Wireless device deployment position optimization method and system based on Wi-Fi signal

Abstract

The invention provides a wireless device deployment position optimization method and system based on Wi-Fi signals, relates to the technical field of wireless communication device deployment and intelligent wireless sensing, and aims to solve the problems that in the prior art, only indoor communication quality and communication safety are considered, a wireless sensing layer is not considered, and the position of a wireless device is not optimized in combination with the surrounding environment. The method comprises the steps of extracting key evaluation parameters of measurement data; constructing a fingerprint matching data set based on position-perception; constructing a two-dimensional quantitative model according to the key evaluation parameters; acquiring fine-grained characteristic parameters by adopting an energy compensation mechanism; and inputting the position factors, the fine-grained characteristic parameters and the position data of the fingerprint matching data set into a spatio-temporal information self-learning model, and determining the optimal position of deployment of the wireless equipment. According to the method, the communication quality and the sensing range of the Wi-Fi signal are combined, so that the position of the wireless equipment in the known indoor environment can be optimized, and the optimal position of the wireless equipment in the unknown environment can be recommended.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless communication equipment deployment and intelligent wireless sensing, and in particular, relates to a method and system for optimizing the deployment position of wireless equipment based on Wi-Fi signals. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] With the application of wireless sensing technology in the fields of smart home and medical care (gesture control of home appliances, fall detection, vital sign estimation, etc.), accurate sensing technology based on indoor wireless signals has received widespread attention from academia and industry. Relying on the perception principle behind wireless signals, the perception ability and perception range of wireless signals are two key indicators for evaluating the maturity of wireless sensing technology. However, the above two key indicators are heavily dependent on the deployment location of existing wireless devices indoors. At present, both the indoor wireless signal detection instruments on the market and the indoor wireless device deployment location optimization methods based on traditional technologies are aimed at indoor communication quality and communication security, and do not consider the wireless perception level.

[0004] Wi-Fi signals have achieved great success in the field of wireless communications, but they face many technical challenges in the field of wireless perception. In the field of wireless perception, due to the time-varying characteristics of Wi-Fi signals themselves, jitter and asynchronization problems within the hardware, and unstable environments, various perception technologies based on Wi-Fi signals cannot maintain high perception accuracy and strong stability in practical applications. According to the results of a survey of published relevant literature, mainstream perception technologies all achieve individual behavior recognition, target object tracking, human vital sign estimation, and complex indoor environment perception by increasing the power of the transmitted signal, reducing noise, weakening the influence of the surrounding environment, and extracting effective characterization parameters. In other words, existing research work mainly achieves high-precision perception effects within a limited range of indoor environments by performing a series of signal processing on the Wi-Fi signal itself, and has not made preliminary explorations in optimizing the location of wireless devices to expand the perception range and improve perception performance.

[0005] Although some research works have quantified the impact of the relative position of transceiver devices on the Wi-Fi perception space / range through the Fresnel zone model theory, and concluded that in a relatively ideal indoor environment, the perception range will be different due to the different distances between transceiver devices; however, no research has been done on how to optimize the position of wireless devices in combination with the indoor environment, improve wireless perception accuracy, and enhance stability. Summary of the invention

[0006] In order to overcome the deficiencies of the above-mentioned prior art, the present invention provides a method and system for optimizing the deployment location of wireless devices based on Wi-Fi signals. Three key evaluation parameters are proposed to describe the propagation and perception characteristics of Wi-Fi signals in indoor environments; the changing rules of Wi-Fi signals in different locations are analyzed, a location-energy-perception Wi-Fi fingerprint matching data set is constructed, and a two-dimensional quantitative model is established; a CNN-LSTM optimization algorithm is designed to achieve adaptive recommendation of the optimal wireless device location in indoor environments.

[0007] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions: A first aspect of the present invention provides a method for optimizing the deployment location of a wireless device based on a Wi-Fi signal, comprising: Collecting measurement data, where the measurement data includes Wi-Fi signal energy and Wi-Fi channel status information in a test environment; extracting key evaluation parameters of the measurement data; Based on key evaluation parameters, a location-aware Wi-Fi fingerprint matching dataset is constructed; The Wi-Fi signal quality of the tested environment is determined based on the key evaluation parameters. If the Wi-Fi signal quality is high, the key interference factors are determined through the constructed interference factor-Wi-Fi data model; the wireless device location factor is determined through the constructed device location-Wi-Fi data model; An energy compensation mechanism is used for Wi-Fi signal energy to obtain fine-grained feature parameters; The wireless device location factor, fine-grained feature parameters and location data of a location-aware Wi-Fi fingerprint matching data set are input into a spatiotemporal information self-learning model to determine the optimal location for wireless device deployment.

[0008] As an implementation method, before the location-aware Wi-Fi fingerprint matching dataset is input into the spatiotemporal information self-learning model, the fingerprint matching dataset is preprocessed to obtain preprocessed location data.

[0009] As an implementation manner, the key evaluation parameters of the measurement data include three types: a signal energy index, a signal perception capability index, and a signal stability index.

[0010] As an implementation mode, three key evaluation parameters of the measurement data are extracted using statistical methods.

[0011] As an implementation method, a location-aware Wi-Fi fingerprint matching dataset is constructed based on key evaluation parameters. Specifically, based on the key evaluation parameters and according to the layout of the indoor space environment, fixed areas and random areas are set to collect corresponding Wi-Fi signal data. The data of each location is constructed into a vector to obtain a location-aware Wi-Fi fingerprint matching dataset.

[0012] As an implementation method, an interference factor-Wi-Fi data model and a device location-Wi-Fi data model are constructed. Specifically, the device location-Wi-Fi data quantization model and the interference factor-Wi-Fi data quantization model are determined by analyzing the Wi-Fi signal energy attenuation and frequency domain distribution law by using a signal reflection and diffraction model.

[0013] As an implementation method, an energy compensation mechanism is used for Wi-Fi signal energy to obtain fine-grained feature parameters. The specific process is as follows: Construct Wi-Fi signal propagation space through the transmitter and receiver; The Wi-Fi signal propagation space produces a multipath effect, which results in energy superposition and perception blind spots. Based on the law of energy superposition and perception blind spots, the Rayleigh and Rice models of Wi-Fi signal propagation are used to compensate Wi-Fi signal energy; Based on the energy compensation mechanism, fine-grained feature parameters are obtained.

[0014] As an implementation method, the wireless device location factor, fine-grained feature parameters, and location data of a location-aware Wi-Fi fingerprint matching data set are input into a spatiotemporal information self-learning model to determine the optimal location for wireless device deployment, wherein the spatiotemporal information self-learning model includes a convolutional neural network and a long short-term memory network, and the specific process is as follows: Use convolutional neural networks to extract spatial information from input data; Use long short-term memory networks to extract the temporal information of input data; Based on the spatial and temporal information of the input data, the optimal location for wireless device deployment is determined.

[0015] A second aspect of the present invention provides a wireless device deployment location optimization system based on Wi-Fi signals, comprising: A data acquisition module, used to collect measurement data, wherein the measurement data includes Wi-Fi signal energy and Wi-Fi channel status information under the test environment; A parameter extraction module, used to extract key evaluation parameters of the measurement data; A dataset construction module, used to construct a location-aware Wi-Fi fingerprint matching dataset based on key evaluation parameters; A quantitative model module is constructed to determine the Wi-Fi signal quality of the tested environment based on key evaluation parameters. If the Wi-Fi signal quality is high, the key interference factor is determined through the constructed interference factor-Wi-Fi data model; the wireless device location factor is determined through the constructed device location-Wi-Fi data model; An energy compensation module is used to adopt an energy compensation mechanism for Wi-Fi signal energy and obtain fine-grained feature parameters; The location optimization module is used to input the wireless device location factor, fine-grained feature parameters and location data of the location-aware Wi-Fi fingerprint matching data set into the spatiotemporal information self-learning model to determine the optimal location for wireless device deployment.

[0016] As an implementation mode, the data acquisition module includes a wireless router and a data acquisition device, wherein the wireless router is used to transmit Wi-Fi signals; and the data acquisition device is used to obtain Wi-Fi signal energy and Wi-Fi channel status information.

[0017] One or more of the above technical solutions have the following beneficial effects: In this embodiment, by constructing a location-energy-aware Wi-Fi fingerprint matching data set, establishing an indoor environment-WiFi signal quantization model, and using a deep learning algorithm, the communication quality of the Wi-Fi signal and the Wi-Fi signal perception range are coordinated to optimize the deployment location of wireless devices in the indoor environment, and the optimal recommended location can be quickly obtained.

[0018] In this embodiment, the original data for evaluating the indoor Wi-Fi signal quality and Wi-Fi sensing range can be obtained by directly utilizing the existing wireless devices in the indoor environment without reinstalling or adding other additional hardware, thereby saving the use of hardware resources and lowering the power consumption cost.

[0019] In this embodiment, by obtaining coarse-grained RSSI and fine-grained CSI measurement data, not only can the quality of the wireless signal in the current indoor environment be evaluated, but also the human behavior can be perceived within the formed coverage perception range. Through the proposed evaluation parameters and the related quantitative models and optimization algorithms, we can optimize the location of the wireless device according to the different indoor environments, providing users with a good communication environment and a highly accurate perception area.

[0020] Advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0022] Figure 1 This is a framework diagram of a method for optimizing the deployment location of wireless devices based on Wi-Fi signals according to the first embodiment of the present invention; Figure 2 This is a flow chart of a method for optimizing the deployment location of wireless devices based on Wi-Fi signals according to the first embodiment of the present invention. DETAILED DESCRIPTION

[0023] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0024] It should be noted that the terms used herein are for describing specific embodiments only and are not intended to be limiting of exemplary embodiments according to the present invention.

[0025] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.

[0026] Embodiment 1 This embodiment discloses a method for optimizing the deployment location of wireless devices based on Wi-Fi signals.

[0027] In order to more clearly illustrate this embodiment, a process for optimizing the deployment location of wireless devices based on Wi-Fi signals can be specifically described as follows: A method for optimizing the deployment location of wireless devices based on Wi-Fi signals, comprising: S1. Collecting measurement data, where the measurement data includes Wi-Fi signal energy and Wi-Fi channel status information in a test environment; S2, extracting key evaluation parameters of the measurement data; S3, based on key evaluation parameters, build a location-aware Wi-Fi fingerprint matching dataset; S4. Determine the Wi-Fi signal quality of the tested environment based on the key evaluation parameters. If the Wi-Fi signal quality is high, determine the key interference factor through the constructed interference factor-Wi-Fi data model; and determine the wireless device location factor through the constructed device location-Wi-Fi data model. S5. Using energy compensation mechanism for Wi-Fi signal energy to obtain fine-grained feature parameters; S6. Input the wireless device location factor, fine-grained feature parameters, and location data of a location-aware Wi-Fi fingerprint matching data set into a spatiotemporal information self-learning model to determine an optimal location for wireless device deployment.

[0028] like Figure 1 , Figure 2 As shown, in step S1, measurement data is collected, and the measurement data includes Wi-Fi signal energy and Wi-Fi channel status information in a test environment.

[0029] S1-1. Collect measurement data.

[0030] In this embodiment, a measurement platform tool based on CSI-TOOL is built to measure the Wi-Fi signal energy RSSI and physical layer information CSI in an indoor environment as raw data.

[0031] In this embodiment, a wireless router deployed indoors is used as an AP, which works in the 2.4GHz and 5GHz frequency bands. The terminal device can be a common laptop or a mobile phone. The terminal device sends a uniform detection frame to the AP through the ping command; after receiving the detection frame, the AP feeds back a response message; the terminal device saves the received response message as the measured Wi-Fi raw data. The packet rate is set to 30bps, and data is collected continuously for 1 minute.

[0032] Among them, the wireless router is used as a wireless device to transmit the required Wi-Fi signal; a microcontroller or notebook with a network card inserted is used as a data acquisition device, and the data acquisition device and the wireless router belong to the same local area network.

[0033] In this embodiment, the acquisition device is a laptop computer with a built-in Intel 5300 network card, and the goal is to popularize this method to ordinary terminal devices.

[0034] S1-2. Preprocess the collected measurement data.

[0035] In this embodiment, the collected Wi-Fi signals are simply pre-processed to detect and remove abnormal data and denoise the signal data.

[0036] Specifically, abnormal data processing includes two stages: (1) Detecting abnormal data According to the transmission rate of the transmitter in the environment, the length of each time segment is determined, and the amplitude mean and standard deviation of the channel state information corresponding to the time segment are solved; the threshold range under the time segment is set to 1.2 to 1.5 times the amplitude mean. Combined with the standard deviation, the specific threshold is determined.

[0037] Secondly, if the amplitude corresponding to a data point in each time series segment is greater than the set threshold, it is considered to be abnormal data; if it is less than the threshold, it is considered to be normal data.

[0038] (2) Abnormal data processing For data points identified as abnormal data, the position of the data point in the time series segment is extracted. The data point is removed or the original amplitude is replaced with the average amplitude of the four data points adjacent to the data point. The amplitude mentioned here is the signal energy.

[0039] Signal data denoising, specific denoising method: After abnormal data processing, use a simple Butterworth low-pass filtering technique to remove high-frequency noise data and retain the low-frequency real perception data.

[0040] like Figure 2 As shown, in step S2, key evaluation parameters of the measurement data are extracted.

[0041] The statistical analysis method is used to analyze the preprocessed Wi-Fi data, and the three key evaluation parameters, signal energy index, signal perception ability index and signal stability index, are extracted by combining statistical parameters, and the effective value range of the corresponding parameters is given.

[0042] Specifically, coarse-grained information RSSI and fine-grained information CSI are extracted from the measured original Wi-Fi data frame. Statistical methods and signal energy attenuation models are used to extract three key evaluation parameters: signal energy index, signal perception capability index, and signal stability index.

[0043] The value range of the signal energy index is directly related to the room area and the gain of the wireless device; the signal perception ability is closely related to the signal energy and the surrounding environment; the signal stability is determined by the distribution of the signal energy mean value, the number of empty packets, and the packet loss rate in a specific time period. After a period of measurement, the corresponding value ranges of the three evaluation parameters can be given.

[0044] (1) The signal energy index comprehensively evaluates the Wi-Fi signal quality based on the size, distribution, and difference of Wi-Fi signal energy. Indicates the standard deviation of the signal energy (RSSI) under the test environment, Represents the average value of signal energy (RSSI) under the test environment).

[0045] (2) The signal perception capability index is determined based on the signal energy index and noise information.

[0046] (3) The signal stability index is , in, Indicates an invalid data packet. It represents the standard deviation of the signal (RSSI and CSI) under the test environment, N represents the number of empty packets and packet loss, and M represents the number of data packets received in a certain period of time.

[0047] After the above steps, three key evaluation parameters are extracted, which can pre-understand the state of the indoor environment and the characteristics of the Wi-Fi signal data in the indoor environment. Through preliminary measurement and analysis of the Wi-Fi signal in the current indoor environment, there is sufficient scientific evidence to show that the key factors affecting the measured signal data come from the indoor environment, rather than sudden failures of the wireless device itself or human factors. Using the measured signal data, a more reliable and realistic location-signal energy-perception data set and subsequent related processing can be established.

[0048] like Figure 2 As shown, in step S3, a location-aware Wi-Fi fingerprint matching dataset is constructed based on key evaluation parameters.

[0049] S3-1. Divide the measurement locations based on the Wi-Fi energy distribution in the indoor environment.

[0050] In this embodiment, according to the key evaluation parameters given in step 2, two forms of Wi-Fi location data collection are set: fixed area and random area; the collected data types include Wi-Fi signal energy and Wi-Fi channel status information.

[0051] (1) A fixed area is set near the LOS path and divided into 1*1m grids. The corresponding Wi-Fi measurement data is collected and the data of each location is constructed into a vector E_Signal{L(x, y),RSSI,CSI}.

[0052] (2) The random area is set in a crowded area. Instead of requiring a fixed-size grid, a fuzzy area is randomly selected for location data collection.

[0053] S3-2. Construct a location-aware Wi-Fi fingerprint matching dataset.

[0054] In this embodiment, 5 samples are collected at each location, each sample lasts for 60 seconds, and a location-aware Wi-Fi fingerprint matching data set is constructed.

[0055] The data packet rate is 1000 Hz, and the acquisition time of 60 seconds corresponds to 60,000 data frames. RSSI and CSI are extracted from each data frame. The RSSI extracted from each data frame is a real number; the extracted CSI is a three-dimensional matrix in complex form, that is, 1*3*30 (the first dimension represents a transmitting antenna; the second dimension represents three receiving antennas; the third dimension represents 30 subcarriers). In other words, each selected position in the indoor environment corresponds to a time series of RSSI and CSI.

[0056] A sample of RSSI is shown below: ; in, Indicates the first RSSI values, Indicates the length of a sample. A sample of CSI can be expressed as:

[0057] in, Indicates the first receiving antenna in a sample. CSI values, Indicates the length of a sample.

[0058] After the above steps, a location-aware Wi-Fi fingerprint matching dataset is obtained, which provides effective training data for the subsequent CNN-LSTM spatiotemporal information self-learning model, thereby making the model more effective.

[0059] like Figure 2 As shown, in step S4, the Wi-Fi signal quality of the tested environment is determined according to the key evaluation parameters. If the Wi-Fi signal quality is high, the key interference factor is determined by the constructed interference factor-Wi-Fi data model; and the wireless device location factor is determined by the constructed device location-Wi-Fi data model.

[0060] Specifically, the signal perception function depends on the signal quality, which is measured by the signal energy index Determine the Wi-Fi signal quality.

[0061] The signal perception function is conditioned.

[0062] A two-dimensional quantitative model, namely the interference factor-Wi-Fi data model and the device location-Wi-Fi data model, is established. The device location-Wi-Fi data quantitative model and the interference factor-Wi-Fi data quantitative model are determined by analyzing the Wi-Fi signal energy attenuation and frequency domain distribution law using the signal reflection and diffraction model.

[0063] Among them, the location-WiFi data quantification model analyzes the reflection model and diffraction model in the indoor environment; the interference factor-WiFi data quantification model is mainly constructed based on the reflection model reasoning of the signal in the indoor environment.

[0064] (1) Interference Factors - Wi-Fi Data Quantification Model Through the analysis of a large amount of sample data, key interference factors such as moving objects and metal objects are determined. For the interference factors of moving objects, the attenuation and phase difference information of Wi-Fi signal energy are quantified through the reflection model theory; for the interference factors of static objects, a mathematical relationship model is constructed through the relationship between the distribution of Wi-Fi signal energy changes and the reflection area of ​​static objects.

[0065] Specifically, the interference factor-WiFi data quantification model is essentially to build a quantitative model to explain how static objects in indoor environments affect Wi-Fi signal changes. This quantification is mainly based on the reflection model theory to explain and analyze the impact of static objects on the reflection path length, thereby changing the change of Wi-Fi signal energy in its time-frequency domain and the change of phase information.

[0066] (2) The location-WiFi data quantization model analyzes the measurement data through the signal distance attenuation theory to determine the law of signal energy attenuation under different distance ranges.

[0067] Specifically, the location-WiFi data quantification model is essentially to build a quantitative model to explain that each location in the indoor environment has unique characteristics at the Wi-Fi signal level. The relationship between the quantified location and WiFi signal changes is specifically: 1) The location is within the line-of-sight path of the transceiver (the line-of-sight path is LOS). The measured Wi-Fi signal data is mainly characterized by changes in which the diffraction model is dominant and the reflection model is auxiliary.

[0068] 2) The location is not within the line-of-sight path of the transceiver. The measured Wi-Fi signal data is mainly formed by the reflection model and the signal change characteristics are formed.

[0069] After the above steps, the two-dimensional quantitative model formed by the location-WiFi data quantitative model and the interference factor-WiFi data quantitative model can accurately describe the impact of the indoor environment on the propagation of Wi-Fi signals. That is, it fully combines the surrounding environment and paves the way for the subsequent optimization of the wireless device location.

[0070] like Figure 2 As shown, in step S5, an energy compensation mechanism is used for Wi-Fi signal energy to obtain fine-grained feature parameters.

[0071] (1) Construct a Wi-Fi signal propagation space through the transmitter and the receiver.

[0072] (2) The energy superposition law and perception blind spot caused by the multipath effect phenomenon generated by the signal propagation space constructed by the transmitter and the receiver.

[0073] (3) Based on the law of energy superposition and perception blind spots, the Rayleigh and Rice model of Wi-Fi signal propagation is adopted, and an energy compensation machine is used for Wi-Fi signal energy.

[0074] Considering the Rayleigh attenuation and Rice attenuation phenomena in the process of indoor wireless signal propagation, a Wi-Fi signal energy compensation mechanism is designed to ensure the accuracy of time-frequency domain information and achieve the purpose of expanding the Wi-Fi perception space / range.

[0075] Among them, indoor signal propagation will cause signal energy attenuation. The attenuation types include Rayleigh attenuation and Rice attenuation.

[0076] 1) Rayleigh attenuation means that there is no line-of-sight path in the indoor environment, only multiple reflected non-line-of-sight paths.

[0077] 2) Rice attenuation refers to the indoor environment consisting of a line-of-sight path and multiple reflected non-line-of-sight paths.

[0078] Specifically, the Wi-Fi signal energy compensation mechanism is mainly aimed at the Rice attenuation model existing in indoor environments. The Rayleigh model is only used as a reference to analyze the differences in the signal degradation patterns generated by line-of-sight paths and non-line-of-sight paths in indoor environments. This is because Rice attenuation includes the attenuation caused by the increase in propagation distance on the line-of-sight path, as well as the attenuation caused by the non-line-of-sight path. Combined with Rayleigh attenuation, the energy attenuation characteristics of the non-line-of-sight path can be used to extract the signal information corresponding to the line-of-sight path.

[0079] Then, based on the Rice attenuation characteristics and diffraction model theory, the starting point of attenuation in the time series data can be accurately defined, thereby performing energy recovery or energy compensation.

[0080] like Figure 2 As shown, in step S6, the wireless device location factor, fine-grained feature parameters and location data of the location-aware Wi-Fi fingerprint matching data set are input into the spatiotemporal information self-learning model to determine the optimal location for wireless device deployment.

[0081] In this embodiment, the spatiotemporal information self-learning model includes a convolutional neural network and a long short-term memory network.

[0082] In an unknown indoor environment, a short training period is required; in a known indoor environment, the fingerprint data set built in the early stage can detect changes in the indoor environment, and training is also required in this case to optimize the device deployment location. Setting a scheduled training self-learning process can adapt to the indoor environment to adjust the trained model.

[0083] The location-aware Wi-Fi fingerprint matching dataset is used to train the spatiotemporal information self-learning model CNN-LSTM.

[0084] Specifically, the initial input data of the CNN module: the location data in the fingerprint dataset is preprocessed as one of the initial inputs; the location-aware Wi-Fi data quantization model obtains the location factor as the second initial input; the Wi-Fi signal energy compensation mechanism is mainly aimed at the energy loss generated when the object position is in the line-of-sight path of the transceiver device; the energy compensation mechanism can make up for the lost energy and obtain more accurate and fine-grained feature parameters (frequency information and phase information) as the third initial input.

[0085] Initial input data of the LSTM module: After preprocessing the location data (CSI) in the fingerprint dataset, the best 5 subcarriers are selected from 30 subcarriers as the initial input of the LSTM. Subcarrier optimization mechanism: Location-aware Wi-Fi data quantization model.

[0086] (1) Use convolutional neural networks to extract spatial information from input data.

[0087] The preprocessed position data, position factors and fine-grained feature parameters are input into the input of the convolutional neural network (CNN) to obtain spatial representation information.

[0088] (2) Use long short-term memory networks to extract the temporal information of input data.

[0089] The optimal five subcarriers are selected and input into the long short-term memory network LSTM to extract the representation information in the time domain.

[0090] (3) Determine the optimal location for wireless device deployment based on the spatial and temporal information of the input data.

[0091] Based on the characterization information at the spatial level and the characterization information in the time domain, the optimal recommended location for deploying wireless devices that expand the perception range while ensuring communication quality is determined.

[0092] In this embodiment, the energy distribution of Wi-Fi signals formed by the wireless routers deployed in the room and the change law of Wi-Fi CSI are analyzed to extract key evaluation parameters and determine key interference factors; the two-dimensional quantitative models of position-WiFi signal and interference factor-WiFi signal are quantified and constructed through the reflection model and diffraction model of the signal; the energy superposition law and perception blind area caused by the multipath effect phenomenon generated in the signal propagation space constructed by the wireless router and the receiving device are used, and the Rayleigh (non-line-of-sight path) and Rice models (line-of-sight path) of indoor wireless signal propagation are combined to design and implement a Wi-Fi signal energy compensation mechanism to ensure the accuracy of time-frequency domain information, so as to achieve the purpose of expanding the Wi-Fi perception space / range; finally, based on the CNN-LSTM spatiotemporal information self-learning model, the optimal area or location for wireless device deployment is determined.

[0093] Embodiment 2 The purpose of this embodiment is to provide a wireless device deployment location optimization system based on Wi-Fi signals, including: A data acquisition module, used to collect measurement data, wherein the measurement data includes Wi-Fi signal energy and Wi-Fi channel status information under the test environment; A parameter extraction module, used to extract key evaluation parameters of the measurement data; A dataset construction module, used to construct a location-aware Wi-Fi fingerprint matching dataset based on key evaluation parameters; A quantitative model module is constructed to determine the Wi-Fi signal quality of the tested environment based on key evaluation parameters. If the Wi-Fi signal quality is high, the key interference factor is determined through the constructed interference factor-Wi-Fi data model; the wireless device location factor is determined through the constructed device location-Wi-Fi data model; An energy compensation module is used to adopt an energy compensation mechanism for Wi-Fi signal energy and obtain fine-grained feature parameters; The location optimization module is used to input the wireless device location factor, fine-grained feature parameters and location data of the location-aware Wi-Fi fingerprint matching data set into the spatiotemporal information self-learning model to determine the optimal location for wireless device deployment.

[0094] Based on providing a wireless device deployment location optimization system based on Wi-Fi signals, the method steps in embodiment 1 are implemented.

[0095] In this embodiment, the data acquisition module includes a wireless router and a data acquisition device, etc. The wireless router is used as a wireless device to transmit the required Wi-Fi signal and can be used as a transmitter; the data acquisition device is generally a microcontroller or a laptop computer with a network card inserted. The data acquisition device and the wireless router belong to the same local area network and can be used as a receiver.

[0096] In the monitor mode, the data acquisition device can obtain the received signal strength indicator (RSSI) of the Wi-Fi signal and the channel state information (CSI) of the physical layer. At the acquisition device end, it is currently a laptop with a built-in Intel 5300 network card. The goal is to popularize this method to ordinary terminal devices.

[0097] In this embodiment, a signal preprocessing module is also included, which is used to preprocess the collected measurement data to obtain preprocessed measurement data.

[0098] Specifically, the collected Wi-Fi signals are simply preprocessed to detect and remove abnormal data and denoise the signal data.

[0099] In this embodiment, the signal preprocessing module, the quantization model building module and the position optimization module are implemented through Matlab+python language hybrid programming.

[0100] The steps involved in the apparatus of the above embodiment correspond to the method embodiment 1, and the specific implementation method can refer to the relevant description part of embodiment 1. The term "computer-readable storage medium" should be understood as a single medium or multiple media including one or more instruction sets; it should also be understood to include any medium that can store, encode or carry an instruction set for execution by a processor and enable the processor to execute any method in the present invention.

[0101] Those skilled in the art should understand that the modules or steps of the present invention described above can be implemented by a general-purpose computer device, or alternatively, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. The present invention is not limited to any specific combination of hardware and software.

[0102] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without creative work are still within the scope of protection of the present invention.

Claims

1. A method for optimizing the deployment location of wireless devices based on Wi-Fi signals, characterized in that: include: Collecting measurement data, where the measurement data includes Wi-Fi signal energy and Wi-Fi channel status information in a test environment; extracting key evaluation parameters of the measurement data; Based on key evaluation parameters, a location-aware Wi-Fi fingerprint matching dataset is constructed; The Wi-Fi signal quality of the tested environment is determined based on the key evaluation parameters. If the Wi-Fi signal quality is high, the key interference factors are determined through the constructed interference factor-Wi-Fi data model; the wireless device location factor is determined through the constructed device location-Wi-Fi data model; An energy compensation mechanism is used for Wi-Fi signal energy to obtain fine-grained feature parameters; The wireless device location factor, fine-grained feature parameters and location data of a location-aware Wi-Fi fingerprint matching data set are input into a spatiotemporal information self-learning model to determine the optimal location for wireless device deployment.

2. A method for optimizing the deployment location of wireless devices based on Wi-Fi signals as claimed in claim 1, characterized in that: Before inputting the location-aware Wi-Fi fingerprint matching dataset into the spatiotemporal information self-learning model, the fingerprint matching dataset is preprocessed to obtain the preprocessed location data.

3. A method for optimizing the deployment location of wireless devices based on Wi-Fi signals as claimed in claim 1, characterized in that: The key evaluation parameters of the measurement data include three types: signal energy index, signal perception capability index and signal stability index.

4. The method for optimizing the deployment location of wireless devices based on Wi-Fi signals according to claim 1, characterized in that: Using statistical methods, three key evaluation parameters of the measurement data are extracted.

5. The method for optimizing the deployment location of wireless devices based on Wi-Fi signals according to claim 1, characterized in that: Based on the key evaluation parameters, a location-aware Wi-Fi fingerprint matching dataset is constructed. Specifically, based on the key evaluation parameters and according to the layout of the indoor space environment, fixed areas and random areas are set to collect corresponding Wi-Fi signal data. The data of each location is constructed into a vector to obtain a location-aware Wi-Fi fingerprint matching dataset.

6. The method for optimizing the deployment location of wireless devices based on Wi-Fi signals according to claim 1, characterized in that: Construct an interference factor-Wi-Fi data model and a device location-Wi-Fi data model. Specifically, determine the device location-Wi-Fi data quantification model and the interference factor-Wi-Fi data quantification model by analyzing the Wi-Fi signal energy attenuation and frequency domain distribution law using signal reflection and diffraction models.

7. The method for optimizing the deployment location of wireless devices based on Wi-Fi signals according to claim 1, characterized in that: An energy compensation mechanism is used for Wi-Fi signal energy to obtain fine-grained feature parameters. The specific process is as follows: Construct Wi-Fi signal propagation space through the transmitter and receiver; The Wi-Fi signal propagation space produces a multipath effect, which results in energy superposition and perception blind spots. Based on the law of energy superposition and perception blind spots, the Rayleigh and Rice models of Wi-Fi signal propagation are used to compensate Wi-Fi signal energy; Based on the energy compensation mechanism, fine-grained feature parameters are obtained.

8. The method for optimizing the deployment location of wireless devices based on Wi-Fi signals according to claim 1, characterized in that: The wireless device location factor, fine-grained feature parameters, and location data of the location-aware Wi-Fi fingerprint matching data set are input into the spatiotemporal information self-learning model to determine the optimal location for wireless device deployment, wherein the spatiotemporal information self-learning model includes a convolutional neural network and a long short-term memory network, and the specific process is as follows: Use convolutional neural networks to extract spatial information from input data; Use long short-term memory networks to extract the temporal information of input data; Based on the spatial and temporal information of the input data, the optimal location for wireless device deployment is determined.

9. A wireless device deployment location optimization system based on Wi-Fi signals, characterized in that: include: A data acquisition module, used to collect measurement data, wherein the measurement data includes Wi-Fi signal energy and Wi-Fi channel status information under the test environment; A parameter extraction module, used to extract key evaluation parameters of the measurement data; A dataset construction module, used to construct a location-aware Wi-Fi fingerprint matching dataset based on key evaluation parameters; A quantitative model module is constructed to determine the Wi-Fi signal quality of the tested environment based on key evaluation parameters. If the Wi-Fi signal quality is high, the key interference factors are determined through the constructed interference factor-Wi-Fi data model. Determine the wireless device location factor through the constructed device location-Wi-Fi data model; An energy compensation module is used to adopt an energy compensation mechanism for Wi-Fi signal energy and obtain fine-grained feature parameters; The location optimization module is used to input the wireless device location factor, fine-grained feature parameters and location data of the location-aware Wi-Fi fingerprint matching data set into the spatiotemporal information self-learning model to determine the optimal location for wireless device deployment.

10. A wireless device deployment location optimization system based on Wi-Fi signals as claimed in claim 9, characterized in that: The data acquisition module includes a wireless router and a data acquisition device, wherein the wireless router is used to transmit Wi-Fi signals; the data acquisition device is used to obtain Wi-Fi signal energy and Wi-Fi channel status information.

Images