Passive coding equipment positioning and tracking method and device

Abstract

The invention provides a passive coding equipment positioning and tracking method and device, and relates to the technical field of computers, and the method comprises the steps: obtaining radio frequency signal original data reflected by all passive coding equipment disposed in a monitoring area; carrying out feature separation according to the original data of the radio frequency signal to obtain a signal baseband feature set; performing propagation path derivation according to the baseband feature set to obtain a channel impulse response describing environmental reflection characteristics; constructing an equipment position probability field according to the channel impulse response to obtain the position probability distribution of the equipment in the monitoring area; performing motion trail function fitting according to the position probability distribution to obtain a space-time trail function of the equipment; and trajectory optimization under physical constraint is carried out according to the space-time trajectory function, and a positioning tracking result is obtained. According to the invention, high-precision, stable and continuous positioning tracking of the passive coding equipment in a complex electromagnetic environment is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer, in particular, to a passive coding device positioning tracking method and device. BACKGROUND

[0002] The passive coding device positioning tracking technology has important application value in the fields of Internet of Things and intelligent warehousing. The technology usually relies on parameter analysis of device reflection signals to realize spatial position solution. However, in a complex indoor environment, factors such as dense multipath effect, dynamic environmental changes and weak device reflection signals seriously restrict the accuracy and robustness of existing positioning methods. How to realize continuous, stable and high-precision tracking of devices under high interference and low signal-to-noise ratio conditions is a core technical challenge. Existing solutions mainly focus on signal arrival time, arrival angle or signal strength, and use filtering, data fusion or geometric positioning methods for processing. However, these methods have the following limitations: first, they usually adopt suppression or rejection strategies for multipath signals, failing to effectively utilize the environmental structure information contained in the multipath; second, the positioning process often relies on ideal propagation models or a large number of preset environmental parameters, and the performance significantly decreases when the model is mismatched or the environment changes; third, trajectory generation relies mainly on smoothing and fitting of discrete positioning points, lacking a mechanism for overall constraint and optimization from the signal level and physical propagation law, resulting in problems such as trajectory jumping or deviation from the actual physical path in complex scenarios.

[0003] Based on the above-mentioned shortcomings of the prior art, there is an urgent need for a passive coding device positioning tracking method and device. SUMMARY

[0004] The purpose of the present application is to provide a passive coding device positioning tracking method to improve the above-mentioned problems. In order to achieve the above-mentioned purpose, the technical solution adopted by the present application is as follows:

[0005] In a first aspect, the present application provides a passive coding device positioning tracking method, comprising:

[0006] Obtaining raw radio frequency signal data reflected by all passive coding devices deployed in a monitoring area, the raw radio frequency signal data including signal amplitude, phase, time difference of arrival and device unique coding identifier;

[0007] Performing feature separation according to the raw radio frequency signal data to obtain a signal baseband feature set;

[0008] Performing propagation path derivation according to the baseband feature set, using the phase and amplitude constraint relationship between the projection components to reversely derive the geometric structure of the electromagnetic wave reflection path of the generated components, and obtaining channel impulse responses describing environmental reflection characteristics;

[0009] constructing a device position probability field according to the channel impulse response, by regarding the end point of each reflection path as a potential signal source emission position, and calculating the joint probability of existence of all the potential signal source emission positions based on path delay and attenuation, to obtain a position probability distribution of the device in the monitoring area;

[0010] fitting a motion trajectory function according to the position probability distribution, to obtain a space-time trajectory function of the device;

[0011] optimizing the trajectory under physical constraints according to the space-time trajectory function, to obtain a positioning tracking result.

[0012] In a second aspect, the present application further provides a passive coded device positioning tracking device, comprising:

[0013] an acquisition module, configured to acquire radio frequency signal original data reflected by all passive coded devices deployed in a monitoring area, wherein the radio frequency signal original data comprises signal amplitude, phase, time difference of arrival and device unique coding identification;

[0014] a separation module, configured to perform feature separation according to the radio frequency signal original data, to obtain a signal baseband feature set;

[0015] a derivation module, configured to perform propagation path derivation according to the baseband feature set, to inversely derive the geometric structure of the electromagnetic wave reflection path of the generated component by using the phase and amplitude constraint relationship between the projection components, to obtain a channel impulse response describing the environmental reflection characteristics;

[0016] a construction module, configured to construct a device position probability field according to the channel impulse response, by regarding the end point of each reflection path as a potential signal source emission position, and calculating the joint probability of existence of all the potential signal source emission positions based on path delay and attenuation, to obtain a position probability distribution of the device in the monitoring area;

[0017] a fitting module, configured to fit a motion trajectory function according to the position probability distribution, to obtain a space-time trajectory function of the device;

[0018] an optimization module, configured to optimize the trajectory under physical constraints according to the space-time trajectory function, to obtain a positioning tracking result.

[0019] The present application has the following beneficial effects:

[0020] The present application converts multipath signal interference into effective information describing environmental reflection characteristics, and realizes signal separation and probability field construction based on device coding characteristics, and finally performs trajectory optimization under strict physical propagation law constraints, so as to realize high-precision, stable and continuous positioning tracking of passive coded devices in a complex electromagnetic environment. BRIEF DESCRIPTION OF DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be considered as limiting the scope. For those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0022] Figure 1 A flow chart of a passive coding device positioning tracking method described in an embodiment of the present application;

[0023] Figure 2 A structure diagram of a passive coding device positioning tracking device described in an embodiment of the present application;

[0024] Figure 3 A structure diagram of a passive coding device positioning tracking device described in an embodiment of the present application.

[0025] In the figure, 800 is a passive coding device positioning tracking device; 801 is a processor; 802 is a memory; 803 is a multimedia component; 804 is an I / O interface; 805 is a communication component; 901 is an acquisition module; 902 is a separation module; 903 is a derivation module; 904 is a construction module; 905 is a fitting module; and 906 is an optimization module. DETAILED DESCRIPTION

[0026] In order to make the objectives, technical solutions and advantages of the embodiments of the present application clearer, the following will combine the drawings in the embodiments of the present application to make a clear and complete description of the technical solutions in the embodiments of the present application. Obviously, the described embodiments are some of the embodiments of the present application, but not all the embodiments. The components of the embodiments of the present application described and shown in the drawings can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0027] It should be noted that: similar labels and letters represent similar items in the following drawings, and therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. Meanwhile, in the description of the present application, the terms "first", "second", etc. are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0028] Embodiment 1:

[0029] The embodiment provides a passive coded device positioning tracking method.

[0030] Referring to Figure 1 , the method comprises steps S100 to S600.

[0031] In step S100, original data of radio frequency signals reflected by all passive coded devices arranged in a monitoring area is acquired, and the original data of the radio frequency signals comprises signal amplitude, phase, time difference of arrival and device unique coding identification.

[0032] It can be understood that in a complex indoor monitoring scene, the radio frequency signals reflected by the passive devices carry all information including device identity, propagation distance and environmental interaction. The original data containing signal amplitude, phase, time difference of arrival and device unique coding identification is acquired to avoid introducing any pre-processing that may cause information loss in the initial stage, and to ensure that subsequent processing can be based on the most real and complete data basis. Especially in the automatic scene of bulk material management, for example, in a mine material warehouse, the basic data acquisition of step S100 is realized in the following specific manner: a plurality of radio frequency reader nodes are regularly arranged on the ceiling and four walls of the warehouse to form a full-coverage monitoring network. When the passive coded devices attached to the material pile or the transportation device enter the area, the radio frequency signals continuously transmitted by the reader excite these passive devices, so that the passive devices reflect the signals carrying their unique coding identification back. Each reader node synchronously receives these reflected signals and accurately measures the amplitude and phase information of the signals by using coherent reception technology. By calculating the time difference of signal arrival at different reader nodes and combining the known node coordinates, basic time delay information is provided for subsequent positioning. This process directly acquires original observation data containing device identity (coding), distance (phase, time delay) and signal strength (amplitude), and lays a data foundation for accurate tracking in a complex bulk material environment full of obstructions and reflections.

[0033] In step S200, feature separation is performed according to the original data of the radio frequency signals to obtain a signal baseband feature set.

[0034] The feature separation processing of step S200 is to extract the essential features closely related to each independent device from the mixed original data of the radio frequency signals. In the typical scene of multiple device concurrency and multipath interference, it is extremely difficult to directly use the original signal for positioning calculation. This step uses the structured information contained in the device unique coding identification to transform the signal from the original time-frequency domain to a feature domain that can better highlight the individual differences of the devices, thereby creating conditions for subsequent fine processing of device and path separation, and essentially completing the purification from “mixed signal” to “device feature”.

[0035] Step S300, propagation path derivation according to the baseband feature set, using the phase and amplitude constraint relationship between each projection component, inversely deriving the geometric structure of the electromagnetic wave reflection path of the generated component to obtain the channel impulse response describing the environmental reflection characteristics;

[0036] It should be noted that the propagation path derivation process of step S300 realizes one-time view conversion, i.e., from analyzing the characteristics of the signal itself to inverting the propagation history of the signal in space. This step realizes that the phase and amplitude constraints between each projection component in the signal baseband feature set are not random, but are determined by the physical law of electromagnetic wave propagation in space. By inversely deriving the reflection path of the electromagnetic wave using these constraint relationships, the method links the interpretation of signal characteristics to specific spatial geometry, and converts the multipath effect that is difficult to use directly into valuable information for building an environmental channel model.

[0037] Step S400, device location probability field construction according to the channel impulse response, by regarding the endpoint of each reflection path as a potential signal source transmission location, and calculating the joint probability of the existence of all potential signal source transmission locations based on path delay and attenuation, to obtain the location probability distribution of the device in the monitoring area;

[0038] Specifically, the device location probability field construction of step S400 introduces a spatial positioning idea based on probability theory. By regarding the endpoint of the path as a potential signal source transmission location, and comprehensively calculating the joint probability based on the delay and attenuation information of all paths, this process converts the positioning problem into an estimation problem of spatial probability distribution, and the result (location probability distribution) can better reflect the inherent uncertainty of location information in a complex environment.

[0039] Step S500, motion trajectory function fitting according to the location probability distribution, to obtain the spatio-temporal trajectory function of the device;

[0040] It can be understood that the motion trajectory function fitting of step S500 aims to improve the discrete and possibly discontinuous instantaneous position estimate to a continuous and smooth spatio-temporal trajectory. This step takes into account the inertial constraint of device motion in the physical world, i.e., the position change of the device should be continuous in the time dimension. By fitting the discrete-time position probability distribution to a continuous spatio-temporal trajectory function, a leap from static "point" estimation to dynamic "path" description is achieved, providing a mathematical expression for analyzing the continuous motion behavior of the device.

[0041] Step S600, trajectory optimization under physical constraints according to the spatio-temporal trajectory function, to obtain the positioning tracking result.

[0042] The trajectory optimization under physical constraints in step S600 is a key link to ensure that the final result conforms to the reality of the physical world. This step introduces a closed-loop verification and correction mechanism: the initially fitted space-time trajectory function is substituted into the electromagnetic wave propagation model to calculate the theoretically observed signal, which is then compared with the channel impulse response derived in practice. By minimizing the difference between the two, the trajectory is optimized so that the final positioning and tracking result is not only smooth in mathematics, but more importantly, it is consistent with the reflection characteristics and wave propagation laws of the physical environment.

[0043] Further, step S200 includes steps S210 to S230.

[0044] Step S210, according to the radio frequency signal original data, the coding feature space construction processing is carried out, the device unique coding mark is parsed into the orthogonal base function, and the low-dimensional subspace is formed by the linear combination of the base function, and the device coding feature space is obtained;

[0045] Step S220, according to the device coding feature space and the radio frequency signal original data, signal interference component stripping processing is carried out, the projection residual of the received signal on the coding feature space is calculated, and the component related to multipath scattering in the residual is modeled as structured noise of the environment, to obtain the preliminary purified signal projection;

[0046] Step S230, according to the preliminary purified signal projection, baseband feature extraction processing is carried out, by solving the optimization problem of maximizing the energy of the projection component, the signal component corresponding to each coding mark is separated out, and the signal baseband feature set is obtained.

[0047] In particular, the encoding feature space construction process of step S210 first parses the unique identifier of each passive encoding device into a set of mutually orthogonal basis functions, which are similar to constructing a mathematical coordinate axis system that is exclusive to the device encoding. All possible linear combinations of the encoding sequences form a low-dimensional subspace, i.e., the device encoding feature space. Next, the signal interference component stripping process of step S220 projects the received mixed radio frequency signal raw data into this feature space. The core of this process is that the signal component matching the device encoding will mainly fall within the subspace, and the residual generated after projection mainly contains the interference related to the environmental multipath scattering. The key process here is not to simply discard the residual, but to model it as an environmental structured noise with certain statistical characteristics. In a bulk material environment full of metal machines and densely stacked materials, multipath interference is not completely random, but shows a certain regularity that can be described, thereby obtaining a preliminarily purified signal projection. On this basis, the baseband feature extraction process of step S230 solves an optimization problem that maximizes the energy of the projection component. The purpose is to further filter out the signal component that best represents the target device and has the highest signal-to-noise ratio from the preliminarily purified signal projection. The optimization process is essentially to find the strongest energy aggregation direction of the signal in the feature space, thereby finally separating out the pure signal baseband feature set that clearly and uniquely corresponds to each encoding identifier. This series of progressive processes effectively overcome the challenge of serious multi-device signal aliasing and distortion in the bulk material environment.

[0048] wherein the device encoding feature space basis function orthogonal formula is represented as:

[0049] ;

[0050] The feature space projection energy maximization optimization formula is represented as:

[0051] ;

[0052] In the formula, is the unique encoding identifier of the device, which represents the ID number of the RFID tag in the bulk material management scene, used to uniquely distinguish different devices or material units; and represent the first and the first basis function, respectively, generated by encoding through a specific parsing function; represents the Kronecker function, which is 1 when , otherwise 0, ensuring that the basis functions are orthogonal to each other, thereby avoiding information redundancy in the feature space; represents the feature subspace corresponding to the encoding , which is spanned by the basis function, used to focus on the signal features of a specific device; For feature space The unit vector in the vector represents the direction of the signal projection; This represents the raw data vector of the received radio frequency signal, which includes information such as amplitude and phase. This represents the transpose operation of a vector.

[0053] Further, step S300 includes steps S310 to S330.

[0054] Step S310: Perform multipath component clustering based on the signal baseband feature set. By analyzing the phase change rate and amplitude attenuation ratio of each projection component under the same coding identifier, the components that satisfy the same spherical wavefront constraint relationship are classified as signals from the same reflection path, thus obtaining a directional multipath signal cluster.

[0055] Step S320: Perform path geometry parameter solving based on the multipath signal clusters, convert the arrival time difference and phase information of each signal cluster into a set of equations for the direction of arrival and relative time delay, and determine the incident angle and path length difference of the signal cluster relative to the receiving node array by solving the set of equations to obtain the reflection path set.

[0056] This system of equations relates the observations (time difference of arrival and phase difference) of the same multipath signal cluster at different receiving nodes to the geometric parameters to be determined (direction of arrival vector and path length difference). The expression is:

[0057] Observation equation:

[0058] ;

[0059] Geometric constraint equations:

[0060] ;

[0061] In the formula, Indicates the signal has arrived at the first The and the first The time difference between each receiving node, measured by signal processing, reflects the difference in the length of the signal propagation path. Indicates the signal has arrived at the first The and the first The received carrier phase difference, measured by signal processing, provides path length difference information with higher precision than the time difference. The speed at which electromagnetic waves propagate in the environment; The carrier wavelength of the radio frequency signal; and This indicates that the multipath signal travels from the virtual source (i.e., the reflection point or device) to the first... The and the first The absolute path length of each receiving node; and is a residual term in the observation equation, representing the measurement uncertainty caused by noise, scattering, etc. in the bulk environment; is a three-dimensional coordinate vector of the virtual source in space; is a known fixed coordinate vector of the th receiving node in space; is a virtual reflection point coordinate on the th main reflecting surface in the environment; is a reference point coordinate of the receiving node array; is the integer ambiguity in the phase difference observation.

[0062] Step S330, according to the reflection path set, performs an environment scatterer position inversion process, takes each path as a ray from the device to the receiving node, and jointly calculates the virtual position of the reflecting surface in the environment by using the path incidence angle and length difference, to obtain a channel impulse response containing the direct path and the reflection path.

[0063] Specifically, the multipath component clustering processing of step S310 takes the signal baseband feature set separated in step S200 as input, and its core operation is to analyze the phase variation and amplitude attenuation patterns of multiple signal projection components under the same device code on multiple receiving nodes; in practical applications, the phase variation and amplitude attenuation of signal components from the same physical reflection path (e.g. reflection through warehouse walls or fixed mechanical surfaces) will satisfy certain spherical wave front propagation constraints, that is, the relationship between them conforms to a spherical wave model with the reflection point or the transmission source as the spherical center, and components with consistent performance are classified into the same category through this constraint, so as to separate the mixed signals into several multipath signal clusters with directional characteristics corresponding to different propagation paths. The path geometry parameter solving processing of step S320 further excavates the spatial information contained in each multipath signal cluster, which converts the time difference of arrival and accurate phase information between different receiving nodes in a signal cluster into a set of mathematical equations about the signal direction of arrival and relative transmission time delay; solving the equation set can quantitatively determine the spatial incident angle of each signal cluster (i.e. each path) relative to the receiving node array and the length difference between the path and the reference path, so as to convert the signal features into a reflection path set describing the geometric properties of the reflection path. The environment scatterer position inversion processing of step S330 abstracts each path in the reflection path set as a ray from the device to the receiving node, and based on the path incident angle and length difference obtained in step S320, uses the ray tracing idea in geometric optics to jointly calculate the virtual position of the medium (such as walls, large equipment and other fixed obstacles) in the environment that causes the reflection; this processing is essentially a process of "backstepping" the environment structure using multipath signals, and finally outputs a channel impulse response that not only contains the direct path but also more completely describes the main reflection path, thereby laying the foundation for accurate positioning in complex scenes.

[0064] Further, step S400 includes step S410 to step S430.

[0065] Step S410, according to the channel impulse response, a virtual source spatial mapping processing is performed, the time delay of arrival of each reflection path is converted into an ellipsoidal surface with the receiving node as the focus, the distance from each point on the ellipsoidal surface to the focus is equal to the path length, and the path is mapped into an ellipsoidal positioning area in space, obtaining an ellipsoidal surface set representing the potential source position;

[0066] Step S420, according to the ellipsoidal surface set, a source position hypothesis conflict resolution processing is performed, the intersection of different ellipsoidal surfaces in the space region formed by two receiving nodes is calculated, and the source existence probability of the intersection region is weighted based on the preset signal attenuation model, and the candidate source position points are selected;

[0067] Step S430, according to the candidate source position point, a device position probability field synthesis processing is performed, each candidate point and its weight are regarded as a kernel function, a continuous probability distribution is synthesized in the monitoring area through a non-parametric density estimation method, and a position probability distribution of the device in the monitoring area is obtained.

[0068] Specifically, the virtual source space mapping processing of step S410 converts the time delay information of each reflection path in the channel impulse response into geometric positioning constraints, and the basic principle is that the total propagation path length of electromagnetic waves from the source to the receiving node is fixed, which defines an ellipsoidal surface with the receiving node as the focus point, and the distance from any point on the ellipsoidal surface to the two foci is equal to the measured path length. Through this conversion, each path is no longer only time delay data, but is mapped to an ellipsoidal positioning area in space, and all possible source positions are located on the ellipsoidal surface, so that a set composed of multiple ellipsoidal surfaces is obtained, which collectively represents the potential device position space distribution. The source position hypothesis conflict resolution processing of step S420 aims to solve the uncertainty of the intersection of these ellipsoidal surfaces. In environments such as bulk storage warehouses where multipath effects are significant, a single ellipsoidal surface may contain a large number of invalid position points. This step calculates the intersection curve or area of the ellipsoidal surfaces corresponding to each pair of receiving nodes in three-dimensional space, and weights these intersection areas according to the signal attenuation model (for example, the path loss model) in the propagation process. The logic is that the intersection points with signal attenuation consistent with the propagation distance are more likely to be the true source position, so that a set of candidate source position points with the highest weighted score is selected from a large number of spatial intersection points. The device position probability field synthesis processing of step S430 finally converts discrete position estimates into a continuous probability distribution, which regards each weighted candidate point obtained in the previous step as the center of a kernel function of a probability density function, and the higher the weight, the greater the probability density contribution at that point. Through a non-parametric density estimation method, the probability contributions of all candidate points are superimposed and smoothed in the entire monitoring space to synthesize a continuous probability distribution field. This probability distribution intuitively reflects the possibility of the existence of the device at any position in space, and provides a probabilistic basis for subsequent trajectory tracking.

[0069] Further, step S500 includes step S510 to step S530.

[0070] Step S510, according to the position probability distribution, a space-time correlation field construction processing is performed, a probability field describing the possibility of the existence of the device in the joint space-time domain is constructed by performing tensor product operation on the position probability distribution at continuous multiple time points along the time axis, and space-time probability volume data is obtained;

[0071] Step S520, path flow extraction processing is performed according to the spatiotemporal probability volume data, a spatial curve that meets the local maximum of the probability distribution and has the minimum change in the tangent direction of the path is solved in the volume data, the trajectory fitting problem is converted into a geodesic line problem of solving the curve on the spatiotemporal manifold, and an initial spatiotemporal path of the device is obtained;

[0072] Step S530, trajectory function regularization processing is performed according to the initial spatiotemporal path, inertia constraint of device motion is introduced as a regularization term to smooth and correct the initial path, the first derivative and the second derivative of the trajectory function are made to conform to the physical law, and a spatiotemporal trajectory function of the device is obtained.

[0073] In the dynamic tracking scene, the spatiotemporal correlation field construction processing of step S510 integrates the discrete time dimension into the positioning model. It fuses the position probability distribution of continuous multiple time points into a four-dimensional probability field composed of three-dimensional space and one-dimensional time through tensor product operation, forming spatiotemporal probability volume data. This data structure completely retains the spatiotemporal evolution characteristics of the position uncertainty of the device in the bulk material transportation process, such as the signal shielding change law caused by the movement of the material pile. The path flow extraction processing of step S520 is optimized for the characteristics of the spatiotemporal probability volume data. The core innovation of this method is to convert the trajectory finding problem into solving the optimal path flow in the probability field: finding a spatial curve that meets the local maximum value constraint of the spatial probability (high-likelihood position) and the minimum change constraint of the path direction (motion smoothness). By establishing a spatiotemporal manifold model and converting the trajectory fitting problem into a geodesic line solving problem, the trajectory breaking problem caused by the temporary shielding of the device (such as being buried in the material pile) is effectively overcome. The trajectory function regularization processing of step S530 introduces physical motion constraints as a mathematical correction mechanism. In the transportation scene, the device usually moves smoothly with the conveyor belt or the material flow, and its acceleration has a physical upper limit. This step performs variational optimization on the initial path through the inertia constraint regularization term, forcing the first derivative (velocity) and the second derivative (acceleration) of the trajectory function to meet the typical motion characteristics of the material conveying device, such as eliminating the non-physical position jump that may occur when the hopper is discharging, and finally generating a spatiotemporal trajectory function that conforms to the motion law of the bulk material automatic system.

[0074] Further, step S600 includes steps S610 to S630.

[0075] Step S610, virtual signal forward generation processing is performed according to the spatiotemporal trajectory function, each point on the trajectory is regarded as a signal source, the theoretical waveform of the signal source should generate, including direct wave and multipath signal reflected by the environment fixed obstacle, is calculated based on the electromagnetic wave propagation principle, and a set of theoretical channel impulse responses corresponding to the trajectory is obtained;

[0076] Step S620, difference field construction processing is performed according to the set of theoretical channel impulse responses and the channel impulse response, a space-time difference distribution field representing the degree of coincidence between the trajectory and the physical reality is generated by comparing the amplitude, phase and time delay residuals of the theoretical waveform and the actual signal point by point;

[0077] Step S630, trajectory function variational correction processing is performed according to the space-time difference distribution field, and a positioning tracking result conforming to the wave propagation law is obtained.

[0078] In the physical verification stage of positioning tracking, the virtual signal forward generation processing of step S610 regards each space-time point on the space-time trajectory function as a virtual signal source. Based on the propagation characteristics of electromagnetic waves in a bulk material environment (such as metal warehouse wall reflection, material shielding attenuation, etc.), the theoretical signal waveform that the virtual source should generate is calculated, including direct wave and multipath signal reflected by known fixed obstacles. This process is essentially to convert the geometric trajectory into a quantifiable physical signal model, generating a set of theoretical channel impulse responses that completely match the current trajectory hypothesis. The difference field construction processing of step S620 establishes a physical verification mechanism. By comparing the set of theoretical channel impulse responses and the channel impulse response derived from the actual signal (output by step S300) point by point, the residuals of the two in signal amplitude, phase and arrival time delay are calculated. In the bulk material transportation scene, the shape change of the material pile will cause the reflection path to change dynamically, and the space-time difference distribution field generated in this step accurately quantifies the deviation between the trajectory hypothesis and the real physical propagation, especially capturing the abnormal deviation of the signal caused by the movement of the material pile. The trajectory function variational correction processing of step S630 realizes dynamic closed-loop optimization. Taking the space-time difference distribution field as the guide field, a target function is constructed by variational method: minimize the global residual of the theoretical signal and the actual signal while maintaining the smoothness of the trajectory. This processing forces the trajectory correction result to simultaneously satisfy the motion continuity constraint and the electromagnetic wave propagation constraint, finally eliminates the trajectory distortion caused by temporary material shielding or dynamic reflection surface, and outputs a high-confidence positioning tracking result consistent with the physical environment of the warehouse.

[0079] Further, step S630 includes steps S631 to S633.

[0080] Step S631, vector field construction processing is performed according to the space-time difference distribution field, a trajectory correction vector field is obtained by calculating the negative gradient of the difference field along the space and time directions, the direction and amplitude of the trajectory correction vector field are aimed at minimizing the signal residual;

[0081] Step S632, space-time manifold evolution processing is performed according to the trajectory correction vector field, the original space-time trajectory function is regarded as a one-dimensional manifold embedded in a space-time coordinate system, and the one-dimensional manifold is evolved along the correction vector field direction to obtain a new space-time manifold after evolution;

[0082] Step S633, optimal trajectory extraction processing under physical constraints is performed according to the new spatiotemporal manifold, and a smooth curve that satisfies the wave equation constraint and is closest to the evolved manifold is obtained as the final positioning tracking result.

[0083] In the final stage of trajectory optimization, the vector field construction processing of step S631 converts the spatiotemporal difference distribution field into a mathematical tool that can drive trajectory deformation. By calculating the negative gradient of the difference field in the three-dimensional space and time dimension, a trajectory correction vector field with clear physical meaning is generated: its spatial component indicates which spatial direction the trajectory points should move to reduce signal residual, and the time component guides the trajectory time scale adjustment, and the overall amplitude is proportional to the residual reduction amplitude. The spatiotemporal manifold evolution processing of step S632 realizes the geometric adaptive adjustment of the trajectory. The original spatiotemporal trajectory is regarded as an elastic curve (one-dimensional manifold) in four-dimensional space-time, and it is continuously deformed along the correction vector field. This process is similar to allowing the trajectory to naturally stretch in the "gravitational field" of signal residual while maintaining the topological continuity of the curve. Especially for the case where the device is temporarily buried in the bulk material environment, this evolution can automatically correct the trajectory break caused by signal interruption, forming a new spatiotemporal manifold that is more physically coherent. The optimal trajectory extraction processing of step S633 imposes decisive physical constraints. From the evolved new spatiotemporal manifold, a smooth curve that meets the double conditions is extracted: it needs to conform to the propagation law of electromagnetic waves in the bulk material environment (achieved by embedding the wave equation constraint), and it needs to maintain the minimum geometric deviation from the manifold. This processing effectively solves the positioning drift problem caused by strong reflection of metal warehouse walls and changes in material dielectric properties, and finally outputs a high-reliability tracking trajectory that meets the continuity of device motion and the explanation of electromagnetic propagation.

[0084] Embodiment 2:

[0085] As shown in the figure, the embodiment provides a passive coded device positioning tracking device, the device comprising: Figure 2 The acquisition module 901 is configured to acquire radio frequency signal original data reflected by all passive coded devices deployed in the monitoring area, and the radio frequency signal original data includes signal amplitude, phase, time difference of arrival, and device unique coding identifier.

[0086] The separation module 902 is configured to separate features according to the radio frequency signal original data to obtain a signal baseband feature set.

[0087] The derivation module 903 is configured to derive a propagation path according to the baseband feature set, and inversely derive the geometric structure of the electromagnetic wave reflection path of each projection component by using the phase and amplitude constraint relationship between the projection components to obtain a channel impulse response describing the reflection characteristics of the environment.

[0088] ​

[0089] The construction module 904 is configured to construct a device position probability field according to the channel impulse response, by regarding the end point of each reflection path as a potential signal source transmission position, and calculating a joint probability of existence of all potential signal source transmission positions based on path delay and attenuation, to obtain a position probability distribution of the device in the monitoring area;

[0090] The fitting module 905 is configured to fit a motion trajectory function according to the position probability distribution, to obtain a space-time trajectory function of the device.

[0091] The optimization module 906 is configured to optimize the trajectory under physical constraints according to the space-time trajectory function, to obtain a positioning tracking result.

[0092] In an embodiment of the present application, the separation module 902 includes:

[0093] The first separation unit is configured to construct an encoding feature space according to the radio frequency signal raw data, by parsing the device unique encoding identifier into orthogonal basis functions, and constructing a low-dimensional subspace by linear combination of the basis functions, to obtain a device encoding feature space.

[0094] The second separation unit is configured to perform signal interference component stripping processing according to the device encoding feature space and the radio frequency signal raw data, by calculating a projection residual of the received signal on the encoding feature space, and modeling the component related to multipath scattering in the residual as an environmental structured noise, to obtain a preliminary purified signal projection.

[0095] The third separation unit is configured to perform baseband feature extraction processing according to the preliminary purified signal projection, by solving an optimization problem that maximizes the energy of the projection component, to separate out a signal component corresponding to each encoding identifier, to obtain a signal baseband feature set.

[0096] In an embodiment of the present application, the derivation module 903 includes:

[0097] The first derivation unit is configured to perform multipath component clustering processing according to the signal baseband feature set, by analyzing the phase change rate and amplitude attenuation ratio of each projection component under the same encoding identifier, to classify the components satisfying the same spherical wave front constraint relationship as signals from the same reflection path, to obtain a directional multipath signal cluster.

[0098] The second derivation unit is configured to perform path geometric parameter solving processing according to the multipath signal cluster, to convert the time difference of arrival and phase information of each signal cluster into an equation set of direction of arrival and relative time delay, and determine the incident angle and path length difference of the signal cluster relative to the receiving node array by solving the equation set, to obtain a reflection path set.

[0099] The third derivation unit is used to perform environmental scatterer position inversion processing based on the set of reflection paths. Each path is regarded as a ray from the device to the receiving node, and the virtual position of the reflecting surface in the environment is calculated by jointly using the path incident angle and length difference, so as to obtain the channel impulse response including the direct path and the reflection path.

[0100] Example 3:

[0101] Corresponding to the above method embodiments, this embodiment also provides a passive coding device positioning and tracking device. The passive coding device positioning and tracking device described below and the passive coding device positioning and tracking method described above can be referred to each other.

[0102] Figure 3 This is a block diagram illustrating a passive coding device positioning and tracking device 800 according to an exemplary embodiment. Figure 3 As shown, the passive coding device positioning and tracking device 800 may include a processor 801 and a memory 802. The passive coding device positioning and tracking device 800 may also include one or more of a multimedia component 803, an I / O interface 804, and a communication component 805.

[0103] The processor 801 is configured to control overall operations of the passive coding device positioning and tracking device 800 to complete all or part of the steps in the above passive coding device positioning and tracking method. The memory 802 is configured to store various types of data to support the operation of the passive coding device positioning and tracking device 800. For example, the data can include instructions for any application or method operating on the passive coding device positioning and tracking device 800, and application-related data, such as contact data, sent and received messages, pictures, audio, video, and the like. The memory 802 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The multimedia component 803 can include a screen and an audio component. The screen can be a touch screen, for example, and the audio component is configured to output and / or input audio signals. For example, the audio component can include a microphone configured to receive external audio signals. The received audio signals can be further stored in the memory 802 or transmitted through the communication component 805. The audio component also includes at least one speaker configured to output audio signals. The I / O interface 804 provides an interface between the processor 801 and other interface modules, which can be a keyboard, a mouse, a button, and the like. The buttons can be virtual buttons or physical buttons. The communication component 805 is configured to perform wired or wireless communication between the passive coding device positioning and tracking device 800 and other devices. The wireless communication, such as Wi-Fi, Bluetooth, near field communication (NFC), 2G, 3G or 4G, or a combination of one or more of them, so the corresponding communication component 805 can include a Wi-Fi module, a Bluetooth module, an NFC module.

[0104] In an exemplary embodiment, a passive coding device positioning and tracking device 800 can be implemented by one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), controllers, micro-controllers, microprocessors, or other electronic elements for executing a passive coding device positioning and tracking method as described above.

[0105] In another exemplary embodiment, a computer readable storage medium including program instructions that, when executed by a processor, implement the steps of a passive coding device positioning and tracking method as described above is also provided. For example, the computer readable storage medium can be the memory 802 as described above including program instructions executable by the processor 801 of a passive coding device positioning and tracking device 800 to perform a passive coding device positioning and tracking method as described above.

[0106] The above description is merely illustrative of the application and is not intended to limit the scope of the application. Any modifications and changes can be made to the application without departing from the spirit and scope of the application.

Claims

1. A passive coded device positioning tracking method, characterized by, The method comprises the following steps: acquiring raw radio frequency signal data reflected by all passive coded devices deployed in a monitoring area, wherein the raw radio frequency signal data comprises signal amplitude, phase, time difference of arrival, and device unique coding identifier; performing feature separation according to the raw radio frequency signal data to obtain a signal baseband feature set; deriving a propagation path according to the baseband feature set, utilizing the phase and amplitude constraint relationship between each projection component to reversely derive the geometric structure of the electromagnetic wave reflection path of the component, and obtaining a channel impulse response describing the reflection characteristics of the environment; constructing a device position probability field according to the channel impulse response, regarding the endpoint of each reflection path as a potential signal source transmission position, and calculating the joint probability of the existence of all the potential signal source transmission positions based on path delay and attenuation to obtain the position probability distribution of the device in the monitoring area; fitting a motion trajectory function according to the position probability distribution to obtain a space-time trajectory function of the device; performing trajectory optimization under physical constraints according to the space-time trajectory function to obtain a positioning tracking result.

2. The passive coded device positioning tracking method of claim 1, wherein, The feature separation according to the raw radio frequency signal data comprises the following steps: performing coding feature space construction processing according to the raw radio frequency signal data, parsing the device unique coding identifier into orthogonal base functions, and constructing a low-dimensional subspace by linear combination of the base functions to obtain a device coding feature space; performing signal interference component stripping processing according to the device coding feature space and the raw radio frequency signal data, calculating the projection residual of the received signal on the coding feature space, and modeling the component related to multipath scattering in the residual as structured noise of the environment to obtain a preliminarily purified signal projection; performing baseband feature extraction processing according to the preliminarily purified signal projection, solving an optimization problem that maximizes the energy of the projection component to separate the signal component corresponding to each coding identifier, and obtaining a signal baseband feature set.

3. The method of claim 1, wherein, The propagation path derivation according to the baseband feature set comprises the following steps: performing multipath component clustering processing according to the signal baseband feature set, analyzing the phase change rate and amplitude attenuation ratio of each projection component under the same coding identifier, classifying the components satisfying the same spherical wave front constraint relationship as signals from the same reflection path to obtain directional multipath signal clusters; performing path geometric parameter solving processing according to the multipath signal clusters, converting the time difference of arrival and phase information of each signal cluster into an equation group of direction of arrival and relative time delay, and determining the incident angle and path length difference of the signal cluster relative to the receiving node array by solving the equation group to obtain a reflection path set; performing environment scatterer position inversion processing according to the reflection path set, regarding each path as a ray from the device to the receiving node, and jointly calculating the virtual position of the reflection surface in the environment by using the path incident angle and length difference to obtain a channel impulse response comprising a direct path and a reflection path.

4. The method of claim 1, wherein, The device position probability field construction according to the channel impulse response comprises the following steps: According to the channel impulse response, a virtual source space mapping process is performed, time delays of each reflection path are converted into ellipsoidal surfaces with the receiving node as the focus, a distance from each point on the ellipsoidal surface to the focus is equal to the path length, and the path is mapped into an ellipsoidal positioning area in space, to obtain an ellipsoidal surface set representing potential source positions; According to the ellipsoidal surface set, a source position hypothesis conflict resolution process is performed, intersections of different ellipsoidal surfaces in a space region formed by two receiving nodes are calculated, and source existence probabilities of the intersection regions are weighted based on a preset signal attenuation model, to screen candidate source position points; According to the candidate source position points, a device position probability field synthesis process is performed, each candidate point and its weight are regarded as a kernel function, a continuous probability distribution in a monitoring region is synthesized by a non-parametric density estimation method, and a position probability distribution of the device in the monitoring region is obtained.

5. The method of claim 1, wherein, According to the position probability distribution, a motion trajectory function fitting process is performed, including: According to the position probability distribution, a space-time correlation field construction process is performed, a position probability distribution at a plurality of continuous time points is subjected to a tensor product operation along a time axis, a probability field describing a possibility of the device existing in a space-time joint domain is constructed, and space-time probability volume data is obtained; According to the space-time probability volume data, a path flow extraction process is performed, a spatial curve in the volume data that satisfies a local maximum of the probability distribution and has a minimum change in a path tangent direction is solved, a trajectory fitting problem is converted into a problem of solving a geodesic line of the curve on a space-time manifold, and an initial space-time path of the device is obtained; According to the initial space-time path, a trajectory function regularization process is performed, an inertial constraint of device motion is introduced as a regularization term to perform smoothing correction on the initial path, so that first and second derivatives of the trajectory function conform to physical laws, and a space-time trajectory function of the device is obtained.

6. The method of claim 1, wherein, According to the space-time trajectory function, a trajectory optimization under physical constraints is performed, including: According to the space-time trajectory function, a virtual signal forward generation process is performed, each point on the trajectory is regarded as a signal emission source, a theoretical waveform of the signal emission source, including a direct wave and a multipath signal reflected by an environment-fixed obstacle, is calculated based on electromagnetic wave propagation principles, and a theoretical channel impulse response set corresponding to the trajectory is obtained; According to the theoretical channel impulse response set and the channel impulse response, a difference field construction process is performed, a residual error of the theoretical waveform and an actual signal in amplitude, phase and time delay is compared point by point, and a space-time difference distribution field representing a degree of consistency of the trajectory with a physical reality is generated; According to the space-time difference distribution field, a trajectory function variational correction process is performed, to obtain a positioning and tracking result conforming to wave propagation laws.

7. The method of claim 6, wherein, According to the space-time difference distribution field, a trajectory function variational correction process is performed, including: According to the space-time difference distribution field, a vector field construction process is performed, a trajectory correction vector field is obtained by calculating a negative gradient of the difference field along a space and time direction, and the trajectory correction vector field is directed to minimize a signal residual error. According to the trajectory correction vector field, a space-time manifold evolution process is performed, the original space-time trajectory function is regarded as a one-dimensional manifold embedded in a space-time coordinate system, and the one-dimensional manifold is evolved along the direction of the correction vector field to obtain a new evolved space-time manifold; According to the new space-time manifold, an optimal trajectory extraction process under physical constraints is performed, a smooth curve that satisfies the wave equation constraint and is closest to the evolved manifold is solved, and the smooth curve is taken as the final positioning tracking result.

8. A passive coded device positioning tracking apparatus, characterized by, Comprise: An acquisition module is configured to acquire original radio frequency signal data reflected by all passive coding devices deployed in a monitoring area, wherein the original radio frequency signal data comprises signal amplitude, phase, time difference of arrival, and device unique coding identifier; A separation module is configured to separate features according to the original radio frequency signal data to obtain a signal baseband feature set; A derivation module is configured to derive a propagation path according to the baseband feature set, utilize the phase and amplitude constraint relationship between each projection component, reversely derive the geometric structure of the electromagnetic wave reflection path of the generated component, and obtain a channel impulse response describing environmental reflection characteristics; A construction module is configured to construct a device location probability field according to the channel impulse response, regard the endpoint of each reflection path as a potential source transmission location, calculate the joint probability of the existence of all potential source transmission locations based on path delay and attenuation, and obtain a location probability distribution of the device in the monitoring area; A fitting module is configured to fit a motion trajectory function according to the location probability distribution to obtain a space-time trajectory function of the device; An optimization module is configured to optimize a trajectory under physical constraints according to the space-time trajectory function to obtain a positioning tracking result.

9. The passive coded device locator tracking apparatus of claim 8, wherein, The separation module comprises: A first separation unit is configured to construct an encoding feature space according to the original radio frequency signal data, parse the device unique coding identifier into orthogonal base functions, and construct a low-dimensional subspace with a linear combination of the base functions to obtain a device encoding feature space; A second separation unit is configured to separate signal interference components according to the device encoding feature space and the original radio frequency signal data, calculate the projection residual of the received signal on the encoding feature space, model the components related to multipath scattering in the residual as environmental structured noise, and obtain a preliminarily purified signal projection; A third separation unit is configured to extract baseband features according to the preliminarily purified signal projection, solve an optimization problem that maximizes the energy of the projection component, separate the signal component corresponding to each coding identifier, and obtain a signal baseband feature set.

10. The passive coded device locator tracking apparatus of claim 8, wherein, The derivation module comprises: A first derivation unit is configured to cluster multipath components according to the signal baseband feature set, analyze the phase change rate and amplitude attenuation ratio of each projection component under the same coding identifier, classify components satisfying the same spherical wave front constraint relationship as signals from the same reflection path, and obtain a directional multipath signal cluster. a second derivation unit configured to perform path geometry parameter solving processing according to the multipath signal clusters, convert the time difference of arrival and phase information of each signal cluster into an equation set of direction of arrival and relative time delay, and determine the incident angle and path length difference of the signal cluster relative to the receiving node array by solving the equation set to obtain a reflection path set; a third derivation unit configured to perform environment scatterer position inversion processing according to the reflection path set, regard each path as a ray from the device to the receiving node, and jointly calculate the virtual position of the reflection surface in the environment by using the path incident angle and length difference to obtain a channel impulse response containing the direct path and the reflection path.

Images