Method for providing correction data, network-RTK network element and network-RTK rover

Abstract

The invention relates to a method for providing correction data, a network-RTK network element and a network-RTK rover. The invention relates to providing atmospheric correction data for correcting GNSS data in a GNSS network-RTK system, wherein a basic triangulation (6) surrounding at least a portion of a reference station (3) of the GNSS network-RTK system is subdivided into sub-triangles (10) by recursively splitting a parent triangle (10) into four sub-triangles, synthetic data (11) is determined for each of the sub-triangles (10) based on a triangulation algorithm applied to the basic data of the reference station (3), so that the synthetic data (11) represents a grid representation of the basic data, and access to the correction data is provided, wherein the correction data includes at least a portion of the synthetic data (11) arranged in a quadtree hierarchy (100).

Description

Technical Field

[0001] The present invention relates to a method of providing correction data in a GNSS Network-RTK system, a Network-RTK network element, a Network-RTK rover, a computer program product and a GNSS Network-RTK system. Background Art

[0002] Positioning performed by Global Navigation Satellite Systems (GNSS) is subject to dispersive and non-dispersive effects of the atmosphere on the positioning signals, wherein, for example, the ionosphere imposes dispersive delays on the GNSS signals exchanged between satellites and rovers of the GNSS system, while the troposphere imposes non-dispersive delays on the same GNSS signals, depending on the varying electron density due to solar radiation and activity.

[0003] A Network-RTK (Real Time Kinematic) system is a differential positioning technique, for example used in conjunction with GNSS positioning to provide high positioning performance, wherein corrections to the GNSS satellite signals, including corrections for atmospheric biases, are provided in conjunction with reference measurements based on known positions of reference stations.

[0004] For example, a GNSS network-RTK system includes a network of reference stations whose positions are known, wherein correction information is derived based on a comparison of GNSS positioning measurements about the reference stations with the known positions of the reference stations. The network of reference stations typically limits the service area, e.g., a province, a country, a continent, or even the entire Earth. The correction information is then sent to one or more users of the GNSS network-RTK system, typically in real time, in order to improve GNSS positioning performance.

[0005] The derived correction data are valid along the ray paths between the network stations and the observed satellites, ie for the case where the user is located at or at least near the reference station, whereas the quality of the correction decreases when the user is located further away from the reference station.

[0006] Therefore, a model is needed to estimate the correction at any point in the network.

[0007] Such models can rely on the assumption that the interference is spatially correlated and continuous over a region or even over the entire Earth. For example, spherical harmonics are a method commonly applied to model the ionosphere over the entire Earth. However, spherical harmonics have some disadvantages in modeling the regional ionosphere, which have led to the development of other modeling methods.

[0008] Many different Network-RTK positioning / processing techniques are known, for example so-called RTK, VRS (Virtual Reference Station), PPP (Precise Point Positioning), PPP-AR (PPP Positioning with Ambiguity) and PPP-RTK.

[0009] One common approach is to represent the continuous interference region using a (eg uniform) grid of values ​​within the service area, where the expected correction for a specific location is interpolated from the surrounding grid points using a linear or non-linear function (especially based on triangulation principles).

[0010] For example, the interpolation may be based on a Delaunay triangulation, such as described by CN 106970404 A relating to a multiple redundant network RTK atmospheric error interpolation method for VRS processing.

[0011] As another example, WO 2016 / 185500 A1 relates to discretizing space into Delaunay triangles and discloses methods for predicting ionospheric total electron content (TEC) and / or scintillation parameters to provide mitigation algorithms aimed at improving the accuracy of GNSS precise positioning techniques (RTK, NRTK and PPP).

[0012] US 9651668 B2 relates to another possible interpolation method, generally known as the TRIN model (Triangulated Interpolation), in which the ionosphere is compared to a thin layer around the Earth in which the entire electron charge of the ionosphere is accumulated. The model is constructed using an elementary regular polyhedron that is refined by successive subdivisions into triangular faces.

[0013] Some precise positioning techniques also require providing not only the atmospheric model itself but also information about the accuracy of the model to the GNSS processing unit. For example, US 2014 / 0292573 A1 relates to GNSS positioning, in which both the ionospheric model and its accuracy are provided to a device for rover GNSS data processing in order to improve the convergence time of PPP, i.e., the waiting time for a floating position solution to converge to centimeter accuracy.

[0014] In theory, i.e. given a substantially uniform and sufficiently dense distribution of available reference stations, correction data can be derived and / or provided for the service area at any spatial resolution. However, in practice, the performance of GNSS Network-RTK systems (i.e. providing adequate correction data at any spatial resolution) is often limited due to limited bandwidth within the Network-RTK system for broadcasting correction data, lack of available reference stations in some areas to adequately measure atmospheric delays, and the inadequacy of the models themselves to represent the true behavior of atmospheric delays.

[0015] For example, the problem of limited memory and processing power of GPS rovers is addressed by US 7031730 B1. The current position of the GPS receiver is determined, and a boundary is created based on the current position. One or more bands and one or more blocks of grid point correction data are analyzed to determine whether they are within the boundary, for example based on the latitude and longitude of the grid points. The GPS device then purges unnecessary data and corrects the initially determined current position using only the appropriate ionospheric grid point corrections that are within the boundary.

[0016] However, there is still a large data overhead sent to the GPS device, which limits the effective information content of the data sent over a given available bandwidth. Summary of the invention

[0017] The object of the present invention is to provide an improved GNSS Network-RTK system which overcomes the deficiencies of the prior art and in particular provides better access to atmospheric correction data and improved quality control of the correction data.

[0018] The present invention relates to a method for providing correction data in a GNSS network-RTK system with reference stations. According to the present invention, the method comprises the following steps: determining basic data about the reference stations, the basic data of each of the reference stations respectively indicating the influence of the atmosphere on the signal propagation exerted on the satellite signals of the GNSS associated with the corresponding reference station; defining a basic triangulation surrounding at least a part of the reference station; subdividing the basic triangulation into sub-triangles by means of recursive partitioning of a parent triangle into four sub-triangles; determining synthetic data for each of the sub-triangles based on a triangulation algorithm applied to the basic data, so that the synthetic data represents a grid representation of the basic data; and providing access to correction data, wherein the correction data comprises at least a part of the synthetic data arranged in a quadtree hierarchy.

[0019] As an example, the method is used within a Network-RTK GNSS framework comprising a GNSS network processing facility that receives formatted raw GNSS measurements of a set of satellites and other auxiliary information from a network of three or more reference stations distributed in a segment surrounded by a basic triangulation. The processing facility includes a data processing module with a mathematical model and algorithm for estimating measurement biases, and the processing facility feeds the measurements to the data processing module, which then estimates line-of-sight atmospheric delays for each satellite and receiver in the network, performs a recursive partitioning of the basic triangulation, determines synthetic data for each sub-triangle in the triangle, and stores the synthetic data in a quadtree hierarchy. The correction data set is then sent to a Network-RTK rover, which is configured to decode the data set and to interpolate the atmospheric delay for any point within the basic triangulation using the same interpolation function as used for the initial generation of the correction data.

[0020] The quadtree data structure is known in the field of image processing and allows, for example, the storage of image files with varying resolution or tree depth of the image pixel index. However, unlike the segmentation into squares or rectangles used in the field of image processing, the segmentation here is based on triangulation.

[0021] For example, a basic triangulation that encompasses at least a portion of a reference station may be calculated or predefined, wherein the vertices do not necessarily correspond to coordinates of the reference station. The basic triangulation forms a basic level of the quadtree. The basic level is iteratively subdivided a number of times as determined by a given set of criteria, wherein the given set of criteria may vary for different applications.

[0022] Providing correction data based on a quadtree hierarchy enables reducing the amount of data to be broadcast within the GNSS network-RTK system. In particular, the overhead of unnecessary information is minimized, thereby enabling appropriate correction data to be provided at a higher spatial resolution without increasing bandwidth.

[0023] For example, when using deterministic recursive partitioning to generate sub-triangles, only the coordinates of the base level need to be sent with the quadtree, i.e., the order of the nodes of each subdivision level is also determined. However, since this order can also be provided with the quadtree, a determined explicit order is not strictly required.

[0024] Additionally, the accuracy achieved for the respective tree level may also be sent as part of the data comprising the correction data, ie the correction model may provide a combination of accuracy and correction information at the same time.

[0025] In particular, variations in the spatial resolution of the correction model due to variable distribution and / or accuracy requirements of the reference stations can be mapped by the quadtree without generating data overhead. For example, the estimated accuracy of the interpolated correction at any point can be inferred from the encoded accuracy at the quadtree level.

[0026] Empty (0 bit) levels may be used to reserve the hierarchy of subsequent levels in the tree without the overhead of encoding any correction information for that level, and wherein the tree hierarchy may be defined at subsequent levels using a single bit level without the overhead of sending correction information.

[0027] In one embodiment, the correction data comprises at least the coordinates of the vertices of the basic triangulation.

[0028] In another embodiment, a quadtree hierarchy provides nodes and child nodes that order synthetic data into different tree levels, wherein for each tree level, each node includes exactly four or zero child nodes, and wherein: the nodes of the first tree level include synthetic data corresponding to the vertices of those sub-triangles created in the first step of the recursive partitioning; the nodes of the second tree level or another tree level respectively include synthetic data corresponding to the vertices of those sub-triangles created in the second step or another step of the recursive partitioning; the synthetic data is partitioned into valid data groups and invalid data groups; and each node is divided into child nodes only if at least one of its child nodes will include synthetic data of the valid data group. Thus, values ​​associated with sub-triangles that should not be provided in the final quadtree can be excluded based on a set of exclusion criteria.

[0029] As an example, the synthetic data is divided into a valid data group and an invalid data group by the following steps: at least a portion of the synthetic data is marked as invalid, especially when the uncertainty of at least a portion of the synthetic data exceeds a predefined uncertainty threshold. For example, the uncertainty of at least a portion of the synthetic data is derived based on at least one of the following items: uncertainty of basic data, interpolation uncertainty of a triangulation algorithm, and a spatial distance from a sub-triangle corresponding to at least a portion of the synthetic data to a nearest reference station.

[0030] According to another embodiment, the correction data is configured such that for each child node, differential data is provided with respect to data of its parent node, ie the node of the previous tree level which is divided into those child nodes including the corresponding child node.

[0031] In another embodiment, the quadtree hierarchy is configured such that the partitioning of nodes into sub-nodes is stopped based on at least one of: a predefined total number of nodes in the quadtree hierarchy, and a predefined minimum surface area of ​​each sub-triangle represented by a node of the quadtree hierarchy. In other words, the quadtree size and / or maximum spatial resolution for providing correction data is fixed a priori.

[0032] According to another embodiment, the correction data comprises metadata indicating an accuracy range of an accuracy of the corresponding synthesis data, in particular according to a tree level of nodes representing the corresponding synthesis data.

[0033] In another embodiment, the correction data comprises basic data about the reference station.In particular, due to the quadtree data format, the basic data of the reference station may be provided as a triangulation formed from the reference station coordinates without the need for subsequent tessellation.

[0034] A further embodiment is characterized in that the step of providing correction data comprises selecting the synthesis data based on an accuracy criterion and / or based on a maximum tree level of the quadtree hierarchy, in particular wherein only synthesis data for one of the tree levels are selected.

[0035] The present invention also relates to a Network-RTK network element, such as a Network-RTK server, which comprises a computing unit and a communication device, wherein the Network-RTK network element is configured to perform the method according to any one of the above-mentioned embodiments.

[0036] The present invention also relates to a computer program product comprising a program code, which is stored on a machine-readable medium or is specifically implemented by an electromagnetic wave comprising a program code segment, and the program code has computer-executable instructions, which are used to perform at least the following steps, especially when running on a computing unit of a network-RTK network element as described above: reading basic data about a reference station of a GNSS network-RTK system, the basic data of each of the reference stations respectively indicating the influence of the atmosphere on the signal propagation exerted by the satellite signals of the GNSS associated with the corresponding reference station; reading definition data of a basic triangulation or determining a basic triangulation, which surrounds at least a part of the reference station; subdividing the basic triangulation into sub-triangles by means of recursively splitting a parent triangle into four sub-triangles; determining synthetic data for each sub-triangle in a triangle based on a triangulation algorithm applied to the basic data, so that the synthetic data represents a grid representation of the basic data; and providing access to correction data, wherein the correction data includes at least a part of the synthetic data arranged in a quadtree hierarchy.

[0037] The invention also relates to a Network-RTK rover, for example a Network-RTK client, having a computing unit and a communication device configured to receive Network-RTK data, wherein the Network-RTK rover is configured to determine a position based on GNSS data associated with satellite signals of the GNSS, wherein corrections to the GNSS data are performed based on the Network-RTK data in order to take into account the effects on signal propagation exerted by the atmosphere on the satellite signals. The Network-RTK rover is configured to receive correction data as described above, which are Network-RTK data, decode a quadtree hierarchy as described above and provide a reconstruction of at least a part of a sub-triangle and corresponding synthetic data as described above from the correction data, and perform corrections based on the reconstruction.

[0038] In particular, the Network-RTK rover may be configured to read metadata within the correction data as described with respect to the above method, and / or read synthetic data and / or elemental data within the correction data as described with respect to the above method.

[0039] In another embodiment, the Network-RTK rover is configured to determine the accuracy of the corrections from a tree level of nodes of a quadtree hierarchy.

[0040] In yet another embodiment, the Network-RTK rover is configured to request correction data based on an accuracy criterion of the synthesized data and / or based on a maximum tree level of the quadtree hierarchy, in particular to request synthesized data only for one of the tree levels.

[0041] The present invention also relates to a computer program product comprising a program code, which is stored on a machine-readable medium or is specifically implemented by an electromagnetic wave comprising a program code segment, and the program code has computer-executable instructions, which are used to perform at least the following steps, especially when running on a computing unit of the network-RTK device described above: reading GNSS data associated with satellite signals of the GNSS; reading correction data as described above; decoding the quadtree hierarchy as described above, and providing a reconstruction of at least a part of the sub-triangle and the corresponding synthetic data as described above from the correction data; and correcting the GNSS data based on the reconstruction, that is, correcting the influence of the atmosphere on the satellite signal on the signal propagation based on the reconstruction.

[0042] The present invention also relates to a GNSS Network-RTK system, which is configured to perform a method according to any one of the above-mentioned embodiments, and the GNSS Network-RTK system has at least one Network-RTK network element as described above and at least one Network-RTK mobile station according to any one of the above-mentioned embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The GNSS network-RTK system according to the present invention is described or illustrated in more detail below, by way of example only, with reference to the operating examples schematically shown in the accompanying drawings. In the drawings, the same elements are marked with the same reference numerals. The described embodiments are generally not shown true to scale, and they should not be interpreted as limiting the present invention. Specifically,

[0044] Figure 1 is an example implementation of a GNSS network-RTK system according to the present invention;

[0045] Figure 2 is an exemplary tessellation of a basic triangle into sub-triangles;

[0046] Figure 3 is corresponding to Figure 2 An exemplary quadtree structure for tessellation of a basic triangle;

[0047] Figure 4 are exemplary triangular faces of three different parent levels with corresponding boundary vertices;

[0048] Figure 5 is corresponding to Figure 4 The resulting sublevel of the parent triangle with the interpolated vertex values. DETAILED DESCRIPTION

[0049] Figure 1 An example implementation of a GNSS Network-RTK system according to the present invention is shown, the GNSS Network-RTK system comprising a Network-RTK network element 1, which receives formatted raw GNSS measurements from a set of satellites 2 and other auxiliary information from three or more reference stations 3 covering a service area 4 (e.g., a province, a country, or a continent). The Network-RTK network element 1 comprises a computing unit 5, which is configured to estimate line-of-sight atmospheric delays associated with each reference station 3 within the service area 4, wherein the computing unit 5 is further configured to define and recursively split a basic triangulation (e.g., based on one or more basic triangles 6), determine synthetic data of the recursively split sub-triangles, and derive correction data 7 based on the synthetic data stored in a quadtree hierarchy.

[0050] The Network-RTK network element 1 further comprises a communication device 8 configured to transmit the correction data 7 to a Network-RTK rover 9, wherein the Network-RTK rover 9 is configured to decode the quadtree hierarchy in order to derive appropriate correction data to be applied in the GNSS positioning performed at its current position. In other words, the Network-RTK rover 9 is configured to decode the correction data and to interpolate the atmospheric delay for any point within the basic triangle 6 using the same interpolation function as used for the initial generation of the correction data.

[0051] Figure 2 and Figure 3 The storage of correction data according to a quadtree hierarchy is exemplarily illustrated, wherein: Figure 2 The subdivision of a single basic triangular surface into sub-triangles 10 is shown, wherein according to Figure 3 The depicted quadtree structure 100 assigns synthesized data points 11 to sub-triangles 10 .

[0052] In a quadtree, each node 12 has exactly four or zero children (child nodes), wherein the quadtree can be used to partition a two-dimensional space by recursively subdividing the two-dimensional space into four regions 10, each region being assigned to a node 12 of the tree structure. The so-called leaf nodes 13 represent information units, which can be defined differently depending on the application (for example, where the information unit is defined as representing correction data of a desired quality and / or correction data associated with a minimum spatial resolution to be achieved). In addition to the leaf nodes 13 and the regions corresponding to the leaf nodes, the regions 10 / nodes 12 are further subdivided a number of times as determined by a given set of criteria, which may vary for different applications.

[0053] The tree directory follows the spatial decomposition of a quadtree, where the data in the tree pyramid can be compactly stored in an array as an implicit data structure, similar to how a full binary tree can be compactly stored in an array.

[0054] According to the present invention, the quadtree 100 is based on a unique basic level triangulation of a primitive (e.g., a triangle, a square, or a polygon with more than four points) defined by a set of arbitrary points. An arbitrary point may represent a reference station 3 ( Figure 1 ) physical coordinates or any set of base-level vertices of mesh data.

[0055] Based on basic primitives, define basic triangulation, for example Figure 1 As shown, a basic triangle 6 is defined within a square area 4 (basic primitive) of service.

[0056] Then as Figure 2As depicted, each basic triangle is recursively subdivided into four sub-triangles 10, wherein data associated with the sub-triangles is stored in a manner such as Figure 3 In the depicted quadtree 100, data at each child level is stored differentially to the parent level. This format allows data at a child level and all subsequent levels to be removed from the data stream on a per-vertex basis. This enables higher compression and also allows for Figure 2 The spatially variable precision is encoded over the modeled service area indicated by the distribution of the depicted data points 11 (ie, information units), wherein the data points 11 correspond to Figure 3 Leaf nodes 13 in the depicted quadtree. Thus, variable distribution and / or accuracy requirements of reference stations can be mapped via the quadtree without generating data overhead.

[0057] Figure 4 and Figure 5 An exemplary manner for defining a quadtree hierarchy is shown, wherein a NaN value for a parent triangle's boundary vertex index is used to remove a child triangle and associated nodes from the quadtree. NaN stands for Not a Number, but a Number data type value representing an undefined value for defining a quadtree hierarchy.

[0058] Figure 4 depicts three triangular faces with different parent levels, and Figure 5 The resulting corresponding sub-level is depicted. Here, tessellation involves uniformly subdividing each of the parent level triangle faces in sequence.

[0059] Each triangular face may be defined by its three boundary vertex indices 14, wherein the data value of the next tessellation level at each unique vertex index 15 is stored as a difference from the average of the data values ​​associated with the parent level vertex 14. The data for each vertex is stored sequentially in the output data stream, first for the base level and then for subsequent child levels.

[0060] The quadtree hierarchy is defined using data values ​​set to NaN. A parent-level vertex 16 with a value of NaN indicates that the child face associated with that vertex 16 is removed from the quadtree. In the case of interpolation between a pair of vertices 14 on a parent level consisting of a single vertex 16 with NaN, the NaN is treated as zero. For interpolation where both parent vertices 16 are NaN, there is no child vertex. Therefore, Figure 5 shows the interpolation of vertex values ​​15 pairs Figure 4 The sub-faces are encoded at three different parent levels, where for some sub-vertex index 17, no data is sent.

[0061] The maximum level at which the data is encoded can be used to provide an estimate of the precision of the data values ​​at each vertex. Note that the ordering of vertices in the encoded stream is unchanged as they exist.

[0062] Although the present invention has been described above in part with reference to some preferred embodiments, it is to be understood that various modifications and combinations of different features of the embodiments are possible, all of which fall within the scope of the appended claims.

Claims

1. A method for providing correction data in a GNSS network-RTK system having a reference station (3), the method comprising the following steps: obtaining basic data about the reference stations (3), the basic data for each of the reference stations (3) respectively indicating the influence on signal propagation exerted by the atmosphere on the satellite signals of the GNSS associated with the corresponding reference station (3), defining a basic triangulation (6) surrounding at least a portion of said reference station (3), Subdividing the basic triangulation (6) into sub-triangles (10) by recursively splitting the parent triangle into four sub-triangles, Determining synthetic data (11) for each of the sub-triangles (10) based on a triangulation algorithm applied to the basic data, such that the synthetic data (11) represents a mesh representation of the basic data; and providing access to correction data, wherein the correction data comprises at least a portion of the synthesized data (11) arranged in a quadtree hierarchy (100), The correction data is configured such that, for each child node, differential data is provided by data on a parent node of each child node, ie, a node of a previous tree level divided into those child nodes including the corresponding child node.

2. The method according to claim 1, Features The correction data comprises at least the coordinates of the vertices of the basic triangulation (6).

3. The method according to claim 1, Features The quadtree hierarchy (100) provides nodes (12, 13) and child nodes (12, 13) that order the synthetic data (11) into different tree levels, wherein for each tree level, each node (12) includes exactly four or zero child nodes (12, 13), and wherein said nodes (12, 13) of the first tree level comprise synthetic data (11) corresponding to the vertices of those sub-triangles (10) created in the first step of said recursive partitioning, said nodes (12, 13) of the second or further tree level respectively comprising synthetic data (11) corresponding to said vertices of those sub-triangles (10) created in the second or further step of said recursive partitioning, The synthesized data (11) is divided into a valid data group and an invalid data group, and Each node (12) is divided into child nodes (12, 13) only when at least one of its child nodes (12, 13) includes the synthetic data (11) of the useful data set.

4. The method according to claim 3, Features The synthesized data (11) is divided into the valid data group and the invalid data group by the following steps: if the uncertainty of at least a part of the synthesized data exceeds a predefined uncertainty threshold, at least a part of the synthesized data is marked as invalid.

5. The method according to claim 4, Features The uncertainty of the at least a portion of the synthetic data is based on at least one of: The uncertainty of the basic data, The interpolation uncertainty of the triangulation algorithm, and The spatial distance of the sub-triangle (10) corresponding to the at least part of the synthetic data to the nearest reference station (3).

6. The method according to claim 3, Features The quadtree hierarchy (100) is configured such that partitioning of the node (12, 13) into child nodes (12, 13) is stopped based on at least one of: a predefined total number of nodes (12, 13) in the quadtree hierarchy, and A predefined minimum surface area of ​​each sub-triangle (10) represented by a node (12, 13) of the quadtree hierarchy (100).

7. The method according to claim 3, Features The correction data includes at least one of the following: metadata indicating an accuracy range of the accuracy of the corresponding synthetic data (11), and Basic data about the reference station.

8. The method according to claim 7, Features The metadata is based on a tree hierarchy of the nodes (12, 13) representing the corresponding synthetic data (11).

9. The method according to any one of claims 1 to 3, Features The step of providing access to the correction data comprises selecting the synthesis data (11) based on an accuracy criterion and / or based on a maximum tree level of the quadtree hierarchy (100).

10. The method according to claim 9, Features Only synthetic data for one of the tree levels is selected.

11. A network-RTK network element (1), comprising a computing unit (5) and a communication device (8), Features The Network-RTK network element (1) is configured to perform the method according to any one of claims 1 to 10.

12. A computer program product comprising a program code, the program code being stored on a machine-readable medium or embodied by an electromagnetic wave comprising program code segments, and the program code having computer executable instructions for executing at least the following steps when run on a computing unit of a network-RTK network element (1) according to claim 11: reading basic data about reference stations (3) of the GNSS network-RTK system, the basic data of each of the reference stations (3) respectively indicating the influence of the atmosphere on signal propagation exerted on the satellite signals of the GNSS associated with the corresponding reference station (3), reading definition data of a basic triangulation (6) or determining a basic triangulation (6), said basic triangulation (6) enclosing at least a part of said reference station (3), Subdividing the basic triangulation (6) into sub-triangles (10) by recursively splitting the parent triangle into four sub-triangles, determining synthetic data (11) for each of the sub-triangles (10) based on a triangulation algorithm applied to the basic data, such that the synthetic data (11) represents a mesh representation of the basic data, and Provides access to calibration data, where The corrected data includes at least a portion of the synthesized data (11) arranged in a quadtree hierarchy (100).

13. A Network-RTK rover (9) for a GNSS Network-RTK system, the Network-RTK rover (9) having a computing unit and a communication device configured to receive Network-RTK data, wherein: The Network-RTK rover (9) is configured to determine a position based on GNSS data associated with satellite signals of the GNSS, wherein a correction of the GNSS data is performed based on the Network-RTK data in order to take into account the influence on signal propagation exerted by the atmosphere on the satellite signals, Features The network-RTK rover (9) is configured to: receiving correction data as defined in claim 12, said correction data being said Network-RTK data, decoding the quadtree hierarchy (100) as defined in claim 12 and providing a reconstruction of at least a portion of the sub-triangle (10) from the correction data and corresponding synthetic data (11) as defined in claim 12, and The correction is performed based on the reconstruction.

14. The network-RTK rover (9) according to claim 13, Features The Network-RTK rover (9) is configured to determine the accuracy of the correction from a tree level of nodes (12, 13) of the quadtree hierarchy (100).

15. The network-RTK rover (9) according to claim 13 or 14, Features The Network-RTK rover (9) is configured to request the correction data based on an accuracy criterion of the synthesized data (11) and / or based on a maximum tree level of the quadtree hierarchy (100).

16. The Network-RTK rover (9) of claim 15, configured to request synthetic data (11) only for one of the tree levels.

17. A computer program product comprising a program code stored on a machine-readable medium or embodied by an electromagnetic wave comprising program code segments, and having computer executable instructions for executing at least the following steps when run on a computing unit of a Network-RTK rover (9) according to claim 13: Read GNSS data associated with GNSS satellite signals, Reading the correction data defined in claim 12, decoding the quadtree hierarchy (100) as defined in claim 12 and providing a reconstruction of at least a portion of the sub-triangle (10) from the correction data and corresponding synthetic data (11) as defined in claim 12, and The GNSS data is corrected based on the reconstruction.

18. A GNSS network-RTK system, the GNSS network-RTK system being configured to execute the method according to any one of claims 1 to 10, the GNSS network-RTK system having: At least a network-RTK network element (1) according to claim 11, and At least according to claim 13, a Network-RTK rover (9).

Images