includes a seating area, where seats are arranged in the open area. The determining module is further configured to: Based on the multiple room areas, the open area, and the at least one seating area in the second building structure diagram, a deployment mode of wireless local area network devices in the building is determined.

19. The device according to any one of claims 12 to 18, characterized in that The room area is a first room area or a second room area, the first room area indicates an independent room in the building, and the second room area indicates a room with sub-rooms in the building; The identification module includes: an identification unit, configured to perform room area identification on the first building structure diagram to obtain a plurality of first room areas; A fusion unit is configured to perform regional fusion on the multiple first room areas in the first building structure diagram based on the overlap of the outlines of the multiple first room areas and the adjacent room areas, to obtain at least one second room area, where the second room area includes two adjacent first room areas.

20. The device according to claim 19, characterized in that The fusion unit is used for: For any room area among the plurality of first room areas, obtaining a degree of outline overlap between the room area and an adjacent room area, where the adjacent room area refers to a first room area adjacent to the room area, and the degree of outline overlap indicates a degree of overlap between an outline of the room area and an outline of the adjacent room area; If the outline overlap is greater than or equal to an overlap threshold, the room area and the adjacent area are fused on the first building structure diagram to obtain a second room area.

21. The device according to any one of claims 12 to 20, characterized in that The device further comprises: A first prediction module is used to predict the signal coverage in the building under the deployment mode to obtain a first prediction result; A second prediction module is used to predict the network rate of the terminal in the building under the deployment mode to obtain a second prediction result; The determination module is further used to determine a network planning report for the building based on the first prediction result, the second prediction result and the deployment method, where the network planning report includes the deployment location of the wireless local area network device in the building.

22. The device according to claim 21, characterized in that The determining module is further configured to: If the first prediction result does not achieve the first expected result and / or the second prediction result does not achieve the second expected result, the deployment method is adjusted based on the first prediction result and the second prediction result. The first expected result refers to the expected signal coverage in the building, and the second expected result refers to the expected network rate of the terminal in the building.

23. A computing device cluster, characterized in that: comprising at least one computing device, each computing device including a processor and a memory; The processor of the at least one computing device is configured to execute instructions stored in a memory of the at least one computing device, so that the computing device cluster executes the method according to any one of claims 1 to 11.

24. A computer program product comprising instructions, characterized in that When the instructions are executed by a computing device cluster, the computing device cluster is caused to perform the method according to any one of claims 1 to 11.

25. A computer-readable storage medium, characterized in that The method comprises computer program instructions, which, when executed by a computing device cluster, perform the method according to any one of claims 1 to 11.

Images