A channel identification method based on composite spherical model
Through the channel recognition method based on the composite spherical model, the spherical model of equal-scale sequence distribution and single-reflective channel theory are used to accurately judge the LOS and NLOS channel types, and the problem of insufficient channel recognition success rate in the prior art is solved, achieving high Accurate channel identification.
Patent Information
- Application Number
- CN202211484323.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-11-24
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2042-11-24
AI Technical Summary
The existing channel recognition methods cannot exceed 98% recognition success rate, and it is difficult to accurately judge the LOS and NLOS channel types.
Using a channel recognition method based on a composite spherical model, the concentric spherical model with equal-scale sequence distribution is constructed, the area to be located is divided, the initial interspherical domain of the terminal is determined, and the LOS diameter of the terminal is judged according to the single-reflection channel theory Delay range, determine channel type by actual delay.
It improves the accuracy of channel recognition and can identify channel types at a success rate of more than 98%, which is simple to operate and clear in principle.
Smart Images

Figure CN115865236B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of wireless channel technology, and in particular to a channel identification method based on a composite spherical model. Background Art
[0002] In wireless communication systems, system performance is mainly affected by the wireless channel, so channel identification is crucial.
[0003] Among the existing channel type identification algorithms, there are many LOS / NLOS path identification algorithms that are implemented for different scenarios and through different parameter indicators.
[0004] Specifically, among the numerous existing technologies, some solutions use RSSI parameters to identify the channel type algorithm for the single base station positioning technology scenario, with a channel identification success rate of about 85%; some solutions use machine learning, also for the single base station positioning technology scenario, with an identification success rate of about 87%; some solutions are channel type identification algorithms for AOA triangulation positioning technology scenarios, with a channel type identification success rate of up to 98%; some solutions use convolutional neural networks to identify the channel type algorithm for indoor UWB positioning technology scenarios (three-sided positioning technology), with a channel type identification success rate of up to 98%.
[0005] However, there is still no channel identification method with an identification success rate of more than 98%. Summary of the invention
[0006] The present application provides a channel identification method based on a composite spherical model, which can solve the problem that the identification success rate of the existing channel identification method cannot exceed 98%.
[0007] The technical solution of the present application is a channel identification method based on a composite spherical model, comprising:
[0008] S1: Taking the base station as the sphere center, constructing a composite spherical model including a number of concentric spheres whose spherical radii are distributed in geometric progression, and dividing the area to be positioned according to the composite spherical model to obtain an initial spherical inter-domain formed by a number of adjacent concentric spheres;
[0009] S2: Based on the single reflection channel theory and according to the positioning of the base station corresponding to the terminal, the initial inter-spherical domain where the terminal is located is determined as the terminal inter-spherical domain, and according to the terminal inter-spherical domain, the terminal LOS path delay range corresponding to the terminal is determined;
[0010] S3: Obtain the actual delay of the terminal, and determine whether the actual delay is within the LOS path delay range of the terminal. If so, determine that the channel between the terminal and the base station is a LOS path; if not, determine that the channel between the terminal and the base station is a NLOS path.
[0011] Optionally, the step S1 includes:
[0012] S11: With the base station as the sphere center, a composite spherical model including a plurality of concentric spheres whose spherical radii are distributed in a geometric progression is constructed;
[0013] The general formula for the spherical radius distributed in a geometric progression is as follows:
[0014]
[0015] In the formula, R n+1 and R n They represent the spherical radii of two adjacent concentric spherical surfaces respectively;
[0016] α0 represents a constant determined according to the channel identification success rate;
[0017] S12: Divide the area to be positioned according to the composite spherical model to obtain initial inter-spherical domains formed by a plurality of adjacent concentric spherical surfaces.
[0018] Optionally, the α0 is 176.4°.
[0019] Optionally, step S2 includes:
[0020] S21: Based on the single reflection channel theory and according to the positioning of the base station corresponding to the terminal, determine the estimated position l0 of the terminal, and determine the estimated time delay τ0 of the estimated position accordingly based on the delay calculation formula, and accordingly fuzzily determine the estimated spherical domain where the estimated position is located;
[0021] The delay calculation formula corresponding to the estimated delay τ0 is as follows:
[0022] τ0=l0 / C;
[0023] In the formula, C represents the air propagation speed of the signal;
[0024] S22: Determine the smaller spherical radius R of two adjacent concentric spherical surfaces separated by the estimated spherical surface region according to the estimated spherical surface region where the estimated position is located. m and the larger spherical radius R m+1 ;
[0025] S23: Based on the time delay calculation formula, the smaller spherical radius R of the estimated spherical domain is calculated according to the interval m and the larger spherical radius R m+1 , determine the estimated LOS path delay range (τ m , τ m+1 );
[0026] Corresponding to the estimated LOS path delay range (τ m , τ m+1 The delay calculation formula of ) is as follows:
[0027] τ m =R m / C;
[0028] τ m+1 =R m+1 / C;
[0029] S24: Determine whether the estimated delay τ0 is within the estimated LOS path delay range (τ m , τ m+1 ), if yes, determine the estimated inter-spherical domain as the terminal inter-spherical domain, and determine the estimated LOS path delay range (τ m , τ m+1 ) is the terminal LOS path delay range (τ MIN , τ MAX ).
[0030] Beneficial effects:
[0031] The present application proposes a channel identification method that is different from the prior art. By setting a composite spherical model, the terminal LOS path delay range corresponding to the terminal is regularly determined, and then according to the actual delay of the terminal, whether the actual delay is within the terminal LOS path delay range is determined, so as to determine whether there is a LOS path between the base station and the terminal. The whole process has a clear principle, convenient operation, and a very high accuracy rate. Therefore, the present application can solve the problem that the recognition success rate of the existing channel identification method cannot exceed 98%. BRIEF DESCRIPTION OF THE DRAWINGS
[0032] In order to more clearly illustrate the technical solution of the present application, the drawings required for use in the embodiments are briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.
[0033] Figure 1 Schematic diagram of the flow of a channel identification method based on a composite spherical model in an embodiment of the present application;
[0034] Figure 2 A schematic diagram of a single reflection channel model in an embodiment of the present application;
[0035] Figure 3 It is a cross-sectional schematic diagram of the composite sphere model in the embodiment of the present application. DETAILED DESCRIPTION
[0036] The following embodiments are described in detail, and examples thereof are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following embodiments do not represent all implementations consistent with the present application. They are only examples of systems and methods consistent with some aspects of the present application as detailed in the claims.
[0037] The present application provides a channel identification method based on a composite spherical model, such as Figure 1 As shown, Figure 1 The flowchart of the channel identification method based on the composite spherical model in the embodiment of the present application includes:
[0038] S1: With the base station as the sphere center, a composite spherical model including a number of concentric spheres whose spherical radii are distributed in geometric progression is constructed, and the area to be positioned is divided according to the composite spherical model to obtain an initial spherical inter-domain formed by a number of adjacent concentric spheres.
[0039] Specifically, the embodiment of the present application uses the time delay τ as a basis for determining whether the LOS path exists.
[0040] Through channel estimation, we can get the path with the shortest delay, the delay is τ, and the LOS path delay is τ LoS ,
[0041] If the channel is a LOS path, the condition τ = τ should be satisfied. LOS ,
[0042] If the channel is a NLOS path, the condition τ>τ should be satisfied LOS .
[0043] The key to the above solution is to obtain τ LOs , since it is impossible to estimate the parameter τ by processing the received signal LOS , and it is also impossible to obtain the τ of each point through testing Los Therefore, a new method is needed to determine whether a channel is a LOS path.
[0044] Wherein, step S1 comprises:
[0045] S11: With the base station as the sphere center, a composite spherical model is constructed, which includes a plurality of concentric spheres whose spherical radii are distributed in a geometric progression.
[0046] The general formula for the spherical radius distributed in a geometric progression is as follows:
[0047]
[0048] In the formula, R n+1 and R nThey respectively represent the spherical radius of two adjacent concentric spherical surfaces.
[0049] α0 represents a constant determined according to the channel identification success rate.
[0050] Specifically, in some embodiments, α0 is 176.4°.
[0051] S12: Divide the area to be positioned according to the composite spherical model to obtain initial inter-spherical domains formed by a plurality of adjacent concentric spherical surfaces.
[0052] Specifically, with reference to the fingerprint positioning technology, a method for dividing the positioning area is provided. The positioning area is divided, and concentric spheres are made in the positioning area with the base station as the sphere center, and the area to be positioned is divided to obtain a number of initial spherical domains.
[0053] For each initial spherical surface, each point on the outer spherical surface is farthest from the base station in a straight line, with a radius of R. MAX , the corresponding direct path delay is τ MAX , each point on the inner sphere is closest to the base station, with a radius of R MIN , the corresponding direct path delay is τ MIN , let the range of the initial spherical domain d = R MAx -R MIN .
[0054] In the initial spherical domain, the direct path delay τ corresponding to all points LOs should satisfy τ MIN ≤τ LOS ≤τ MAX ,
[0055] When the channel is a LOS path, τ = τ LOs , then the condition τ is satisfied MIN ≤τ≤τ MAX , the former is a sufficient but not necessary condition for the latter, and when the interval range d is small enough, lim d→0 τ MAX =τ MIN =τ LOs , at this time τ=τ Los With τ MIN ≤τ≤τ max They are necessary and sufficient conditions for each other. Similarly, when the interval range is small enough, when the channel is an NLOS path, τ>τ MAX With τ>τ LOs They are also mutually necessary and sufficient conditions.
[0056] The smaller the interval range, the more accurate the judgment result. When the interval range approaches 0, the accuracy is 100%. However, in actual operation, the interval range cannot be 0. Therefore, how to determine a reasonable interval range so that the use of τMIN ≤τ≤τ MAX The core issue of this study is to achieve a high degree of accuracy when used as a basis for determining whether a channel is a LOS path.
[0057] To determine a reasonable interval division scheme, we can use the channel model to analyze the channel model and determine the τ MAX With τ MIN Appropriate value.
[0058] like Figure 2 As shown, Figure 2 Schematic diagram of a single reflection channel model in an embodiment of the present application. A triangle is formed between the single reflection path and the LOS path, and the vertex angle of the triangle is α.
[0059] When α=180°, then τ=τ1+τ2=τ Los , the channel is considered to be a LOS path;
[0060] When α<180°, then τ=τ1+τ2>τ LOs , the channel is considered to be a NLOS path.
[0061] Therefore, we can take τ = τ when α = 180° LOS =τ MIN ; Take α less than 180° but close enough to 180°, τ=τ1+τ2=τ MAX .
[0062] In the process of using the millimeter wave system for positioning, it can be assumed that its multipath situation only has a direct path and a single reflection path. The reason is that the signal attenuation of the millimeter wave system is large after reflection. When the number of reflections is greater than or equal to 2 times, the signal power is negligible compared with the single reflection path and the direct path.
[0063] Therefore, when determining the area division scheme, the single reflection channel model can be used.
[0064] The next step is to determine τ MAX With τ MIN The relationship between the two can be used to obtain a specific regional division plan.
[0065] Assume that in the single reflection channel model, when τ MIN ≤τ≤τ MAX When determining whether a channel is a LOS path, any channel whose vertex angle α satisfies the condition α0≤α≤180° will be identified as a LOS path, where α0 is a constant very close to 180°.
[0066] like Figure 2 As shown, when the positions of the base station and the mobile station are fixed, τ LosThe value of is fixed, and when α = 1 80°, the actual propagation path delay τ = τ1 + τ2, and the theoretical LOS path delay τ at the terminal location LOS The values are equal, and the channel at this time is the channel with the smallest propagation delay among the channels that satisfy the condition α0≤α≤180°, and τ is the minimum delay. Therefore, when α=180°, τ can be taken MIN Equal to τ, which is the lower limit of the judgment basis interval.
[0067] The upper limit τ MAX The basic condition that needs to be met is τ MAX It should be greater than or equal to the delay τ of all channels that satisfy the condition α0≤α≤180°, that is, greater than or equal to the maximum value of all channel delays. If you want to know when the maximum delay is obtained and what the maximum delay is, you can deduce it as follows:
[0068] When α=α0, take τ=τ MAX .
[0069] like Figure 2 As shown, by the cosine theorem:
[0070]
[0071] After a slight deformation, we can get the mean inequality:
[0072]
[0073] τ MAX The value of τ should ensure MAX The maximum value of τ when α is greater than α0 can be obtained from the mean inequality. The equal sign is taken when and only when τ1=τ2=τ / 2, and τ reaches its maximum value. At this time, τ=τ MAX It can be guaranteed that when α=α0, all τ≤τ MAX :
[0074]
[0075]
[0076] From this result, we know that τ MAX With τ MIN Linear correlation.
[0077] Therefore, in the area division scheme, the spherical radius of the concentric spheres is taken in the form of a geometric progression, such as Figure 3 As shown, Figure 3 This is a cross-sectional schematic diagram of a composite sphere model in an embodiment of the present application. The circular structure in the center represents a base station, and the square structure between the circles represents a terminal. The common ratio is as follows:
[0078]
[0079] The general formula of the sequence is as follows:
[0080]
[0081] Where R n Q is the sphere n Corresponding radius, for Q n+1 and Q n In the area between them, Q n+1 The corresponding delay τ n+1 That is τ MAX , Q n The corresponding delay τ n That is τ MIN . °
[0082] Finally, we need to determine that when τ = τ MAX When , the appropriate value of α0.
[0083] The value of α0 directly affects the algorithm's success rate in identifying channel types.
[0084] In the process of identifying the channel type, when α<α0, the channel type can be correctly identified as a NLOS channel. Therefore, the closer α0 is to 180°, the higher the recognition success rate.
[0085] Without adding an additional channel type identification algorithm, when estimating the arrival angle and transmission angle in single base station positioning, there will be an error of ±5° between the estimated result and the actual result. This also means that when 170°≤α≤180°, the channel type cannot be determined and the channel can only be roughly considered to be a LOS path.
[0086] If we want to improve the result of single-base station positioning technology to more than 98%, the value of α0 should meet the condition α0 ≥ 180° × 98% = 176.4°. Substituting α0 = 176.4°, we can get the general formula that theoretically satisfies the recognition success rate of 98%:
[0087]
[0088] S2: Based on the single reflection channel theory and the positioning of the base station corresponding to the terminal, the initial inter-spherical domain where the terminal is located is determined as the terminal inter-spherical domain, and the terminal LOS path delay range corresponding to the terminal is determined based on the terminal inter-spherical domain.
[0089] Wherein, step S2 comprises:
[0090] S21: Based on the single reflection channel theory and according to the positioning of the base station corresponding to the terminal, the estimated position l0 of the terminal is determined, and the estimated delay τ0 of the estimated position is determined accordingly based on the delay calculation formula, and the estimated inter-spherical domain where the estimated position is located is fuzzily determined accordingly.
[0091] The delay calculation formula corresponding to the estimated delay τ0 is as follows:
[0092] τ0=l0 / C.
[0093] Where C represents the air propagation speed of the signal.
[0094] S22: Determine the smaller spherical radius R of two adjacent concentric spherical surfaces separated by the estimated spherical surface region according to the estimated spherical surface region where the estimated position is located. m and the larger spherical radius R m+1 .
[0095] S23: Based on the time delay calculation formula, the smaller spherical radius R of the estimated spherical domain is calculated according to the interval m and the larger spherical radius R m+1 , determine the estimated LOS path delay range (τ m , τ m+1 ).
[0096] Corresponding to the estimated LOS path delay range (τ m , τ m+1 The delay calculation formula of ) is as follows:
[0097] τ m =R m / C.
[0098] τ m+1 =R m+1 / C.
[0099] S24: Determine whether the estimated delay τ0 is within the estimated LOS path delay range (τ m , τ m+1 ), if yes, determine the estimated inter-spherical domain as the terminal inter-spherical domain, and determine the estimated LOS path delay range (τ m , τ m+1 ) is the terminal LOS path delay range (τ MIN , τ MAX ).
[0100] S3: Obtain the actual delay of the terminal, and determine whether the actual delay is within the LOS path delay range of the terminal. If so, determine that the channel between the terminal and the base station is a LOS path; if not, determine that the channel between the terminal and the base station is a NLOS path.
[0101] Specifically, this solution is suitable for high-power base station positioning using large-scale antennas in relatively open indoor venues, such as gymnasiums; it can also be used in environments with many obstacles, such as offices, factories, libraries, etc.
[0102] In these scenarios, multiple base stations are usually deployed to cover all positioning areas according to actual conditions. Within the coverage area of a single base station, the above solution can be used to determine the channel type.
[0103] The above detailed description of the embodiments of the present application is only a preferred embodiment of the present application and cannot be considered to limit the scope of implementation of the present application. All equivalent changes and improvements made within the scope of the present application should still fall within the scope of the patent coverage of the present application.
Claims
1. A channel identification method based on a composite spherical model, characterized in that: include: S1: Taking the base station as the sphere center, constructing a composite spherical model including a number of concentric spheres whose spherical radii are distributed in geometric progression, and dividing the area to be positioned according to the composite spherical model to obtain an initial spherical inter-domain formed by a number of adjacent concentric spheres; S2: Based on the single reflection channel theory and according to the positioning of the base station corresponding to the terminal, the initial inter-spherical domain where the terminal is located is determined as the terminal inter-spherical domain, and according to the terminal inter-spherical domain, the terminal LOS path delay range corresponding to the terminal is determined; S3: Obtain the actual delay of the terminal, and determine whether the actual delay is within the LOS path delay range of the terminal. If so, determine that the channel between the terminal and the base station is a LOS path; if not, determine that the channel between the terminal and the base station is a NLOS path; The step S1 comprises: S11: With the base station as the sphere center, a composite spherical model including a plurality of concentric spheres whose spherical radii are distributed in a geometric progression is constructed; The general formula for the spherical radius distributed in a geometric progression is as follows: In the formula, R n+1 and R n They represent the spherical radii of two adjacent concentric spherical surfaces respectively; α0 represents a constant determined according to the channel identification success rate; S12: dividing the area to be positioned according to the composite spherical model to obtain initial spherical surface regions separated by a plurality of adjacent two concentric spherical surfaces; The α0 is 176.4°.
2. A channel identification method based on a composite spherical model according to claim 1, characterized in that: The step S2 comprises: S21: Based on the single reflection channel theory and according to the positioning of the base station corresponding to the terminal, determine the estimated position l0 of the terminal, and determine the estimated time delay τ0 of the estimated position accordingly based on the delay calculation formula, and accordingly fuzzily determine the estimated spherical domain where the estimated position is located; The delay calculation formula corresponding to the estimated delay τ0 is as follows: τ0=l0 / C; In the formula, C represents the air propagation speed of the signal; S22: Determine the smaller spherical radius R of two adjacent concentric spherical surfaces separated by the estimated spherical surface region according to the estimated spherical surface region where the estimated position is located. m and the larger spherical radius R m+1 ; S23: Based on the time delay calculation formula, the smaller spherical radius R of the estimated spherical domain is calculated according to the interval m and the larger spherical radius R m+1 , determine the estimated LOS path delay range (τ m ,τ m+1 ); Corresponding to the estimated LOS path delay range (τ m , τ m+j The delay calculation formula of ) is as follows: t m =R m / C; t m+1 =R m+1 / C; S24: Determine whether the estimated delay τ0 is within the estimated LOS path delay range (τ m , τ m+1 ), if yes, determine the estimated inter-spherical domain as the terminal inter-spherical domain, and determine the estimated LOS path delay range (τ m , τ m+1 ) is the terminal LOS path delay range (T MIN , T MAx ).
Citation Information
Patent Citations
Direct path extraction method and device
CN108337197A
Mine moving target locating method and system capable of restraining non-line-of-sight delay error
CN110045325A
