A method and apparatus for sampling test data for calibration of an indoor radio propagation model

Abstract

The application provides a kind of indoor wireless propagation model correction test data sampling method and device, it is related to indoor wireless communication technical field.The method is by spatial domain division eigen length grid, time domain is according to frequency band specific frequency sampling, eliminate fast fading in combination with both, specifically includes: determining to be centered with signal source, 2R is the sampling area of radius, according to spatial domain eigen length 2L division grid;In each grid center point, according to time domain sampling frequency sampling, each grid sampling is not less than 50 data;Processing sampling data, eliminating exception and taking average to obtain wireless signal strength data.The device corresponds to the functional module of the above method.The application proposes the sampling method of combination of spatial domain and time domain, breaks through the limitation of only relying on spatial sampling of Li's theorem, solves the problem of spatial domain sampling difficulty and insufficient sampling points caused by limited indoor space, improves the accuracy and reliability of indoor propagation model correction.

Description

Technical Field

[0001] This application relates to the field of wireless communication technology, and more specifically, to a method and apparatus for sampling test data for indoor wireless propagation model correction. Background Technology

[0002] With the rapid development of 5G-A vertical industries and the widespread application of passive IoT technology, the demand for indoor services is increasing, driving the improvement of indoor planning accuracy and making indoor propagation model calibration a key task. The main goal of propagation model calibration is to correct the theoretical propagation model using measured data, so that it can more accurately reflect the propagation characteristics of signals in a specific environment, especially the median propagation loss.

[0003] In propagation model calibration, fast fading causes drastic instantaneous fluctuations in signal strength, masking the median signal decay trend. If this is not eliminated, the calibrated model will be inaccurate. Currently, indoor wireless propagation model calibration methods mainly follow those used for outdoor macro base stations, adhering to "Li's Theorem": when the number of sampling points is between 30 and 50, and they are distributed within an interval of 40 wavelengths, slow fading characteristics can be effectively distinguished and preserved, while the effects of fast fading are eliminated.

[0004] However, "Li's Theorem" was proposed based on outdoor test data from over 50 years ago, using the 800MHz band. Today, with rapid urbanization, indoor and outdoor wireless environments differ significantly. For indoor coverage, taking common frequency bands as examples, the average coverage radius for the 700MHz band is approximately 50 meters, and for the 2600MHz band, it's approximately 20 meters. The 40 wavelengths account for approximately 31% and 23% of their average coverage radii, respectively. Dividing the grid according to 40 wavelengths would result in insufficient sampling points for indoor propagation model calibration test data, compromising calibration accuracy. Specifically, for the 700MHz band, 40 wavelengths cover 15.6 meters, representing 31% of its average indoor coverage radius of 50 meters; for the 2600MHz band, 40 wavelengths cover 4.6 meters, representing 23% of its average indoor coverage radius of 20 meters. For outdoor macro base stations, the average coverage radius of the 700MHz band is about 400 meters, and the 40 wavelengths account for only about 4%; the average coverage radius of the 2600MHz band is about 240 meters, and the 40 wavelengths account for only about 2%, ensuring the number of outdoor sampling points.

[0005] Furthermore, in indoor environments, fast fading is mainly caused by multipath effects such as reflection, diffraction, and scattering from obstacles like walls and furniture. Its spatial characteristics differ from those outdoors, rendering Lee's theorem inapplicable. Therefore, there is an urgent need for a wireless propagation model calibration test data sampling method and device suitable for indoor environments to address these issues. Summary of the Invention

[0006] To achieve the above objectives, this application provides a method and apparatus for sampling test data for indoor wireless propagation model correction, in order to solve the problems of difficulty in eliminating the effects of fast fading and insufficient sampling points in indoor propagation model correction.

[0007] This invention proposes a method for sampling test data for indoor wireless propagation model calibration, comprising the following steps: The indoor signal source coverage radius is defined as R. The area with a radius of 2R centered on the signal source is used as the test data sampling area for indoor wireless propagation model correction. Within the sampling area, rectangular grids are divided with the spatial domain intrinsic length 2L as the side length; Test data is sampled at the center point of each rectangular grid according to the time-domain sampling frequency. The sampled data in each grid is processed, and outliers are removed. The sampled data in each grid is then averaged to obtain the wireless signal strength data for each grid.

[0008] Optionally, the spatial domain intrinsic length 2L is determined according to different frequency bands. The process of determining the spatial domain intrinsic length 2L includes: comparing the standard deviation σ of the sample mean of test data under different spatial lengths for different frequency bands. When the standard deviation σ drops to 1dB, the corresponding spatial length is determined as the spatial domain intrinsic length 2L of that frequency band.

[0009] Optionally, the number of sampling points N in each grid is not less than 50.

[0010] Optionally, the time-domain sampling frequency is determined according to different frequency bands. The process of determining the time-domain sampling frequency includes: determining the coherence time Tc of different frequency bands in a static state, with the value of the coherence time Tc set between 0.2μs and 0.5μs; and setting the sampling time length. When the number of sampling points N in each grid is 50, according to the formula The time-domain sampling frequencies of different frequency bands were calculated. .

[0011] Optionally, when processing the sampled data in each grid, the 3σ criterion is used to remove outlier data. The process of removing outlier data includes: when the absolute value of the difference between a sampled data and the mean of all sampled data in the grid is greater than 3 times the standard deviation, the data is determined to be outlier data and removed.

[0012] This invention also proposes an indoor wireless propagation model calibration test data sampling device, comprising: The grid division module is used to determine the indoor signal source coverage radius as R, and the area with the signal source as the center and a radius of 2R as the sampling area for indoor wireless propagation model correction; within the sampling area, a rectangular grid is divided with a spatial domain intrinsic length of 2L as the side length; The sampling module is used to sample test data at the center point within each rectangular grid according to the time-domain sampling frequency. The data processing module processes the sampled data in each grid, removes abnormal data, and averages the sampled data in each grid to obtain the wireless signal strength data for each grid.

[0013] Optionally, the device further includes an intrinsic length determination module, which is used to determine the spatial length as the intrinsic length 2L of the spatial domain when the standard deviation σ of the sample mean of test data in different frequency bands drops to 1dB by comparing the standard deviation σ of the sample mean of test data in different spatial lengths.

[0014] Optionally, the device further includes a sampling frequency determination module, which includes: a coherence time determination unit for determining the coherence time Tc of different frequency bands in a static state, wherein the value of the coherence time Tc is set between 0.2 μs and 0.5 μs; and a frequency calculation unit for setting the sampling time length. When the number of sampling points N in each grid is 50, according to the formula The sampling frequencies of different frequency bands were calculated. .

[0015] Optionally, the data processing module uses the 3σ criterion when removing abnormal data. That is, when the absolute value of the difference between a sampled data and the mean of all sampled data in the grid is greater than 3 times the standard deviation, the data is determined to be abnormal data and removed.

[0016] The present invention also proposes an electronic device, comprising: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the above method.

[0017] The beneficial effects of this invention are as follows: A sampling method combining spatial and temporal domains is proposed, which overcomes the limitation of Li's theorem relying solely on spatial sampling. It solves the problems of spatial domain sampling difficulties and insufficient sampling points caused by limited indoor space, thereby improving the accuracy and reliability of indoor propagation model correction. Attached Figure Description

[0018] Figure 1 This is a flowchart of a method for sampling test data for indoor wireless propagation model calibration according to the present invention; Figure 2 This is a schematic diagram of the sampling area and grid division of the present invention; Figure 3 This is a structural block diagram of an indoor wireless propagation model calibration test data sampling device according to the present invention; Figure 4This is a schematic diagram of an electronic device according to the present invention. Detailed Implementation

[0019] Signal fading due to multipath propagation and other factors results in two parts in the received signal: long-term fading and short-term fading. Within a certain range, long-term fading variations can be ignored and can be considered as a local mean of the wireless signal. To achieve propagation model correction, methods are needed to eliminate the effects of fast fading, thereby extracting the median propagation loss characteristics of the signal.

[0020] There are two main methods to eliminate fast fading: The first is spatial domain averaging, which involves evenly distributing multiple measurement points within a certain area and averaging the signal strength at these points. This method can eliminate the effects of fast fading caused by local environmental differences and extract the average propagation loss characteristics within that area. The second method is time domain averaging, which involves measuring the signal strength at the same location multiple times and taking the time average. Since fast fading is rapidly changing, time averaging can smooth out these instantaneous fluctuations and extract the median characteristics of the signal.

[0021] 1. Spatial Domain Test Data Sampling If measured in the spatial domain, the received signal can be expressed as:

[0022] Spatial location The received signal strength (or power) at a certain point is the superposition of long-term fading and short-term fading; Spatial location variables (such as distance, two-dimensional / three-dimensional coordinates, etc., describing the signal's location in space); Long-term fading in the spatial domain reflects large-scale signal changes (such as slow changes in path loss with distance and signal fluctuations caused by shadowing effects), with long periods and large ranges of change. Short-term fading in the spatial domain reflects small-scale signal changes (such as Rayleigh / Rice fading caused by multipath effects), with rapid fluctuations in signal strength over short distances / times (short change period and small range).

[0023] Points along the spatial axis The expression for the local mean estimate is:

[0024] Space point Local mean estimate at (i.e., for) (Averaging the signal in the nearby area in an attempt to "extract" local features of long-term fading); A specific point on a spatial axis (such as a test location or reference point); The "half-length" of a local interval (i.e., the total length of the interval is...) (Controls the size of the "local area"); Integration interval :right Received signal within a spatial region of 2L in the vicinity The purpose of integral averaging is to filter out the rapid fluctuations of short-term decline through "local smoothing" while preserving the slow trend of long-term decline.

[0025] If in By selecting an appropriate time length L nearby, the estimated value of the local mean equals the actual value of the local mean, i.e. Then the following relationship holds:

[0026] Short-term fading is caused by rapid changes in signal strength due to multipath effects; these changes occur within a short time or distance. The integral average within a local interval approaches 0. This is due to long-term fading. It is "slowly changing" within a local range. The internal approximation is a constant (i.e., the actual local mean). ; and short-term decline It is "rapidly changing" (with dramatic positive and negative fluctuations). If the interval L is chosen reasonably, its integral average will approach 0 due to the "cancellation of positive and negative values"—at this point, the local mean estimate... It is equal to the actual local mean. .

[0027] In actual testing, discrete sampling (rather than continuous integration) is used, combined with Li's theorem and engineering test conclusions (sampling points). When the difference between the test value and the actual local mean is <1dB, we have:

[0028] Discrete sampling points (in) Nearby local area Inside).

[0029] The number of sampling points must meet the following requirements. —At this point, the sum of the sampled values ​​of the short-term fading approaches 0 (due to the "positive and negative cancellation" of the fast fluctuations). Discrete summation can approximate continuous integration, ensuring the accuracy of local mean estimation.

[0030] By employing a "local averaging" operation (integration or discrete sampling summation), and leveraging the characteristics of short-term fading—namely, "rapid changes and easy cancellation"—the "local interval length" can be appropriately selected. "and number of sampling points" Under the premise of filtering out interference from short-term fading, the local mean of long-term fading (i.e., the characteristics of large-scale signal changes) can be accurately estimated.

[0031] By comparing the average sampling values ​​of test data from different frequency bands at different spatial lengths (2L), The standard deviation σ can be reduced to 1 dB when the frequency range is 2L = 2~4λ. At this time, the length of the average sampling interval 2L is called the intrinsic length. If 2L is less than the intrinsic length, the local mean will be affected by residual Rayleigh fading; if 2L is greater than the intrinsic length, the local mean will be affected by smoothing beyond the length.

[0032] Taking the frequency bands commonly used in domestic mobile communication networks as an example, typical values ​​of the spatial intrinsic length 2L for different frequency bands are shown in Table 1.

[0033] Table 1 Typical values ​​of spatial intrinsic length 2L in different frequency bands

[0034] 2. Time-domain test data sampling If measured in the time domain, the received signal can be represented as:

[0035] :time The received signal strength (or power) at a certain point is the superposition of long-term fading and short-term fading; Time variable (describes the changes of a signal over time); Long-term fading in the time domain reflects large-scale signal changes (such as the slow change of road shadow effect over time and the slow fluctuation of path loss caused by user movement), with a long change period and a large range. Short-term fading in the time domain reflects small-scale signal changes (such as Rayleigh / Rice fading caused by multipath effects), and the signal strength fluctuates rapidly over short distances / times (short change period and small range).

[0036] Points along the time axis The expression for the local mean estimate is:

[0037] Time point Local mean estimate at (i.e., for) (Averaging signals within a nearby time window in an attempt to "extract" local features of long-term fading); : A specific point on the timeline (such as a test time point or a reference time); The "half-length" of the time window (i.e., the total window duration is...) (Controlling the range of "local time"); Integration interval :right Received signals within a 2T time window in the vicinity The purpose of integral averaging is to filter out the rapid fluctuations of short-term decline through "time smoothing" while preserving the slow trend of long-term decline.

[0038] If in By selecting an appropriate time length T, the estimated value of the local mean equals the actual value of the local mean, i.e. Then the following relationship holds:

[0039] Short-term decline The integral over the time window averages close to 0. This is due to long-term decay. It is "slowly changing," within a time window. The internal approximation is a constant (i.e., the actual local mean). ; and short-term decline It is "rapidly changing" (with dramatic positive and negative fluctuations). If the window duration T is chosen reasonably, its time average will approach 0 due to the "cancellation of positive and negative values"—at this point, the local mean estimate... It is equal to the actual local mean. .

[0040] In actual testing, discrete sampling (rather than continuous integration) is used, combined with Li's theorem and engineering test conclusions (sampling points). When the difference between the test value and the actual local mean is <1dB, we have:

[0041] Discrete sampling points (in) Nearby local area Inside).

[0042] The number of sampling points must meet the following requirements. —At this point, the sum of the sampled values ​​of the short-term fading approaches 0 (due to the "positive and negative cancellation" of the fast fluctuations). Discrete summation can approximate continuous integration, ensuring the accuracy of local mean estimation.

[0043] By employing a "local averaging" operation (integration or discrete sampling summation), and leveraging the characteristics of short-term fading—namely, "rapid changes and easy cancellation"—and by appropriately selecting the "time window length," "and number of sampling points" Under the premise of filtering out interference from short-term fading, the local mean of long-term fading (i.e., the characteristics of large-scale signal changes) can be accurately estimated.

[0044] The key to eliminating fast fading in the time domain is that the sampling time is much longer than the channel coherence time. If the sampling time is much longer than If the channel has already experienced "fast fading" (independent fading characteristics) during the two sampling times, then the random fluctuations of fast fading can be offset by "local averaging"; if the sampling time is much shorter than... The channel is approximately "slow fading" (its characteristics remain unchanged), requiring no additional processing.

[0045] Root mean square delay spread is a quantitative indicator of multipath effect, describing the "dispersion" of the delay of multipath signals (the more severe the multipath effect, the greater the delay spread).

[0046] Formula 1:

[0047] Root mean square delay spread (unit: seconds) measures the "standard deviation" of multipath signal delay and reflects the strength of multipath effects; Expectation operator (statistical average); The time delay of a multipath signal (i.e., the time difference between the arrival time of the signal on that path and the time of the "reference path"). The average delay of a multipath signal (the statistical average of the delays of all paths). ).

[0048] Formula 2 (expanded form, for easier calculation):

[0049] The second moment of the delay (the statistical average of the squared delays of all multipath signals). )).

[0050] Using the mathematical relationship "variance = second moment - squared mean", we can... Transformed into statistical measures ( and The form of calculation.

[0051] Coherence time is the length of time during which the channel characteristics are "relatively stable" (beyond this time, the channel changes significantly due to fast fading).

[0052] Formula 3:

[0053] Channel coherence time (unit: seconds) describes the duration of the "time correlation" of the channel impulse response.

[0054] and Inversely proportional – the more severe the multipath effect ( The larger the channel, the faster its characteristics change. The smaller).

[0055] Multipath propagation refers to the phenomenon where a signal travels through different paths (reflection, diffraction, scattering) to reach the receiver. The difference in path length leads to different time delays, resulting in "delay spread." ).

[0056] High-frequency channels (e.g., 2GHz~3GHz): Shorter wavelengths mean even small obstacles / reflectors can trigger multipath propagation, resulting in a large number of multipath paths and significant time delay differences. Big → Small (channel changes rapidly); Low-frequency channels (e.g., Sub 1 GHz): longer wavelengths, relatively weaker multipath effects → Small → Large (slow channel change).

[0057] Typical coherence time values ​​for different frequency bands (reflecting differences in multipath effects in real-world environments): Sub 1GHz (<1GHz):

[0058] 1GHz~2GHz:

[0059] 2GHz~3GHz:

[0060] Through root mean square delay expansion ( Quantify the strength of the multipath effect, and then through Derivation of coherence time ( (Channel stability duration). The engineering method for eliminating fast fading in the time domain is: the sampling time is much longer than... This allows the random fluctuations of fast fading to be fully "cancelled" within the sampling interval, thereby extracting the characteristics of long-term fading.

[0061] By comparing the average sampling values ​​of test data at different time lengths (T) for different frequency bands The standard deviation σ is taken as At that time, the standard deviation σ can be reduced to 1 dB. By designing a time window length of 2T and a sampling frequency, it is possible to achieve this. (To ensure the accuracy of the sampling mean). Among them, : Local mean estimate in the time domain (averaging the received signal within a time window to extract long-term fading features); : Sample mean The standard deviation measures the degree of deviation between the sampled mean and the actual local mean.

[0062] If the total length of the time window Much larger If the window contains enough fast fading fluctuations (positive and negative cancel each other out), then the standard deviation of the sampling mean will be such that... It dropped to below 1 dB.

[0063] Assuming the number of test data samples within a 2T cycle is N, then the required sampling frequency is:

[0064] Sampling frequency (unit: Hz, representing the number of samples per second); Number of sampling points within a 2T time window (in engineering practice) To ensure that the error between the sampled mean and the actual mean is <1dB.

[0065] Coherence time of different frequency bands Different (multipath effects are more pronounced in the high-frequency band), Smaller; Low frequency band Larger), substitute into the formula to calculate the sampling frequency ( Take 50), as shown in Table 2.

[0066] Table 2 Sampling frequency of different frequency bands

[0067] 3. Indoor Propagation Model Calibration Test Data Sampling Method Because of the limited indoor space, sampling 50 data points within an intrinsic distance of 2L presents significant challenges. The distance between two adjacent samples is only a few centimeters or eleven centimeters, and factors such as antenna size and measurement errors can all affect the accuracy of the test data. Therefore, propagation model calibration test data sampling within each intrinsic length needs to be performed in the time domain.

[0068] The sampling of indoor propagation model calibration test data needs to combine time and spatial domains to solve the problem of insufficient sampling points and eliminate the influence of fast signal fading.

[0069] like Figure 1As shown, this invention proposes a method for sampling test data for indoor wireless propagation model calibration, comprising the following steps: 1) Determine the indoor signal source coverage radius as R, and take the signal source as the center and the range with a radius of 2R as the sampling area for indoor wireless propagation model correction; within the sampling area, divide the area into rectangular grids with the spatial domain intrinsic length 2L as the side length.

[0070] 2) At the center point of each rectangular grid, test data is sampled according to the time domain sampling frequency, and the number of sampling points N in each grid is not less than 50.

[0071] 3) Process the sampled data in each grid, remove abnormal data, average the sampled data in each grid to eliminate the effect of fast fading, and obtain the wireless signal strength data of each grid.

[0072] Furthermore, the process of determining the spatial domain intrinsic length 2L is as follows: by comparing the standard deviation σ of the sample mean of test data at different spatial lengths for different frequency bands, when the standard deviation σ drops to 1dB, the corresponding spatial length is the spatial domain intrinsic length 2L of that frequency band; typical values ​​of the spatial intrinsic length 2L for different frequency bands are 2λ for the Sub 1GHz band, 3λ for the 1GHz~2GHz band, and 4λ for the 2GHz~3GHz band, where λ is the wavelength of the signal in the corresponding frequency band, including 0.8 meters for the 700MHz band, 0.7 meters for the 800MHz band, and 0.5 meters for the 2600MHz band, etc.

[0073] Eliminating fast fading in the time domain requires considering the channel's coherence time; the sampling time should be much longer than the channel's coherence time. In a stationary state, the channel's coherence time is related to multipath effects, which cause signals to arrive at the receiver along different paths with varying delays, thus leading to signal spread.

[0074] Furthermore, the process of determining the time-domain sampling frequency includes: 1) Determine the coherence time Tc of different frequency bands in the static state. The value is set between 0.2μs and 0.5μs. The typical values ​​of the correlation time Tc of different frequency bands are 0.2μs for the Sub 1GHz band, 0.3μs for the 1GHz~2GHz band, and 0.45μs for the 2GHz~3GHz band. 2) Set the sampling time length When the number of sampling points N=50, according to the formula The time-domain sampling frequencies of different frequency bands were calculated. Typical time-domain sampling frequencies for different frequency bands are 125Hz for the Sub 1GHz band, 83Hz for the 1GHz~2GHz band, and 55Hz for the 2GHz~3GHz band.

[0075] Furthermore, when processing the sampled data within each grid, the 3σ criterion is used to remove outliers. That is, when the absolute value of the difference between a sampled data and the mean of all sampled data in that grid is greater than 3 times the standard deviation, the data is determined to be outlier and removed.

[0076] The present invention also provides an indoor wireless propagation model calibration test data sampling device, comprising: 1) Grid division module: used to determine the indoor signal source coverage radius as R, with the signal source as the center, and the area with a radius of 2R as the sampling area for indoor wireless propagation model correction; within the sampling area, rectangular grids are divided with the spatial domain intrinsic length 2L as the side length.

[0077] 2) Sampling module: used to sample test data at the center point of each rectangular grid according to the time domain sampling frequency, and the sampling data scale in each grid is not less than 50.

[0078] 3) Data processing module: This module processes the sampled data in each grid cell, removes abnormal data, averages the sampled data in each grid cell to eliminate the effects of fast fading, and obtains the wireless signal strength data for each grid cell.

[0079] Furthermore, the device also includes an intrinsic length determination module, which compares the standard deviation σ of the sample mean of test data for different frequency bands under different spatial lengths. When the standard deviation σ drops to 1dB, the corresponding spatial length is determined as the spatial domain intrinsic length 2L of that frequency band. Typical values ​​of the spatial intrinsic length 2L for different frequency bands are 2λ for the Sub 1GHz band, 3λ for the 1GHz~2GHz band, and 4λ for the 2GHz~3GHz band, where λ is the wavelength of the signal in the corresponding frequency band. Specifically, it is 0.8 meters for the 700MHz band, 0.7 meters for the 800MHz band, and 0.5 meters for the 2600MHz band, etc.

[0080] Furthermore, the device also includes a sampling frequency determination module, which includes: 1) Coherence Time Determination Unit: Used to determine the coherence time Tc of different frequency bands in a static state, with a value set between 0.2μs and 0.5μs; typical values ​​of the correlation time Tc for different frequency bands are 0.2μs for the Sub 1GHz band, 0.3μs for the 1GHz~2GHz band, and 0.45μs for the 2GHz~3GHz band; 2) Frequency calculation unit: used to set the time length. When the number of sampling points N=50, according to the formula The time-domain sampling frequencies of different frequency bands were calculated. Typical time-domain sampling frequencies for different frequency bands are 125Hz for the Sub-1GHz band, 83Hz for the 1GHz~2GHz band, and 55Hz for the 2GHz~3GHz band.

[0081] Furthermore, when removing outlier data, the data processing module adopts the 3σ criterion, which means that when the absolute value of the difference between a sampled data and the mean of all sampled data in the grid is greater than 3 times the standard deviation, the data is determined to be outlier and removed.

[0082] The present invention also proposes an electronic device, comprising: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the above-described method or apparatus.

[0083] The application of specific embodiments will be described below with reference to specific examples.

[0084] Example 1 This embodiment provides a method for sampling test data for indoor wireless propagation model correction, which is applied to the propagation model correction of the 2.6GHz band (belonging to the 2GHz~3GHz band) in the indoor environment of office buildings.

[0085] Includes the following steps: 1) Determine the coverage radius R=20 meters of the 2.6 GHz band signal source in the office building. With the signal source as the center, the area with a radius of 2R=40 meters is used as the test data sampling area.

[0086] 2) For the 2.6GHz band, its wavelength The intrinsic length of the spatial domain is 2L=4λ=4×0.115=0.46 meters. Rectangular grids with a side length of 0.46 meters are divided within the sampling area.

[0087] 3) This frequency band belongs to the 2GHz~3GHz frequency band, the time domain sampling frequency is 55Hz, sampling is performed at the center point of each rectangular grid, and the sampling data scale in each grid is 50.

[0088] 4) Process the 50 sampled data points within each grid cell, calculate the mean and standard deviation, and use the 3σ criterion to remove outliers (if the absolute value of the difference between a data point and the mean is greater than 3 times the standard deviation, it is removed). Then, average the remaining data to obtain the wireless signal strength data for that grid cell, thus eliminating the influence of fast fading. For example, taking the 50 sampled data points of a grid cell in the 2.6GHz band as an example, the calculated mean μ = -75dBm and standard deviation σ = 2dB; if a data point is -83dBm, the absolute value of its difference from μ is 8dB (>3σ = 6dB), it is determined to be outlier and removed; the average value of the remaining 49 data points is -74.8dBm, which is taken as the signal strength data for that grid cell.

[0089] like Figure 2 As shown, the asterisk indicates the source location, 2R represents the test data sampling area (40 meters for the 2.6 GHz band; the figure is only partially shown, displaying a square area with a side length of 40 meters and a circular area with a radius of 20 meters), and the side length of the rectangular grid is the spatial domain intrinsic length 2L (0.46 meters for the 2.6 GHz band). The rectangular grid is a schematic diagram of sampling units divided by intrinsic length.

[0090] Example 2 This embodiment provides a method for sampling test data for indoor wireless propagation model correction, which is applied to the propagation model correction of the 700MHz band (belonging to the Sub 1GHz band) in residential indoor environments.

[0091] The method includes the following steps: 1) Determine the coverage radius R=50 meters of the 700MHz band signal source in the residential area. With the signal source as the center, the area with a radius of 2R=100 meters will be used as the test data sampling area.

[0092] 2) For the 700MHz band, its wavelength The intrinsic length of the spatial domain is 2L = 2λ = 2 × 0.39 = 0.78 meters. Rectangular grids with a side length of 0.78 meters are divided within the sampling area.

[0093] 3) This frequency band belongs to the Sub 1GHz band, the time domain sampling frequency is 125Hz, sampling is performed at the center point of each rectangular grid, and the sampling data scale in each grid is 50.

[0094] 4) Process the 50 sampled data in each grid, calculate the mean and standard deviation, use the 3σ criterion to remove outliers, and then average the remaining data to obtain the wireless signal strength data of the grid, thus eliminating the influence of fast fading.

[0095] Example 3 like Figure 3 As shown, this embodiment provides an indoor wireless propagation model calibration test data sampling device, corresponding to the method of Embodiment 1, wherein solid lines connect to represent the system's constituent modules, and the device includes: 1) Grid division module: Determine the coverage radius R=20 meters for the 2.6 GHz band signal source. With the signal source as the center, take the area with a radius of 40 meters as the sampling area and divide it into rectangular grids with a side length of 0.46 meters.

[0096] 2) Sampling module: Sampling is performed at a sampling frequency of 55Hz at the center point of each grid, and 50 data points are sampled per grid.

[0097] 3) Data processing module: Processes 50 data points for each grid, removes outlier data using the 3σ criterion, and then averages the results to obtain the wireless signal strength data for each grid.

[0098] 4) Intrinsic length determination module: By comparing the standard deviation σ of the sampling mean of the 2.6GHz band under different spatial lengths, it is determined that when the spatial length is 0.46 meters, σ drops to 1dB, and this is taken as the spatial domain intrinsic length 2L of the band.

[0099] 5) Sampling frequency determination module: The coherence time determination unit determines the coherence time of the 2.6GHz band. Frequency calculation unit settings When N=50, the sampling frequency .

[0100] The aforementioned device can effectively collect and process data, providing accurate sampling data for the correction of the 2.6GHz indoor propagation model.

[0101] like Figure 3 As shown, the dashed lines connect the methods of Embodiment 1 and Embodiment 2. For example, the output 2L of the intrinsic length determination module is used as the input of the grid division module, and the output of the sampling frequency determination module is used as the input of the grid division module. As input to the sampling module.

[0102] Example 4 Figure 4 This is a schematic diagram of another electronic device 20 provided in an embodiment of this application. The device 20 includes a processor 21, which is used to execute computer programs or instructions stored in a memory 22, or to read data / signaling stored in the memory 22, to perform the methods in the above-described method embodiments. Optionally, there may be one or more processors 21.

[0103] Optionally, such as Figure 4 As shown, the device 20 also includes a memory 22 for storing computer programs or instructions and / or data. The memory 22 may be integrated with the processor 21 or may be disposed separately. Optionally, there may be one or more memories 22.

[0104] Optionally, such as Figure 4 As shown, the device 20 also includes a transceiver 23 for receiving and / or transmitting signals. For example, the processor 21 controls the transceiver 23 to receive and / or transmit signals.

[0105] As one option, the device 20 is used to implement the operations performed by the first communication device or the second communication device in the various method embodiments described above.

[0106] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0107] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0108] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0109] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0110] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0111] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0112] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for sampling test data for indoor wireless propagation model calibration, characterized in that, Includes the following steps: The indoor signal source coverage radius is defined as R. The area with a radius of 2R centered on the signal source is used as the test data sampling area for indoor wireless propagation model correction. Within the sampling area, a rectangular grid is divided with a spatial domain intrinsic length of 2L as the side length; Test data is sampled at the center point of each rectangular grid according to the time-domain sampling frequency. The sampled data in each grid is processed, and after removing outliers, the sampled data in each grid is averaged to obtain the wireless signal strength data of each grid. The time-domain sampling frequency is determined according to different frequency bands. The process of determining the time-domain sampling frequency includes: determining the coherence time Tc of different frequency bands in a static state, wherein the value of the coherence time Tc is set between 0.2 μs and 0.5 μs; and setting the sampling time length. When the number of sampling points N in each grid is 50, according to the formula The time-domain sampling frequencies of different frequency bands were calculated. .

2. The method according to claim 1, characterized in that, The intrinsic length 2L of the spatial domain is determined according to different frequency bands. The process of determining the intrinsic length 2L of the spatial domain includes: comparing the standard deviation σ of the sample mean of test data at different spatial lengths for different frequency bands. When the standard deviation σ drops to 1dB, the corresponding spatial length is determined as the intrinsic length 2L of the spatial domain for that frequency band.

3. The method according to claim 1, characterized in that, The number of sampling points N in each grid is not less than 50.

4. The method according to any one of claims 1 to 3, characterized in that, When processing the sampled data in each grid, the 3σ criterion is used to remove outlier data. The process of removing outlier data includes: when the absolute value of the difference between a sampled data and the mean of all sampled data in the grid is greater than 3 times the standard deviation, the data is determined to be outlier data and removed.

5. An indoor wireless propagation model calibration test data sampling device, characterized in that, include: The grid division module is used to determine the indoor signal source coverage radius as R, and the area with the signal source as the center and a radius of 2R as the test data sampling area for indoor wireless propagation model correction; within the sampling area, a rectangular grid is divided with a spatial domain intrinsic length of 2L as the side length; the sampling module is used to sample test data at the center point of each rectangular grid according to the time domain sampling frequency. The data processing module is used to process the sampled data in each grid, remove abnormal data, and average the sampled data in each grid to obtain the wireless signal strength data of each grid. The sampling frequency determination module includes: a coherence time determination unit, used to determine the coherence time Tc of different frequency bands in a static state, wherein the value of the coherence time Tc is set between 0.2 μs and 0.5 μs; and a frequency calculation unit, used to set the sampling time length. When the number of sampling points N in each grid is 50, according to the formula The sampling frequencies of different frequency bands were calculated. .

6. The apparatus according to claim 5, characterized in that, It also includes an intrinsic length determination module, which is used to determine the spatial length 2L of the frequency band by comparing the standard deviation σ of the sample mean of test data at different spatial lengths for different frequency bands. When the standard deviation σ drops to 1dB, the corresponding spatial length is determined as the spatial domain intrinsic length 2L of that frequency band.

7. The apparatus according to any one of claims 5 to 6, characterized in that, When removing abnormal data, the data processing module adopts the 3σ criterion, which means that when the absolute value of the difference between a sampled data and the mean of all sampled data in the grid is greater than 3 times the standard deviation, the data is determined to be abnormal and removed.

8. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the method as described in any one of claims 1-4.

Images