A three-dimensional positioning method based on time division activation of multi-leakage cable multi-clamping antenna

Abstract

The application discloses a three-dimensional positioning method based on time division activation of a multi-leaky cable and a multi-clamping antenna, and relates to the technical field of wireless communication and positioning. The method constructs a spatial positioning system composed of a plurality of leaky cables and a plurality of clamping antennas, activates each clamping antenna in turn through a time division activation mechanism, and obtains a multi-leaky cable and multi-time slot joint observation vector under limited radio frequency receiving link conditions. Global coarse positioning is performed based on received signal strength to obtain an initial position estimate of a user; local area searching is performed by using spatial phase information, and a plurality of mutually independent candidate positions are obtained through screening; a multi-start continuous optimization method is used to perform continuous spatial iterative optimization on the candidate positions to obtain a three-dimensional position estimate result of the user. The application utilizes high-dimensional spatial observation capability formed by time division activation, and the distributed radiation characteristics and spatial phase information of the leaky cable, reduces system hardware complexity, overcomes grid resolution limitation and local optimization problems, and realizes high-precision three-dimensional positioning in a complex environment.

Description

Technical Field

[0001] This invention relates to the field of wireless communication and positioning technology, specifically to a three-dimensional positioning method based on time-division activation of multiple leaky cable and multiple clamping antennas. Background Technology

[0002] With the rapid development of technologies such as 6G mobile communication, the Internet of Things (IoT), smart buildings, and the Industrial Internet, the demand for high-precision positioning in indoor environments is increasing. In applications such as intelligent manufacturing, unmanned inspection, robot navigation, personnel management, and asset tracking, systems not only need to achieve wireless communication coverage but also provide real-time and accurate user location awareness. Therefore, researching low-cost, high-precision, and highly reliable indoor positioning technologies has become an important research direction in the field of wireless communication and sensing.

[0003] Currently, common indoor positioning technologies mainly include wireless positioning methods based on wireless LAN, Bluetooth, ultra-wideband, and cellular networks. Among them, positioning methods based on received signal strength are simple to implement, but their positioning accuracy is easily affected by multipath fading and environmental changes. Positioning methods based on time of arrival or angle of arrival have higher positioning accuracy, but they usually require high-cost synchronization equipment, large-scale antenna arrays, or complex radio frequency hardware support, resulting in high system deployment costs and making it difficult to meet the needs of large-scale indoor applications.

[0004] Leaky cable is a special transmission medium capable of continuously radiating and receiving electromagnetic signals along its length. It offers advantages such as uniform coverage, flexible deployment, strong resistance to obstruction, and suitability for confined spaces, and has been widely used in wireless coverage scenarios in subways, tunnels, underground utility tunnels, and large buildings. In recent years, researchers have increasingly applied leaky cables to indoor positioning, estimating location by analyzing the propagation characteristics of user signals within the leaky cable.

[0005] However, existing leaky cable-based positioning technologies still have the following limitations: most positioning systems use a single leaky cable or a small number of fixed radiating units, resulting in limited spatial observation information and difficulty in meeting the requirements of three-dimensional positioning; most existing methods only utilize signal amplitude information and do not fully explore the spatial features contained in phase information, leading to positioning accuracy being limited by grid resolution; some methods use nonlinear optimization algorithms with a single starting point, which are prone to getting trapped in local optima when the objective function has multiple local extrema; existing methods are mostly limited to one-dimensional or two-dimensional positioning in the horizontal plane, lacking effective perception of the user's height direction and unable to achieve complete three-dimensional positioning; if each radiating unit is equipped with a separate radio frequency receiving link, the system hardware cost will increase linearly with the number of antennas, making it difficult to balance high-dimensional spatial observation capabilities with low-cost deployment requirements.

[0006] Therefore, there is an urgent need to propose a high-precision positioning method that can fully utilize the spatial observation capabilities of multi-leaky cable multi-clamp antennas, fuse received signal strength and spatial phase information, effectively avoid local optima problems, support complete three-dimensional positioning, and have controllable system hardware costs, so as to meet the actual needs of high-precision three-dimensional positioning in complex environments. Summary of the Invention

[0007] The purpose of this invention is to provide a three-dimensional positioning method based on time-division activation of multiple leaky cable and multiple clamping antennas to solve the problems mentioned in the background art.

[0008] To achieve the above objectives, the present invention adopts the following technical solution: a three-dimensional positioning method based on time-division activation of multiple leaky cable and multiple clamping antennas, comprising the following steps:

[0009] S1: Will A parallel leaky cable is laid at a height of The plane, with each leaky cable set Each clamping antenna; each leaky cable is connected to an RF receiving link at at least one end; the spatial three-dimensional physical coordinates of all clamping antennas, the electrical signal attenuation constant inside the cable, and the leaky cable propagation constant are obtained at the system base station;

[0010] S2: Using a time-division activation mechanism, the clamping antennas with corresponding numbers on each leaky cable are activated sequentially in each time slot to receive uplink pilot signals sent by multiple users. Multi-leaky cable multi-time slot spatial observation information is obtained through multi-time slot sequential activation, which is equivalent to constructing a spatial virtual antenna array.

[0011] S3: Demodulate the received signals in each time slot and separate the complex observation scalars of each user; cascade the complex observation scalars obtained by sampling all leaky cables in all time slots in spatiotemporal order to construct a global equivalent spatial virtual aperture observation vector.

[0012] S4: Within the three-dimensional physical boundary confinement area, establish an amplitude fitting cost function without phase information, and obtain the user's three-dimensional initial position estimation result through global search;

[0013] S5: Establish a three-dimensional local fine search region centered on the user's three-dimensional initial position estimation result obtained in step S4, calculate the normalized spatial phase coherence spectrum projection measure in the local region, and use a spatial suppression mechanism to screen several candidate positions that are physically isolated from each other.

[0014] S6: Using each candidate position as a parallel initial value, continuously optimize each candidate position in a continuous three-dimensional physical space so that each candidate position converges to its local residual minimum.

[0015] S7: Calculate the sum of squared residuals of the channel fitting at each convergence point, and output the convergence coordinate with the smallest residual as the final user's three-dimensional coordinate estimation result.

[0016] Preferably, in step S1, The root leaky cable is parallel to Erected along the axial direction, and along The axial direction is evenly spaced, and the erection length is The erection height is On each leaky cable One clamping antenna edge Arranged along the axis, the spacing between two adjacent clamping antennas is The first edge distance is set to .

[0017] Preferably, in step S2, the time-division activation mechanism satisfies the following: the positioning period includes Each time slot, in any time slot Below, the base station control terminal only activates the first [number] on each leaky cable. Each clamp-on antenna transmits radio frequency signals, so the entire base station only needs to be equipped with one of these antennas. Parallel RF receiving links, and in After each time slot, the equivalent construction includes... A virtual antenna array with spatial sampling locations.

[0018] Preferably, in step S3, when multiple users simultaneously transmit uplink pilot signals, the base station uses a dedicated orthogonal pilot sequence corresponding to each user to perform matched filtering on the received signals in each time slot, separating the complex observation scalars of each user; and concatenates the complex observation scalars obtained by each user in all leaky cables and all time slots in spatiotemporal order to construct a global equivalent spatial virtual aperture observation vector:

[0019]

[0020] In the formula, Number the user; This refers to the number of leaky cables; The number of clamping antennas installed on each leaky cable; Indicates the first The user in the first The first leaky cable Complex observation scalars obtained at each clamped antenna.

[0021] Preferably, in step S3, the user's three-dimensional position coordinates are: , No. One clamping antenna (of which) Theoretical Synthetic Channel Response at () Electrical transmission channel within the leaky cable and the spatial wireless propagation channel between the clamped antenna and the user Product construction:

[0022]

[0023] in, The model is as follows:

[0024]

[0025] The model is as follows:

[0026]

[0027] In the formula, and These are the attenuation constant within the leaky cable and the propagation constant of the leaky cable, respectively. To hold the antenna spacing, This is the edge distance at the beginning of the leaky cable. This is the normalized path loss coefficient. The phase constant in free space. For users to the first The Euclidean distance between the clamped antennas For users relative to the first The elevation angle of the horizontal plane holding the antenna. The height at which the parallel cable is installed. User's three-dimensional position coordinates The height coordinate component, is the Lambert cosine radiation factor. Number the leaky cable , The number of the antenna clamped on the leaky cable .

[0028] Preferably, in step S4, the established amplitude fitting cost function without phase information is a function based on the amplitude least squares criterion. The user's initial position estimation result is obtained by performing a global traversal search of discrete coarse grid points within the restricted area. :

[0029]

[0030] in, It is a set of coarse uniform grid points. For transmitting pilot power, Global equivalent spatial virtual aperture observation vector The One element, For grid points The theoretical physical channel at that location.

[0031] Preferably, in step S5, the formula for calculating the normalized spatial phase coherence spectrum projection measure is:

[0032]

[0033] in, For local fine grid points, The global theoretical synthesized channel response vector At grid points The response vector at that location; Defined as all A concatenated vector of theoretically synthesized channel responses from clamped antennas:

[0034]

[0035] In fine-grid retrieval, a non-maximum suppression mechanism is used for screening to obtain multiple spatially independent candidate positions.

[0036] Preferably, in step S6, the multiple candidate positions are used as parallel initial values, and continuous spatial iterative optimization is performed on each candidate position. The iterative optimization process is implemented using a nonlinear least squares iterative optimization algorithm based on gradient information, and an adaptive step size control mechanism and a damping factor adjustment mechanism are combined during the iteration process. By minimizing the objective function of the sum of squared residuals of the channel fitting based on the maximum likelihood criterion, each candidate position converges to its local residual minimum.

[0037]

[0038] In the formula, This represents the global equivalent spatial virtual aperture observation vector. all A concatenated vector of theoretically synthesized channel responses from clamped antennas. For the user's three-dimensional position coordinates, This is the transmit pilot power.

[0039] The present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory, characterized in that the processor executes the computer program to implement the steps of the above method.

[0040] The beneficial effects are as follows: Compared with the prior art, the present invention has the following significant advantages:

[0041] (1) The present invention divides the positioning process into three progressive stages: coarse amplitude estimation based on received signal strength, candidate location search based on spatial phase coherence spectrum, and refined estimation based on continuous optimization of multiple starting points. This effectively overcomes the limitations of grid resolution and local optima, and achieves high-precision three-dimensional positioning.

[0042] (2) The present invention adopts a time-division activation mechanism, and the entire base station only needs to be equipped with A parallel RF receiving link activates each clamped antenna sequentially through multiple time slots, effectively constructing a system containing... A virtual antenna array with spatial sampling positions improves positioning accuracy while reducing hardware complexity;

[0043] (3) The present invention establishes a comprehensive channel model that couples the internal electrical transmission channel of the leaky cable with the spatial wireless propagation channel, and introduces a radiation direction factor related to the user's spatial location into the spatial wireless propagation channel, so that the comprehensive channel response can reflect the user's height direction change characteristics, thereby giving the system three-dimensional positioning capability. Attached Figure Description

[0044] Figure 1 This is a flowchart of a three-dimensional positioning method based on time-division activation of multiple leaky cable and multiple clamping antennas according to the present invention;

[0045] Figure 2 This is a schematic diagram of the spatial deployment structure of multiple leaky cables and clamping antennas in an embodiment of the present invention;

[0046] Figure 3 This is a performance comparison curve of the root mean square error of different positioning algorithms as a function of signal-to-noise ratio in the embodiments of the present invention. Detailed Implementation

[0047] To explain the technical content, objectives, and effects of the present invention in detail, the following description is provided in conjunction with the embodiments and accompanying drawings.

[0048] like Figure 1 As shown, the present invention provides a three-dimensional positioning method based on time-division activation of multiple leaky cable and multiple clamped antennas, including system deployment, time-division activation observation, observation vector construction, amplitude coarse positioning, phase coherence spectrum candidate position search, multi-starting point continuous optimization, and final position determination.

[0049] Step S1, in this embodiment, as follows: Figure 2 As shown, the size of the space to be positioned is At a high altitude Ceiling plan installation Root parallel to Leaking cables along the axial direction, each leaking cable along Equally spaced along the axial direction, with an erection length of [missing information]. The upper edge of each leaky cable Uniform installation along the axial direction There are one clamping antenna, and the spacing between two adjacent clamping antennas is... The first edge distance is set to .

[0050] Space co-deployment The clamping antenna, the first The first leaky cable The three-dimensional physical coordinates of the clamped antenna are as follows:

[0051]

[0052] in, For the first The leaky cable Axis physical coordinates. Each leaky cable's left end connects to a single RF receiving link; the base station needs to be equipped with a total of [number missing] [units missing]. RF receiving link.

[0053] The physical prior information configured before the system went live included the coordinates of all 40 antennas. Attenuation constant of electrical signal inside the leaky cable dB / m, carrier operating frequency From this, the free space wavelength can be calculated. The relative permittivity of the leaky cable medium is... The signal propagation constant inside the cable is then calculated as follows:

[0054] rad / m

[0055] Step S2, the system positioning cycle includes Each time slot. In each time slot Inside, the base station control terminal synchronously activates all four leaky cables with the corresponding numbers via an electronic control switch. The clamping antenna.

[0056] When space When multiple users simultaneously transmit uplink signals, each user uses a mutually orthogonal dedicated pilot sequence. Satisfies the orthogonality condition: (when If the value is 0, then the value is 0; otherwise, it is 0.

[0057] The entire base station only needs to be equipped with A parallel RF receiving link is constructed, and after 10 time slots, an equivalent network with [missing information] is built. A virtual antenna array with 3D spatial resolution.

[0058] Step S3: The base station receiver uses the user The dedicated pilot sequence is used to perform cross-correlation matched filtering on the received signals of the four channels, separating the complex observation scalars of each user. (The last part, "user," appears to be incomplete and lacks context. It's unclear what "user" refers to.) The complex observation scalars sampled from all 4 leaky cables and all 10 time slots are concatenated in spatiotemporal order to construct a globally equivalent spatial virtual aperture observation vector:

[0059]

[0060] The theoretical physical model corresponding to this observation vector can be constructed by the following equation: the user's three-dimensional coordinates are... , No. Theoretical Synthetic Channel Response at Each Antenna Electrical transmission channel within the leaky cable and the spatial wireless propagation channel between the clamped antenna and the user Product construction:

[0061]

[0062] in, The model is as follows:

[0063]

[0064] The model is as follows:

[0065]

[0066] In the formula, rad / m is the free space phase constant. This is the normalized path loss coefficient. For users to the first The Euclidean distance between the clamped antennas The Lambert cosine radiation factor reflects the physical characteristics of downward directional radiation from the leaky cable.

[0067] Step S4: The established amplitude fitting cost function without phase information is a function based on the amplitude least squares criterion. By performing a global traversal search of discrete coarse grid points within the restricted area, the user's initial position estimation result is obtained. :

[0068]

[0069] in, This represents the transmit pilot power. In this embodiment, the coarse mesh set... With a step size of 0.2m, a global traversal and matching process is performed within the space to be located. Global equivalent spatial virtual aperture observation vector The One element, For grid points The theoretical physical channel at that location.

[0070] Step S5: Using the user's initial 3D position estimation result obtained in step S4. A three-dimensional local fine search region is established centered on [the target region]. The half-width is set to .

[0071] Within a local area, a high-density fine grid with a spacing less than a quarter wavelength is created. For each fine grid point... The formula for calculating the normalized spatial phase coherence spectrum projection measure is:

[0072]

[0073] in, For local fine grid points, The global theoretical synthesized channel response vector At grid points The response vector at that location; Defined as all A concatenated vector of theoretically synthesized channel responses from clamped antennas:

[0074]

[0075] In fine-grid retrieval, a non-maximum suppression mechanism is used for filtering. In this example, the following results are obtained: A set of spatially independent candidate locations .

[0076] Step S6: Using the five selected candidate positions as parallel initial values, perform nonlinear least squares iterative optimization in continuous space. The objective function of the iteration is defined as the sum of squared residuals of the channel fitting:

[0077]

[0078] In the In the next iteration, the fitted residual vector is calculated. And calculate the Jacobian matrix using the following analytical closed-form partial derivatives. (No. The row corresponds to the first (One clamping antenna):

[0079]

[0080]

[0081]

[0082] in , The first The horizontal and vertical coordinates of the clamped antenna on the horizontal plane; the height partial derivative component is strictly coupled to the variable height gradient effect of the Lambert cosine radiation factor through the product differentiation rule.

[0083] Based on the Jacobian matrix mentioned above, a noise power scale is constructed. gradient vector and approximate Hessian matrix :

[0084]

[0085]

[0086] Combined with damping factor With numerical regularization parameters Calculate the LM update vector:

[0087]

[0088] In this example, if the vector magnitude is updated... Exceeding step size limit Then, the scaling is truncated proportionally to prevent excessively large step jumps in the early stages of iteration.

[0089] Introducing a backtracking search mechanism to ensure iterative monotonic convergence: setting an adaptive reduction factor. ,in In this example The updated vector after step-size limiting will be processed. As a search direction, gradient probing generates new estimated points and counts. And project it onto the feasible physical space boundary. If the new residual satisfies Then accept the step and let At the same time, reduce the damping factor In this embodiment, we take If the residual cannot be reduced after backtracking, then reject the step and increase the damping. In this embodiment This reduces the update distance and redirects the update direction towards a safe gradient descent direction. In this example, the iteration stopping condition is... Or it may reach the maximum number of iterations of 50.

[0090] Step S7: After each candidate starting point converges to a local minimum, the convergent coordinates are obtained. Calculate the sum of squared residuals of the channel fitting at each convergence point, and output the one with the smallest absolute value of the residuals as the final user's 3D coordinate estimation result:

[0091]

[0092] like Figure 3 In the simulation results of the above specific embodiments, as the signal-to-noise ratio increases from 0dB to 35dB, the root mean square error of positioning based on the amplitude coarse estimation of the received signal strength decreases from 3.80m to 0.26m; the root mean square error of positioning based on the fine grid estimation of the spatial phase coherence spectrum decreases from 3.84m to 0.12m; and the root mean square error of positioning based on the multi-starting point maximum likelihood iterative refinement estimation decreases from 3.83m to 0.021m. This fully verifies the high-precision positioning capability of the three-stage progressive positioning method of the present invention in high signal-to-noise ratio scenarios.

[0093] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the principles and spirit of the present invention are included within the protection scope of the present invention.

Claims

1. A three-dimensional positioning method based on time-division activation of multi-leaked cable multi-clamping antenna, characterized in that, Includes the following steps: S1: Will A parallel leaky cable is laid at a height of The plane, with each leaky cable set Each clamping antenna; each leaky cable is connected to an RF receiving link at at least one end; the spatial three-dimensional physical coordinates of all clamping antennas, the electrical signal attenuation constant inside the cable, and the leaky cable propagation constant are obtained at the system base station; S2: Using a time-division activation mechanism, the clamping antennas with corresponding numbers on each leaky cable are activated sequentially in each time slot to receive uplink pilot signals sent by multiple users. Multi-leaky cable multi-time slot spatial observation information is obtained through multi-time slot sequential activation, which is equivalent to constructing a spatial virtual antenna array. S3: Demodulate the received signals in each time slot and separate the complex observation scalars of each user; cascade the complex observation scalars obtained by sampling all leaky cables in all time slots in spatiotemporal order to construct a global equivalent spatial virtual aperture observation vector. S4: Within the three-dimensional physical boundary confinement area, establish an amplitude fitting cost function without phase information, and obtain the user's three-dimensional initial position estimation result through global search; S5: Establish a three-dimensional local fine search region centered on the user's three-dimensional initial position estimation result obtained in step S4, calculate the normalized spatial phase coherence spectrum projection measure in the local region, and use a spatial suppression mechanism to screen several candidate positions that are physically isolated from each other. S6: Using each candidate position as a parallel initial value, continuously optimize each candidate position in a continuous three-dimensional physical space so that each candidate position converges to its local residual minimum. S7: Calculate the sum of squared residuals of the channel fitting at each convergence point, and output the convergence coordinate with the smallest residual as the final user's three-dimensional coordinate estimation result.

2. The three-dimensional positioning method based on time-division activation of multiple leaky cable multi-clamping antennas according to claim 1, characterized in that: In step S1 The root leaky cable is parallel to Erected along the axial direction, and along The axial direction is evenly spaced, and the erection length is The erection height is ; On each leaky cable One clamping antenna edge Arranged along the axis, the spacing between two adjacent clamping antennas is The first edge distance is set to .

3. The three-dimensional positioning method based on time-division activation of multi-leaky cable multi-clamping antenna according to claim 1, characterized in that: In step S2, the time-division activation mechanism satisfies the following: the positioning period includes Each time slot, in any time slot Below, the base station control terminal only activates the first [number] on each leaky cable. Each clamp-on antenna transmits radio frequency signals, so the entire base station only needs to be equipped with one of these antennas. Parallel RF receiving links, and in After each time slot, the equivalent construction includes... A virtual antenna array with spatial sampling locations.

4. The three-dimensional positioning method based on time-division activation of multiple leaky cable multi-clamping antennas according to claim 1, characterized in that: In step S3, when multiple users simultaneously transmit uplink pilot signals, the base station uses a dedicated orthogonal pilot sequence corresponding to each user to perform matched filtering on the received signals in each time slot, separating the complex observation scalars of each user; and concatenates the complex observation scalars obtained by each user in all leaky cables and all time slots in spatiotemporal order to construct a global equivalent spatial virtual aperture observation vector: ； In the formula, Number the user; This refers to the number of leaky cables; The number of clamping antennas installed on each leaky cable; Indicates the first The user in the first The first leaky cable Complex observation scalars obtained at each clamped antenna.

5. The three-dimensional positioning method based on time-division activation of multi-leaky cable multi-clamping antenna according to claim 1, characterized in that: In step S3, the user's three-dimensional position coordinates are: , No. Theoretical Synthetic Channel Response at Each Clamped Antenna Electrical transmission channel within the leaky cable and the spatial wireless propagation channel between the clamped antenna and the user Product construction: ； in, The model is as follows: ； The model is as follows: ； ； In the formula, and These are the attenuation constant within the leaky cable and the propagation constant of the leaky cable, respectively. To hold the antenna spacing, This is the edge distance at the beginning of the leaky cable. This is the normalized path loss coefficient. The phase constant in free space. For users to the first The Euclidean distance between the clamped antennas For users relative to the first The elevation angle of the horizontal plane holding the antenna. The height at which the parallel cable is installed. User's three-dimensional position coordinates The height coordinate component, is the Lambert cosine radiation factor. Number the leaky cable , The number of the antenna clamped on the leaky cable .

6. The three-dimensional positioning method based on time-division activation of multi-leaky cable multi-clamping antenna according to claim 1, characterized in that: In step S4, the established amplitude fitting cost function without phase information is a function based on the amplitude least squares criterion. The user's initial position estimation result is obtained by performing a global traversal search on the discrete coarse grid points within the restricted area. : ； in, It is a set of coarse uniform grid points. For transmitting pilot power, Global equivalent spatial virtual aperture observation vector The One element, For grid points The theoretical physical channel at that location.

7. A three-dimensional positioning method based on time-division activation of a multi-leaky cable multi-clamping antenna according to claim 1, characterized in that: In step S5, the formula for calculating the normalized spatial phase coherence spectrum projection measure is: ； in, For local fine grid points, The global theoretical synthesis channel response vector At grid points The response vector at that location; Defined as all A concatenated vector of theoretically synthesized channel responses from clamped antennas: ； In fine-grid retrieval, a non-maximum suppression mechanism is used for screening to obtain multiple spatially independent candidate positions.

8. A three-dimensional positioning method based on time-division activation of a multi-leaky cable multi-clamping antenna according to claim 1, characterized in that: In step S6, the multiple candidate positions are used as parallel initial values, and continuous spatial iterative optimization is performed on each candidate position. The iterative optimization process is implemented using a nonlinear least squares iterative optimization algorithm based on gradient information. During the iteration process, an adaptive step size control mechanism and a damping factor adjustment mechanism are combined. By minimizing the objective function of the sum of squared residuals of the channel fitting based on the maximum likelihood criterion, each candidate position converges to its local residual minimum. ； In the formula, This represents the global equivalent spatial virtual aperture observation vector. all A concatenated vector of theoretically synthesized channel responses from clamped antennas. For the user's three-dimensional position coordinates, This is the transmit pilot power.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 8.

Images