A method and system for optimizing UWB base station deployment based on the starfish optimization algorithm
By using the starfish optimization algorithm to delineate obstacle areas in a two-dimensional spatial map and combining it with a multi-dimensional fitness function and mobility strategy to optimize UWB base station deployment, the problem of positioning deviation in complex environments in traditional methods is solved, achieving efficient base station coverage and network connectivity.
Patent Information
- Application Number
- CN202611540235.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-30
- Publication Date
- 2026-11-17
AI Technical Summary
Traditional UWB base station deployment methods struggle to guarantee high-precision, seamless positioning in complex environments, especially in the narrow spaces and equipment-obstructed areas of large intelligent warehousing and logistics centers, leading to a high risk of positioning deviation.
The starfish optimization algorithm is used to divide the obstacle and non-obstacle areas in the two-dimensional spatial map. The coverage is evaluated by a multi-dimensional fitness function, and the base station deployment is optimized by combining centripetal force, random wandering and interactive foraging movement strategies. An undirected graph model of base station backhaul communication is constructed to ensure network connectivity.
It improves the coverage and network connectivity of UWB base station deployment, reduces the risk of positioning deviation in complex environments, reduces signal blind spots, and lowers network construction costs.
Smart Images

Figure CN123123922A_ABST
Abstract
Description
Technical Field
[0001] This application belongs to the field of wireless communication networks, and in particular relates to a method and system for optimizing the deployment of UWB base stations based on the starfish optimization algorithm. Background Technology
[0002] Ultra-wideband (UWB) technology is currently the mainstream solution for high-precision indoor positioning, but its positioning accuracy, coverage, and other system performance are highly dependent on the deployment location of base stations. Traditional UWB base station deployment often relies on manual experience to initially place points according to drawings. This method is highly subjective and, when facing large indoor scenes with complex structures and numerous obstacles, it is difficult to take into account the line-of-sight propagation conditions in all areas, resulting in positioning blind spots or large errors, making it difficult to meet the requirements of high-precision seamless positioning.
[0003] Related technologies often employ swarm intelligence algorithms such as Particle Swarm Optimization (PSO) to assist in deciding on base station locations. This method treats the coordinates of multiple base stations to be deployed as particles in a swarm, defines an objective function to measure the merits of different deployment schemes, and simulates bird foraging behavior to drive the particles to iteratively optimize within the search space, ultimately converging to a better coordinate combination. This process achieves automated optimization of deployment schemes, and compared to manual deployment, it can more effectively explore space and improve the overall coverage of the target area.
[0004] When the aforementioned technologies are deployed as base stations in large intelligent warehousing and logistics centers filled with tall metal shelves and mobile AGVs, in order to maximize the overall coverage across the entire factory space, the algorithm iteration process tends to prioritize base stations occupying open areas where wide coverage is easily achievable. Conversely, relatively fewer base station resources are allocated to narrow spaces divided by partition walls or areas where equipment routinely obstructs the view. Under this deployment pattern, when the target being located moves to these complex local areas, the lack of a sufficient number of redundant effective line-of-sight base stations reduces the stability of the UWB positioning system's calculations, thereby increasing the risk of positioning errors in these specific complex areas. Summary of the Invention
[0005] This application provides a method and system for optimizing UWB base station deployment based on the starfish optimization algorithm, which can reduce the risk of positioning deviation of UWB positioning system in local areas of complex environment.
[0006] In the first aspect, this application provides a UWB base station deployment optimization method based on the starfish optimization algorithm, which establishes a two-dimensional spatial map model of the area to be deployed, and divides the obstacle area and non-obstacle area in the two-dimensional spatial map model;
[0007] Identify several sampling points within the non-obstacle area for assessing coverage;
[0008] Obtain the parameters required for deployment optimization, including the total number of UWB base stations to be deployed, the effective signal coverage radius of a single UWB base station, the population size of the starfish optimization algorithm, and the maximum number of iterations.
[0009] A fitness function is constructed based on indicators such as effective coverage, obstacle coverage, UWB dispersion, and coverage uniformity.
[0010] An initial population containing several starfish individuals is created, and the state vector of each starfish individual is composed of a set of two-dimensional coordinates of several randomly generated UWB base stations to be deployed;
[0011] Perform boundary projection operation on the two-dimensional coordinates of each UWB base station that falls into the obstacle area in the initial population, so that the corresponding UWB base station to be deployed is moved to the boundary of the nearest non-obstacle area, and the current population is obtained.
[0012] The evaluation value of each starfish individual in the current population is calculated based on the fitness function;
[0013] The starfish individual with the lowest evaluation value is determined as the globally optimal individual;
[0014] For each starfish individual in the current population, the state vector is updated by selecting one of centripetal movement, random wandering movement, or interactive foraging movement according to a preset probability rule, thus obtaining the updated population.
[0015] Perform boundary projection operation on the two-dimensional coordinates of each population that falls into the obstacle region to obtain the current population for the next evaluation;
[0016] Determine if the current iteration count has reached the maximum iteration count;
[0017] If not, increase the iteration count by one and perform the step of calculating the evaluation value of each starfish individual in the current population based on the fitness function;
[0018] If so, the state vector of the globally optimal individual is output as the final optimized UWB base station deployment coordinate scheme.
[0019] By employing the aforementioned technical solution, obstacle and non-obstacle regions are divided in a two-dimensional spatial map model, and sampling points are set to evaluate coverage. Combined with the population iterative optimization mechanism of the starfish optimization algorithm, a multi-dimensional fitness function is used to evaluate the performance of the UWB base station deployment scheme. During the iterative optimization process, boundary projection operations ensure that base stations are deployed in non-obstacle regions, and diverse update strategies such as centripetal movement, random wandering movement, and interactive foraging movement are used to explore better deployment schemes. This method can improve the coverage effect of UWB base station deployment while considering actual environmental constraints, making the deployment scheme more reasonable. Since the fitness function comprehensively considers multiple indicators such as effective coverage, obstacle coverage, UWB dispersion, and coverage uniformity, it can improve the overall performance of the deployment scheme and reduce the generation of signal blind spots. The combination of the starfish optimization algorithm's diverse update strategies and boundary projection operations improves the algorithm's ability to find optimal solutions in complex environments and reduces the risk of positioning deviation in local areas of complex environments.
[0020] In conjunction with some implementation methods of the first aspect, in some implementation methods, a fitness function is constructed based on indices such as effective coverage, obstacle coverage, UWB dispersion, and coverage uniformity, specifically including:
[0021] The function for calculating effective coverage is:
[0022]
[0023] For effective coverage, The number of sampling points that are covered by at least a minimum number of UWBs and are not located in an obstacle area. This represents the number of sampling points within the area to be deployed that are not located in obstacle zones.
[0024] The function for calculating obstacle coverage is:
[0025]
[0026] For obstacle coverage, The number of sampling points that are covered by at least one UWB and located in an obstacle area. This represents the total number of sampling points within the obstacle area;
[0027] The function for calculating UWB dispersion is:
[0028]
[0029] in, ;
[0030] In the above function, D represents the UWB dispersion. Let be the total length of the minimum spanning tree formed between UWBs. This is the theoretical maximum length of the minimum spanning tree when all UWBs are distributed to the maximum extent. For the number of UWB, The width of the two-dimensional spatial map of the area to be deployed. The length of the two-dimensional spatial map of the area to be deployed;
[0031] The function for calculating the coverage uniformity index is:
[0032]
[0033] In the above function, To cover uniformity indicators, The standard deviation of the number of UWB coverage points at each sampling point in the unobstructed area. This represents the average number of UWB coverage points at each sampling point in the unobstructed area.
[0034] The fitness function is: ;
[0035] In the above function, To effectively cover weights, For obstacle penalty weights, For dispersion weights, For uniformity weights.
[0036] By adopting the above technical solution, a comprehensive fitness function is constructed that includes indices for effective coverage, obstacle coverage, UWB dispersion, and coverage uniformity to comprehensively evaluate UWB base station deployment schemes. The calculation of effective coverage considers the coverage of sampling points in non-obstacle areas; the calculation of obstacle coverage reflects the degree of obstacle avoidance; the calculation of UWB dispersion assesses the dispersion of base station distribution based on minimum spanning tree; and the coverage uniformity index measures the uniformity of coverage by the standard deviation and average of the number of coverage points. This multi-dimensional evaluation method improves the scientific rigor and rationality of the deployment scheme, and the weighting of each index can balance different performance requirements. Because the calculation function considers and processes various boundary conditions, the robustness of the evaluation process is improved, the probability of invalid solutions is reduced, and the optimization results are more reliable.
[0037] In conjunction with some implementations of the first aspect, in some implementations, centripetal movement is as follows: the current starfish individual moves toward the position of the current global best individual, and its state vector is updated by adding a vector pointing to the state vector of the global best individual, the magnitude of which is affected by random factors.
[0038] Random wandering movement means that the current starfish individual conducts a small-range random exploration around its current position. Its state vector is updated by adding a perturbation vector whose components are randomly generated within a preset small range.
[0039] Interactive foraging movement is as follows: The current starfish individual randomly selects another individual in the current population as the interaction object and moves towards that object. The state vector is updated by adding a vector pointing to the state vector of the selected individual. The magnitude of this vector is affected by random factors.
[0040] By employing the aforementioned technical solution and introducing three different movement modes—centripetal movement, random wandering movement, and interactive foraging movement—the search behavior of starfish individuals in the solution space is enriched. Centripetal movement encourages individuals to move closer to the currently discovered optimal solution, increasing the utilization of known high-quality solutions; random wandering movement enhances local search capabilities through small-scale random exploration; and interactive foraging movement increases population diversity through mutual influence among individuals. This diversified movement strategy improves the algorithm's exploration ability in the solution space and reduces the possibility of getting trapped in local optima. The introduction of random factors during movement increases the algorithm's randomness and uncertainty, improving its ability to escape local optima and making the final deployment scheme closer to the global optimum.
[0041] In conjunction with some implementations of the first aspect, in some implementations, after outputting the state vector of the globally optimal individual as the final optimized UWB base station deployment coordinate scheme, the method further includes:
[0042] Obtain the two-dimensional coordinates of at least one data aggregation gateway pre-defined in the area to be deployed;
[0043] Determine the maximum line-of-sight communication radius of the communication module used for wireless data backhaul in UWB base stations;
[0044] Based on the two-dimensional coordinates of a preset number of UWB base stations and the two-dimensional coordinates of the data aggregation gateway in the final optimized UWB base station deployment coordinate scheme, an undirected graph model of base station backhaul communication is constructed. The vertex set of the undirected graph model of base station backhaul communication consists of all UWB base stations and all data aggregation gateways.
[0045] For any two vertices in the vertex set, if the Euclidean distance between the two vertices is less than or equal to the maximum line-of-sight communication radius and the path connecting the two vertices does not intersect with the obstacle area in the two-dimensional spatial map model, an edge between the two vertices is established in the undirected graph model of the base station backhaul communication to obtain the updated undirected graph model.
[0046] If it is determined that there are connected components in the updated undirected graph model that do not contain any data aggregation gateways, all UWB base stations contained in the connected components are marked as network unreachable nodes, and the network unreachable nodes are added to the final optimized UWB base station deployment coordinate scheme.
[0047] By employing the aforementioned technical solution, an undirected graph model of base station backhaul communication is constructed, considering the effects of line-of-sight communication radius and obstacle occlusion, to model and analyze the communication connection relationship between UWB base stations and data aggregation gateways. In the undirected graph model, edge establishment rules reflect actual communication constraints, and unreachable nodes are identified through connected component analysis. This graph theory-based analysis method improves the accuracy of network connectivity problem analysis. By adding unreachable nodes to the deployment scheme, the degree to which the deployment scheme meets network connectivity requirements is enhanced. Since distance constraints and obstacle occlusion are considered simultaneously when establishing edges, the accuracy of communication link reliability assessment is improved, allowing the deployment scheme to better adapt to the limitations of the actual communication environment.
[0048] In conjunction with some implementations of the first aspect, in some implementations, when it is determined that there are connected components in the updated undirected graph model that do not contain any data aggregation gateways, the specific implementation includes:
[0049] Apply a graph traversal algorithm to traverse and update the undirected graph model, and determine all connected components;
[0050] If there are no vertices belonging to a data aggregation gateway in the vertex set of any connected component, then it is determined that there are connected components in the undirected graph model that do not contain any data aggregation gateway.
[0051] By employing the above technical solution, and by applying a graph traversal algorithm to traverse and update the undirected graph model and determine all connected components, the system can accurately identify different connected regions in the network. Furthermore, it determines network connectivity by checking whether the vertex set of a connected component contains a data aggregation gateway. This systematic analysis method based on graph theory improves the accuracy and efficiency of network connectivity determination.
[0052] In some embodiments, in conjunction with some implementations of the first aspect, after adding network-unreachable nodes to the final optimized UWB base station deployment coordinate scheme, the method further includes:
[0053] Initialize all vertices contained in a connected component that contains at least one data aggregation gateway into the main network vertex set; mark connected components that do not contain any data aggregation gateway as isolated components to be processed, and form an isolated component list of all isolated components to be processed.
[0054] Determine if the list of isolated components is empty;
[0055] If not, select any isolated component to be processed from the list of isolated components as the current processing component;
[0056] Call the relay path calculation subroutine to calculate the new relay node coordinate sequence between the current processing component and the main network vertex set;
[0057] All coordinate points in the newly added relay node coordinate sequence, as well as all vertices in the current processing component, are merged and added to the main network vertex set;
[0058] The newly added relay node coordinate sequence will be added to the final optimized UWB base station deployment coordinate scheme;
[0059] Remove the currently processed component from the list of isolated components and return to the step of determining whether the list of isolated components is empty;
[0060] If so, output the final optimized UWB base station deployment coordinate scheme.
[0061] By adopting the above technical solution, which initializes the connected components, including the data aggregation gateway, into the main network vertex set and processes isolated components iteratively, the systematic nature and coherence of network optimization are improved. When processing each isolated component, a dedicated relay path calculation subroutine is invoked to calculate the optimal connection method, and new nodes are dynamically added to the main network vertex set. This progressive network expansion method improves the rationality of network planning. By maintaining and updating the main network vertex set, the solution improves the continuity of network topology and reduces network fragmentation. By dynamically updating the final deployment coordinate scheme, this method improves the integrity of network coverage and reduces network construction costs. This incremental expansion-based optimization approach enhances the flexibility of network planning, enabling the system to make more adaptive adjustments based on actual conditions.
[0062] In conjunction with some implementations of the first aspect, in some implementations, the relay path calculation subroutine is as follows:
[0063] Traverse the current processing component and the set of vertices in the main network, and determine the vertex pair with the smallest Euclidean distance as the best connection endpoint pair;
[0064] In a two-dimensional spatial map model, a relay deployment path is constructed to connect the optimal connection endpoint pairs. If the relay deployment path does not pass through the obstacle area, the relay deployment path is a straight line segment. If it passes through the obstacle area, the relay deployment path is the shortest broken line that avoids the obstacle.
[0065] Based on the maximum line-of-sight communication radius, several insertion points are determined on the relay deployment path, such that the path distance between any two adjacent nodes on the relay deployment path is no greater than the maximum line-of-sight communication radius.
[0066] Obtain the coordinates of all insertion points and generate a sequence of coordinates for the new relay node.
[0067] By adopting the above technical solution, and finding the vertex pair with the minimum Euclidean distance between the current processing component and the main network vertex set as the optimal connection endpoint, this scheme reduces the number of relay nodes deployed. When constructing relay deployment paths, the system selects straight segments or the shortest polyline avoiding obstacles based on the presence of obstacles; this adaptive path planning method improves the practicality of the deployment scheme. Determining the insertion point on the deployment path based on the maximum line-of-sight communication radius improves the reliability of network connections. Calculating the optimal relay node location reduces network construction costs while simultaneously improving network connectivity.
[0068] In a second aspect, embodiments of this application provide a UWB base station deployment optimization system based on the starfish optimization algorithm. The UWB base station deployment optimization system based on the starfish optimization algorithm includes: one or more processors and a memory; the memory is coupled to one or more processors, and the memory is used to store computer program code, the computer program code including computer instructions, and the one or more processors call the computer instructions to cause the system to perform the method as described in the first aspect and any possible implementation thereof.
[0069] Thirdly, embodiments of this application provide a computer-readable storage medium including instructions that, when executed on a system, cause the system to perform the method described in the first aspect and any possible implementation thereof.
[0070] Fourthly, embodiments of this application provide a computer program product that, when run on a system, causes the system to execute the method described in any possible implementation of the first aspect.
[0071] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:
[0072] 1. This application provides a UWB base station deployment optimization method based on the starfish optimization algorithm. It divides a two-dimensional spatial map model into obstacle and non-obstacle regions, sets sampling points to evaluate coverage, and combines the population iterative optimization mechanism of the starfish optimization algorithm with a multi-dimensional fitness function to evaluate the performance of UWB base station deployment schemes. During the iterative optimization process, boundary projection operations ensure that base stations are deployed in non-obstacle regions, and diverse update strategies such as centripetal movement, random wandering movement, and interactive foraging movement are used to explore better deployment schemes. This method can improve the coverage effect of UWB base station deployment while considering actual environmental constraints, making the deployment scheme more reasonable. Since the fitness function comprehensively considers multiple indicators such as effective coverage, obstacle coverage, UWB dispersion, and coverage uniformity, it can improve the overall performance of the deployment scheme and reduce the generation of signal blind spots. The combination of the diverse update strategies of the starfish optimization algorithm and the boundary projection operation improves the algorithm's ability to find optimal solutions in complex environments and reduces the risk of positioning deviation in local areas of complex environments.
[0073] 2. This application provides a UWB base station deployment optimization method based on the starfish optimization algorithm. By constructing an undirected graph model of base station backhaul communication and considering the effects of line-of-sight communication radius and obstacle occlusion, the communication connection relationship between the UWB base station and the data aggregation gateway is modeled and analyzed. In the undirected graph model, the edge establishment rules reflect the actual communication constraints, and unreachable nodes are identified through connected component analysis. This graph theory-based analysis method improves the accuracy of network connectivity problem analysis. By adding unreachable nodes to the deployment scheme, the degree to which the deployment scheme meets network connectivity requirements is improved. Since distance constraints and obstacle occlusion are considered simultaneously when establishing edges, the accuracy of communication link reliability assessment is improved, allowing the deployment scheme to better adapt to the limitations of the actual communication environment.
[0074] 3. This application provides a UWB base station deployment optimization method based on the starfish optimization algorithm. By initializing the connected components, including the data aggregation gateway, into the main network vertex set and processing isolated components iteratively, this scheme improves the systematicity and coherence of network optimization. When processing each isolated component, a dedicated relay path calculation subroutine is called to calculate the optimal connection method, and new nodes are dynamically added to the main network vertex set. This gradual network expansion method improves the rationality of network planning. By maintaining and updating the main network vertex set, this scheme improves the continuity of network topology and reduces network segmentation. By dynamically updating the final deployment coordinate scheme, this method improves the integrity of network coverage and reduces network construction costs. This step-by-step expansion-based optimization method improves the flexibility of network planning, enabling the system to make more adaptive adjustments based on actual conditions. Attached Figure Description
[0075] Figure 1 This is a flowchart illustrating a UWB base station deployment optimization method based on the starfish optimization algorithm in an embodiment of this application.
[0076] Figure 2 This is another flowchart illustrating a UWB base station deployment optimization method based on the starfish optimization algorithm in an embodiment of this application.
[0077] Figure 3 This is another flowchart illustrating a UWB base station deployment optimization method based on the starfish optimization algorithm in this application embodiment.
[0078] Figure 4 This is a schematic diagram of the physical device structure of a UWB base station deployment optimization system based on the starfish optimization algorithm provided in this application embodiment. Detailed Implementation
[0079] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification and appended claims of this application, the singular expressions “a,” “an,” “the,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this application refers to any or all possible combinations including one or more of the listed items.
[0080] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more.
[0081] The following example is used in conjunction with Figure 1 The following describes a method for optimizing UWB base station deployment based on the starfish optimization algorithm in an embodiment of this application:
[0082] Please see Figure 1 This is a flowchart illustrating a UWB base station deployment optimization method based on the starfish optimization algorithm in an embodiment of this application.
[0083] S101. Establish a two-dimensional spatial map model of the area to be deployed, and divide the obstacle area and non-obstacle area into the two-dimensional spatial map model;
[0084] The system first needs to abstract the real-world deployment environment and establish a two-dimensional spatial map model. This model is not just an image, but a data structure containing precise coordinate information and regional attributes. Within this model, the system must clearly define obstacle areas and non-obstacle areas. Obstacle areas refer to physical spaces occupied by entities such as walls, load-bearing columns, large fixed equipment, and tall shelves, which have characteristics that obstruct radio signal propagation or prevent the installation of base station equipment. Non-obstacle areas refer to free space outside of obstacle areas, allowing free propagation of radio signals and meeting the conditions for installing UWB base stations. During the partitioning process, the system may set specific safety thresholds or expansion thresholds. For example, considering the actual volume of the installed equipment and the impact of signal reflection, the boundary of the obstacle area may be extended outward by a certain distance (e.g., 10-50 cm) from the physical boundary to form a more stringent obstacle avoidance zone.
[0085] In implementing this step, the system can employ grid map construction technology. The system divides the deployment area into equal-sized square grids (e.g., 10cm x 10cm grids), with each grid serving as a basic data unit. Through on-site measurements or imported architectural CAD drawings, grids containing obstacles are marked as 1 (occupied), while grids without obstacles are marked as 0 (empty). Alternatively, a vector map construction technology can be used. The system uses geometric primitives such as points, lines, and polygons to accurately describe the outline boundaries of obstacles. The system reads CAD files or BIM model data from the environment, extracts the corner coordinates of walls and columns, and constructs closed polygons to represent the obstacle area.
[0086] S102. Determine several sampling points within the non-obstacle area for assessing coverage;
[0087] Since radio signals are continuously distributed in space, but computer simulations cannot calculate an infinite number of points, a finite number of sampling points must be selected to approximate the coverage of the entire area. These sampling points should only be distributed within non-obstacle areas, as signal coverage inside obstacles is meaningless and inaccessible. The number and distribution density of sampling points directly affect the accuracy and computational efficiency of the evaluation results. Too few sampling points may lead to missed coverage blind spots; too many sampling points will significantly increase the computational load on the fitness function. When determining sampling points, the system needs to follow the principles of comprehensiveness and representativeness to ensure that the sampling points can uniformly reflect the spatial characteristics of the area to be deployed. In addition, the system may set a sampling interval threshold, which determines the minimum distance between adjacent sampling points to ensure a moderate sparsity of the sampling distribution.
[0088] In practical implementation, the system can employ a uniform grid sampling method. Within the non-obstacle area, the system constructs a virtual grid according to a preset step size, selecting the grid's center point or intersection as sampling points. For grids at boundaries, if the center point falls into an obstacle area, it is discarded or moved to the nearest non-obstacle area boundary. Another implementation method is a random point sampling method. The system randomly generates a large number of coordinate points within the coordinate range of the non-obstacle area and performs validity checks on each generated point, discarding points falling into obstacle areas until the number of valid sampling points reaches a preset total.
[0089] S103. Obtain the parameters required for deployment optimization;
[0090] The system acquires the parameters required for deployment optimization, including the total number of UWB base stations to be deployed, the effective signal coverage radius of a single UWB base station, the population size of the starfish optimization algorithm, and the maximum number of iterations. The system needs to read key control variables from external inputs or internal presets. These parameters directly determine the constraints of the optimization problem and the algorithm's search behavior. The parameters are mainly divided into two categories: one is physical constraint parameters, including the total number of UWB base stations to be deployed and the effective signal coverage radius of a single UWB base station. The number of base stations is usually limited by the project budget, while the coverage radius depends on the transmit power, antenna gain, and path loss characteristics of the selected UWB hardware. The other category is algorithm control parameters, including the population size and maximum number of iterations of the starfish optimization algorithm. The population size determines the number of solutions searched in parallel in each iteration; a larger population size results in a stronger global search capability but a longer iteration time. The maximum number of iterations defines the termination condition of the algorithm and determines the optimization depth. When acquiring these parameters, the system needs to perform validity checks, such as ensuring that the number of base stations is a positive integer and the coverage radius is a positive real number.
[0091] S104. Construct a fitness function based on effective coverage, obstacle coverage, UWB dispersion, and coverage uniformity indices;
[0092] The system constructs a fitness function based on indices such as effective coverage, obstacle coverage, UWB dispersion, and coverage uniformity, specifically including:
[0093] The function for calculating effective coverage is:
[0094]
[0095] For effective coverage, The number of sampling points that are covered by at least a minimum number of UWBs and are not located in an obstacle area. This represents the number of sampling points within the area to be deployed that are not located in obstacle zones.
[0096] The function for calculating obstacle coverage is:
[0097]
[0098] For obstacle coverage, The number of sampling points that are covered by at least one UWB and located in an obstacle area. This represents the total number of sampling points within the obstacle area;
[0099] The function for calculating UWB dispersion is:
[0100]
[0101] in, ;
[0102] In the above function, D represents the UWB dispersion. Let be the total length of the minimum spanning tree formed between UWBs. This is the theoretical maximum length of the minimum spanning tree when all UWBs are distributed to the maximum extent. For the number of UWB, The width of the two-dimensional spatial map of the area to be deployed. The length of the two-dimensional spatial map of the area to be deployed;
[0103] The function for calculating the coverage uniformity index is:
[0104]
[0105] In the above function, To cover uniformity indicators, The standard deviation of the number of UWB coverage points at each sampling point in the unobstructed area. This represents the average number of UWB coverage points at each sampling point in the unobstructed area.
[0106] The fitness function is: ;
[0107] In the above function, To effectively cover weights, For obstacle penalty weights, For dispersion weights, For uniformity weights.
[0108] In practical implementation, for calculating effective coverage, the system traverses all sampling points in non-obstacle areas and calculates the Euclidean distance from each sampling point to all base stations. If the distance is less than the effective coverage radius and the connecting line does not pass through an obstacle, the count is incremented. The proportion of sampling points that meet the minimum coverage requirement is statistically analyzed. For calculating UWB dispersion, the system constructs a complete graph with all base station coordinates as vertices, and the edge weights are the distances between base stations. Then, the Prim algorithm or Kruskal algorithm is used to calculate the total length of the minimum spanning tree and normalize it. For coverage uniformity, the system counts the number of covered base stations for each sampling point, calculates the mean and standard deviation of these values, and uses the reciprocal of the ratio of the standard deviation to the mean (coefficient of variation) to construct an index. The smaller the coefficient of variation, the better the uniformity.
[0109] S105. Create an initial population containing several starfish individuals;
[0110] The system creates an initial population containing several starfish individuals. Each starfish individual's state vector consists of a set of randomly generated two-dimensional coordinates of several UWB base stations to be deployed. In the starfish optimization algorithm, each starfish individual represents a possible UWB base station deployment scheme. Each individual contains a state vector composed of the two-dimensional coordinates of all the UWB base stations to be deployed. For example, if 5 base stations are to be deployed, then the state vector of a starfish individual contains 5 pairs of (x, y) coordinates, for a total of 10 values. All these individuals together constitute the initial population. The quality of initialization has a significant impact on the algorithm's convergence speed and the quality of the final solution. The system needs to generate these coordinates within the two-dimensional space of the area to be deployed, ensuring they fall within the map boundaries.
[0111] In practical implementation, the system can employ a completely random initialization method. Based on the length and width of the map, the system uses a pseudo-random number generator (such as the Mason swivel algorithm) to generate X and Y coordinates for each base station. The generated coordinates are uniformly distributed within the map area. This method is simple and fast, but uneven distribution may occur when the sample size is small. Another implementation approach is to use a chaotic mapping initialization method (such as the Logistic mapping or the Tent mapping). Chaotic sequences are ergodic and non-repeating. Mapping chaotic variables to the map coordinate space can make the generated initial population more evenly distributed in the search space, reducing the possibility of initial population clustering and thus improving the algorithm's ability to escape local optima.
[0112] S106. Perform boundary projection operation on the two-dimensional coordinates of each UWB base station that falls into the obstacle area in the initial population, so that the corresponding UWB base station to be deployed is moved to the boundary of the nearest non-obstacle area, and the current population is obtained.
[0113] Since the random generation process in S105 may not completely avoid obstacles, or invalid solutions may be allowed to be generated in order to maintain the randomness of the algorithm, S106 must force correction of these solutions. The system traverses every starfish individual in the initial population and checks every base station coordinate in its state vector. If a coordinate point is found to be within the obstacle region defined in S101, the system performs a boundary projection operation. The geometric meaning of this operation is to find the boundary point of the non-obstacle region that is closest to the invalid point in Euclidean distance and move the base station coordinate to this boundary point. This simulates the operation in actual engineering where, if the design location is inside a pillar, the engineer would install the base station on the surface of the pillar.
[0114] In practical implementation, the system can employ a polygon nearest-point projection method. If the obstacle is represented as a polygon, the system calculates the perpendicular distance from invalid points to each edge of the polygon or to a vertex, and finds the point with the minimum distance as the projection target. If the perpendicular's foot lies on the line segment, the projection is made to the foot of the perpendicular; if the foot of the perpendicular lies on the extension of the line segment, the projection is made to the nearest endpoint. Another implementation method is a grid gradient-based search method. In the grid map, the system uses the grid cell containing the obstacle as the center and performs a spiral or breadth-first search outwards until it finds the first grid cell marked as a non-obstacle. The base station coordinates are then updated to the center coordinates of this non-obstacle grid cell.
[0115] S107. Calculate the evaluation value of each starfish individual in the current population based on the fitness function;
[0116] The system calls the fitness function constructed by S104, substituting the UWB base station coordinates carried by each starfish individual into the function for calculation. The calculation process involves a large number of geometric calculations, including the distance between points, the containment relationship between points and polygons, the construction of the minimum spanning tree, and the calculation of statistical variance. Finally, each starfish individual will obtain a unique value—the evaluation value. According to the definition of the fitness function (usually containing a negative sign or calculating the maximum / minimum value), the smaller (or larger) this value, the better the performance of the deployment scheme.
[0117] S108. The starfish individual with the smallest evaluation value is determined as the global optimal individual;
[0118] In the starfish optimization algorithm, the globally optimal individual holds a special position. It not only represents the current highest level but also serves as a guide (target for centripetal force movement) for the movement of other individuals in subsequent iterations. The system iterates through the evaluation values of all individuals calculated by S108 and performs comparison operations.
[0119] S109. For each starfish individual in the current population, according to the preset probability rules, select one of centripetal movement, random wandering movement or interactive foraging movement to update the state vector and obtain the updated population.
[0120] The system updates the state vector for each starfish individual in the current population by performing one of three actions based on preset probability rules: centripetal movement, random wandering movement, or interactive foraging movement, thus updating the population. Centripetal movement involves the current starfish individual moving towards the position of the current globally optimal individual; the state vector is updated by adding a vector pointing to the state vector of the globally optimal individual. Random wandering movement involves the current starfish individual randomly exploring within a preset range near its current position; the state vector is updated by adding a perturbation vector whose components are randomly generated within a preset small range. Interactive foraging movement involves the current starfish individual randomly selecting another individual in the current population as its interaction target and moving towards it; the state vector is updated by adding a vector representing the state vector of that other individual.
[0121] S110. Perform boundary projection operation on the two-dimensional coordinates of each population that falls into the obstacle region to obtain the current population for the next evaluation.
[0122] Since starfish movement (especially random wandering and interactive foraging) is based on mathematical vector operations, the updated coordinates are highly likely to cross obstacle boundaries again or even completely exit the map. To ensure the algorithm always searches within the valid solution space, the system must perform boundary projection again. The logic of this step is exactly the same as S106, but it applies to a new population that has undergone one round of evolution. The system checks the validity of each new location, forcibly pulling illegal locations back to the nearest non-obstacle boundary. This ensures that each iteration evaluates a physically feasible deployment plan, avoiding waste of computational resources.
[0123] In practical implementation, the system can reuse the polygon nearest point projection method or the grid gradient search method in S106. To improve efficiency, the system can first perform a fast bounding box detection. If the point is outside the bounding rectangle of the obstacle, there is no need for complex geometric projection calculations; fine detection is only performed within the bounding box. Another implementation method is to use a bounce strategy. When the calculated new position falls into the obstacle, the system does not project it onto the boundary. Instead, it assumes that the obstacle boundary is elastic, calculates the incident angle and reflection angle, and bounces the point back to the non-obstacle region.
[0124] S111. Determine whether the current iteration count has reached the maximum iteration count;
[0125] The system reads the current iteration counter value and compares it with the maximum number of iterations set in S103. This decision determines whether the algorithm continues with the next round of population update and evaluation, or terminates and outputs the result. The maximum number of iterations is an empirical value, usually set based on the complexity of the problem and the real-time requirements.
[0126] S112, Increase the number of iterations by one;
[0127] If not, increase the iteration count by one and perform the step of calculating the evaluation value of each starfish individual in the current population based on the fitness function.
[0128] S113. Output the state vector of the globally optimal individual as the final optimized UWB base station deployment coordinate scheme.
[0129] If so, the state vector of the globally optimal individual is output as the final optimized UWB base station deployment coordinate scheme.
[0130] In the above embodiments, obstacle and non-obstacle regions are divided in a two-dimensional spatial map model, and sampling points are set to evaluate coverage. Combined with the population iterative optimization mechanism of the starfish optimization algorithm, a multi-dimensional fitness function is used to evaluate the performance of the UWB base station deployment scheme. During the iterative optimization process, boundary projection operations ensure that base stations are deployed in non-obstacle regions, and diverse update strategies such as centripetal movement, random wandering movement, and interactive foraging movement are used to explore better deployment schemes. This method can improve the coverage effect of UWB base station deployment while considering actual environmental constraints, making the deployment scheme more reasonable. Since the fitness function comprehensively considers multiple indicators such as effective coverage, obstacle coverage, UWB dispersion, and coverage uniformity, it can improve the overall performance of the deployment scheme and reduce the generation of signal blind spots. The combination of the diverse update strategies of the starfish optimization algorithm and the boundary projection operation improves the algorithm's ability to find optimal solutions in complex environments and reduces the risk of positioning deviation in local areas of complex environments.
[0131] After completing the initial deployment optimization of the base stations, in order to ensure the reliable backhaul of data from the UWB positioning system, it is also necessary to consider the communication connectivity between the UWB base stations and the data aggregation gateway. Therefore, this application also provides another UWB base station deployment optimization method based on the starfish optimization algorithm, which evaluates and optimizes network connectivity by constructing an undirected graph model of base station backhaul communication. The following section combines... Figure 2 Another UWB base station deployment optimization method based on the starfish optimization algorithm in the embodiments of this application is described:
[0132] Please see Figure 2 This is another flowchart illustrating a UWB base station deployment optimization method based on the starfish optimization algorithm in this application.
[0133] S201. Obtain the two-dimensional coordinates of at least one data aggregation gateway preset in the area to be deployed;
[0134] A data aggregation gateway is a core network device in a UWB positioning system responsible for collecting raw positioning data (such as TDOA timestamps and TOF distance values) transmitted wirelessly or wired from various UWB base stations (Anchors) and forwarding this data to the upper-layer positioning calculation server. Gateways typically connect to the backbone network (such as Ethernet or fiber optics) and act as a bridge between the wireless sensor network and the backend server. The presence of at least one gateway indicates system scalability; in large and complex scenarios, multiple gateways are usually deployed to ensure coverage and load balancing, and the system must be able to handle the topology of multiple gateways simultaneously. The system identifies the physical locations of these gateways on the map by reading pre-configured parameters or files. These locations are usually limited by the location of weak current wells, network interfaces, or power outlets on site, and are therefore typically considered known and fixed constants (Fixed Nodes) before the base station optimization algorithm runs, but can also be used as variable parameters in certain specific embodiments.
[0135] S202. Determine the maximum line-of-sight communication radius of the communication module used for wireless data backhaul of UWB base stations;
[0136] A communication module refers to a wireless transceiver unit integrated within a UWB base station for building a backhaul network. This is typically different from a UWB RF chip used for ranging; it may be a Wi-Fi (e.g., 802.11n / ac / ax), ZigBee, LoRa, or a proprietary 2.4GHz / 5.8GHz band wireless module. Of course, in some mesh networking schemes, UWB chips can also be reused for data communication. The maximum line-of-sight communication radius refers to the farthest distance at which a communication module can maintain a stable connection while meeting minimum throughput and maximum packet loss rate requirements under ideal, unobstructed, straight-line propagation conditions. This is a crucial threshold parameter that directly determines whether two nodes in a graph theory model have the potential to establish an edge. This radius depends not only on the transmit power and receive sensitivity but also on the operating frequency band, antenna gain, and the path loss exponent of the environment. The system must determine a specific value (e.g., 50 meters or 100 meters) as a geometric constraint for subsequent connectivity assessments. It should be noted that this radius should be a reliable radius that takes into account a certain engineering margin, rather than the theoretical limit of transmission distance, in order to ensure the robustness of the system.
[0137] S203. Based on the two-dimensional coordinates of a preset number of UWB base stations and the two-dimensional coordinates of the data aggregation gateway in the final optimized UWB base station deployment coordinate scheme, construct an undirected graph model for base station backhaul communication.
[0138] Based on the two-dimensional coordinates of a preset number of UWB base stations and the two-dimensional coordinates of the data aggregation gateway in the final optimized UWB base station deployment coordinate scheme, the system constructs an undirected graph model for base station backhaul communication. The vertex set of the undirected graph model for base station backhaul communication consists of all UWB base stations and all data aggregation gateways.
[0139] The preset number is the total number of base stations determined in S103. The undirected graph model of base station backhaul communication is a mathematical structure G=(V,E), where V represents the vertex set and E represents the edge set. In this step, the system mainly focuses on constructing the vertex set V. The vertex set consists of two parts: one part is all UWB base station nodes, and the other part is all data aggregation gateway nodes. The reason for constructing an undirected graph is that in wireless backhaul networks, communication is usually bidirectional (base stations send data, and gateways reply with ACK), and the physical connectivity of the link is usually symmetrical in both directions. The system uniformly maps these scattered coordinate points into node objects in the graph model, with each node carrying its ID, type, and two-dimensional coordinate attributes. This step has not yet established the connection relationships between nodes; it has only completed the initialization of the graph structure, equivalent to drawing all the points on a blank sheet of paper, ready to connect them.
[0140] In implementing this step, the system can employ object-oriented data structure construction techniques. The system defines a `Node` class containing member variables such as `id`, `type` (enumeration values: `ANCHOR` / `GATEWAY`), and `position` (x, y). The system iterates through the base station coordinate list output by S114, instantiating the corresponding number of `Node` objects and adding them to a list or set; then it iterates through the gateway coordinate list obtained by S201, similarly instantiating `Node` objects and adding them to a set. Finally, the system instantiates a `Graph` object and assigns the above node set to its vertex attributes. This approach has a clear structure and is easy to extend with other node attributes (such as remaining battery power and channel number). Another implementation method is to use an adjacency matrix initialization technique. Assuming the number of base stations is N and the number of gateways is M, the system creates a two-dimensional matrix of dimension (N+M) times (N+M). The row and column indices of the matrix correspond to each device. In this step, the system first establishes a mapping table between indices and device coordinates and initializes all elements of the matrix to 0 or infinity (indicating no connection yet).
[0141] S204. For any two vertices in the vertex set, if the Euclidean distance between the two vertices is less than or equal to the maximum line-of-sight communication radius and the path connecting the two vertices does not intersect with the obstacle area in the two-dimensional spatial map model, establish the edge between the two vertices in the undirected graph model of the base station backhaul communication to obtain the updated undirected graph model.
[0142] The system needs to traverse every pair of nodes in the vertex set to determine whether a valid wireless communication link exists between them. Two necessary conditions must be met simultaneously: First, the Euclidean distance constraint, meaning the straight-line distance between the coordinates (x_1, y_1) and (x_2, y_2) of two nodes, sqrt{(x_1-x_2)^2+(y_1-y_2)^2}, must be less than or equal to the maximum line-of-sight communication radius determined in S202. This ensures sufficient signal strength. Second, the line-of-sight unobstructed constraint, meaning the line segment connecting these two nodes cannot pass through any pixels or polygons marked as obstacles in the two-dimensional spatial map established in S101. This ensures the signal is not blocked by walls or metal devices. Only when both conditions are true will the system add an edge between these two vertices in the undirected graph model. This edge represents a physically feasible direct communication link. After traversing and judging all node pairs, the originally discrete points are connected by edges into a mesh structure, thus updating the undirected graph model.
[0143] In implementing this step, the system can employ a grid collision detection technique based on the Bresenham line algorithm. For any two points, the system first calculates their Euclidean distance and compares it with the radius. If the distance condition is met, the system uses the Bresenham algorithm to calculate the coordinate sequence of all grid cells traversed by the line segment connecting the two points. The system sequentially checks the attribute values of these grid cells in the map model. Once any grid cell is found to belong to an obstacle region, the link is immediately deemed invalid, and the check stops; if all traversed grid cells are non-obstacles, the link is deemed valid, and the corresponding position in the adjacency matrix is set to 1, or a neighbor node is added to the adjacency list. This method is suitable for grid maps and has a fast calculation speed. Another implementation method is to use a line segment and polygon intersection detection algorithm (such as the Separating Axis Theorem (SAT) or ray method). In a vector map, the system treats the line connecting two nodes as a line segment and obstacles as a set of polygons. The system traverses all obstacle polygons and determines whether the line segment intersects any edge of the polygon. If an intersection point exists, the link is blocked. This method offers higher geometric accuracy and is suitable for scenarios with strict requirements for obstacle edge detection.
[0144] S205. If it is determined that there is a connected component in the updated undirected graph model that does not contain any data aggregation gateway, mark all UWB base stations contained in the connected component as network unreachable nodes, and add the network unreachable nodes to the final optimized UWB base station deployment coordinate scheme.
[0145] The graph traversal algorithm is applied to traverse and update the undirected graph model to determine all connected components. If there are no vertices belonging to the data aggregation gateway in the vertex set of any connected component, it is determined that there are connected components in the updated undirected graph model that do not contain any data aggregation gateway. All UWB base stations contained in the connected component are marked as network unreachable nodes, and the network unreachable nodes are added to the final optimized UWB base station deployment coordinate scheme.
[0146] A connected component is a maximal subgraph in an undirected graph, where any two vertices are connected by a path. A complete network graph may contain one or more connected components. The system needs to examine the vertex set of each connected component to determine if it contains nodes of type data aggregation gateway. If a connected component consists entirely of UWB base station nodes and has no gateway nodes, it means that although all base stations within that component may be connected to each other, as a whole, they are disconnected from the outside world (the server). Data generated by these base stations cannot be transmitted back, and therefore they are called unreachable nodes. The system specially marks these nodes and incorporates this information into the final deployment plan. This not only outputs coordinates but also the network health status of that coordinate point, indicating the remedial measures that need to be taken.
[0147] In implementing this step, the system can use a breadth-first search (BFS) traversal algorithm. The system creates a visit marker array `visited`, initialized to `false`. The system traverses all nodes in the graph. For each unvisited node, it initiates a BFS search starting from that node to find the entire connected component to which that node belongs. During the search, the system maintains a boolean variable `has_gateway`. If a gateway node is encountered while traversing the component, it is set to `true`. After the traversal, if `has_gateway` is `false`, all nodes in the component are marked as unreachable. The system repeats this process until all nodes have been visited. Alternatively, a disjoint-set data structure (Union-Find / Disjoint SetUnion) can be used. The system initializes each node as an independent set. The system traverses each edge in the graph, merging the sets containing the nodes at both ends of the edge. After traversing all edges, several disjoint sets are formed in the graph. The system checks the representative element (Root) of each set and traverses all members of the set. If no member of a set is a gateway, then all base stations in that set are unreachable nodes. The disjoint-set data structure (Union-Set Union) algorithm is highly efficient in handling dynamic connectivity problems and is suitable for situations with a large number of nodes.
[0148] In the above embodiments, an undirected graph model of base station backhaul communication is constructed, and the effects of line-of-sight communication radius and obstacle occlusion are considered to model and analyze the communication connection relationship between the UWB base station and the data aggregation gateway. The edge establishment rules in the undirected graph model reflect actual communication constraints, and unreachable nodes are identified through connected component analysis. This graph theory-based analysis method improves the accuracy of network connectivity problem analysis. By adding unreachable nodes to the deployment scheme, the degree to which the deployment scheme meets network connectivity requirements is improved. Since distance constraints and obstacle occlusion are considered simultaneously when establishing edges, the accuracy of communication link reliability assessment is improved, allowing the deployment scheme to better adapt to the limitations of the actual communication environment.
[0149] After identifying unreachable nodes in the network, to improve overall network connectivity, it is necessary to address the network segmentation problem by rationally deploying relay nodes. To this end, this application also provides another UWB base station deployment optimization method based on the starfish optimization algorithm, which achieves complete network connectivity by gradually processing isolated network components and adding necessary relay nodes. The following section combines... Figure 3 This application describes another method for optimizing UWB base station deployment based on the starfish optimization algorithm:
[0150] Please see Figure 3 This is another flowchart illustrating a UWB base station deployment optimization method based on the starfish optimization algorithm in this application.
[0151] S301. Initialize all vertices contained in a connected component that contains at least one data aggregation gateway into a main network vertex set; mark connected components that do not contain any data aggregation gateway as isolated components to be processed, and form an isolated component list of all isolated components to be processed.
[0152] Isolated components to be processed refer to groups of base stations that, although internally interconnected, cannot communicate with any data aggregation gateway as a whole. An isolated component list is a data structure used to store information about these isolated components, typically in the form of a queue or list. The system traverses the entire network topology and, based on the connectivity relationships between nodes, uses graph theory algorithms to divide the network into several independent connected regions. Subsequently, the system checks whether a data aggregation gateway exists within each connected region. If it exists, all nodes within that region are added to the main network vertex set as the basis for subsequent expansion; if not, the region is marked as an isolated component and added to the pending processing list, awaiting subsequent connection via relay nodes. This process ensures that the optimization algorithm can clearly distinguish between connected regions and regions requiring repair, providing a clear target for subsequent iterative optimization.
[0153] The system can implement this step using a breadth-first search (BFS) algorithm. Starting with all known data aggregation gateway nodes as the initial search layer, the system expands outwards to search all adjacent nodes, marking all visited nodes as main network members. After completing all traversals starting from the gateways, the remaining unmarked nodes in the system are considered isolated components. The system then performs a second traversal on these unmarked nodes. For each unvisited node, a new traversal (such as BFS or DFS) is initiated to find the entire connected subgraph to which the node belongs, storing it as an independent isolated component object in a list. Alternatively, a union-find data structure can be used. The system first initializes each node as an independent set, then traverses all existing edges in the network, merging sets containing two nodes connected by an edge. After the traversal is complete, the system checks the root node or set attributes of each set to determine if the set contains the ID of the data aggregation gateway. Sets containing gateways are merged into the main network vertex set, while sets not containing gateways are encapsulated as isolated component objects and added to the isolated component list.
[0154] S302. Determine if the isolated component list is empty;
[0155] After the system completes initialization or completes one isolated component processing, it needs to perform a loop control judgment to decide whether to continue the optimization process or terminate the algorithm and output the result.
[0156] S303. Select any isolated component to be processed from the list of isolated components as the current processing component.
[0157] If not, select any isolated component from the isolated component list as the current processing component. After confirming the existence of isolated components to be processed, the system needs to extract a specific component from the list as the operation object for the current algorithm cycle. In this step, the selection action is not just a simple data reading; it usually involves marking the element from the pending state to the processing state, or temporarily locking the element to prevent concurrent conflicts. The currently processed component refers to the target subgraph that the system attempts to connect to the main network in this iteration. The system accesses the isolated component list and obtains a component object according to the preset selection strategy. This object contains information such as the coordinates, IDs, and internal connection relationships of all base station nodes in the isolated network. The selection process is key to serializing parallel problems. By focusing on only one isolated component at a time, the system can decompose the complex network connectivity problem into a series of simple point-to-point or point-to-set path planning subproblems.
[0158] S304. Call the relay path calculation subroutine to calculate the new relay node coordinate sequence between the current processing component and the main network vertex set;
[0159] The relay path calculation subroutine is an independent functional module encapsulating complex optimization logic. In the context of this application, it specifically refers to a program that uses the Starfish Optimization Algorithm (SFOA) or similar intelligent biomimetic algorithms to find the optimal location. The newly added relay node coordinate sequence refers to a set of ordered two-dimensional or three-dimensional coordinate points output by the calculation, representing the locations of the newly deployed UWB relay base stations. The system inputs all nodes in the currently processed component as the source set and all nodes in the main network vertex set as the target set (or vice versa) into the subroutine. The subroutine needs to consider constraints such as signal transmission distance limitations, environmental obstacles, and relay hop count limitations. The goal is to find the minimum number of relay nodes, or, with a fixed number of nodes, the path with the best signal quality, so that at least one node in the isolated component can establish connectivity with any node in the main network through these newly added relay nodes. This step is the most technically demanding part of the entire method, directly determining the cost (number of relays) and quality (connectivity stability) of network construction.
[0160] The system can implement this step using a search strategy based on the Starfish Optimization Algorithm (SFOA). The system initializes a group of starfish as search agents, each representing a potential relay node deployment scheme (i.e., a set of coordinates). The fitness function is designed to evaluate the merits of the scheme, primarily considering the number of relay nodes used (the fewer the better), whether the distance between adjacent nodes is within the effective communication range, and whether the path avoids known obstacle areas. During iteration, the starfish agents update their positions (i.e., adjust their relay coordinates) according to the SFOA-specific attachment, predation, and regeneration behavior patterns, gradually converging to the fitness-optimal solution. The final optimal solution output is the required sequence of new relay node coordinates. Another implementation combines geometric computation with graph search algorithms. The system first calculates the nearest pair of nodes between the isolated component and the main network, connecting these two points to form a straight line segment. Then, the system segments this line segment according to the maximum effective communication radius of the UWB base station, placing a candidate relay node at intervals. If there are obstacles on the direct path, the system uses the A* (A-Star) algorithm or the RRT (Rapid Expanding Random Tree) algorithm to bypass the obstacles and find a path, and samples the generated path trajectory as needed to generate a sequence of relay node coordinates.
[0161] S305. Merge all coordinate points in the newly added relay node coordinate sequence and all vertices in the current processing component into the main network vertex set;
[0162] Merging and adding refers to the union operation in set theory, which adds new elements to an existing set while ensuring the uniqueness of the elements. At this point, the main network vertex set expands, absorbing not only the newly generated relay nodes but also all the original base station nodes that belonged to the currently processed component. This means that the previously isolated component is now logically and physically part of the main network through the newly added relay nodes. The system performs this operation to maintain the consistency of the network state, ensuring that subsequent iterations (i.e., when processing the next isolated component) can utilize these newly added nodes as potential connection targets. In other words, the newly added nodes may become bridges connecting to the next, more distant isolated component. This step demonstrates the dynamic expansion characteristic of the algorithm; the main network, like a snowball, continuously absorbs isolated components and relay nodes as iterations progress, until it covers the entire network.
[0163] The system can implement this step using a HashSet data structure. The system iterates through the sequence of coordinates of the newly added relay node and each node object in the currently processed component. For each node, the system generates a hash value based on its coordinates or unique ID. The system calls the `add` or `insert` method of the main network set to add these nodes to the set. Because the HashSet has automatic deduplication, if a node (e.g., a boundary node) is already in the main network, the system automatically ignores duplicate additions, thus ensuring the uniqueness of elements in the set. After the operation is complete, the metadata of the main network set (such as the total number of nodes, bounding box range, etc.) will be updated synchronously. Another implementation method is a graph structure update based on an adjacency matrix or adjacency list. The system not only adds the ID of the new node to the node list but also adds the corresponding row and column to the adjacency matrix. Based on the new node's location and communication radius, the system calculates the connectivity between the new node and existing nodes in the main network, as well as the connectivity between the new node and existing nodes, and sets the corresponding position in the adjacency matrix to 1 (indicating connectivity). Simultaneously, the system also updates the matrix with the connection relationships between existing nodes in the currently processed component and the new relay node. This approach not only updates the node set but also completely updates the network's edge information, providing comprehensive data for potential full-network path analysis later on.
[0164] S306. Add the newly added relay node coordinate sequence to the final optimized UWB base station deployment coordinate scheme;
[0165] Unlike S305, which focuses on updating the main network set (which focuses on the internal logical connectivity of the algorithm), S306 focuses on accumulating output results. The system extracts the coordinate sequence of the newly added relay nodes calculated in S304 and appends it to this result scheme. This scheme typically includes not only coordinates (X, Y, Z), but may also include node type markers (e.g., relay nodes), the connection segment ID to which they belong, and estimated signal coverage parameters. This step ensures that every valid computational result of the algorithm is securely saved and forms part of the final deliverable. It serves as a bridge connecting the logical world of the algorithm with the physical implementation world, ensuring the traceability of the optimization process and the integrity of the results.
[0166] S307. Remove the currently processed component from the list of isolated components;
[0167] After completing path calculation, network merging, and result recording for the current isolated component, the system must update the list of tasks to be processed to reflect the current progress. In this step, the removal operation refers to completely deleting the isolated component object that was just processed from the isolated component list or marking it as completed. This is a critical state update step, directly affecting the loop condition in step S302. If the system does not execute this step, the size of the isolated component list will never decrease, causing the S302 condition to always be negative, leading to an infinite loop where the algorithm processes the same component repeatedly. Through the removal operation, the system recognizes that the component is no longer isolated (because it has been merged into the main network in S305), thus directing the algorithm's attention to the remaining unprocessed components. This step ensures the convergence of the algorithm, making the problem size monotonically decrease with iterations until all tasks are completed.
[0168] S308, Output the final optimized UWB base station deployment coordinate scheme.
[0169] If the judgment result of S302 is yes, the final optimized UWB base station deployment coordinate scheme is output. After confirming that the isolated component list is empty, meaning that all known network islands have been successfully connected to the main network, the system performs the final delivery action. In this step, the output refers to presenting the deployment scheme, which has been continuously accumulated and improved in step S306, in a user-readable or machine-readable format. This final optimized UWB base station deployment coordinate scheme not only includes the original base station layout but also the coordinates of all newly added relay nodes calculated by the algorithm, forming a complete and fully connected network planning diagram. The output can be a graphical interface display or a file export. This step marks the formal end of the entire base station deployment optimization process based on the Starfish optimization algorithm. The output data will serve as a guide for actual engineering construction. Construction personnel will install UWB base stations on the physical site based on these coordinates, or simulation software will use this data for final signal coverage simulation verification. Therefore, the accuracy and format standardization of the output data are crucial to the practical value of the algorithm.
[0170] The system can implement this step by generating a visual report. The system calls a plotting library (such as Matplotlib, OpenGL, or Web Canvas) to plot the locations of all base stations in a two-dimensional or three-dimensional coordinate system. Different colors or icons are used to distinguish original base stations, data aggregation gateways, and newly added relay nodes, and lines connect interconnected nodes to illustrate the network topology. The final generated image file (JPG / PNG) or interactive webpage is displayed to the user, visually presenting the optimization effect. Alternatively, the system can export a standard format data file. The system serializes the final coordinate scheme into JSON, XML, or CSV format. The file details the ID, type (gateway / base station / relay), X / Y / Z coordinates, and possible parent node IDs for each node. The system saves this file to a specified disk path or sends it to the downstream engineering management system via a network API interface.
[0171] In the above embodiments, by initializing the connected components, including the data aggregation gateway, into the main network vertex set and processing isolated components iteratively, this scheme improves the systematicness and coherence of network optimization. When processing each isolated component, a dedicated relay path calculation subroutine is invoked to calculate the optimal connection method, and new nodes are dynamically added to the main network vertex set. This progressive network expansion method improves the rationality of network planning. By maintaining and updating the main network vertex set, this scheme improves the continuity of network topology and reduces network fragmentation. By dynamically updating the final deployment coordinate scheme, this method improves the integrity of network coverage and reduces network construction costs. This incremental expansion-based optimization approach improves the flexibility of network planning, enabling the system to make more adaptive adjustments based on actual conditions.
[0172] The system in the embodiments of this invention is described below from the perspective of hardware processing. Please refer to [link / reference needed]. Figure 4 This is a schematic diagram of the physical device structure of a UWB base station deployment optimization system based on the starfish optimization algorithm provided in an embodiment of this application.
[0173] It should be noted that, Figure 4 The structure of the system shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.
[0174] like Figure 4 As shown, the system includes a Central Processing Unit (CPU) 401, which can perform various appropriate actions and processes based on a program stored in Read-Only Memory (ROM) 402 or a program loaded from storage portion 408 into Random Access Memory (RAM) 403, such as executing the methods described in the above embodiments. The RAM 403 also stores various programs and data required for system operation. The CPU 401, ROM 402, and RAM 403 are interconnected via a bus 404. An Input / Output (I / O) interface 405 is also connected to the bus 404.
[0175] The following components are connected to I / O interface 405: input section 406 including a camera, infrared sensor, etc.; output section 407 including a liquid crystal display (LCD) and speakers, etc.; storage section 408 including a hard disk, etc.; and communication section 409 including a network interface card such as a LAN (Local Area Network) card and a modem, etc. Communication section 409 performs communication processing via a network such as the Internet. Drive 410 is also connected to I / O interface 405 as needed. Removable media 411, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., are installed on drive 410 as needed so that computer programs read from it can be installed into storage section 408 as needed.
[0176] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing computer programs for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 409, and / or installed from removable medium 411. When the computer program is executed by central processing unit (CPU) 401, it performs the various functions defined in the present invention.
[0177] It should be noted that the computer-readable medium shown in the embodiments of the present invention can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In the present invention, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, wherein a computer-readable computer program is carried. The transmitted data signal can take many forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof.
[0178] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. Each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.
[0179] In another aspect, the present invention also provides a computer-readable storage medium, which may be included in the system described in the above embodiments; or it may exist independently and not assembled into the system. The storage medium carries one or more computer programs that, when executed by a processor of a system, cause the system to implement the methods provided in the above embodiments.
[0180] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.
[0181] As used in the above embodiments, depending on the context, the term "when..." can be interpreted as "if...", "after...", "in response to determining...", or "in response to detecting...". Similarly, depending on the context, the phrase "when determining..." or "if (the stated condition or event) is interpreted as "if determining...", "in response to determining...", "when (the stated condition or event) is detected", or "in response to detecting (the stated condition or event)".
[0182] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive), etc.
[0183] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.
Claims
1. A UWB base station deployment optimization method based on a starfish optimization algorithm, characterized in that, include: Establish a two-dimensional spatial map model of the area to be deployed, and divide the two-dimensional spatial map model into obstacle areas and non-obstacle areas; Determine several sampling points within the non-obstacle area for assessing coverage; Obtain the parameters required for deployment optimization, including the total number of UWB base stations to be deployed, the effective signal coverage radius of a single UWB base station, the population size of the starfish optimization algorithm, and the maximum number of iterations. A fitness function is constructed based on indicators such as effective coverage, obstacle coverage, UWB dispersion, and coverage uniformity. An initial population containing several starfish individuals is created, and the state vector of each starfish individual is composed of several randomly generated two-dimensional coordinate sets of the UWB base stations to be deployed; Perform a boundary projection operation on the two-dimensional coordinates of each UWB base station that falls into the obstacle area in the initial population, so that the corresponding UWB base station to be deployed is moved to the boundary of the nearest non-obstacle area, and the current population is obtained. The evaluation value of each starfish individual in the current population is calculated based on the fitness function; The starfish individual with the lowest evaluation value is determined as the globally optimal individual; For each starfish individual in the current population, the state vector is updated by performing one of centripetal movement, random wandering movement, or interactive foraging movement according to a preset probability rule, thus obtaining the updated population. Perform a boundary projection operation on the two-dimensional coordinates of each population that falls into the obstacle region to obtain the current population for the next evaluation; Determine whether the current iteration count has reached the maximum iteration count; If not, the iteration count is increased by one, and the step of calculating the evaluation value of each starfish individual in the current population based on the fitness function is performed. If so, the state vector of the globally optimal individual is output as the final optimized UWB base station deployment coordinate scheme.
2. The method according to claim 1, characterized in that, The fitness function constructed based on indicators such as effective coverage, obstacle coverage, UWB dispersion, and coverage uniformity specifically includes: The function for calculating the effective coverage rate is: The For the effective coverage, the The number of sampling points that are covered by at least a minimum number of UWBs and are not located in the obstacle region. The number of sampling points within the area to be deployed that are not located in the obstacle area; The function for calculating the obstacle coverage rate is: The The obstacle coverage rate, the The number of sampling points that are covered by at least one of the UWBs and are located in the obstacle region. This represents the total number of sampling points within the obstacle area; The function for calculating the UWB dispersion is: Among them, the ; In the above function, D represents the UWB dispersion. The total length of the minimum spanning tree formed by the UWBs is the value of the UWBs. Assuming that all UWBs are distributed to the maximum extent, the theoretical maximum length of the minimum spanning tree is... The number of UWBs, the The width of the two-dimensional spatial map of the area to be deployed, the The length of the two-dimensional spatial map of the area to be deployed; The calculation function for the coverage uniformity index is: In the above function, the For the coverage uniformity index, the The standard deviation of the number of UWB coverage points at each sampling point in the non-obstruction area is given by the following: This represents the average number of UWB coverage points at each sampling point in the non-barrier area. The fitness function is: ; In the above function, the To effectively cover the weights, the As the obstacle penalty weight, the As the dispersion weight, the For uniformity weights.
3. The method according to claim 1, characterized in that, The centripetal movement is as follows: the current starfish individual moves toward the position of the current global best individual, and the state vector is updated by adding a vector pointing to the state vector of the global best individual; The random wandering movement is as follows: the current starfish individual randomly explores within a preset range near its current position, and the state vector is updated by adding a perturbation vector whose components in each dimension are randomly generated within a preset small range. The interactive foraging movement is as follows: the current starfish individual randomly selects another individual in the current population as the interaction object and moves there. The state vector is updated by adding a vector of the state vector of the other individual.
4. The method according to claim 1, characterized in that, After outputting the state vector of the globally optimal individual as the final optimized UWB base station deployment coordinate scheme, the method further includes: Obtain the two-dimensional coordinates of at least one data aggregation gateway preset in the area to be deployed; Determine the maximum line-of-sight communication radius of the communication module used for wireless data backhaul in UWB base stations; Based on the two-dimensional coordinates of a preset number of UWB base stations and the two-dimensional coordinates of the data aggregation gateway in the final optimized UWB base station deployment coordinate scheme, a base station backhaul communication undirected graph model is constructed. The vertex set of the base station backhaul communication undirected graph model consists of all the UWB base stations and all the data aggregation gateways. For any two vertices in the vertex set, if the Euclidean distance between the two vertices is less than or equal to the maximum line-of-sight communication radius and the path connecting the two vertices does not intersect the obstacle region in the two-dimensional spatial map model, an edge is established between the two vertices in the undirected graph model of the base station backhaul communication to obtain an updated undirected graph model. If it is determined that there is a connected component in the updated undirected graph model that does not contain any data aggregation gateway, all UWB base stations contained in the connected component are marked as network unreachable nodes, and the network unreachable nodes are added to the final optimized UWB base station deployment coordinate scheme.
5. The method according to claim 4, characterized in that, The step of determining that there exists a connected component in the updated undirected graph model that does not contain any data aggregation gateway specifically includes: The updated undirected graph model is traversed using a graph traversal algorithm to determine all connected components. If no vertex in the vertex set of any of the connected components belongs to the data aggregation gateway, it is determined that there exists a connected component in the updated undirected graph model that does not contain any data aggregation gateway.
6. The method according to claim 4, characterized in that, After adding the network-unreachable node to the final optimized UWB base station deployment coordinate scheme, the method further includes: All vertices contained in a connected component that includes at least one of the data aggregation gateways are initialized into a main network vertex set; connected components that do not contain any of the data aggregation gateways are marked as isolated components to be processed, and all isolated components to be processed are compiled into an isolated component list. Determine whether the list of isolated components is empty; If not, then select any isolated component to be processed from the list of isolated components as the current processing component; Invoke the relay path calculation subroutine to calculate the new relay node coordinate sequence between the current processing component and the main network vertex set; All coordinate points in the newly added relay node coordinate sequence, and all vertices in the current processing component, are merged and added to the main network vertex set; The newly added relay node coordinate sequence is added to the final optimized UWB base station deployment coordinate scheme; Remove the currently processed component from the list of isolated components, and return to the step of determining whether the list of isolated components is empty; If so, output the final optimized UWB base station deployment coordinate scheme.
7. The method according to claim 6, characterized in that, The relay path calculation subroutine is as follows: Traverse the current processing component and the set of vertices in the main network, and determine the vertex pair with the smallest Euclidean distance as the best connection endpoint pair; In the two-dimensional spatial map model, a relay deployment path is constructed to connect the optimal connection endpoint pair. If the relay deployment path does not pass through the obstacle area, the relay deployment path is a straight line segment. If it passes through the obstacle area, the relay deployment path is the shortest broken line that avoids the obstacle. Based on the maximum line-of-sight communication radius, several insertion points are determined on the relay deployment path, such that the path distance between any two adjacent nodes on the relay deployment path is no greater than the maximum line-of-sight communication radius. Obtain the coordinates of all the insertion points and generate the coordinate sequence of the newly added relay node.
8. A UWB base station deployment optimization system based on the starfish optimization algorithm, characterized in that, The system includes: One or more processors and a memory; the memory is coupled to the one or more processors, the memory being used to store computer program code, the computer program code including computer instructions, the one or more processors invoking the computer instructions to cause the system to perform the method as described in any one of claims 1-7.
9. A computer-readable storage medium comprising instructions, characterized in that, When the instructions are executed on the system, the system performs the method as described in any one of claims 1-7.
10. A computer program product, characterized in that, When the computer program product is run on the system, the system performs the method as described in any one of claims 1-7.
