Intelligent optimization method and system for deployment of ray tracing and positioning base station for passenger positioning system in cruise ship room

Abstract

The invention discloses a ray tracing positioning base station deployment intelligent optimization method for a passenger positioning system in a cruise ship. The method comprises the following steps: modeling a scene in the cruise ship, determining basic conditions of UWB positioning base station deployment, dividing the scene in the cruise ship into two-dimensional grids, and obtaining a basic deployment scheme based on a genetic algorithm; performing simulation modeling on an indoor scene of the cruise ship by utilizing three-dimensional modeling software, and performing electromagnetic simulation based on ray tracing simulation software; analyzing and optimizing the basic deployment scheme according to an electromagnetic simulation result, and adjusting the number and positions of UWB positioning base stations to obtain a final base station deployment scheme; acquiring channel pulse response waveforms of different propagation paths of a cruise ship indoor scene, and extracting channel pulse response features; and inputting channel pulse response features into the constructed dichotomy decision tree model, identifying and rejecting sight distance shielding signals, and retaining a direct path. According to the method, the optimization of the positioning precision of indoor positioning of the cruise ship and the reduction of the positioning cost can be realized.

Description

Technical Field

[0001] This invention relates to the field of maritime traffic safety technology, specifically to an intelligent optimization method and system for deploying ray-tracing positioning base stations for indoor personnel positioning systems on cruise ships. Background Technology

[0002] As a new form of entertainment, cruise tourism has gained immense popularity in recent years due to its abundant entertainment facilities and excellent travel experience. However, with the increasing popularity of this tourism activity, the incidence of dangerous incidents such as missing persons and accidents has also been rising year by year. Cruise ships contain numerous indoor environments, and the commonly used satellite-based navigation systems suffer significant signal loss during propagation due to the complex indoor environment. This greatly reduces the signal power received by the receiving equipment, significantly impacting positioning accuracy and rendering traditional satellite positioning solutions infeasible in indoor scenarios. Existing indoor positioning technologies such as Zigbee, Wi-Fi, Bluetooth, UWB (Ultra Wideband), and RFID are more commonly used in large land-based indoor environments. When facing cruise ships with multiple decks, complex indoor structures, numerous sensitive areas, significant differences between cabins, and extremely high passenger flow, wireless signals are prone to reflection during propagation, causing multipath and non-line-of-sight noise, which can easily lead to decreased positioning accuracy and safety hazards. Between 2014 and 2019, EMSA received a total of 15,989 reports of maritime incidents, averaging 2,664 incidents per year. Among these, 2,081 were reports of missing persons. These data demonstrate that improving the accuracy of indoor positioning on cruise ships is of great significance in reducing travel risks and protecting the safety of passengers on board.

[0003] Cruise ships, as a unique indoor environment, have a more complex structure than ordinary land-based buildings. Various entertainment facilities frequently interfere with signal transmission, and walls between different rooms can significantly attenuate signal propagation. Simultaneously, with the rapid development of artificial intelligence and the Internet of Things (IoT), intelligent interconnection between people and things inside cruise ships has become an urgent need for the development of large passenger vessels. Establishing a large cruise ship personnel positioning network and developing positioning and navigation equipment for ocean-going passenger ships can greatly improve the status and location awareness capabilities of crew members on large vessels. Considering that cruise ships are large tourist vessels, providing passengers with accurate and real-time navigation and positioning is essential. Through indoor navigation technology, boarding passengers can quickly find the resources and facilities they need and determine their own location, greatly enhancing the passenger experience. Furthermore, in the event of emergencies such as fires requiring evacuation, accurate indoor personnel positioning services can quickly determine the distribution of people, assisting in the rational planning of evacuation routes and rescue plans, which is of great significance for ensuring the safety of passengers and crew. In recent years, countries around the world have been promoting the intelligent development of ships. Effective perception of the location information of people and cargo on cruise ships not only helps improve the level of internal supervision but also has significant practical implications for preventing and controlling major accidents on large passenger ships, enhancing the automation and intelligence levels of cruise ships, and building smart logistics platforms. Cruise ship indoor positioning technology is a crucial foundation for the intelligent development of cruise ships. With the rapid development of the global economy, more and more tourists are choosing cruise ships as their mode of travel, leading to significant changes in the supply and demand relationship in the passenger ship market and a substantial increase in demand for cruise ships. Therefore, the need to build and develop indoor positioning systems for cruise ship scenarios is even more urgent.

[0004] Current research on cruise ship indoor positioning technology mainly focuses on Wi-Fi and Bluetooth technologies, with limited and less mature research on UWB technology. For location fingerprinting, the constant updating of the fingerprint database during use is typically labor-intensive and cumbersome. Current research indicates that ship indoor positioning systems are generally not mature enough, with issues such as robustness, accuracy, and positioning cost remaining to be resolved. Many problems also exist in indoor positioning technology and node deployment. Considering the unique environment of cruise ships and the varying levels of maturity of different indoor positioning technologies, UWB positioning technology, compared to Wi-Fi, offers advantages such as higher data transmission speed, stronger resistance to multipath interference, higher positioning accuracy, insensitivity to channel fading, stronger penetration, and lower interception rate, making it the core of this project. Meanwhile, in using UWB for indoor positioning, the non-line-of-sight propagation of UWB signals is a crucial factor, affecting positioning accuracy. Due to potential reflections, refractions, or even complete attenuation or blockage during signal propagation, signal reception and positioning can be significantly impacted. Currently, there are two main methods for addressing the impact of non-line-of-sight (NLOS) errors: the first method involves collecting distance data using appropriate ranging techniques, followed by filtering to optimize the data and mitigate or eliminate the adverse effects of the NLOS environment; the second method, based on the characteristic that signal propagation time is typically longer in NLOS environments than in line-of-sight environments, reduces the impact of NLOS errors by correcting the signal propagation time. However, in the indoor environment of cruise ships, with frequent passenger movement and rapid and complex environmental changes, eliminating the impact of NLOS errors remains a key issue for improving the accuracy of indoor positioning on cruise ships.

[0005] In conclusion, although existing technologies exist for indoor positioning on cruise ships, the economic challenges of deploying positioning base stations due to the complex environment and the interference from non-line-of-sight propagation during UWB positioning still pose significant challenges. Therefore, achieving a highly economical and error-optimized indoor positioning system for cruise ships remains extremely difficult. Summary of the Invention

[0006] The technical problem to be solved by the present invention is to provide an intelligent optimization method and system for the deployment of ray-tracing positioning base stations for an indoor personnel positioning system on cruise ships, which addresses the problems existing in the prior art. This method can optimize the positioning accuracy and reduce the positioning cost of indoor positioning on cruise ships.

[0007] To achieve the above objectives, according to one aspect of the present invention, a method for intelligent optimization of ray-tracing positioning base station deployment for an indoor passenger positioning system on a cruise ship is provided, comprising: The cruise ship indoor scene is modeled to clarify the basic conditions for the deployment of UWB positioning base stations. The cruise ship indoor scene is divided into a two-dimensional grid, and the basic deployment scheme of the UWB positioning base stations in the two-dimensional grid is obtained based on the genetic algorithm. The interior scene of the cruise ship was simulated using 3D modeling software, and electromagnetic simulation was performed using ray tracing simulation software to obtain electromagnetic simulation results. Based on the electromagnetic simulation results, the basic deployment scheme is analyzed and optimized, and the number and location of the UWB positioning base stations are adjusted to obtain the final base station deployment scheme. Obtain the channel impulse response (CIR) waveforms of different propagation paths in the cruise ship indoor scene of the final base station deployment scheme, and extract the channel impulse response (CIR) features; The channel impulse response (CIR) features are input into the constructed binary classification decision tree model to identify and eliminate line-of-sight (NLOS) signals and retain direct path (LOS) signals.

[0008] In the above scheme, the genetic algorithm is a multi-objective genetic algorithm, a reference point-based non-dominated sorting genetic algorithm, or a degree Pareto evolutionary algorithm.

[0009] In the above scheme, the signal communication distance of the UWB positioning base station is 50-200m, and the positioning accuracy is between 5cm and 50cm. The basic conditions for deploying the UWB positioning base stations include: there is a direct path of sight (LOS) between the UWB positioning base stations; the distance between the UWB positioning base stations does not exceed 35 meters; the tag signal can be received by at least three of the UWB positioning base stations; when the UWB positioning base stations are installed against a wall, they are spaced 20-30 centimeters away from the wall; and the UWB positioning base stations need to be deployed separately in different rooms.

[0010] In the above scheme, the step of using 3D modeling software to simulate and model the interior scene of the cruise ship, and performing electromagnetic simulation based on ray tracing simulation software, includes: The modeling process considers the reflectivity, resistivity, and dielectric constant of different materials, and performs scene-specific electromagnetic simulations for each area of ​​the cruise ship's interior.

[0011] In the above scheme, the 3D modeling software is SketchUp 3D modeling software, AutoCAD, or Revit.

[0012] In the above scheme, the ray tracing simulation software is Lauraycs, Wireless InSite, or WinProp.

[0013] In the above scheme, the analysis and optimization of the basic deployment scheme based on the electromagnetic simulation results includes: Channel quality and communication feasibility are analyzed based on multipath quantity, received power, path loss, shadow fading, K-factor, and signal delay. Coverage blind spots are analyzed to form optimization decisions, and the number and location of the UWB positioning base stations are adjusted.

[0014] In the above scheme, the channel impulse response (CIR) characteristics include: First path location, peak path location, index of first path and peak path, maximum peak amplitude, first peak slope.

[0015] In the above scheme, the input of the binary classification decision tree model is the channel impulse response (CIR) feature, and the output is the classification result of whether the signal belongs to line-of-sight obstruction (NLOS) or direct path (LOS). The binary classification decision tree model filters signals based on the maximum peak value, the first path index, or the slope of the first peak, and according to the corresponding threshold.

[0016] According to another aspect of the present invention, a ray-tracing positioning base station deployment intelligent optimization system for an indoor passenger positioning system on a cruise ship is provided, comprising: The modeling module is used to model the indoor scene of the cruise ship, clarify the basic conditions for the deployment of UWB positioning base stations, divide the indoor scene of the cruise ship into a two-dimensional grid, and obtain the basic deployment scheme of the UWB positioning base stations in the two-dimensional grid based on the genetic algorithm. The simulation module is used to simulate and model the interior scene of the cruise ship using 3D modeling software, and to perform electromagnetic simulation based on ray tracing simulation software to obtain electromagnetic simulation results. The optimization analysis module is used to analyze and optimize the basic deployment scheme based on the electromagnetic simulation results, adjust the number and location of the UWB positioning base stations, and obtain the final base station deployment scheme. The extraction module is used to obtain the channel impulse response waveforms of different propagation paths in the cruise ship indoor scene in the final base station deployment scheme, and extract the channel impulse response features; The identification module is used to input the channel impulse response features into the constructed binary classification decision tree model to identify and eliminate line-of-sight obstruction signals and retain the direct path.

[0017] In summary, compared with the prior art, the above-described technical solutions conceived by this invention can achieve the following beneficial effects: This invention provides an intelligent optimization method for deploying ray-tracing positioning base stations in cruise ship indoor personnel positioning systems. This method deeply integrates Building Information Modeling (BIM) with Lauray ray-tracing simulation technology to perform precise electromagnetic simulation and iterative optimization of the deployment scheme, significantly improving signal coverage and positioning accuracy while reducing system deployment and maintenance costs, achieving an overall leap in positioning system performance. It innovatively applies a binary classification decision tree model to identify and filter non-direct (NLOS) ranging data, significantly improving ranging accuracy and laying the foundation for high-precision positioning. This effectively shortens positioning response time in emergency situations, enabling optimization of positioning accuracy and reduction of positioning costs for cruise ship indoor positioning. Attached Figure Description

[0018] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 This is a flowchart illustrating an intelligent optimization method for deploying ray-tracing positioning base stations for an indoor personnel positioning system on a cruise ship, as described in Embodiment 1 of the present invention.

[0019] Figure 2 This is a schematic diagram of a specific scenario on a cruise ship in Embodiment 1 of the present invention.

[0020] Figure 3 This is a floor plan of a specific scene on a cruise ship in Embodiment 1 of the present invention.

[0021] Figure 4 This is a grid diagram of a specific area of ​​a cruise ship in Embodiment 1 of the present invention.

[0022] Figure 5 This is a schematic diagram of converting an indoor two-dimensional planar grid into a genetic algorithm gene chain in Embodiment 1 of the present invention.

[0023] Figure 6 This is a schematic diagram of the deployment scheme of UWB positioning base stations on a two-dimensional grid in Embodiment 1 of the present invention.

[0024] Figure 7 This is a schematic diagram of scene modeling in Embodiment 1 of the present invention.

[0025] Figure 8 This is a schematic diagram of a panoramic modeling of a specific area of ​​a cruise ship in Embodiment 1 of the present invention.

[0026] Figure 9 This is a schematic diagram of the bar counter modeling and simulation in Embodiment 1 of the present invention.

[0027] Figure 10 This is a real-life image of the bar counter in Embodiment 1 of the present invention.

[0028] Figure 11 This is a schematic diagram of the modeling and simulation of furniture in the visitor activity area in Embodiment 1 of the present invention.

[0029] Figure 12 This is a real-life image of the tourist activity area in Embodiment 1 of the present invention.

[0030] Figure 13 This is a diagram of the UWB signal propagation path of the base station with coordinates (29,5) in the bar area in Embodiment 1 of the present invention.

[0031] Figure 14 This represents the multipath quantity distribution between transceivers during the target's movement in Embodiment 1 of the present invention.

[0032] Figure 15 This is a schematic diagram of the received power of the LOS path when the target is located inside the room in Embodiment 1 of the present invention.

[0033] Figure 16 This is a schematic diagram of the received power of all paths in Embodiment 1 of the present invention.

[0034] Figure 17 This is a schematic diagram of distance fading at various locations in Embodiment 1 of the present invention.

[0035] Figure 18 This is a schematic diagram of the shadow fading probability density distribution of the base station in Embodiment 1 of the present invention.

[0036] Figure 19 This is the CDF plot of the Ricean K factor in Embodiment 1 of the present invention.

[0037] Figure 20 This is a schematic diagram of signal delay in Embodiment 1 of the present invention.

[0038] Figure 21 This is a heatmap showing the LOS path coverage of the base station signal in Embodiment 1 of the present invention.

[0039] Figure 22 This is a schematic diagram of the basic deployment scheme in Embodiment 1 of the present invention.

[0040] Figure 23 This is a schematic diagram of the final base station deployment scheme in Embodiment 1 of the present invention.

[0041] Figure 24 This is a CIR waveform diagram under the direct path in Embodiment 1 of the present invention.

[0042] Figure 25 This is a CIR waveform diagram under line-of-sight obstruction in Embodiment 1 of the present invention.

[0043] Figure 26This is an example diagram of CIR waveform and feature selection in Embodiment 1 of the present invention. Detailed Implementation

[0044] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0045] It should be understood that the sequence number of each step in the embodiment does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0046] Example 1 This application provides an intelligent optimization method for deploying ray-tracing positioning base stations in an indoor passenger positioning system for cruise ships. Please refer to [link to relevant documentation]. Figure 1 ,include: S1. Model the indoor scene of the cruise ship, clarify the basic conditions for the deployment of UWB positioning base stations, divide the indoor scene of the cruise ship into a two-dimensional grid, and obtain the basic deployment scheme of UWB positioning base stations in the two-dimensional grid based on the genetic algorithm.

[0047] Specifically, in this embodiment, it is first necessary to construct a base station deployment model for the cruise ship indoor positioning system: a schematic diagram of a specific indoor scenario on a cruise ship in this embodiment is shown below. Figure 2 and Figure 3 As shown, the scene is divided into six main areas: a KTV room (top left in the image), a restaurant (top center in the image), a card room (top right in the image), an L-shaped corridor, a bar (bottom right in the image), and an equipment room (bottom left in the image). The KTV room measures 8m x 8.75m, the restaurant measures 8m x 8.75m, the equipment room measures 16m x 8m, and the corridor is 2m wide. Since the card room and bar have irregular shapes, they are approximated as trapezoids. The card room measures 3m at the top, 11m at the bottom, and 8.75m high, while the bar measures 9m at the top, 12m at the bottom, and 10m high.

[0048] The signal communication distance of commonly available UWB positioning base stations is typically 50-200m, with positioning accuracy between 5cm and 50cm. To ensure positioning accuracy, the base station deployment in this embodiment needs to adhere to the following basic rules: 1. There needs to be a LOS path between different base stations; 2. The distance between base stations should not exceed 35m, and it should be ensured that the signal emitted by the tag can be received by at least three base stations regardless of the location; 3. Consider the possibility of the tag being blocked when people move around indoors on the cruise ship; 4. When the base station is installed against a wall, a bracket needs to be used to separate the base station from the wall by 20cm to 30cm to prevent noise from reflections caused by the wall; 5. The base station locations should be as evenly and squarely as possible; 6. Since walls have a significant obstruction effect on the propagation of UWB wireless signals, the signal attenuation caused when the signal passes through the wall is huge, and the positioning accuracy will drop significantly. Therefore, a positioning base station needs to be deployed in each room.

[0049] Since the given cruise ship scenario is relatively fixed, this embodiment, for ease of discussion regarding base station deployment, divides the test scenario into a two-dimensional grid with the x-axis as the horizontal axis and the y-axis as the vertical axis, as follows: Figure 4 As shown.

[0050] This embodiment uses a cruise ship scenario as a specific fixed scenario, and the base station deployment needs to meet the above conditions. The various indicators considered in this embodiment are typical multi-objective optimization problems. The most direct approach to solving such problems is to weight multiple optimization objectives and then normalize them into a constrained single-objective optimization function before processing. However, this method has drawbacks such as difficulty in adjusting weight coefficients and difficulty in obtaining a non-convex solution set. The Nondominated Sorting Genetic Algorithm II (NSGA-II) is suitable for solving complex multi-objective optimization problems and obtaining a globally optimal solution set. It is a mature and efficient method for solving multi-objective optimization problems. This embodiment is based on the NSGA-II algorithm and uses binary encoding to convert the coordinate values ​​of the indoor two-dimensional planar grid into the chromosome gene chain of the genetic algorithm. A schematic diagram of the conversion of the indoor two-dimensional planar grid into the genetic algorithm gene chain is shown below. Figure 5 As shown.

[0051] In this embodiment, the specific steps for converting indoor two-dimensional planar grid coordinate values ​​into chromosome gene chains of a genetic algorithm are as follows: Step 1: Initialization. Define the population size pop=626, crossover rate, mutation rate, and maximum population generation N=626. Chromosome gene encoding is binary. Set the evolutionary generation gen=1. Based on the actual situation in the L3 Mockup region, permanently set the gene values ​​of locations where base stations cannot be set up, such as guest seats and tables, and locations where electrical boxes and equipment are located on the ship, to 0. That is, base stations will not be set up in these locations. Randomly generate an initial solution set of pop individuals.

[0052] Step two, topology structure discrimination, that is, gene discrimination for each individual in the population, that is, to determine whether the adjacent grid intersections of each grid intersection with a base station have been deployed with a base station. If so, the gene value corresponding to the grid node is modified to 0.

[0053] Step 3: Perform multi-objective calculation, fast non-dominated sorting, selection operation, and crossover operation in sequence, then call Step 2, perform mutation operation, and then call Step 2 again.

[0054] Step four: Determine if the termination condition has been met. If it is gen+1, proceed to step five; otherwise, proceed to step three.

[0055] Step 5: Merge the parent and offspring populations.

[0056] Step 6: Determine if a new population has been generated. If it has, proceed to Step 7; otherwise, proceed to Step 9.

[0057] Step seven, after calling step three, proceed to step eight.

[0058] Step 8: Determine if gen is less than N. If so, increment gen by 1 and go to step 5. Otherwise, go to step 10.

[0059] Step nine involves performing multi-objective calculations, fast non-dominated sorting, calculating crowding distance, and selecting suitable individuals to form a new population. After that, proceed to step six.

[0060] Step 10, End.

[0061] Since the KTV area and the card room area can be considered private areas when used by tourists, it is necessary to detect the presence of tourists, i.e., determine whether a tourist is located within the KTV area or the card room area. The UWB positioning base station deployment scheme on a two-dimensional grid obtained in this embodiment is as follows: Figure 6 As shown.

[0062] S2. The interior scene of the cruise ship was simulated using 3D modeling software, and electromagnetic simulation was performed using ray tracing simulation software to obtain the electromagnetic simulation results.

[0063] Specifically, in this embodiment, the room area is first simulated and divided. Based on the floor plan and schematic diagram of the cruise ship scene, the area is simulated and modeled using SketchUp 3D modeling software. The cruise ship scene is divided into six areas: KTV room (upper left), restaurant (upper middle), chess and card room (upper right), L-shaped corridor, bar (lower right), and equipment room (lower left). Based on the analysis and measurement data, the scene areas are modeled, such as... Figure 7 and Figure 8 As shown. The wall is mainly composed of an outer metal plate and an inner asbestos filling. Since the metal plate is relatively thin, typically less than 1 cm, and the 15 cm thick wall is mostly composed of asbestos, the wall's reflectivity during modeling is determined by the outermost metal layer, which is 64%. However, the thin metal plate's ability to block wireless signals is relatively limited, far less than the internal asbestos filling. Therefore, the dielectric constant and resistivity are determined by the asbestos, specifically a dielectric constant of 4.2 pF / cm and a resistivity of 2.3 × 10⁻⁶. 8 Ω·cm. The electromagnetic simulation coefficients of the wall are shown in Table 1: Table 1 Wall Modeling Parameters

[0064] Secondly, detailed modeling and simulation of each area are required. Since the modeling steps are similar for different areas, this embodiment will only use the modeling and simulation of the bar as an example: This embodiment divides the bar area into a bar counter area and a visitor activity area. The bar counter area mainly includes three parts: a bar table, seats, and a liquor rack. The bar table is made of marble with a reflectivity of 47%, a dielectric constant of 8.5 pF / cm, and a resistivity of 10¹³ Ω·cm. The seat base is made of metal, and the main body is made of leather with a reflectivity of 51%, a dielectric constant of 10.8 pF / cm, and a resistivity of 2.22 × 10⁻⁶ Ω·cm. 8 Ω·cm; The wine rack is made of glass with a reflectivity of 45%, a dielectric constant of 8 pF / cm, and a resistivity of 1.015 × 10⁻⁶. 6 Ω·cm. The material and electromagnetic parameters of the bar area are shown in Table 2. The simulation diagram and actual scene image of the bar in this embodiment are shown below. Figure 9 and Figure 10 As shown.

[0065] Table 2 Bar Counter Modeling Parameters

[0066] The visitor activity area mainly consists of five parts: seating, a central bar table, a metal bar table, a fixed interactive terminal for the ship's networked service platform, and wall-mounted wine racks. The metal bar table is an all-metal structure with a reflectivity of 64%, a dielectric constant of 9.93 pF / cm, and a resistivity of ∞ (conductor). The seat base is made of wood with a reflectivity of 38%, a dielectric constant of 2.67 pF / cm, and a resistivity of 107 Ω·cm. The main body of the seat is made of leather with a reflectivity of 51%, a dielectric constant of 10.8 pF / cm, and a resistivity of 2.22 × 10⁻⁶. 8 Ω·cm; The wine rack is made of glass with a reflectivity of 45%, a dielectric constant of 8 pF / cm, and a resistivity of 1.015 × 10⁻⁶. 6 Ω·cm; The central table and the fixed interactive terminal of the onboard network service platform are both made of translucent plastic. For tourist safety, the screen of the interactive terminal is also made of plastic, with a reflectivity of 31%, a dielectric constant of 2.3 pF / cm, and a resistivity of 1.017 × 10⁻⁶. 6 Ω·cm; The wall-mounted wine rack is made of the same material as the bulkhead, with both the support and the wine bottles made of glass. It has a reflectivity of 45%, a dielectric constant of 8 pF / cm, and a resistivity of 1.015 × 10⁻⁶. 6 Ω·cm. The material and electromagnetic parameters of the visitor activity area are shown in Table 3, and the modeling results are as follows: Figure 11 As shown in the image, the actual scene is as follows: Figure 12 As shown.

[0067] Table 3 Modeling parameters for visitor activity areas

[0068] Next, Lauraycs was used to perform ray-tracing electromagnetic simulations on each region to obtain simulation results.

[0069] S3. Based on the electromagnetic simulation results, the basic deployment scheme is analyzed and optimized, and the number and location of UWB positioning base stations are adjusted to obtain the final base station deployment scheme.

[0070] In this embodiment, the analysis and optimization of the basic deployment scheme based on the electromagnetic simulation results includes: Channel quality and communication feasibility are analyzed based on multipath quantity, received power, path loss, shadow fading, K-factor, and signal delay. Coverage blind spots are analyzed to form optimization decisions, and the number and location of the UWB positioning base stations are adjusted.

[0071] Specifically, since the electromagnetic simulation steps for each major area are the same, this embodiment only uses the electromagnetic simulation optimization of the bar area as an example for relevant description.

[0072] In this embodiment, four base stations were deployed in the bar area, with coordinates (22,11), (24,5), (29,5), and (28,11). The target movement route was set to walk from (17,12) to (22,12) and then to (28,4). In this embodiment, base station (29,5) was selected for detailed demonstration.

[0073] The UWB signal propagation path diagram of the base station at coordinates (29,5) in the bar area is shown below. Figure 13 As shown, Figure 13 The thickest red line represents the direct reflection (LoS) path, the blue line represents the specular reflection path, and the orange line represents the diffuse reflection path. Figure 14 This is a schematic diagram showing the distribution of the number of multipaths between transceivers during the target's movement corresponding to the base station. Figure 15 This diagram illustrates the received power of the LOS path when the target is inside the room (the X-axis represents the sampling points as the target moves along the trajectory, and the Y-axis represents the received power of the LOS path). Figure 16 This diagram illustrates the received power for all paths, where the X-axis represents the sampling points as the target moves along the trajectory, the Y-axis represents the signal arrival time for different paths, and the received power is indicated by color, with lighter colors indicating higher power and darker colors indicating lower power. Figure 17 This diagram illustrates the distance fading at various locations. The vertical axis represents path loss, the horizontal axis represents the sampling points during the target's movement along the trajectory, the blue hollow circles represent environmental loss, the red dots represent free space loss, and the yellow markers represent the CI fitting curve. Figure 18 This diagram illustrates the probability density distribution of shadow fading at a base station, where the X-axis represents the value of shadow fading, the Y-axis represents the probability of that value occurring, and the red line is the fitted curve of shadow fading. Figure 19 The image shows the CDF plot of the Ricean K-factor, where the closer the image is to the top left corner, the greater the probability that the received signal includes a Loss of Sight (LoS) path. The transmission and reception delays during the target's movement along the defined path are as follows: Figure 20 As shown. Figure 21 This is a heatmap showing the LOS path coverage of the base station signal. Red indicates the strongest signal, and dark blue indicates the weakest signal.

[0074] (1) Channel quality and communication feasibility analysis: Reference Figures 14 to 20 The simulation results provide a comprehensive evaluation of multipath quantity, received power, path loss, shadow fading, K-factor, and signal delay. Figures 14 to 16 The display shows that the receiving device can effectively identify high-power LOS signals (-62dBm to -70dBm), which are significantly higher than the reflected signals (approximately -90dBm). Figure 17 The path loss CI fitting indicates that the bar's facility environment is complex, and signal attenuation is mainly affected by environmental obstacles. Figure 20The signal latency is controlled below 30ns, meeting the requirements for real-time positioning. This data verifies the feasibility of deploying a UWB base station in the complex environment of a bar.

[0075] (2) Coverage blind spot analysis: combined with Figure 13 Signal propagation path and Figure 21 The coverage heatmap of the base station signal LOS path is analyzed. From the coverage heatmap ( Figure 21 As can be clearly seen from the data, the base station deployed at the original coordinates (24,5) inside the bar counter experiences severe signal obstruction from the marble bar counter structure when propagating towards the main activity area (corresponding to...). Figure 13 The path obstruction in the signal path leads to a decrease in signal coverage along the direct path in that direction, creating a coverage blind spot. The other three base stations, however, meet all positioning requirements.

[0076] (3) Final optimization decision: Based on the coverage blind spot analysis in the above electromagnetic simulation results, targeted adjustments are made to the obscured base stations.

[0077] After analysis, the base station (24,5) at the bar was adjusted to (26,5). This new position avoids the bar counter, preventing it from blocking the base station's signal and ensuring that the target has three or more base stations in any area of ​​the bar that can receive wireless signals.

[0078] This embodiment optimizes the basic deployment schemes for different regions using the same method to obtain the final base station deployment scheme. Details are as follows: Based on the simulation results analysis of the initial basic deployment plan, such as Figure 22 As shown, this paper discusses the improvements to the initial scheme that did not meet the positioning requirements. Based on meeting the requirements for positioning accuracy and presence detection, the base station locations are modified. Base stations are added in the necessary areas, and redundant base stations are removed, resulting in the final optimized base station deployment scheme: three base stations are deployed in the KTV area, with coordinates (2,18), (7,14), and (7,22); four base stations are deployed in the restaurant area, with coordinates (9,15), (9,22), (15,15), and (15,22); and three base stations are deployed in the chess and card room area, with coordinates (18,15), (18,14), (18,15 ... ,21)(25,15); Four base stations are deployed in the bar area, with coordinates (22,11), (26,5), (29,5), (28,11); Five base stations are deployed in the corridor area, with coordinates (4,12), (11,14), (18,14), (16,8), (18,4); Eight base stations are deployed in the equipment room area, with coordinates (1,9), (5,9), (9,9), (13,9), (1,5), (5,5), (9,5), (13,5). The final base station deployment scheme is shown in Figure 23.

[0079] Compared to the initial base station deployment scheme, the optimized final base station deployment scheme in this embodiment achieves 100% base station coverage. Regarding LOS paths, the optimized scheme ensures that in rooms requiring real-time positioning, each base station can receive LOS signals with 20dBm higher power than other signals when the target is in different areas. Furthermore, the number of LOS paths is increased compared to the initial base station deployment scheme, allowing the device to quickly and accurately identify LOS signals and improve positioning accuracy. Appropriately increasing the number of base stations in the equipment room area also achieves 100% signal coverage in that area, with signal power fully meeting the requirements for visitor presence detection. Adjusting the base station positions in the corridor area to reduce the number of base stations further improves corridor positioning accuracy and saves positioning costs. In summary, the optimized final base station deployment scheme can meet the positioning and detection needs of specific cruise ship scenarios, reasonably saves positioning costs, and has excellent positioning performance.

[0080] S4. Obtain the channel impulse response waveforms for different propagation paths in the cruise ship indoor scene of the final base station deployment scheme, and extract the channel impulse response features.

[0081] Understandably, Channel Impulse Response (CIR) feature classification is a widely used signal processing method in wireless communication systems, involving the identification and classification of specific attributes of the channel impulse response. In the complex and dynamic environment of a cruise ship's indoor positioning system, the direct path signal may be weakened or blocked, causing diffuse reflection signals from longer paths to reach the receiver, resulting in a ranging value greater than the actual distance and severely impacting the accuracy of UWB positioning. To overcome these challenges, in-depth data processing and analysis of the acquired CIR data are necessary during indoor positioning to address this ranging error caused by line-of-sight obstruction.

[0082] Specifically, in this embodiment, it is first necessary to obtain the CIR waveforms of different propagation paths in the cruise ship.

[0083] This embodiment employs a positioning module integrating the Decawave DW1000 ultra-wideband (UWB) wireless transceiver chip. After receiving the latest frame, this chip analyzes and measures the received signal to generate a CIR (Continuous Indicator Response). This response data accurately reflects the pulse characteristics of the communication channel between the transmitter and receiver. This data can be used to analyze the CIR waveforms under two conditions: direct path and line-of-sight obstruction. The impact of different obstruction environments on the waveform can be observed, and then the characteristics of the CIR waveform under obstruction can be extracted for non-line-of-sight signal identification. This embodiment processes a set of ranging data from an indoor environment, including CIR waveforms under three conditions: no obstruction between the transmitter and receiver, metal obstruction between the transmitter and receiver, and human body obstruction between the transmitter and receiver. Downsampling is used to process the original 1016 CIR sampling points, preserving the overall waveform curve. This is to facilitate a more intuitive display of the overall waveform and CIR characteristics. The final processed CIR waveform is shown below. Figure 24 , Figure 25 As shown.

[0084] Secondly, it is necessary to summarize the characteristics of the CIR waveform under different paths. According to the CIR waveform diagram above, it can be observed that under the direct path, the first peak of the CIR waveform diagram is the largest peak of the overall pattern. However, in the case of occlusion, a "pseudo-peak" will appear before the first peak.

[0085] Secondly, code needs to be designed based on the characteristics of the CIR waveform to identify different paths. To identify the number of peaks in the CIR waveform, we calculate the difference between consecutive data points. A peak occurs when the difference changes from positive to negative. So, if d(n) is the difference between the value at index n-1 and the value at index n, and d(n+1) is the difference between the value at index n and the value at index n+1, then if d(n) is positive and d(n+1) is negative, a peak is identified.

[0086] This embodiment uses the index of the first peak as a feature to determine whether the signal is in a direct path. Furthermore, "pseudo-peaks" typically do not reach the height of the maximum peak in a direct path. In this case, this embodiment sets a threshold for LOS / NLOS determination: if the maximum peak value does not reach the specified threshold, the signal may be blocked; alternatively, this embodiment can also calculate the slope of the first peak by looking up the first path index to perform LOS / NLOS determination.

[0087] S5, input the channel impulse response features into the constructed binary classification decision tree model, identify and remove line-of-sight occlusion signals, and retain the direct path.

[0088] Specifically, in this embodiment, CIR features are extracted to distinguish the characteristics of the channel impulse response under obstruction from the signal under a direct path. The CIR is divided into three regions for labeling, such as... Figure 26 As shown: (1) the region before the first path (Pre FP), (2) the region during the first path (FP), (3) the region after the first path (Post FP), where the first path is the most important feature.

[0089] A comparison of the CIR waveforms under direct path and under line-of-sight occlusion reveals that the distance between the first path and the peak path is significantly closer under direct path conditions than under non-line-of-sight conditions. By comparing the positions of the first path and the peak path, a metric is proposed to help distinguish between line-of-sight and non-line-of-sight signals. In this embodiment, this metric is called the Probability of NLOS (prNLOS), and the difference between these two path positions is termed the magnitude of IDdiff, resulting in the following formula:

[0090] Then, based on the calculated IDiff amplitude, a threshold discrimination method is used to assess the probability of NLOS occurrence. The specific processing logic is as follows: a discrimination threshold is set. If the IDiff value is greater than the threshold, it indicates a significant difference between the first path position and the peak path position, meaning the strongest signal is not the first path, and the signal is more likely to be a line-of-sight occlusion (NLOS) signal (i.e., the prNLOS value increases). Conversely, if the IDiff value is less than the threshold, it indicates that the first path and the peak path are close, and the signal is classified as a direct path (LOS) signal. This evaluation result is used as the judgment condition for a binary classification decision tree to classify the signal as LOS / NLOS.

[0091] If a peak is detected before the first path being identified, it can be concluded that the signal is strongly attenuated due to refraction and reflection in the indoor environment, causing the device to fail to detect the real first path and instead regard the reflected path as the first path. This ranging information should be excluded from the ranging data used by UWB positioning.

[0092] Traditional UWB ranging error compensation based on channel impulse response (CIR) is built upon this foundation. It analyzes CIR features under both occluded and unoccluded conditions to establish a binary classification decision tree model. Input CIR features, such as the maximum peak-to-peak value, the first path index, and the slope of the first peak, are used as the main features for non-line-of-sight (NLS) signal discrimination. The output is either a line-of-sight or NLS signal. After establishing the binary classification model, the ranging errors of NLS signals are statistically summarized based on actual surveys or experiments to establish a mean value for the ranging error of occluded signals. This mean value is then used to compensate for the measured value, improving the accuracy of UWB ranging under NLS conditions.

[0093] The method in this application embodiment further includes: verifying the effect of the binary classification decision tree model on improving positioning accuracy through experiments. Specifically, in this embodiment, this is as follows: Test Procedure: In the indoor positioning experiment, the test environment was a closed space of approximately 12m × 6m. Five fixed base stations were deployed with location coordinates of (2,2), (2,6), (6,4), (10,2), and (10,6). Three positioning tag points were selected in the space. The EVB1000 device was connected to the computer as a base station and a ranging signal was sent at a frequency of once every 50 milliseconds. At the same time, ranging data and channel impulse response (CIR) data from the tags were collected. The data was collected continuously for 1 minute, and the first 500 sets of data were used for subsequent processing.

[0094] In the data processing stage, a binary classification decision tree model is used to classify CIR data into line-of-sight (LOS) and non-line-of-sight (NLOS) signals. Only the ranging data identified as LOS are retained. If there are at least three usable data sets, the average is used as the input to the positioning algorithm; otherwise, the measurements are remeasured. Finally, the positioning results before and after NLOS filtering are plotted and compared with the actual locations.

[0095] Experimental results show that the positioning results after removing NLOS data using the binary classification decision tree model are closer to the actual tag location, with errors concentrated around 10 cm; while the positioning error using the original ranging data (including NLOS) is larger, around 35 cm. Error cumulative probability distribution analysis verifies that the proposed NL identification method effectively improves indoor positioning accuracy and achieves the expected goals.

[0096] In summary, this embodiment provides an intelligent optimization method for the deployment of ray-tracing positioning base stations for an indoor passenger positioning system on cruise ships. This method can optimize the positioning cost and positioning accuracy of indoor positioning on cruise ships.

[0097] Example 2 This application provides, in one aspect, a smart optimization method for deploying ray-tracing positioning base stations in an indoor personnel positioning system for cruise ships. This method is basically the same as the method in Embodiment 1, except that: In this embodiment, the 3D modeling software is Revit; the ray tracing simulation software is Wireless InSite.

[0098] Another embodiment of this application provides an intelligent optimization system for the deployment of ray-tracing positioning base stations for an indoor passenger positioning system on cruise ships, including: The modeling module is used to model the indoor scene of the cruise ship, clarify the basic conditions for the deployment of UWB positioning base stations, divide the indoor scene of the cruise ship into a two-dimensional grid, and obtain the basic deployment scheme of UWB positioning base stations in the two-dimensional grid based on the genetic algorithm. The simulation module is used to simulate and model the interior scene of the cruise ship using 3D modeling software, and to perform electromagnetic simulation based on ray tracing simulation software to obtain electromagnetic simulation results. The optimization analysis module is used to analyze and optimize the basic deployment scheme based on the electromagnetic simulation results, adjust the number and location of UWB positioning base stations, and obtain the final base station deployment scheme. The extraction module is used to obtain the channel impulse response waveforms of different propagation paths in the cruise ship indoor scene in the final base station deployment scheme, and extract the channel impulse response features; The identification module is used to input the channel impulse response features into the constructed binary classification decision tree model to identify and eliminate line-of-sight obstruction signals and retain the direct path.

[0099] It should be noted that, depending on the implementation needs, the various steps described in this application can be broken down into more steps, or two or more steps or parts of the steps can be combined into new steps to achieve the purpose of this invention.

[0100] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for intelligent optimization of ray-tracing positioning base station deployment for an indoor passenger positioning system on a cruise ship, characterized in that, include: The cruise ship indoor scene is modeled to clarify the basic conditions for the deployment of UWB positioning base stations. The cruise ship indoor scene is divided into a two-dimensional grid, and the basic deployment scheme of the UWB positioning base stations in the two-dimensional grid is obtained based on the genetic algorithm. The interior scene of the cruise ship was simulated using 3D modeling software, and electromagnetic simulation was performed using ray tracing simulation software to obtain electromagnetic simulation results. Based on the electromagnetic simulation results, the basic deployment scheme is analyzed and optimized, and the number and location of the UWB positioning base stations are adjusted to obtain the final base station deployment scheme. Obtain the channel impulse response waveforms of different propagation paths in the cruise ship indoor scene of the final base station deployment scheme, and extract the channel impulse response features; The channel impulse response features are input into the constructed binary classification decision tree model to identify and eliminate line-of-sight obstruction signals while retaining the direct path.

2. The intelligent optimization method for deploying ray-tracing positioning base stations for an indoor personnel positioning system on a cruise ship, as described in claim 1, is characterized in that... The genetic algorithm is a multi-objective genetic algorithm, a reference-point-based non-dominated sorting genetic algorithm, or a degree Pareto evolutionary algorithm.

3. The intelligent optimization method for deploying ray-tracing positioning base stations for an indoor personnel positioning system on a cruise ship, as described in claim 1, is characterized in that... The signal communication distance of the UWB positioning base station is 50-200m, and the positioning accuracy is between 5cm and 50cm. The basic conditions for deploying the UWB positioning base stations include: there is a direct path between the UWB positioning base stations; the distance between the UWB positioning base stations does not exceed 35 meters; the tag signal can be received by at least three of the UWB positioning base stations; when the UWB positioning base stations are installed against the wall, they are spaced 20-30 centimeters apart from the wall; and the UWB positioning base stations need to be deployed separately in different rooms.

4. The intelligent optimization method for deploying ray-tracing positioning base stations for an indoor personnel positioning system on a cruise ship, as described in claim 1, is characterized in that... The process of using 3D modeling software to simulate the interior scene of the cruise ship and performing electromagnetic simulation based on ray tracing simulation software includes: The modeling process considers the reflectivity, resistivity, and dielectric constant of different materials, and performs scene-specific electromagnetic simulations for each area of ​​the cruise ship's interior.

5. The intelligent optimization method for deploying ray-tracing positioning base stations for an indoor personnel positioning system on a cruise ship, as described in claim 1, is characterized in that... The 3D modeling software is SketchUp 3D, AutoCAD, or Revit.

6. The intelligent optimization method for deploying ray-tracing positioning base stations for an indoor personnel positioning system on a cruise ship, as described in claim 1, is characterized in that... The ray tracing simulation software is Lauraycs, Wireless InSite, or WinProp.

7. The intelligent optimization method for deploying ray-tracing positioning base stations for an indoor personnel positioning system on a cruise ship, as described in claim 1, is characterized in that... The analysis and optimization of the basic deployment scheme based on the electromagnetic simulation results includes: Channel quality and communication feasibility are analyzed based on multipath quantity, received power, path loss, shadow fading, K-factor, and signal delay. Coverage blind spots are analyzed to form optimization decisions, and the number and location of the UWB positioning base stations are adjusted.

8. The intelligent optimization method for deploying ray-tracing positioning base stations for an indoor personnel positioning system on a cruise ship, as described in claim 1, is characterized in that... Channel impulse response characteristics include: First path location, peak path location, index of first path and peak path, maximum peak amplitude, first peak slope.

9. The intelligent optimization method for deploying ray-tracing positioning base stations for an indoor personnel positioning system on a cruise ship, as described in claim 1, is characterized in that... The input to the binary classification decision tree model is the channel impulse response characteristics, and the output is the classification result of whether the signal belongs to line-of-sight obstruction or direct path. The binary classification decision tree model filters signals based on the maximum peak value, the first path index, or the slope of the first peak, and according to the corresponding threshold.

10. A ray-tracing positioning base station deployment intelligent optimization system for an indoor passenger positioning system on a cruise ship, characterized in that, include: The modeling module is used to model the indoor scene of the cruise ship, clarify the basic conditions for the deployment of UWB positioning base stations, divide the indoor scene of the cruise ship into a two-dimensional grid, and obtain the basic deployment scheme of the UWB positioning base stations in the two-dimensional grid based on the genetic algorithm. The simulation module is used to simulate and model the interior scene of the cruise ship using 3D modeling software, and to perform electromagnetic simulation based on ray tracing simulation software to obtain electromagnetic simulation results. The optimization analysis module is used to analyze and optimize the basic deployment scheme based on the electromagnetic simulation results, adjust the number and location of the UWB positioning base stations, and obtain the final base station deployment scheme. The extraction module is used to obtain the channel impulse response waveforms of different propagation paths in the cruise ship indoor scene in the final base station deployment scheme, and extract the channel impulse response features; The identification module is used to input the channel impulse response features into the constructed binary classification decision tree model to identify and eliminate line-of-sight obstruction signals and retain the direct path.

Images