Intelligent reflecting surface assisted ultra-wideband indoor positioning method

Abstract

The application provides a kind of intelligent reflecting surface assisted ultra-wideband indoor positioning method, comprising the following steps: S1, constructing path encoding atlas;S2, calculating multi-user positioning difficulty index and sorting;S3, ranging resource scheduling;S4, positioning solution and atlas updating.The application can identify difficult terminals that are long-term in non-line-of-sight occlusion area or poor geometric condition area based on the linkage mechanism of path encoding atlas and multi-user positioning difficulty index, and preferentially allocate higher quality path cluster and ranging time slot to the terminals in ranging resource scheduling, so that the limited ranging resources are more reasonably distributed among different terminals, reducing the long-term low positioning accuracy and large fluctuation of some terminals.

Description

Technical Field

[0001] This invention belongs to the field of wireless communication network and indoor positioning technology, specifically relating to an ultra-wideband indoor positioning method assisted by an intelligent reflective surface. Background Technology

[0002] In the field of indoor positioning technology, ultra-wideband signal positioning is widely used for fine-grained location perception of personnel and equipment in scenarios such as warehousing and logistics, industrial production, and rail transportation due to its high time resolution and strong anti-multipath capability. With the increasing demand for multi-user, high-density terminal positioning, relying solely on the geometric layout of fixed ultra-wideband base stations is often insufficient to maintain stable positioning geometry and a sufficient number of effective ranging paths in complex obstructed environments. To improve signal accessibility and path diversity in non-line-of-sight environments, an auxiliary positioning scheme combining intelligent reflective surfaces has emerged. By deploying controllable reflective surfaces in indoor spaces, the signal propagation path structure is reconstructed to enhance coverage blind spots and improve ranging conditions in some areas.

[0003] In indoor positioning scenarios where multiple users share ultra-wideband base stations and smart reflectors, there is a mismatch between positioning resource allocation and path selection. Specifically: On the one hand, in complex environments with dynamically changing terminal locations, existing methods lack a multi-user positioning difficulty index based on the number of available paths, geometric conditions, and historical positioning errors. This makes it difficult to identify difficult terminals that have been in non-line-of-sight or geometrically degraded areas for a long time, resulting in uneven distribution of ranging resources and smart reflector configuration opportunities among different terminals. On the other hand, under the constraints of smart reflector array size, number of controllable reflective units, and path coding mode switching frequency, existing technologies lack path-level knowledge representation combined with historical positioning statistics and a ranging resource scheduling mechanism based on this representation. This makes it difficult to effectively constrain the number of paths that can be served simultaneously within a single ranging time slot and the number of path coding mode switching times within a positioning cycle. As a result, smart reflector resources are repeatedly occupied by paths with poor historical performance or low value, while paths with stable historical performance are difficult to be prioritized. Summary of the Invention

[0004] The main objective of this invention is to provide an ultra-wideband indoor positioning method assisted by an intelligent reflective surface, addressing the aforementioned problems.

[0005] Therefore, the above-mentioned objective of the present invention is achieved through the following technical solution:

[0006] A smart reflective surface-assisted ultra-wideband indoor positioning method includes the following steps:

[0007] S1. Constructing a path coding map: Obtain spatial configuration information of the ultra-wideband base station, smart reflector, and reference area; control the ultra-wideband base station and smart reflector to perform signal transmission and reflection switching according to the preset path coding mode during the training phase; collect training observation data and perform path clustering processing to generate multiple path clusters; construct a path coding map based on each path cluster and its corresponding path coding mode, ultra-wideband base station, smart reflector, and reference area.

[0008] Among them, a path cluster is a set of multiple signal propagation paths with similar propagation delay and energy distribution characteristics under the conditions of a fixed ultra-wideband base station, smart reflector, and reference area; the path coding map is a data graph structure with path clusters as nodes and records the relationship between each path cluster and its path coding mode, as well as the associated ultra-wideband base station, smart reflector, and reference area.

[0009] S2. Calculate and sort the multi-user positioning difficulty index: Calculate the multi-user positioning difficulty index based on the positioning observation data and tag status information of the ultra-wideband terminal tags, and sort the ultra-wideband terminal tags according to the multi-user positioning difficulty index to obtain the multi-user positioning priority queue.

[0010] S3. Ranging resource scheduling: Based on the multi-user positioning priority queue, the target UWB terminal tag is determined. Candidate path clusters corresponding to the location of the target UWB terminal tag are retrieved from the path coding map. The target path cluster is selected from the candidate path clusters. A ranging resource scheduling plan is generated based on the target path cluster. The UWB base station and the intelligent reflector are combined with the intelligent reflector resource constraint control to execute the ranging resource scheduling plan.

[0011] S4. Positioning Calculation and Map Update: Obtain the positioning observation data of the target UWB terminal tag, perform positioning calculation in combination with the spatial configuration information of the corresponding UWB base station and smart reflector, obtain the updated positioning result of the target UWB terminal tag, and update the path coding map based on the updated positioning result and the positioning observation data corresponding to the target path cluster.

[0012] While adopting the above technical solutions, the present invention may also adopt or combine the following technical solutions:

[0013] As a preferred technical solution of the present invention: In step S1, the training observation data consists of multi-base station ranging data, signal arrival time data, and channel impulse response data obtained when the ultra-wideband terminal tag performs signal transmission and reception interaction with multiple ultra-wideband base stations and smart reflectors in accordance with a preset path coding mode during the training phase; the path cluster is registered in the path coding map with a path cluster identifier, and serves as a collection index for candidate path clusters, target path cluster identifiers, and historical statistical information of path clusters. The path cluster identifier corresponds to the set of signal propagation paths belonging to that path cluster in the training observation data.

[0014] As a preferred technical solution of the present invention: In step S1, the path coding map is a data map structure constructed with the set of nodes composed of path cluster identifiers as nodes and the connection relationship between each path cluster and its corresponding path coding mode, ultra-wideband base station, smart reflector and reference area as the edge set, and serves as the unified index basis for path clusters.

[0015] As a preferred technical solution of the present invention: in step S2, the multi-user positioning difficulty index includes: the number of available path clusters, the ranging geometric distribution index and signal-to-noise ratio index corresponding to the available path clusters, the historical positioning error statistics of the path clusters, and the motion state parameters.

[0016] As a preferred technical solution of the present invention: In step S3, the candidate path cluster is a set of path clusters selected from the path coding map that are associated with the reference area to which the current position and the predicted position of the target ultra-wideband terminal tag belong within the target positioning period. The method for retrieving candidate path clusters from the path coding map is as follows: determine the reference area identifier based on the current position and the predicted position of the target ultra-wideband terminal tag, search for path clusters in the path coding map that correspond to the reference area identifier and whose historical positioning statistics meet the preset reliability conditions, and form candidate path clusters from the path clusters that meet the preset reliability conditions; the target path cluster is a set of path clusters selected from the candidate path clusters for ranging and positioning calculation of the target ultra-wideband terminal tag.

[0017] As a preferred technical solution of the present invention: In step S3, the ranging resource scheduling plan is a scheduling scheme based on the target path cluster allocating ultra-wideband base station ranging operations and intelligent reflector path coding mode configuration status to each ranging time slot within the target positioning period. It is used to coordinate each ultra-wideband base station and each intelligent reflector to perform the ranging process according to the predetermined time slot sequence and path coding mode within the target positioning period, and to determine that the signal propagation path associated with each target path cluster is in an available state within the corresponding ranging time slot. The ranging resource scheduling plan includes a ranging time slot allocation scheme and an intelligent reflector path coding mode switching scheme. The ranging time slot allocation scheme is to allocate each ranging time slot to the combination of ultra-wideband base station and intelligent reflector associated with the target path cluster within the target positioning period. The intelligent reflector path coding mode switching scheme is to determine the switching time and switching target of the intelligent reflector path coding mode between adjacent ranging time slots.

[0018] As a preferred technical solution of the present invention: In step S3, the intelligent reflector resource constraint is a resource restriction condition determined based on the array size of the intelligent reflector, the number of independently controllable reflector units, and the control update rate. It is used to limit the number of target path clusters that the intelligent reflector can simultaneously serve within the target positioning period and the allowed path coding mode switching frequency. Specifically, the intelligent reflector resource constraint is as follows: the number of target path clusters allocated to the same intelligent reflector in any ranging time slot does not exceed a first preset upper limit, and the number of path coding mode switching times of the intelligent reflector within a target positioning period does not exceed a second preset upper limit.

[0019] As a preferred technical solution of the present invention: In step S4, the positioning observation data of the target ultra-wideband terminal tag is the set of observation data collected by the target ultra-wideband terminal tag during the target positioning period when it performs the ranging process with multiple ultra-wideband base stations and smart reflectors according to the ranging resource scheduling plan; the positioning observation data corresponding to the target path cluster is a subset of observation data selected from the positioning observation data of the target ultra-wideband terminal tag; wherein, the subset of observation data is associated with the signal propagation path corresponding to the ultra-wideband base station and smart reflector associated with the target path cluster in the path coding map, and corresponds one-to-one with the ranging time slot allocated to the target path cluster in the ranging resource scheduling plan.

[0020] As a preferred technical solution of the present invention: In step S4, the updated positioning result of the target ultra-wideband terminal tag is the three-dimensional position estimation result of the target ultra-wideband terminal tag within the target positioning period and its corresponding positioning error estimation; the three-dimensional position estimation result is based on the positioning observation data corresponding to the target path cluster, and combined with the spatial configuration information of the ultra-wideband base station and smart reflector corresponding to the target path cluster recorded in the path coding map; the specific content of updating the path coding map based on the updated positioning result and the positioning observation data includes: calculating the positioning error statistics of the target path cluster based on the updated positioning result and the positioning observation data corresponding to the target path cluster; updating the usage statistics of the target path cluster in the path coding map; updating the reliability score of the target path cluster; and recording the positioning error statistics, usage statistics, and reliability score as historical positioning statistics associated with the target path cluster in the path coding map.

[0021] Compared with the prior art, the present invention has the following beneficial effects:

[0022] 1) Based on the linkage mechanism of path coding map and multi-user positioning difficulty index, this invention can identify difficult terminals that are in non-line-of-sight obstruction area or poor geometric condition area for a long time when multiple users share ultra-wideband base station and intelligent reflector, according to the differences in the number of available paths of each ultra-wideband terminal tag, historical positioning error and motion state. When scheduling ranging resources, higher quality path clusters and ranging time slots are allocated to such terminals first, so that the limited ranging resources are allocated more reasonably among different terminals, and the situation of some terminals having low positioning accuracy and large fluctuations for a long time is reduced.

[0023] 2) This invention introduces intelligent reflective surface resource constraints into the ranging resource scheduling plan, and combines the historical positioning statistics of path clusters recorded in the path coding map to perform candidate path cluster screening, target path cluster selection, and path coding mode switching control. This enables controllability of the number of paths that the intelligent reflective surface can serve simultaneously within a single ranging time slot and the number of path coding mode switching times within a positioning cycle. This avoids the intelligent reflective surface frequently providing reflection resources to paths with poor historical performance or low value, and encourages more resources to be concentrated on high-quality paths with small historical positioning errors and stable usage records. As a result, this improves the utilization efficiency of intelligent reflective surface resources and maintains the long-term stability of positioning results in complex indoor environments. Attached Figure Description

[0024] Figure 1 The flowchart illustrates the intelligent reflective surface-assisted ultra-wideband indoor positioning method provided by this invention. Detailed Implementation

[0025] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0026] like Figure 1 As shown, an ultra-wideband indoor positioning method assisted by an intelligent reflective surface specifically includes the following steps:

[0027] S1. Constructing a path coding map: Obtain spatial configuration information of the ultra-wideband base station, smart reflector, and reference area; control the ultra-wideband base station and smart reflector to perform signal transmission and reflection switching according to the preset path coding mode during the training phase; collect training observation data and perform path clustering processing to generate multiple path clusters; construct a path coding map based on each path cluster and its corresponding path coding mode, ultra-wideband base station, smart reflector, and reference area.

[0028] Training observation data consists of multi-base station ranging data, signal arrival time data, and channel impulse response data obtained by the UWTS tag during the training phase when it interacts with multiple UWTS base stations and smart reflectors according to a preset path coding mode for signal transmission and reception. The training observation data is used to perform statistical analysis on the signal propagation characteristics during the training phase and to implement path clustering processing. Path clusters are registered in the path coding map with path cluster identifiers and serve as a collection index for candidate path clusters, target path cluster identifiers, and historical statistical information of path clusters. Among them, the path cluster identifier corresponds to the set of signal propagation paths belonging to that path cluster in the training observation data.

[0029] The spatial configuration information of the ultra-wideband base station includes parameters such as installation location coordinates, installation height, and antenna pointing angle. The spatial configuration information of the intelligent reflector includes the geometric center coordinates of the reflector, the normal direction of the reflector plane, the array size, and the arrangement of the controllable reflective units. The spatial configuration information of the reference area includes the spatial boundary range of each reference area in a unified three-dimensional coordinate system and the corresponding reference area identifier. The unified three-dimensional coordinate system can be a coordinate representation of the building plane coordinate system superimposed with the height axis. By reading the building design drawings and combining on-site measurements with a laser rangefinder or total station, the spatial positions of the ultra-wideband base station, intelligent reflector, and reference area are registered with this unified coordinate system. After deployment, the measured configuration information is registered as the basic spatial data of the indoor scene for use in subsequent training and positioning phases.

[0030] The training phase refers to a dedicated channel probing period period period periodd periodically inserted before the system is put into formal positioning service or during system operation to update the path coding map. During the training phase, the UWTH terminal tag moves statically or slowly according to a preset trajectory or within a preset reference area, exposing the channel response characteristics of different propagation paths through signal interaction with multiple UWTH base stations. The signal propagation characteristics during the training phase refer to the propagation delay distribution, received energy distribution, and multipath structure characteristics of each signal propagation path during the training period. These characteristics reflect the typical signal propagation environment under the combined conditions of a specific base station, smart reflector, and reference area. They can be statistically described by the arrival time and channel impulse response in the training observation data and serve as the input basis for subsequent path clustering processing.

[0031] A preset path coding mode refers to a set of distinguishable signal transmission and reflection configuration modes pre-defined for the combined state of an ultra-wideband base station and a smart reflector. Each path coding mode includes a set of ultra-wideband base stations that are active in a training time slot, the corresponding transmission timing configuration, and the reflection phase, switching state, or reflection gain configuration information of each reflection unit or reflection area of ​​the smart reflector. The path coding mode can be determined during the system design phase through simulation analysis or field trial operation. For example, a set of configurations that allows different path coding modes to form differentiated responses in time, space, or angle can be selected, and a unique path coding mode identifier can be assigned to each path coding mode. The signal transmission and reflection switching during the training phase refers to changing the active set of ultra-wideband base stations and the reflection configuration state of the smart reflector sequentially in adjacent training time slots according to the path coding mode identifier index. Transmission control commands are sent to the ultra-wideband base stations and reflection configuration commands are sent to the smart reflector controller through the control interface, so that a series of distinguishable signal propagation path combinations are formed during the training process.

[0032] The training observation data is a set of observation data collected and uploaded by UWTP tags in training mode. Each training time slot, the UWTP tag receives signals from each active UWTP base station propagating via a direct path or a path reflected by a smart reflector. The local signal processing module estimates the multi-base station ranging data, signal arrival time data, and channel impulse response data between the tag and each base station. The multi-base station ranging data refers to the set of distance estimates from each base station to the UWTP tag, calculated based on the relationship between the received signal propagation time and the speed of light, under known signal transmission times or known reference timing conditions. The signal arrival time data refers to the estimated time of arrival of the first path component or significant path component of each received signal at the receiver. The channel impulse response data refers to the complex amplitude sequence at discrete time delay sampling points obtained by correlation processing or matched filtering of the received signal. This sequence reflects the energy distribution and relative phase information of different time delay components in the multipath structure. The training observation data is organized according to the training time slot number, path coding mode identifier, base station identifier, and smart reflector configuration status, and is used to extract propagation delay and energy characteristics for each signal propagation path in subsequent processing.

[0033] Path clustering is a clustering analysis process based on signal propagation path features extracted from training observation data. These features can include any one or a combination of the following: the arrival time of the first path, the energy distribution at multiple delay sampling points, and the estimated angle of arrival. Depending on the specific implementation, existing clustering algorithms such as K-means clustering, hierarchical clustering, density-based clustering, or Gaussian mixture model clustering can be selected. The feature vector of each signal propagation path is mapped to a feature space. Based on distance or similarity metrics in the feature space, signal propagation paths with similar features are grouped into the same class. After clustering, multiple path clusters are obtained, each corresponding to a set of signal propagation paths similar in propagation delay and energy distribution characteristics. During the clustering process, parameters such as the maximum intra-cluster delay difference threshold, the intra-cluster energy distribution difference threshold, and the minimum cluster size can be set to control the compactness of the path clusters and the number of effective path clusters.

[0034] In the path coding map, each path cluster is uniquely identified as a set of signal propagation paths that have similar propagation delay and energy distribution characteristics under the same ultra-wideband base station, smart reflector, and reference area configuration, obtained through path clustering. The unique path cluster identifier assigned to each path cluster serves as a primary key field in the path coding map, used to uniquely locate the corresponding signal propagation path set and the historical positioning statistics associated with that path cluster. These historical positioning statistics include the cumulative positioning error statistics, usage frequency statistics, and reliability calculated based on these statistics during subsequent positioning operations. Information such as reliability score; Candidate path cluster identifiers refer to a set of path cluster identifiers selected from the path coding map under the condition of a given reference area or a given target label location, used to mark a set of path clusters with potential positioning value. Target path cluster identifiers refer to a set of path cluster identifiers further selected from the candidate path cluster set to support the ranging and positioning calculation of the current target label. The path cluster identifiers are used as fields for collecting and indexing the historical positioning statistics of the path clusters through the correspondence with the historical positioning statistics of the path clusters, so that the positioning error statistics, usage statistics and reliability scores obtained at different time periods are uniformly associated with the same set of path clusters.

[0035] The path coding map is a data graph structure constructed with a set of nodes composed of path cluster identifiers as nodes and a set of edges consisting of the connection relationships between each path cluster and its corresponding path coding mode, ultra-wideband base station, smart reflector and reference area. It serves as a unified index basis for path clusters and supports candidate path cluster retrieval based on the target ultra-wideband terminal tag location, target path cluster selection, and storage and updating of historical location statistics of path clusters.

[0036] The path coding map is a data graph structure built after path clustering processing is completed and all path clusters are determined. Each node in the map corresponds to a path cluster identifier. The node attribute fields may include the path cluster identifier, the identifier of the reference area to which it belongs, the set of path coding pattern identifiers associated with the path cluster, the set of ultra-wideband base station identifiers associated with the path cluster, the set of smart reflector identifiers associated with the path cluster, and historical positioning statistics of the path cluster. The historical positioning statistics of the path cluster can be initialized to default values ​​when the path coding map is built. For example, the positioning error statistics are initialized to zero or null values, the usage count statistics are initialized to zero, and the reliability score is initialized to a preset neutral value. It is then updated according to the new positioning results during the positioning operation phase. The path coding map can be implemented using an adjacency list structure or a relational data structure based on key-value mapping. For example, the path cluster identifier can be used as the primary key, and the node record stores the identifier fields of the path coding pattern, ultra-wideband base station, smart reflector, and reference area associated with the path cluster, as well as the statistical information field.

[0037] The connections in the path coding graph record the association information between path clusters and path coding patterns, ultra-wideband base stations, smart reflectors, and reference areas through edge sets or association tables. Each connection can consist of a path cluster identifier and an associated object identifier pair. For example, the connection between a path cluster and an ultra-wideband base station records the set of ultra-wideband base station identifiers traversed by the signal propagation path contained in the path cluster; the connection between a path cluster and a smart reflector records the smart reflector identifiers traversed by the signal propagation path corresponding to the path cluster and their reflection configuration state indexes used during the training phase; the connection between a path cluster and a reference area records the reference area identifiers into which the signal propagation path corresponding to the path cluster spatially falls; and the connection between a path cluster and a path coding pattern records the set of path coding pattern identifiers used by the signal propagation path that forms the path cluster during the training phase. Through these connections, given a path cluster identifier, information on path coding patterns, ultra-wideband base stations, smart reflectors, and reference areas related to that path cluster can be quickly retrieved.

[0038] The path coding map serves as the unified storage and retrieval foundation for path clusters and their associated parameters. In terms of data organization, it utilizes node records indexed by path cluster identifiers and various types of connection relationships to centrally manage the spatial distribution information, transmission and reflection configuration information, and statistical information of path clusters. When retrieving path clusters based on a reference area, the path coding map can search for all path cluster identifier sets associated with that reference area identifier by inputting the reference area identifier based on the connection relationship between the path cluster and the reference area. When retrieving associated path coding patterns, ultra-wideband base stations, or intelligent reflectors based on path cluster identifiers, the corresponding identifier set can be located through the connection relationship between the path cluster and the corresponding object, and further access can be made to the historical location statistics of the path cluster in the node attribute fields, achieving unified query and management of path cluster-related parameters and historical statistical information. The storage medium for the path coding map can be a database system or an in-memory data structure. During system initialization, the path coding map is loaded into the runtime environment, and after the training phase, it is reconstructed or incrementally updated based on the new training results.

[0039] Among them, a path cluster is a set of multiple signal propagation paths with similar propagation delay and energy distribution characteristics under the conditions of a fixed ultra-wideband base station, smart reflector, and reference area; the path coding map is a data graph structure with path clusters as nodes and records the relationship between each path cluster and its path coding mode, as well as the associated ultra-wideband base station, smart reflector, and reference area.

[0040] S2. Calculate and sort the multi-user positioning difficulty index: Calculate the multi-user positioning difficulty index based on the positioning observation data and tag status information of the ultra-wideband terminal tags, and sort the ultra-wideband terminal tags according to the multi-user positioning difficulty index to obtain the multi-user positioning priority queue.

[0041] The multi-user positioning difficulty index is a comprehensive measure of the ease with which each ultra-wideband terminal tag achieves a predetermined positioning accuracy under the current intelligent reflector resource configuration. It serves as the ranking criterion for sorting ultra-wideband terminal tags and also reflects the difficulty of obtaining high-precision positioning for the ultra-wideband terminal under the current intelligent reflector resource constraints. The multi-user positioning difficulty index is specifically composed of: the number of available path clusters, the ranging geometric distribution index and signal-to-noise ratio index corresponding to the available path clusters, the historical positioning error statistics of the path clusters, and motion state parameters. Among them, the multi-user positioning difficulty index increases with the decrease of the number of available path clusters, the increase of the historical positioning error statistics of the path clusters, and the increase of the motion intensity represented by the motion state parameters.

[0042] The positioning observation data of the ultra-wideband terminal tags is received by the positioning server from each ultra-wideband terminal tag within the target positioning period. Each ultra-wideband terminal tag interacts with multiple ultra-wideband base stations and smart reflectors according to the existing ranging process in the current positioning period. After completing the ranging and positioning-related signal processing, the observation record containing the ranging result, the signal arrival time estimation result, and the channel impulse response estimation result is uploaded to the positioning server. The positioning server collects these observation records according to the terminal tag identifier and time index to form the positioning observation data set of each terminal tag in the current positioning period. The tag status information is obtained by the terminal tag through the local sensing module or by the positioning server based on the continuous positioning results. The tag status information includes at least parameters reflecting the current motion state, such as the velocity vector calculated based on the three-dimensional position estimation result of the terminal tag in adjacent positioning periods, the approximate value of acceleration calculated based on the velocity change, and the rate of change of the motion trajectory direction, or measurement data such as linear acceleration and angular velocity uploaded from the inertial measurement unit of the terminal tag. After unified time alignment and coordinate transformation, the tag status information is used for the calculation of the multi-user positioning difficulty index.

[0043] The multi-user positioning difficulty index is used to measure the ease with which each ultra-wideband terminal tag achieves a predetermined positioning accuracy under the current intelligent reflector resource configuration. The predetermined positioning accuracy can be defined by a preset positioning error threshold, such as the maximum allowable value of the 3D position estimation error under a given probability confidence level, or the upper limit of the trace value of the 3D position error covariance matrix. When a terminal tag is unable to meet the predetermined positioning accuracy under the current resource conditions, its multi-user positioning difficulty index value is relatively large. The specific components of the multi-user positioning difficulty index include the number of available path clusters, the ranging geometric distribution index corresponding to the available path clusters, the signal-to-noise ratio index corresponding to the available path clusters, the historical positioning error statistics of the path clusters, and motion state parameters. Among them, the number of available path clusters reflects the richness of path resources that the current terminal tag can call at the current location; the ranging geometric distribution index reflects the impact of the base station geometric layout corresponding to the available path clusters on the amplification effect of positioning errors; the signal-to-noise ratio index reflects the signal quality level corresponding to the available path clusters; the historical positioning error statistics of the path clusters reflect the stability and reliability of the path clusters in the past positioning process; and the motion state parameters reflect the motion intensity and motion uncertainty of the terminal tag in the current period.

[0044] The number of available path clusters is determined by querying the path coding map for each ultra-wideband terminal tag within the current positioning period. The search identifies the number of path clusters associated with the reference area of ​​the current location and whose historical positioning statistics meet preset reliability conditions. The historical positioning statistics for path clusters may include the aforementioned historical positioning error statistics, usage frequency statistics, and a reliability score calculated from these statistics. When the reliability score is below a preset threshold, the corresponding path cluster is not included in the available path cluster set. The ranging geometric distribution index corresponding to the available path clusters is based on the spatial geometric relationship between the ultra-wideband base station locations involved in the available path clusters and the current location of the terminal tag. The calculation involves analyzing whether the distribution of these base stations relative to the tag positions in the three-dimensional coordinate system is concentrated within a narrow angular range, and whether there is geometric degradation such as near-collinearity or coplanarity. A geometric distribution metric reflecting the quality of ranging geometry is constructed, and the value of this metric increases when geometric degradation is severe. The signal-to-noise ratio (SNR) index corresponding to the available path cluster is calculated based on the estimated received signal energy and noise power associated with the available path cluster in the positioning observation data. A single SNR index value can be generated by averaging, weighted averaging, or taking the minimum value of the SNR corresponding to each available path cluster. When the SNR decreases, the signal quality reflected by this index weakens.

[0045] The historical positioning error statistics of the path cluster are read from the path coding map for the set of available path clusters for the current terminal tag. These statistics can include the mean, variance, or quantiles of the 3D position errors obtained from positioning calculations based on this path cluster over multiple past positioning cycles, reflecting the stability of the path cluster's positioning performance in historical operation. Motion state parameters are calculated from tag status information, including the terminal tag's current velocity magnitude, acceleration magnitude, and rate of change of motion direction. When the velocity magnitude is large or the motion trajectory direction changes frequently, the motion state parameter value increases, indicating that the terminal tag is in a fast or unstable motion state. The multi-user positioning difficulty index is calculated by normalizing and weighting the number of available path clusters, ranging geometric distribution index, signal-to-noise ratio index, historical positioning error statistics of the path cluster, and motion state parameters according to preset rules. When the number of available path clusters decreases, the historical positioning error statistics of the path cluster increase, and the motion intensity represented by the motion state parameters increases, the multi-user positioning difficulty index increases accordingly, reflecting the difficulty for the terminal tag to achieve the predetermined positioning accuracy under the current constraints of intelligent reflective surface resources.

[0046] After the multi-user positioning difficulty index is calculated, the positioning server sorts all ultra-wideband terminal tags participating in the positioning process according to the value of the multi-user positioning difficulty index. Terminal tags with larger difficulty index values ​​are ranked higher in the sorting results, and terminal tags with smaller difficulty index values ​​are ranked lower. The sorting method can be implemented using standard sorting algorithms, such as quicksort, mergesort, or heapsort based on comparison operations. Alternatively, a priority queue data structure with the multi-user positioning difficulty index as the key can be maintained during operation. For each terminal tag, a corresponding priority node is inserted or updated. After the difficulty index calculation of all terminal tags is completed, the first tag of the priority queue is read sequentially to form the multi-user positioning priority queue. When multiple terminal tags have the same multi-user positioning difficulty index value or are close within a preset tolerance range, the sorting can be based on the terminal tag identifier, the priority order in the previous positioning cycle, or the access time order. This maintains the order stability of the multi-user positioning priority queue when the values ​​are the same, ensuring a monotonic correspondence between the queue position in the multi-user positioning priority queue and the value of the multi-user positioning difficulty index.

[0047] Candidate path clusters are sets of path clusters selected from the path coding map that are associated with the reference area to which the current and predicted positions of the target ultra-wideband terminal tag belong within the target positioning period. These clusters serve as candidate sets for the selection of target path clusters. The method for retrieving candidate path clusters from the path coding map is as follows: determine the reference area identifier based on the current and predicted positions of the target ultra-wideband terminal tag; search the path coding map for path clusters that correspond to the reference area identifier and whose historical positioning statistics meet preset reliability conditions; and construct candidate path clusters from the path clusters that meet the preset reliability conditions. The target path cluster is a set of path clusters selected from the candidate path clusters for ranging and positioning calculation of the target ultra-wideband terminal tag.

[0048] S3. Ranging resource scheduling: Based on the multi-user positioning priority queue, the target UWB terminal tag is determined. Candidate path clusters corresponding to the location of the target UWB terminal tag are retrieved from the path coding map. The target path cluster is selected from the candidate path clusters. A ranging resource scheduling plan is generated based on the target path cluster. The UWB base station and the intelligent reflector are combined with the intelligent reflector resource constraint control to execute the ranging resource scheduling plan.

[0049] The multi-user positioning priority queue has been calculated based on the multi-user positioning difficulty index of the previous step at the beginning of the target positioning period. Each element in the queue consists of an ultra-wideband terminal tag identifier and the corresponding multi-user positioning difficulty index. Within a target positioning period, the positioning server selects the first few ultra-wideband terminal tags in the queue order as the target ultra-wideband terminal tags for the current period. Each time a tag identifier is selected, it is taken from the head of the queue and used as the processing object for subsequent candidate path cluster retrieval, target path cluster selection and ranging resource scheduling plan generation in S3 of this embodiment. When multiple target ultra-wideband terminal tags need to be supported, the above process is executed sequentially on the first few tags in the queue.

[0050] The current position of the target ultra-wideband terminal tag within the target positioning cycle is the three-dimensional position estimation result obtained by positioning calculation at the end of the previous positioning cycle. The predicted position is the prediction result obtained by position extrapolation based on the current position and motion state parameters. Position extrapolation can be carried out using a linear motion model, which superimposes the three-dimensional position estimation result of the previous positioning cycle with the velocity vector in the tag's state information over the target positioning cycle time length to obtain the predicted position. Alternatively, a state estimation filtering method can be used, which estimates the current position and velocity based on the positioning results and motion state parameters of several historical cycles and then makes a further prediction. After obtaining the current position and the predicted position, the current position and the predicted position are mapped to the corresponding reference areas according to the spatial division rules under a unified three-dimensional coordinate system, and the corresponding reference area identifiers are obtained.

[0051] Candidate path clusters are sets of path clusters retrieved from the path coding map based on the current and predicted positions of the target ultra-wideband terminal tag within the target positioning period. The retrieval process includes: determining the corresponding reference area identifiers based on the current and predicted positions; searching for a set of path cluster identifiers that are associated with these reference area identifiers in the path coding map; and using the path clusters associated with any reference area identifier as the initial candidate set. For each path cluster in the initial candidate set, the historical positioning statistics of the path cluster recorded in the path coding map are read to determine whether the positioning error statistics, usage frequency statistics, and reliability score of the path cluster simultaneously meet the preset reliability conditions. Only path clusters that meet the preset reliability conditions are retained as candidate path clusters, and duplicate path cluster identifiers are deduplicated to form a set of candidate path clusters associated with the current and predicted positions for subsequent target path cluster selection.

[0052] The target path cluster is selected from the candidate path cluster set. The target path cluster set is used to support the ranging and positioning calculation of the target ultra-wideband terminal tag within the target positioning cycle. The selection process of the target path cluster includes: evaluating the path clusters in the candidate path cluster set according to the ranging geometric distribution index, signal-to-noise ratio index, and historical positioning error statistics of each path cluster. The path clusters with better geometric conditions, higher signal quality, and smaller historical errors are selected first through preset sorting or scoring rules. At the same time, a lower limit is set for the number of target path clusters according to the positioning dimension requirements. For example, at least three target path clusters are selected in a two-dimensional positioning scenario and at least four target path clusters are selected in a three-dimensional positioning scenario to avoid geometric degradation. In addition, an upper limit is set for the number of target path clusters based on the constraints of ranging resources and intelligent reflector resources. When the number of candidate path clusters exceeds the upper limit, the path clusters with the highest scores are selected to form the target path cluster set. Through the above limitations, the number of path clusters participating in ranging and positioning calculation is controlled under the premise of ensuring positioning geometric constraints, so as to reduce scheduling complexity and reduce the occupation of intelligent reflector resources.

[0053] The ranging resource scheduling plan is a scheduling scheme based on the target path cluster, which allocates ranging operations of ultra-wideband base stations and the path coding mode configuration status of intelligent reflectors to each ranging time slot within the target positioning period. It is used to coordinate the execution of the ranging process by each ultra-wideband base station and each intelligent reflector according to a predetermined time slot sequence and path coding mode within the target positioning period, and to determine that the signal propagation path associated with each target path cluster is available in the corresponding ranging time slot. The ranging resource scheduling plan includes a ranging time slot allocation scheme and an intelligent reflector path coding mode switching scheme. The ranging time slot allocation scheme allocates each ranging time slot to the combination of ultra-wideband base stations and intelligent reflectors associated with the target path cluster within the target positioning period. The intelligent reflector path coding mode switching scheme determines the switching time and switching target of the intelligent reflector path coding mode between adjacent ranging time slots.

[0054] The ranging resource scheduling plan is constructed for the target positioning period, which is divided into several consecutive ranging time slots. Each ranging time slot has a fixed duration and includes a complete ranging process of transmitting, propagating and receiving ultra-wideband signals. The ranging resource scheduling plan consists of a ranging time slot allocation scheme and a path coding mode switching scheme for intelligent reflectors. It is used to determine the set of ultra-wideband base stations to be activated in each ranging time slot and the path coding mode configuration status of intelligent reflectors within the target positioning period.

[0055] The ranging time slot allocation scheme is generated based on a set of target path clusters. Each target path cluster has a set of associated ultra-wideband base station identifiers, a set of smart reflector identifiers, and a set of path coding mode identifiers recorded in the path coding map. When generating the ranging time slot allocation scheme, at least one ranging time slot is first allocated to each target path cluster. Within the ranging time slot, the ultra-wideband base station associated with the path cluster is activated and the path coding mode of the smart reflector is set, so that the signal propagation path corresponding to the path cluster has the conditions for signal transmission and reflection within the time slot. When the number of target path clusters is less than the number of ranging time slots, some target path clusters can be repeatedly allocated in different ranging time slots to improve redundant observation. When the number of target path clusters is greater than the number of ranging time slots, target path clusters with the same or similar base station combinations and that can share the same path coding mode can be merged and allocated to the same ranging time slot, or some target path clusters can be discarded based on their importance, so that each ranging time slot corresponds to one or more target path clusters.

[0056] The path coding mode switching scheme for the intelligent reflector is generated after the ranging time slot allocation scheme is determined. For each ranging time slot, based on the target path cluster set allocated to that time slot, the path coding mode identifier set associated with these target path clusters recorded in the path coding map is searched, and the path coding mode that can simultaneously satisfy the requirements of as many target path clusters as possible is selected as the target path coding mode for that ranging time slot. For adjacent ranging time slots, when the target path coding modes of the two time slots are different and the intelligent reflector resource constraints allow for configuration updates, a path coding mode switching operation is inserted between the two ranging time slots, and the path coding mode is recorded. The switching time and target of the code mode from the current state to the next ranging time slot target state; through the combination of ranging time slot allocation scheme and path coding mode switching scheme, the signal propagation path associated with each target path cluster is in an available state within the allocated ranging time slot. The available state means that within the ranging time slot, the ultra-wideband base station corresponding to the target path cluster is in the transmitting state, the intelligent reflector is in the target path coding mode configuration state, and the signal-to-noise ratio of the signal propagation on the path meets the preset ranging quality conditions, so that the target ultra-wideband terminal tag can obtain the effective ranging observation data corresponding to the target path cluster in the ranging time slot.

[0057] The intelligent reflector resource constraint is a resource limitation condition determined based on the array size of the intelligent reflector, the number of independently controllable reflector units, and the control update rate. It is used to limit the number of target path clusters that the intelligent reflector can simultaneously serve within a target positioning cycle and the allowed path coding mode switching frequency. Specifically, the intelligent reflector resource constraint is that the number of target path clusters allocated to the same intelligent reflector in any ranging time slot does not exceed a first preset upper limit, and the number of path coding mode switching times of the intelligent reflector within a target positioning cycle does not exceed a second preset upper limit.

[0058] Resource constraints for intelligent reflectors are determined based on the array size, the number of independently controllable reflective elements, and the control update rate. The array size includes the total number of reflective elements, the array physical dimensions, and the array topology. The number of independently controllable reflective elements is determined by the number of channels and control protocol capabilities of the intelligent reflector control hardware. The control update rate is jointly limited by the control link bandwidth, control command processing time, and reflective element physical response time. When modeling resource constraints, the upper limit of the number of target path clusters that the intelligent reflector can simultaneously serve in a single ranging time slot is determined as the first preset upper limit, based on the number of effective reflected beams that the intelligent reflector can simultaneously form in a ranging time slot and the number of path coding mode configuration updates that can be executed in a target positioning cycle. The upper limit of the number of path coding mode switching times in a target positioning cycle is determined as the second preset upper limit.

[0059] The first preset upper limit is used to limit the number of target path clusters that can be simultaneously allocated to the same intelligent reflector in any ranging time slot. When the number of target path clusters allocated to a certain intelligent reflector in a ranging time slot exceeds the preset upper limit, the scheduling algorithm adjusts the target path cluster set for that ranging time slot. For example, it merges path clusters with the same or similar emission directions, prioritizes retaining target path clusters with higher multi-user positioning difficulty indicators or better geometric conditions, and reallocates other path clusters to adjacent ranging time slots or removes them from the current target positioning cycle, so that each intelligent reflector in each ranging time slot can achieve the desired balance. The number of target path clusters associated with the reflector does not exceed the first preset upper limit; the second preset upper limit is used to limit the number of times the intelligent reflector switches the path coding mode within a target positioning cycle. When the number of path coding mode switches of a certain intelligent reflector exceeds the second preset upper limit according to the statistics of the initial ranging resource scheduling plan, the scheduling algorithm reduces the number of path coding mode switches by merging the path coding mode configurations of adjacent ranging time slots, extending the duration of the same path coding mode, or reducing the number of services to some low-priority target path clusters, so that the number of path coding mode switches of the intelligent reflector does not exceed the second preset upper limit.

[0060] During the generation of the ranging resource scheduling plan, constraint checks are always performed based on the intelligent reflector resource constraints. When allocating ranging time slots to target path clusters and determining the path coding mode for each ranging time slot, the number of target path clusters allocated in the ranging time slot for each intelligent reflector is calculated and compared with the first preset upper limit. At the same time, the number of path coding mode switching times is counted in the target positioning cycle dimension and compared with the second preset upper limit. When resource constraints are detected to be triggered, the scheduling algorithm adjusts the target path cluster allocation or merges the path coding mode configuration in the ranging time slots according to the preset priority rules, so that the number of target path clusters and the number of path coding mode switching times for all intelligent reflectors meet the intelligent reflector resource constraints. Under the premise of meeting the hardware capability and control capability limitations, an executable ranging resource scheduling plan is provided for the target ultra-wideband terminal tag.

[0061] S4. Positioning Calculation and Map Update: Obtain the positioning observation data of the target UWB terminal tag, perform positioning calculation in combination with the spatial configuration information of the corresponding UWB base station and smart reflector, obtain the updated positioning result of the target UWB terminal tag, and update the path coding map based on the updated positioning result and the positioning observation data corresponding to the target path cluster.

[0062] The positioning observation data of the target ultra-wideband terminal tag is the set of observation data collected by the target ultra-wideband terminal tag during the target positioning period when it performs ranging with multiple ultra-wideband base stations and smart reflectors according to the ranging resource scheduling plan. The positioning observation data corresponding to the target path cluster is a subset of the observation data selected from the positioning observation data of the target ultra-wideband terminal tag. Among them, the subset of observation data is associated with the signal propagation path corresponding to the ultra-wideband base station and smart reflector associated with the target path cluster in the path coding map, and corresponds one-to-one with the ranging time slot allocated to the target path cluster in the ranging resource scheduling plan.

[0063] The target positioning period is a time window for performing one round of ranging resource scheduling and positioning calculation on the target ultra-wideband terminal tag. This time window is consistent with the period of the aforementioned ranging resource scheduling plan and includes multiple consecutive ranging time slots. The positioning observation data of the target ultra-wideband terminal tag is a set of observation records collected by the tag during the ranging process with multiple ultra-wideband base stations and smart reflectors according to the ranging resource scheduling plan within the target positioning period. Each observation record includes at least the ranging result or estimated time of arrival, the received signal strength or signal-to-noise ratio estimate, the identifier of the ultra-wideband base station participating in the ranging, the identifier of the smart reflector participating in the reflection, the current path coding mode identifier, and the corresponding ranging time slot index. After the target positioning period ends, the positioning server collects all the observation records received in the period according to the terminal tag identifier to form the positioning observation data set of the target ultra-wideband terminal tag, which serves as the basic data source for the positioning calculation and path coding map update of the current period.

[0064] The positioning observation data corresponding to the target path cluster is a subset of observation data obtained by filtering the positioning observation data of the target ultra-wideband terminal tag. In the path coding map, each target path cluster is recorded with a set of ultra-wideband base station identifiers, a set of smart reflector identifiers, and a set of path coding pattern identifiers associated with that path cluster. When constructing the observation data subset, the positioning server matches the observation records in the positioning observation data subset for each signal propagation path in the target path cluster. The matching conditions are that the ultra-wideband base station identifier, smart reflector identifier, and path coding pattern identifier in the observation record are consistent with the corresponding fields recorded in the path coding map, and the ranging time slot index of the observation record falls within the ranging time slot set pre-allocated for that target path cluster. The observation records that meet the above matching conditions are assigned to the observation data subset corresponding to the target path cluster, thereby realizing the mapping of the association relationship between the observation data subset and the signal propagation paths of the ultra-wideband base stations and smart reflectors associated with the target path cluster in the path coding map.

[0065] A one-to-one correspondence is established between the subset of observation data and the ranging time slots allocated for the target path cluster in the ranging resource scheduling plan. Specifically, when generating the ranging resource scheduling plan, several target ranging time slots are determined for each target path cluster within the target positioning period. Each target ranging time slot is expected to generate a set of ranging observation data corresponding to the target path cluster. In the observation data screening stage, for each target ranging time slot of the target path cluster, the base station identifier and smart reflector marker associated with the target path cluster are selected from the positioning observation data of the target ultra-wideband terminal tag. These markers have the same index as the ranging time slot and match the target path cluster's associated base station identifier and smart reflector marker. The system identifies observation records with path coding patterns and selects a single observation record as the target observation record corresponding to the ranging time slot based on preset selection rules (e.g., prioritizing the observation record with the highest signal-to-noise ratio). This establishes a one-to-one correspondence between each target ranging time slot and the unique observation record within the subset of observation data corresponding to the target path cluster. This one-to-one correspondence ensures that the observation data used in each target ranging time slot and its corresponding spatial path configuration can be accurately tracked during subsequent positioning solutions and positioning error statistics calculations, avoiding the problem of difficulty in distinguishing path contributions caused by the mixing of multiple observation records.

[0066] The updated positioning result of the target ultra-wideband terminal tag is the three-dimensional position estimation result of the target ultra-wideband terminal tag within the target positioning period and its corresponding positioning error estimation. The three-dimensional position estimation result is based on the positioning observation data corresponding to the target path cluster, and combined with the spatial configuration information of the ultra-wideband base station and smart reflector corresponding to the target path cluster recorded in the path coding map. The specific content of updating the path coding map based on the updated positioning result and positioning observation data includes: calculating the positioning error statistics of the target path cluster based on the updated positioning result and the positioning observation data corresponding to the target path cluster; updating the usage statistics of the target path cluster in the path coding map; updating the reliability score of the target path cluster; and recording the positioning error statistics, usage statistics, and reliability score as historical positioning statistics associated with the target path cluster in the path coding map.

[0067] The updated positioning result of the target ultra-wideband terminal tag includes the three-dimensional position estimation result of the tag within the target positioning cycle and the corresponding positioning error estimate. The three-dimensional position estimation result is the estimated value of the target ultra-wideband terminal tag's position in a unified indoor coordinate system, given in three-dimensional coordinate form, used to represent the position of the target ultra-wideband terminal tag in space at the end of the target positioning cycle or at the cycle reference time. The positioning error estimate is used to quantify the estimation uncertainty of the above three-dimensional position estimation result, and can be represented by a scalar error bound index, such as the standard deviation or confidence radius of the three-dimensional position estimation error, or described by error characteristics in vector or matrix form, such as the estimation variance or the main diagonal element of the position estimation error covariance in each coordinate axis direction. In this embodiment, the target positioning cycle is consistent with the cycle of the aforementioned ranging resource scheduling plan, ensuring that an independent positioning calculation is completed and a set of updated positioning results are output using the positioning observation data corresponding to the target path cluster within one target positioning cycle.

[0068] The positioning calculation process takes the positioning observation data corresponding to the target path cluster and the spatial configuration information of the ultra-wideband base stations and smart reflectors recorded in the path coding map as input. It solves for the three-dimensional position estimation result and the positioning error estimate by constructing ranging constraints and geometric constraints. Specifically, it includes: establishing a set of ranging equations between the unknown coordinates of the target ultra-wideband terminal tag and the known base station and reflector positions based on the ranging results or signal arrival time estimates in the subset of observation data corresponding to the target path cluster, combined with the spatial coordinates of each participating ultra-wideband base station recorded in the path coding map, the position and orientation information of the smart reflector in the reference coordinate system, and the geometric relationship of the reflection path corresponding to the path coding mode; solving the above equations using nonlinear solution methods, such as iterative weighted least squares method or other numerical optimization methods, to obtain the estimated position of the target ultra-wideband terminal tag in the three-dimensional coordinate system; after obtaining the position estimation result, calculating the propagation result of the position estimation error based on the local linear approximation of the ranging equations and the statistical characteristics of ranging noise, and obtaining the positioning error estimate, which is used to reflect the accuracy level of the position estimation result within the current target positioning cycle.

[0069] The positioning error statistics of the target path cluster are calculated based on the updated positioning results and the positioning observation data corresponding to the target path cluster. For the target path cluster, the positioning server evaluates the geometric consistency between the observation data participating in the ranging of the path cluster and the updated positioning results, and calculates the difference between the estimated distance obtained from the ranging equation and the theoretical distance derived from the updated three-dimensional position estimation results. By statistically processing the above difference, the positioning error metric corresponding to the target path cluster in the current target positioning cycle is obtained. For example, the average of the absolute error, the average of the squared error, or the maximum error value can be used as the positioning error metric of the path cluster in this cycle. After obtaining the positioning error metric of the path cluster in this cycle, the metric is combined with the historical positioning error statistics recorded in the path coding map for updating. The historical positioning error statistics of the path cluster can be gradually adjusted by means of moving average, weighted average, or exponential weighted moving average, so that the statistics reflect the recent changes in positioning performance while retaining a certain historical stability information.

[0070] The usage count of a target path cluster in the path coding map is a count of the number of times the path cluster participates in the positioning calculation. Each time a target path cluster is selected as a target path cluster and participates in the positioning calculation within a target positioning cycle, the usage count of the path cluster is incremented by one, reflecting the frequency of actual use of the path cluster in long-term operation. The reliability score of the target path cluster is based on a comprehensive evaluation of the historical positioning error statistics and usage count statistics of the path cluster. The reliability score can be calculated using a preset scoring function, such as assigning different scoring levels based on the combination range of the positioning error statistics and usage count statistics, or obtaining a continuous score value by normalizing the positioning error statistics and weighting them with the usage count statistics. After updating the positioning error statistics and usage count statistics of the path cluster in the current cycle, the positioning server calculates the reliability score of the target path cluster based on the new statistics and writes the updated positioning error statistics, usage count statistics, and reliability score as historical positioning statistics into the record entry corresponding to the target path cluster identifier in the path coding map, which is used to characterize the historical positioning performance and reliability level of the path cluster in subsequent cycles.

[0071] The technical solution of the present invention has been described in conjunction with the specific experimental procedures shown in the accompanying drawings. However, the scope of protection of the present invention is not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions resulting from such changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A smart reflective surface-assisted ultra-wideband indoor positioning method, characterized in that, Includes the following steps: S1. Constructing a path coding map: Obtain spatial configuration information of the ultra-wideband base station, smart reflector, and reference area; control the ultra-wideband base station and smart reflector to perform signal transmission and reflection switching according to the preset path coding mode during the training phase; collect training observation data and perform path clustering processing to generate multiple path clusters; construct a path coding map based on each path cluster and its corresponding path coding mode, ultra-wideband base station, smart reflector, and reference area. Among them, a path cluster is a set of multiple signal propagation paths with similar propagation delay and energy distribution characteristics under the conditions of a fixed ultra-wideband base station, smart reflector, and reference area; the path coding map is a data graph structure with path clusters as nodes and records the relationship between each path cluster and its path coding mode, as well as the associated ultra-wideband base station, smart reflector, and reference area. S2. Calculate and sort the multi-user positioning difficulty index: Calculate the multi-user positioning difficulty index based on the positioning observation data and tag status information of the ultra-wideband terminal tags, and sort the ultra-wideband terminal tags according to the multi-user positioning difficulty index to obtain the multi-user positioning priority queue. S3. Ranging resource scheduling: Based on the multi-user positioning priority queue, the target UWB terminal tag is determined. Candidate path clusters corresponding to the location of the target UWB terminal tag are retrieved from the path coding map. The target path cluster is selected from the candidate path clusters. A ranging resource scheduling plan is generated based on the target path cluster. The UWB base station and the intelligent reflector are combined with the intelligent reflector resource constraint control to execute the ranging resource scheduling plan. S4. Positioning Calculation and Map Update: Obtain the positioning observation data of the target UWB terminal tag, perform positioning calculation in combination with the spatial configuration information of the corresponding UWB base station and smart reflector, obtain the updated positioning result of the target UWB terminal tag, and update the path coding map based on the updated positioning result and the positioning observation data corresponding to the target path cluster.

2. The intelligent reflective surface-assisted ultra-wideband indoor positioning method according to claim 1, characterized in that: In step S1, the training observation data consists of multi-base station ranging data, signal arrival time data, and channel impulse response data obtained when the ultra-wideband terminal tag performs signal transmission and reception interactions with multiple ultra-wideband base stations and smart reflectors according to a preset path coding mode during the training phase. Training observation data is used for statistical analysis of signal propagation characteristics during the training phase and for path clustering processing. Path clusters are registered in the path coding map with path cluster identifiers, which serve as indexes for candidate path clusters, target path cluster identifiers, and historical statistical information of path clusters. The path cluster identifier corresponds to the set of signal propagation paths belonging to that path cluster in the training observation data.

3. The ultra-wideband indoor positioning method assisted by an intelligent reflective surface according to claim 1, characterized in that: In step S1, the path coding graph is a data graph structure constructed with the set of nodes composed of path cluster identifiers as nodes and the set of edges composed of the connection relationships between each path cluster and its corresponding path coding mode, ultra-wideband base station, smart reflector and reference area. It serves as the unified index basis for path clusters.

4. The ultra-wideband indoor positioning method assisted by an intelligent reflective surface according to claim 1, characterized in that: In step S2, the multi-user positioning difficulty index is a comprehensive metric used to characterize the ease with which each ultra-wideband terminal tag achieves a predetermined positioning accuracy under the current intelligent reflector resource configuration. It serves as the sorting basis when ranking ultra-wideband terminal tags and also reflects the difficulty of the ultra-wideband terminal obtaining high-precision positioning under the current intelligent reflector resource constraints. The specific components of the multi-user positioning difficulty index include: the number of available path clusters, the ranging geometric distribution index and signal-to-noise ratio index corresponding to the available path clusters, the historical positioning error statistics of the path clusters, and motion state parameters.

5. The ultra-wideband indoor positioning method assisted by an intelligent reflective surface according to claim 1, characterized in that: In step S3, the candidate path cluster is a set of path clusters selected from the path coding map that are associated with the reference area to which the current and predicted positions of the target ultra-wideband terminal tag belong within the target positioning period. These path clusters serve as the candidate set for the target path cluster selection. The method for retrieving candidate path clusters from the path coding map is as follows: determine the reference area identifier based on the current and predicted positions of the target ultra-wideband terminal tag, search the path coding map for path clusters that correspond to the reference area identifier and whose historical positioning statistics meet preset reliability conditions, and form candidate path clusters from the path clusters that meet the preset reliability conditions. The target path cluster is a set of path clusters selected from the candidate path clusters for ranging and positioning calculation of the target ultra-wideband terminal tag.

6. The ultra-wideband indoor positioning method assisted by an intelligent reflective surface according to claim 1, characterized in that: In step S3, the ranging resource scheduling plan is a scheduling scheme based on the target path cluster, which allocates ranging operations of ultra-wideband base stations and the path coding mode configuration status of smart reflectors to each ranging time slot within the target positioning period. It is used to coordinate each ultra-wideband base station and each smart reflector to perform the ranging process according to the predetermined time slot sequence and path coding mode within the target positioning period, and to determine that the signal propagation path associated with each target path cluster is in an available state within the corresponding ranging time slot. The ranging resource scheduling plan includes a ranging time slot allocation scheme and a path coding mode switching scheme for smart reflectors. The ranging time slot allocation scheme allocates each ranging time slot to the combination of ultra-wideband base stations and smart reflectors associated with the target path cluster within the target positioning period. The path coding mode switching scheme for smart reflectors determines the switching time and switching target of the path coding mode of smart reflectors between adjacent ranging time slots.

7. The ultra-wideband indoor positioning method assisted by an intelligent reflective surface according to claim 1, characterized in that: In step S3, the intelligent reflector resource constraint is a resource limitation condition determined based on the array size of the intelligent reflector, the number of independently controllable reflective units, and the control update rate. It is used to limit the number of target path clusters that the intelligent reflector can simultaneously serve within the target positioning cycle and the allowed path coding mode switching frequency. Specifically, the intelligent reflector resource constraint is that the number of target path clusters allocated to the same intelligent reflector in any ranging time slot does not exceed a first preset upper limit, and the number of path coding mode switching times of the intelligent reflector within a target positioning cycle does not exceed a second preset upper limit.

8. The ultra-wideband indoor positioning method assisted by an intelligent reflective surface according to claim 1, characterized in that: In step S4, the positioning observation data of the target ultra-wideband terminal tag is the set of observation data collected by the target ultra-wideband terminal tag during the target positioning period when it performs ranging with multiple ultra-wideband base stations and smart reflectors according to the ranging resource scheduling plan; the positioning observation data corresponding to the target path cluster is a subset of observation data selected from the positioning observation data of the target ultra-wideband terminal tag; wherein, the subset of observation data is associated with the signal propagation path corresponding to the ultra-wideband base station and smart reflector associated with the target path cluster in the path coding map, and corresponds one-to-one with the ranging time slot allocated to the target path cluster in the ranging resource scheduling plan.

9. The intelligent reflective surface-assisted ultra-wideband indoor positioning method according to claim 1, characterized in that: In step S4, the updated positioning result of the target ultra-wideband terminal tag is the three-dimensional position estimation result of the target ultra-wideband terminal tag within the target positioning period and its corresponding positioning error estimation. The three-dimensional position estimation result is based on the positioning observation data corresponding to the target path cluster, and combined with the spatial configuration information of the ultra-wideband base station and smart reflector corresponding to the target path cluster recorded in the path coding map. The specific content of updating the path coding map based on the updated positioning results and positioning observation data includes: calculating the positioning error statistics of the target path cluster based on the updated positioning results and the positioning observation data corresponding to the target path cluster; Update the usage statistics of the target path cluster in the path coding map; update the reliability score of the target path cluster; and record the positioning error statistics, usage statistics, and reliability score as historical positioning statistics associated with the target path cluster in the path coding map.

Images