Intelligent garden light linkage control system and method based on wireless networking

Abstract

The present application relates to the field of intelligent control, and specifically discloses an intelligent garden light linkage control system and method based on wireless networking. When any lamp in the garden senses that someone is passing by, it will not only be activated and lit, but will also actively use the wireless network to communicate with the surrounding neighboring lamps to collect their real-time signal strength. Subsequently, these dynamically changing wireless signal information is cleverly combined with the physical space layout information pre-set for the garden lamps for in-depth analysis to determine which neighboring lamps are not only accessible by signal, but also closely related to the current trigger point in spatial position, and are most suitable for lighting up at the same time. Based on this intelligent analysis that combines signal quality and spatial relationships, the system can dynamically calculate a most appropriate signal strength screening threshold, and accurately screen out the target neighbor lights that should participate in this linkage based on this threshold. Ultimately, only these intelligently selected target neighbor lights will receive instructions and be linked and lit.

Description

Technical Field

[0001] The present application relates to the field of intelligent control, and more specifically, to an intelligent garden light linkage control system and method based on wireless networking. Background Art

[0002] Existing garden lights are widely used in cities, communities, and private gardens, providing nighttime illumination and decorative features. However, traditional garden lights are typically controlled independently, meaning each light is turned on and off individually, or at most automated through simple time or light control. When pedestrians move through the courtyard, relying solely on a single light directly below or nearby often fails to provide sufficient illumination, impacting pedestrian safety and convenience. To improve user experience, enhance safety, and achieve intelligent energy management, it is necessary to build an intelligent garden light linkage control solution. When a specific area in the courtyard is triggered, multiple garden lights can be coordinated to illuminate simultaneously as needed, creating an illuminated area that "follows the person" as they move.

[0003] Although some smart lighting systems attempt to achieve linkage between lamps, existing linkage control solutions for garden lights often have limitations. Some solutions use preset static linkage rules, such as binding lamps in a specific area together. Such static rules cannot adapt to complex environmental changes and user needs, and may cause unnecessary lights to turn on, wasting energy, or failing to activate lights that should turn on. Furthermore, in wireless networking environments, signal strength (RSSI) varies dynamically due to factors such as distance, obstacles, and interference. Existing simple linkage solutions may only determine whether to link up based on a fixed communication distance threshold or a static signal strength threshold. This does not accurately reflect the actual communication reliability and optimal linkage range between lamps at the given moment, and is prone to misjudgments or omissions, resulting in unstable or inefficient linkage.

[0004] Therefore, an optimized linkage control scheme for garden lights is desired. Summary of the Invention

[0005] To address the aforementioned technical problems, the present application is proposed. The embodiments of this application provide a wirelessly networked intelligent garden light linkage control system and method. When any garden light senses a person passing by, it not only activates and illuminates itself but also proactively communicates with neighboring lights via wireless networks, collecting their real-time signal strengths (RSSIs). The system then cleverly combines this dynamically changing wireless signal information with the pre-defined physical spatial layout of the garden lights for in-depth analysis. This analysis determines which neighboring lights are not only reachable but also closely spatially associated with the current trigger point, making them ideal for simultaneous illumination. Based on this intelligent analysis, which integrates signal quality and spatial relationships, the system dynamically calculates an optimal signal strength screening threshold and, based on this threshold, accurately selects the "target neighbor lights" that should participate in the linkage. Ultimately, only these intelligently selected target neighbor lights receive the command and are linked to illuminate. This entire process dynamically adjusts in real time as the user moves, ensuring that the lighting area accurately follows the pedestrian's trajectory. This significantly reduces unnecessary long-term light on time while meeting lighting needs, maximizing energy efficiency and efficiency.

[0006] According to one aspect of the present application, a method for controlling an intelligent garden light linkage based on wireless networking is provided, which includes:

[0007] In response to the MCU control unit detecting an event that the PIR sensor is activated, marking the event that the PIR sensor is activated as a high priority, and extracting the ID of the current garden light node;

[0008] Generate a unique event ID based on the current timestamp;

[0009] Obtain a list of neighboring light nodes containing RSSIs, and generate a dynamic RSSI screening threshold based on the list of neighboring light nodes containing RSSIs;

[0010] Extracting a target neighborhood node list from the neighbor light node list containing RSSI based on the dynamic RSSI screening threshold;

[0011] The high priority, the ID of the current garden light node, the unique event ID and the target neighborhood node list are encapsulated to obtain a linkage trigger message to be sent.

[0012] According to another aspect of the present application, a wireless networking-based intelligent garden light linkage control system is provided, which includes:

[0013] Light node RSSI data acquisition module, used to obtain the RSSI value of each light node;

[0014] A light node spatial distribution topology construction module, used to construct a spatial distribution topology matrix between the light nodes;

[0015] a light node RSSI distribution feature extraction module, configured to arrange the RSSI values ​​of the light nodes into a light node RSSI distribution sequence, and then perform sequence encoding on the sequence to obtain a light node RSSI distribution feature encoding vector;

[0016] a light node spatial distribution activation module, configured to input the spatial distribution topology matrix into a nonlinear activation function to obtain a spatial distribution modulation activation matrix;

[0017] A light node RSSI distribution feature space mapping module is used to map the light node RSSI distribution feature coding vector to the feature space defined by the spatial distribution modulation activation matrix and perform spatial fine-grained feature modulation enhancement processing to obtain an enhanced light node RSSI distribution spatial modulation feature coding vector;

[0018] The dynamic RSSI screening threshold decoding module is used to perform feature decoding on the spatial modulation feature coding vector of the enhanced light node RSSI distribution to obtain the dynamic RSSI screening threshold.

[0019] Compared to existing technologies, this application provides a wireless networking-based intelligent garden light linkage control system and method. When any garden light senses a person passing by, it not only activates and illuminates itself, but also proactively communicates with neighboring lights via wireless networks, collecting their real-time signal strength (RSSI) data. The system then cleverly combines this dynamically changing wireless signal information with the pre-set physical spatial layout of the garden lights for in-depth analysis. This analysis determines which neighboring lights are not only reachable but also closely spatially associated with the current trigger point, making them ideal for simultaneous activation. Based on this intelligent analysis, which integrates signal quality and spatial relationships, the system dynamically calculates an optimal signal strength screening threshold and, based on this threshold, accurately selects the "target neighbor lights" that should participate in the linkage. Ultimately, only these intelligently selected target neighbor lights receive the command and are linked to the lights. This entire process dynamically adjusts in real time as the user moves, ensuring that the lighting area accurately follows the pedestrian's trajectory. This significantly reduces unnecessary long-term light on time while meeting lighting needs, maximizing energy efficiency and efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The above and other purposes, features, and advantages of the present application will become more apparent through a more detailed description of the embodiments of the present application in conjunction with the accompanying drawings. The accompanying drawings are intended to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the drawings, the same reference numerals generally represent the same components or steps.

[0021] Figure 1 Flowchart of a method for controlling linked intelligent garden lights based on wireless networking according to an embodiment of the present application;

[0022] Figure 2 A schematic diagram of data flow for obtaining a list of neighboring light nodes containing RSSIs and generating a dynamic RSSI screening threshold based on the list of neighboring light nodes containing RSSIs in a wireless networking-based intelligent garden light linkage control method according to an embodiment of the present application;

[0023] Figure 3 A flowchart of obtaining a list of neighboring light nodes containing RSSIs and generating a dynamic RSSI screening threshold based on the list of neighboring light nodes containing RSSIs according to a wireless networking-based intelligent garden light linkage control method according to an embodiment of the present application;

[0024] Figure 4 A flowchart of performing spatial fine-grained feature modulation enhancement on the RSSI distribution spatial modulation feature coding vector of the light node to obtain the enhanced RSSI distribution spatial modulation feature coding vector of the light node according to the wireless networking-based intelligent garden light linkage control method of an embodiment of the present application;

[0025] Figure 5 This is a block diagram of an intelligent garden light linkage control system based on wireless networking according to an embodiment of the present application. DETAILED DESCRIPTION

[0026] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described herein.

[0027] As used in this application and the claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" are not intended to refer to the singular but may include the plural. Generally speaking, the terms "comprises" and "include" only indicate the inclusion of the steps and elements specifically identified, and these steps and elements do not constitute an exclusive list. A method or apparatus may also include other steps or elements.

[0028] Although the present application makes various references to certain modules in the system according to embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are illustrative only, and different aspects of the system and method can use different modules.

[0029] Flowcharts are used in this application to illustrate the operations performed by the systems according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, the various steps may be processed in reverse order or simultaneously, as needed. Furthermore, other operations may be added to these processes, or one or more operations may be removed from these processes.

[0030] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described herein.

[0031] Existing linkage control solutions for garden lights rely on pre-set static linkage rules, such as binding all lights within a specific area together. These static rules are unable to adapt to complex environmental changes and user needs, potentially leading to unnecessary lights being activated, wasting energy, or failing to activate lights that should be activated. Existing simple linkage solutions may only determine whether to initiate linkage based on fixed communication distance thresholds or static signal strength thresholds. These solutions fail to accurately reflect the actual communication reliability and optimal linkage range between lights at the given moment, and are prone to misjudgments or missed detections, resulting in unstable or inefficient linkage.

[0032] To address the above technical issues, the present application proposes a wireless networking-based intelligent garden light linkage control method, achieving energy-efficient, user-friendly lighting. Specifically, when a garden light is activated by detecting human motion, such as through an infrared sensor, it not only illuminates itself but also quickly acts as a "trigger" to communicate with its neighboring lights via wireless network. Key to this process is that the system comprehensively considers the current wireless signal strength (RSSI) between the neighboring lights and the triggering light, combined with pre-defined physical spatial distribution information for all lights in the garden. This system not only determines which lights have strong signals but also spatially understands whether they are located in the areas most needed by the user. Based on this in-depth analysis combining wireless signal quality and physical spatial layout, the system dynamically calculates the most appropriate signal reception strength "qualification threshold" (dynamic RSSI screening threshold) for the current environment. This system then accurately identifies "target neighbor lights" that are physically adjacent, have reliable wireless communication, and are most suitable for illuminating simultaneously with the currently triggered light. Only these target neighbor lights that pass the intelligent screening will receive the linkage command and illuminate accordingly. This process occurs dynamically and in real time as the user moves and triggers new lamps, ensuring that the lighting area extends around the user's path and avoiding unnecessary lights being on for long periods of time, thereby maximizing energy savings while ensuring lighting needs.

[0033] In the technical solution of the present application, a method for controlling linkage of intelligent garden lights based on wireless networking is proposed. Figure 1 FIG is a flow chart of a method for controlling the linkage of intelligent garden lights based on wireless networking according to an embodiment of the present application. Figure 1 As shown, according to the embodiment of the present application, the intelligent garden light linkage control method based on wireless networking includes the following steps: S100, in response to the MCU control unit detecting the event that the PIR sensor is activated, marking the event that the PIR sensor is activated as a high priority, and extracting the ID of the current garden light node; S200, based on the current timestamp, generating a unique event ID; S300, obtaining a list of neighbor light nodes containing RSSI, and generating a dynamic RSSI screening threshold based on the list of neighbor light nodes containing RSSI; S400, based on the dynamic RSSI screening threshold, extracting a target neighborhood node list from the list of neighbor light nodes containing RSSI; S500, encapsulating the high priority, the ID of the current garden light node, the unique event ID and the target neighborhood node list to obtain a linkage trigger message to be sent.

[0034] Specifically, in step S100, in response to the MCU control unit detecting the activation of the PIR sensor, the PIR sensor activation event is marked as high priority, and the ID of the current garden light node is extracted. It should be understood that human movement events (detected by the PIR sensor) directly represent a user's immediate and urgent need for lighting in the garden. Marking such events as high priority ensures that the system can quickly and preferentially respond to the user's presence, avoiding lighting delays caused by processing other lower-priority tasks. This is crucial for ensuring user safety and user experience. Furthermore, extracting the ID of the current garden light node clarifies the starting point or trigger point for this linkage trigger. In other words, the linkage process begins with the light that detected movement. This process quickly and accurately identifies the time and location of the "human presence" event (specifically, the specific light node) and immediately assigns it the highest processing priority, providing a clear start signal and spatial positioning information for the subsequent intelligent linkage control process. This ensures that the system responds immediately and effectively to user movement, avoiding the lighting lag that can occur with traditional solutions. By identifying the trigger point ID, the system can specifically start from that node, communicate with neighboring lamps using the wireless network, and collect the necessary RSSI data. This lays a solid foundation for subsequent in-depth analysis based on RSSI and spatial information, the generation of dynamic RSSI screening thresholds, and ultimately the precise extraction of target neighborhood nodes.

[0035] Specifically, in step S200, a unique event ID is generated based on the current timestamp. It should be understood that in a wireless network environment consisting of multiple garden light nodes, multiple sensors may be activated simultaneously or nearly simultaneously, or the same user's movement within the garden may trigger different sensors sequentially. Without a unique identifier, the subsequent system would have difficulty accurately distinguishing and correlating the various information streams received and processed from different luminaires for different triggering events (such as neighbor RSSI reports and linkage instructions). This could easily lead to confusion and even the incorrect application of processing results from one triggering event to another. Therefore, by generating a unique event ID, a unique "identity" is created for each linkage request triggered by human motion detection. This unique event ID acts as a tag, organically linking all relevant data, communication messages, and processing processes generated since that specific triggering event. This ensures that all subsequent operations based on this event, such as collecting neighbor RSSIs, calculating dynamic thresholds, and sending linkage instructions, accurately target and serve the original triggering behavior, avoiding crosstalk and logical errors between different events. Specifically, in the embodiments of this application, uniqueness is achieved by generating a composite value that combines time information and a node-specific identifier. Overall, the implementation relies on triggering the light node's own internal clock source or timer, and combining its value with the light node's unique hardware identifier to form a globally unique event ID with a very high probability.

[0036] Specifically, in step S300, a list of neighboring light nodes containing RSSI is obtained, and a dynamic RSSI screening threshold is generated based on the list of neighboring light nodes containing RSSI. It should be understood that in a wireless network, in the process of generating a dynamic RSSI screening threshold, since the communication connection status and signal quality between lamps are dynamically changing, the direct use of real-time RSSI information can evaluate the communication capabilities and relative position relationships of neighboring nodes, providing an objective basis for deciding which lamps should participate in the linkage. Unlike the use of a fixed threshold, by analyzing the overall RSSI distribution of the current network, a dynamically adjusted screening threshold can be generated, thereby more flexibly and accurately selecting effective linkage targets. Furthermore, in order to make the generation of dynamic RSSI thresholds more intelligent and adaptable, this solution proposes to arrange the RSSI values ​​of each light node into a sequence and perform feature encoding, and at the same time construct a spatial distribution topology matrix that describes the physical position relationship between the light nodes. Mapping the RSSI distribution feature encoding vector of the lamp node to the feature space defined by the spatial distribution topology matrix can combine the signal strength information with the actual physical layout and distance information of the lamps, so that the system not only considers the distance of the signal, but also understands the meaning of the signal in the spatial structure, thereby more accurately judging which lamps are spatially related and communication is feasible, and suitable for linkage. The final derived dynamic RSSI screening threshold can more accurately reflect the optimal linkage node set under the current network state, thereby realizing more intelligent, more reliable and energy-saving garden lamp linkage control.

[0037] Figure 2 A data flow diagram for obtaining a list of neighboring light nodes containing RSSIs and generating a dynamic RSSI screening threshold based on the list of neighboring light nodes containing RSSIs according to the wireless networking-based intelligent garden light linkage control method according to an embodiment of the present application. Figure 3 The present invention provides a flowchart of obtaining a list of neighboring light nodes containing RSSI and generating a dynamic RSSI screening threshold based on the list of neighboring light nodes containing RSSI according to the wireless networking-based intelligent garden light linkage control method of the embodiment of the present invention. Figure 2 and 3As shown, according to the intelligent garden light linkage control method based on wireless networking according to the embodiment of the present application, step S300 includes: S310, obtaining the RSSI value of each light node; S320, constructing the spatial distribution topology matrix between the light nodes; S330, arranging the RSSI values ​​of the light nodes into a light node RSSI distribution sequence, and then performing sequence encoding to obtain a light node RSSI distribution feature coding vector; S340, inputting the spatial distribution topology matrix into a nonlinear activation function to obtain a spatial distribution modulation activation matrix; S350, mapping the light node RSSI distribution feature coding vector to the feature space defined by the spatial distribution modulation activation matrix and performing spatial fine-grained feature modulation enhancement processing to obtain an enhanced light node RSSI distribution spatial modulation feature coding vector; S360, feature decoding the enhanced light node RSSI distribution spatial modulation feature coding vector to obtain the dynamic RSSI screening threshold.

[0038] Specifically, in steps S310 and S330, the RSSI values ​​of each light node are obtained and arranged into a light node RSSI distribution sequence. This sequence is then encoded to produce a light node RSSI distribution feature encoding vector. It should be understood that because wireless signal strength (RSSI) in a courtyard environment is dynamic and affected by multiple factors, a simple fixed threshold cannot accurately reflect the actual communication reliability between lamps at the current moment. Simply obtaining the RSSI value between the triggering light and a neighboring light is local and instantaneous, failing to capture the overall state and relative strength patterns of the entire network signal environment. However, aggregating the RSSI values ​​of all light nodes (or those within a relevant range) and analyzing and encoding them as a holistic "distribution sequence" provides more comprehensive and macroscopic information about the wireless signal environment. By encoding this sequence, more stable and representative features can be extracted compared to the raw RSSI values, such as the relative relationship between signal strengths, signal attenuation trends, and the presence of signal dead zones. These distribution features can better reflect the health of the network and the potential feasibility of linkage than a single RSSI value, providing high-quality input for subsequent in-depth analysis combined with spatial information and the generation of accurate dynamic thresholds.

[0039] That is, because signal strength (RSSI) in wireless networks is highly susceptible to dynamic changes due to factors such as distance, obstacles, and interference, simply relying on the real-time RSSI value of a single neighboring node or a fixed signal strength threshold for judgment fails to accurately reflect the overall state of signal connectivity across the entire courtyard network, nor does it capture the complex interactions and relative communication potential between nodes. Therefore, after further arranging the RSSI values ​​of each light node into a light node RSSI distribution sequence, sequence encoding is performed to obtain a light node RSSI distribution feature encoding vector. Arranging these discrete RSSI values ​​of each light node into a sequence and then performing feature encoding is intended to transform this raw, dynamically changing signal data into a more abstract, stable, and informative overall distribution feature vector. This vector no longer simply represents the RSSI value emitted by a single light, but rather a compact representation of the current signal strength landscape of the entire network. In particular, in a specific example of this application, sequence encoding can be performed using an LSTM-based sequence encoder to generate a feature-rich representation of the RSSI distribution characteristics of all light nodes, laying the foundation for subsequent more advanced and intelligent analysis. Through sequence encoding, the high-dimensional raw RSSI data sequence can be reduced in dimension and its core features extracted, such as implicit information such as fluctuations in signal strength, the presence of signal anomalies in specific areas, and which lights have strong signal correlations. This encoded light node RSSI distribution feature encoding vector serves as the global input for the system's understanding of the current wireless signal environment. In subsequent steps, it will be combined and modulated with the spatial distribution topology matrix representing the physical space layout. This aims to break away from the limitations of pure signal strength judgment and link signal information with actual physical location, achieving more accurate spatial perception.

[0040] Specifically, in one specific example of this application, when a garden light node (acting as a trigger) detects a PIR activation event and generates a unique event ID, it uses its wireless communication module to broadcast an information request message to its neighboring nodes. This message typically includes the triggering node's ID and the unique event ID for this event, requesting that the receiving neighboring nodes measure the signal strength (RSSI) received from the triggering node and transmit this RSSI value along with their own ID back to the triggering node. Alternatively, under certain networking protocols, light nodes periodically broadcast "heartbeat" or status reports containing their own ID and the signal strengths received from other nodes. The triggering node then retrieves the latest RSSI data from its cache when needed. The triggered light node then receives the RSSI values ​​and their corresponding node IDs from each neighboring node. After receiving a sufficient number of RSSI reports (for example, from all neighboring nodes within a preset range, or from all responding nodes within a specified timeframe), the triggering node compiles this data. The triggering node then arranges the collected RSSI values ​​of each neighboring light node into an ordered "light node RSSI distribution sequence" based on a predetermined rule (for example, by node ID or by the time the messages were received). This sequence intuitively represents the signal strength distribution of the surrounding neighboring nodes from the perspective of the current triggering node.

[0041] In particular, in an embodiment of the present application, step S330 includes: arranging the RSSI values ​​of the respective light nodes into a light node RSSI distribution sequence; and passing the light node RSSI distribution sequence through an LSTM-based sequence encoder to obtain the light node RSSI distribution feature encoding vector.

[0042] Specifically, in steps S320 and S340, a spatial distribution topology matrix is ​​constructed between the various light nodes, and this spatial distribution topology matrix is ​​input into a nonlinear activation function to obtain a spatial distribution modulation activation matrix. It is worth noting that the values ​​at each off-diagonal position in the spatial distribution topology matrix represent the spatial distance between the corresponding two light nodes. It should be understood that while the spatial distribution topology matrix can directly describe the physical distance relationship between light fixtures, in actual application scenarios, the relationship between physical distance and linkage response requirements is not a simple linear function. For example, even if a light fixture within a specific area is slightly farther away, it may be more important than a closer light fixture but outside the path based on overall spatial layout or lighting design considerations. Furthermore, factors such as terrain and obstacles may also nonlinearly affect the need for or feasibility of linkage effects. Therefore, in the technical solution of the present application, the spatial distribution topology matrix is ​​input into a nonlinear activation function to obtain a spatial distribution modulation activation matrix. By introducing a nonlinear activation function, the system can learn or model the more complex, nonlinear effects of physical distance on linkage priority, response range, or collaboration mode, thereby converting simple distance information into more practical spatial activation or modulation weights. The resulting spatially distributed modulation activation matrix doesn't simply store distances; it also includes weights or influencing factors, after nonlinear transformation, that better reflect linkage needs and spatial importance. Subsequently, this spatially distributed modulation activation matrix will serve as a high-dimensional feature space, used to "map" or "modulate" the feature vectors previously encoded by the RSSI distribution sequence of the light nodes. This means that the purpose of obtaining this matrix is ​​to provide a spatial "filter" or "enhancer," so that RSSI information is no longer viewed in isolation, but rather interpreted in conjunction with the actual physical location of the lights and their nonlinear correlation importance in the spatial layout.

[0043] More specifically, in one example of this application, during the deployment phase of the garden lighting system, it is necessary to obtain the precise physical location information of all smart luminaires in the garden. This can be achieved in various ways, such as manually measuring and entering the geographic coordinates of each luminaire (such as latitude and longitude or two-dimensional X, Y coordinates) during installation; or utilizing more advanced technologies such as GPS, UWB (ultra-wideband) positioning, or image recognition / SLAM (simultaneous localization and mapping) to automatically obtain the relative or absolute location of the luminaires. This location information is typically stored in the system's configuration parameters or in local storage on each luminaire node. Next, based on the stored physical location information of each luminaire node, the system calculates the spatial distance between any two luminaire nodes or determines their relative positional relationship, thereby constructing the spatial distribution topology matrix. This matrix is ​​typically a two-dimensional square matrix, with rows and columns corresponding to each luminaire node in the garden. The values ​​of the off-diagonal elements (i.e., positions ij) in the matrix can represent the Euclidean distance, Manhattan distance, or some other metric calculated based on their relative coordinates between luminaire nodes i and j. The diagonal elements usually represent the fixtures themselves, and their values ​​are zero or some don't matter value.

[0044] The constructed spatial distribution topology matrix is ​​then fed into a nonlinear activation function for transformation. This nonlinear activation function can be a preset fixed function (such as Sigmoid, ReLU, or Tanh) or a learned function. This function applies a nonlinear transformation to each element in the matrix (i.e., a measure of the spatial correlation between luminaires). For example, it can map physical distance values ​​to an activation weight between 0 and 1, or assign different activation gains to different distance ranges. Choosing an appropriate nonlinear function can better capture the complex relationship between physical distance and linkage importance. For example, while closer luminaires receive higher activation weights, there may be a threshold distance above which the activation weights decay rapidly, rather than simply decreasing linearly. Ultimately, the matrix processed by the nonlinear activation function yields the spatially distributed modulated activation matrix. The values ​​in this matrix are no longer simply the raw physical distances, but rather nonlinearly modulated activation weights that better reflect the strength of the spatial correlation between luminaires and the potential importance of their potential linkages. This spatially distributed modulation activation matrix, as input to subsequent steps, will be used to modulate the RSSI distribution characteristics, guiding the system to more intelligently determine the linkage range and target lamps based on the physical space layout while considering signal strength.

[0045] Specifically, in step S350, the light node RSSI distribution feature coding vector is mapped to the feature space defined by the spatial distribution modulation activation matrix and spatial fine-grained feature modulation enhancement processing is performed to obtain an enhanced light node RSSI distribution spatial modulation feature coding vector.

[0046] Accordingly, according to an embodiment of the present application, step S350 includes: S351, matrix multiplying the light node RSSI distribution feature coding vector with the spatial distribution modulation activation matrix to map the light node RSSI distribution feature coding vector to the feature space defined by the spatial distribution modulation activation matrix to obtain the light node RSSI distribution spatial modulation feature coding vector; S352, performing spatial fine-grained feature modulation enhancement on the light node RSSI distribution spatial modulation feature coding vector to obtain the enhanced light node RSSI distribution spatial modulation feature coding vector.

[0047] Specifically, in step S351, the light node RSSI distribution feature encoding vector is matrix-multiplied by the spatially distributed modulation activation matrix to map the light node RSSI distribution feature encoding vector to the feature space defined by the spatially distributed modulation activation matrix, thereby obtaining the light node RSSI distribution spatial modulation feature encoding vector. It should be understood that both the light node RSSI distribution feature encoding vector obtained by obtaining and encoding the original multi-node RSSI values, and the spatially distributed modulation activation matrix obtained by subjecting the physical space topology matrix to nonlinear activation processing, each only represents one dimension of the information required for linkage decision-making—the former reflects the overall state and relative strength pattern of the wireless signal, while the latter reflects the importance of geographical location and nonlinear spatial associations between lamps. However, relying solely on either of these two methods cannot lead to optimal linkage judgment. Therefore, in the technical solution of the present application, the light node RSSI distribution feature encoding vector is further mapped to the feature space defined by the spatially distributed modulation activation matrix to obtain the light node RSSI distribution spatial modulation feature encoding vector. This allows for a deep fusion of the dynamic changes in multi-node wireless signals with fixed physical space layout information, ensuring that linkage decisions consider both signal reliability and the rationality of the spatial layout. Specifically, in one example of this application, through mapping operations such as matrix multiplication, the spatially distributed modulation activation matrix can be used as a "weighing" or "attention" mechanism to weight or adjust the signal information contained in the node RSSI distribution feature encoding vector based on the relative position and importance of the lamps in the physical space. This is equivalent to using spatial relationship information to "filter" or "enhance" the pure signal information, so that the resulting feature vector is no longer just an abstract representation of signal strength, but incorporates spatial context information such as "which signal strengths come from physically important locations."

[0048] Specifically, in step S352, the spatial fine-grained feature modulation enhancement is performed on the spatial modulation feature coding vector of the RSSI distribution of the light node to obtain the enhanced spatial modulation feature coding vector of the RSSI distribution of the light node. It should be understood that even if the initial "spatial modulation feature coding vector of the RSSI distribution of the light node" has combined information in both signal and space dimensions, this preliminary fusion feature representation may still not be refined enough, may contain redundant information, noise, or fail to fully reveal the deeper and more subtle intrinsic correlation structure between the signal and the spatial layout, which are crucial for accurately judging the linkage requirements. Therefore, in the technical solution of the present application, the spatial fine-grained feature modulation enhancement is further performed on the spatial fine-grained feature modulation enhancement of the RSSI distribution of the light node to obtain the enhanced spatial modulation feature coding vector of the RSSI distribution of the light node. Through the spatial fine-grained feature modulation enhancement processing, the existing spatial modulation features of the RSSI distribution of the light node can be deeply mined and optimized to cope with the complex and nonlinear signal-space interaction in the courtyard environment, and to ensure the accuracy and robustness of the feature expression. Specifically, the process of spatial fine-grained feature modulation and enhancement involves more than just simple feature screening; it also involves adjusting and optimizing the internal structure of the features. For example, this involves performing correlation screening and focusing through a gated mask function, or introducing gradient field terms and mean field tuning to correct for local nonlinear geometric structural inhomogeneities and compensate for the discretization of sub-geometric correlation structures caused by correlation polarization enhancement. In the context of courtyard lights, this means more finely adjusting and enhancing those "signal-space" combination patterns that accurately reflect the actual linkage requirements, while suppressing misleading feature combinations that may be caused by environmental interference or unimportant spatial locations. This allows the final feature vector to more deeply and accurately express which lighting combinations are most suitable for linkage in the current environment. This aims to address the potential problems of initially fused features, such as insufficient understanding of complex scenes and insufficient sensitivity to subtle changes, in order to obtain a feature representation that contributes more to linkage decisions and has higher information entropy.

[0049] Figure 4 The flowchart of the method for controlling the linkage of intelligent garden lights based on wireless networking according to the embodiment of the present application is to perform spatial fine-grained feature modulation enhancement on the RSSI distribution spatial modulation feature coding vector of the light node to obtain the enhanced RSSI distribution spatial modulation feature coding vector of the light node. Figure 4As shown, according to the wireless networking-based intelligent garden light linkage control method of the embodiment of the present application, step S352 includes: S3521, performing feature decomposition based on one-dimensional convolution coding on the modulation feature coding vector of the RSSI distribution space of the light node to obtain a set of initial local implicit feature vectors of the RSSI distribution space of the light node; S3522, based on the manifold structure correlation coefficient between any two initial local implicit feature vectors of the RSSI distribution space of the light node in the set of initial local implicit feature vectors of the RSSI distribution space of the light node, generating a manifold structure fine-grained correlation mask of the RSSI distribution space of the light node. code topology matrix; S3523, based on the fine-grained associative mask topology matrix of the manifold structure of the light node RSSI distribution space, feedback distillation is performed on each initial local implicit feature vector of the light node RSSI distribution space in the set of initial local implicit feature vectors of the light node RSSI distribution space to obtain a set of initial local implicit feature vectors of the distillation of the light node RSSI distribution space; S3524, feature reconstruction is performed based on the self-attention mechanism on the set of initial local implicit feature vectors of the distillation of the light node RSSI distribution space to obtain the enhanced light node RSSI distribution space modulation feature coding vector.

[0050] More specifically, in step S3521, the modulation feature coding vector of the light node RSSI distribution space is subjected to feature decomposition based on one-dimensional convolution coding to obtain a set of initial local implicit feature vectors of the light node RSSI distribution space, which is expressed as follows:

[0051]

[0052] in, is the spatial modulation feature coding vector of the RSSI distribution of the light node, The step length is One-dimensional convolutional coding processing, are the 1st, 2nd, i-th, j-th and n-th initial local implicit feature vectors of the RSSI distribution space of the light node in the set of initial local implicit feature vectors of the RSSI distribution space of the light node respectively.

[0053] It should be understood that although the previous steps have integrated the real-time RSSI distribution information with the preset spatial layout information to generate a "light node RSSI distribution spatial modulation feature coding vector", this vector is a high-dimensional, complex overall representation, and the signals and spatial interaction patterns contained therein may be intertwined, making it difficult to directly analyze and utilize. These patterns may show local correlations or specific combinations at different positions of the vector (which can be understood as feature dimensions). In order to deeply understand these complex, fused features and provide a basis for subsequent refinement, it is necessary to parse this overall, composite feature vector into more basic and easier to handle "local components" or "basic patterns". Based on this, the light node RSSI distribution spatial modulation feature coding vector is subjected to feature decomposition based on one-dimensional convolutional coding to obtain a set of initial local implicit feature vectors of the light node RSSI distribution space. By applying a one-dimensional convolutional neural network (1D CNN)-based coding technique, we achieve "deep structured processing" of the original spatial modulation feature encoding vector of the RSSI distribution of light nodes. This decouples the multiple potential signal-space interaction patterns hidden within the entire vector and makes them explicit as independent local feature units. This transforms the overall representation of the original features into a distributed local representation, resulting in a set of initial local implicit feature vectors for the RSSI distribution space of light nodes. This process aims to reveal and isolate the fundamental components that make up the complex signal-space features, paving the way for subsequent modeling of the relationships between these local components.

[0054] Accordingly, according to an embodiment of the present application, step S3522 includes: calculating the manifold structure correlation coefficient between any two initial local implicit feature vectors of the light node RSSI distribution space in the set of initial local implicit feature vectors of the light node RSSI distribution space to obtain a light node RSSI distribution space manifold structure correlation topology matrix composed of multiple light node RSSI distribution space manifold structure correlation coefficients; inputting the light node RSSI distribution space manifold structure correlation topology matrix into a gated mask function to obtain a light node RSSI distribution space manifold structure fine-grained correlation mask topology matrix.

[0055] More specifically, the manifold structure correlation coefficient between any two initial local implicit eigenvectors of the light node RSSI distribution space in the set of the initial local implicit eigenvectors of the light node RSSI distribution space is calculated to obtain a light node RSSI distribution space manifold structure correlation topology matrix composed of multiple light node RSSI distribution space manifold structure correlation coefficients, which is expressed as follows:

[0056]

[0057] in, represents the one-norm of a vector, are trainable association weights, for and The correlation coefficient of the manifold structure of the RSSI distribution space of the light nodes is the manifold structure correlation topology matrix of the RSSI distribution space of the light nodes. The correlation coefficient of the RSSI distribution space manifold structure of the light nodes at various locations in .

[0058] It should be understood that in the garden light scenario, the initial local latent eigenvectors of the RSSI distribution space of these light nodes represent local patterns generated by the interaction between RSSI distribution and spatial layout at different locations and scales. These local patterns possess complex internal connections and interdependencies. For example, signal patterns of a certain intensity may tend to appear near power sources or specific spatial structures. These different local eigenvectors may not be linearly independent in a higher-dimensional feature space, but instead exhibit specific proximity or correlation structures on a nonlinear manifold. To understand and quantify the mutual influence, structural relationships, or "proximity" between these local feature components on the underlying manifold, it is necessary to explicitly calculate their correlations. By calculating the manifold structural correlation coefficient between any two initial local latent eigenvectors of the RSSI distribution space of light nodes, it is possible to quantify their mutual correlation strength in the feature space or their relative positional relationship on the underlying manifold. Organizing these correlation coefficients into the manifold structural correlation topology matrix is ​​the core purpose of explicitly modeling and structurally analyzing the inherent correlations and geometric relationships between the local features of the RSSI distribution space of each light node, as decomposed. This spatial manifold structure of the RSSI distribution of light nodes demonstrates the connection strength and interdependence between different local signal-spatial patterns in a structured manner. It not only describes the existence of local patterns but also how these patterns are interconnected, providing clear guidance and foundation for subsequent refinement (distillation) of local features based on this correlation information.

[0059] More specifically, the light node RSSI distribution space manifold structure association topology matrix is ​​input into the gated mask function to obtain the light node RSSI distribution space manifold structure fine-grained association mask topology matrix, which is expressed as follows:

[0060]

[0061] in, is the light node RSSI distribution space manifold structure association topology matrix composed of multiple light node RSSI distribution space manifold structure association coefficients, is the fine-grained association mask topology matrix of the light node RSSI distribution space manifold structure, is the gated mask weight matrix, is the gated mask bias matrix, for function.

[0062] It should be understood that although the spatial manifold structure correlation topology matrix of the light node RSSI distribution constructed in the previous step explicitly models the correlation between local signal-space feature patterns, this original spatial manifold structure correlation topology matrix may contain noise, redundant information, or not all correlations are equally important to the final linkage decision. Some calculated correlations may simply be statistical coincidences or weak correlations caused by irrelevant physical factors in the courtyard environment. These should not significantly impact subsequent feature refinement. In order to extract truly meaningful, fine-grained correlation information that can guide linkage decisions, a mechanism is required to filter and modulate the raw correlation strengths. Based on this, the spatial manifold structure correlation topology matrix of the light node RSSI distribution is further input into a gated mask function to obtain the spatial manifold structure fine-grained correlation mask topology matrix of the light node RSSI distribution. By introducing this gated mask function, it acts as an "adaptive correlation screening mechanism," performing nonlinear transformation and modulation on the original manifold structure correlation topology matrix. This function can dynamically generate a mask based on the numerical values ​​or other contextual information in the association topology matrix of the manifold structure of the light node RSSI distribution space. This mask adjusts the original spatial correlation strength of the light node RSSI distribution space. The core idea is to amplify important associations and suppress noisy associations or irrelevant connections, which is equivalent to focusing attention on the associations between the spatial features of the light node RSSI distribution space. In the courtyard light scenario, this means that the system can learn or determine which associations between local signal-spatial patterns are truly important, such as the strong association between the signal attenuation pattern in a certain area and a specific spatial layout, or the association between the signal strength pattern of a certain lamp and the spatial relationship of nearby obstacles. Through the gated mask function, these important associations are strengthened, while those unimportant or misleading associations are weakened or even suppressed, thereby generating a "cleaner" and more focused "fine-grained association" matrix that focuses on key information.

[0063] More specifically, in step S3523, based on the fine-grained associative mask topology matrix of the manifold structure of the light node RSSI distribution space, feedback distillation is performed on each of the initial local implicit feature vectors of the light node RSSI distribution space in the set of initial local implicit feature vectors to obtain a set of distilled initial local implicit feature vectors of the light node RSSI distribution space, which is expressed as follows:

[0064]

[0065] in, is the distillation weight matrix, for The scale, for activation function, is matrix multiplication, is the point product by position, The first set of initial local latent feature vectors for the spatial distillation of RSSI distribution of light nodes The initial local latent feature vector is distilled from the RSSI distribution space of each light node.

[0066] It should be understood that although the gated mask function has filtered and modulated the correlation strength between local features in the light node RSSI distribution space, highlighting important connections, the original initial local implicit feature vectors of the light node RSSI distribution space may still contain a certain amount of noise or incomplete information, insufficient to accurately represent a specific signal-space pattern independently. The true meaning and value of these local patterns often need to be understood in conjunction with other related local patterns. Therefore, it is necessary to enable each initial local implicit feature vector in the light node RSSI distribution space to "learn" or "borrow" information from its associated neighbors based on the identified and meaningful associations (provided by the fine-grained correlation mask matrix), thereby correcting and enhancing itself. Based on this, based on the fine-grained correlation mask topology matrix of the light node RSSI distribution space manifold structure, feedback distillation is performed on each initial local implicit feature vector in the set of light node RSSI distribution space initial local implicit feature vectors to obtain a set of distilled initial local implicit feature vectors of the light node RSSI distribution space. Specifically, using the fine-grained correlation mask topology matrix of the manifold structure of the light node RSSI distribution space as a guide, information from the initial local implicit feature vectors of the RSSI distribution space of other light nodes that are strongly correlated with the initial local implicit feature vector of the current light node RSSI distribution space is incorporated into the representation of the current vector through mechanisms such as weighted aggregation or message passing. The goal of this process is to fuse the perspectives of its associated neighbors, thereby eliminating ambiguity, supplementing information, and promoting consistent and coordinated representation. In the context of garden lights, this means that a local feature describing the signal attenuation and spatial relationships in a specific area can be made more accurate and robust by referencing local features describing the signal patterns or spatial characteristics of neighboring areas. For example, if a local feature exhibits anomalies due to transient interference, but its multiple associated neighboring local features all stably point to a certain signal or spatial pattern, the representation of the anomalous local feature can be corrected through information fusion.

[0067] More specifically, in step S3524, the set of initial local implicit feature vectors of the spatial distillation of the light node RSSI distribution is subjected to feature reconstruction based on the self-attention mechanism to obtain the enhanced light node RSSI distribution spatial modulation feature coding vector, which is expressed as:

[0068]

[0069]

[0070] in, are the first, second, and third in the set of initial local implicit feature vectors of the spatial distillation of the RSSI distribution of the light node. and The initial local latent feature vector of the RSSI distribution space of each light node is distilled. is vector concatenation, Distill the implicit aggregate feature vector for the RSSI distribution space of the light node, 、 and They are the light node RSSI distribution space query weight matrix, the light node RSSI distribution space key weight matrix and the light node RSSI distribution space value weight matrix, 、 and They are respectively the query vector of the light node RSSI distribution space, the key vector of the light node RSSI distribution space, and the value vector of the light node RSSI distribution space. for The scale, for activation function, To enhance the RSSI distribution of the light node RSS point spatial modulation feature coding vector.

[0071] It should be understood that the feedback distillation step has improved the I distribution spatial modulation feature coding vector.

[0072] It should be understandable that although the feedback distillation step has improved the quality and coordination of the initial local implicit feature vectors of the RSSI distribution space of each light node, the final linkage decision often requires a single feature vector that can comprehensively summarize the current signal-space environment state as input, rather than a scattered set of local features. More importantly, the contribution of different local features to the overall linkage decision may be uneven, and some local patterns (such as signal strength in areas close to pedestrians) may be more important than other patterns (such as weak signal patterns at the edge of the courtyard). Simply averaging or simply splicing all the distilled local vectors cannot capture this difference in importance and the complex global dependencies between local features. Therefore, in the technical solution of the present application, the set of initial local implicit feature vectors of the distillation of the light node RSSI distribution space is further subjected to feature reconstruction based on the self-attention mechanism to obtain the enhanced light node RSSI distribution space modulation feature coding vector.

[0073] The self-attention mechanism leverages its powerful global information integration capabilities to dynamically capture long-range dependencies within the set of initial local latent feature vectors derived from the spatial distillation of the RSSI distribution of the light nodes. It then automatically determines, based on context, which distilled local features are most critical to the final overall representation. This allows the refined local detail information and the complex global dependencies between them to be dynamically and intelligently reconstructed into a single, unified, high-order enhanced feature vector. In the context of garden lights, this means the self-attention mechanism can automatically identify which distilled local signal-spatial feature patterns are crucial for determining which lights should be activated, based on the current signal distribution and spatial layout context. For example, even if the local features corresponding to a light fixture are far from the trigger point but located in the direction of pedestrian travel, if their signal and spatial patterns indicate that the light may need to be activated, the self-attention mechanism can assign them a higher weight, significantly incorporating their information into the final enhanced feature vector.

[0074] Preferably, in a specific example of the present application, here, the geometric correlation distribution of the spatial manifold structure correlation topology matrix of the light node RSSI distribution on the potential low-dimensional geometric structure will be nonlinearly undersaturated, so that the overall correlation topology distribution of the spatial manifold structure correlation topology matrix of the light node RSSI distribution will produce a geometric correlation structure field stable state contraction due to nonlinear interaction, which will be more prominent due to the correlation focusing enhancement effect of the gated mask function, affecting the intrinsic geometric structure expression ability of the fine-grained correlation mask topology matrix of the spatial manifold structure of the light node RSSI distribution.

[0075] Based on this, for the light node RSSI distribution space manifold structure fine-grained association mask topology matrix Each eigenvalue of , first generate the gradient field term:

[0076]

[0077] in, 、 The fine-grained association mask topology matrix of the light node RSSI distribution space manifold structure is Different position eigenvalues ​​in Represents the gradient operation.

[0078] It is used to correct the local nonlinear geometric structure inhomogeneity and thus achieve refined homogenization of the geometric correlation field.

[0079] Then, the gradient field term is used as the external field driving term to perform mean field control of each eigenvalue:

[0080]

[0081] in The light node RSSI distribution space manifold structure fine-grained association mask topology matrix The eigenmean of all eigenvalues ​​of , and As the normalization coefficient of the constrained strong driving factor, it is to modulate the scaling weight of the excessive external field driving term. To compensate for the RSSI distribution of rear light nodes, a fine-grained correlation mask topology matrix is ​​constructed based on the spatial manifold structure. Each eigenvalue of .

[0082] In this way, under the action of the external field driving term as a high-order gradient, the nonlinear saturation of the geometric correlation distribution under the mean field is reversely promoted (i.e., the response to strong gradients is reduced), thereby compensating for the substructure discretization effect caused by the correlation polarization enhancement through the average response compensation under the mean field, thereby improving the accuracy of the geometric structure expression of the fine-grained correlation mask topology matrix of the spatial manifold structure of the RSSI distribution of the lamp node.

[0083] Specifically, in step S360, the spatial modulation feature coding vector of the enhanced lamp node RSSI distribution is feature decoded to obtain the dynamic RSSI screening threshold. Accordingly, in a specific example of the present application, the spatial modulation feature coding vector of the enhanced lamp node RSSI distribution is passed through a decoder-based dynamic threshold generator to obtain the dynamic RSSI screening threshold. It should be understood that by understanding the current RSSI distribution status of the entire network and associating it with the nonlinear spatial importance of the lamps in the courtyard, a dynamic RSSI screening threshold that is more in line with the actual scene requirements can be generated. This threshold is derived based on the analysis of joint features such as spatial importance and signal feasibility. This enables the system to more accurately screen out lamps that are not only signal-reachable but also physically located in the area that needs to be linked and illuminated when extracting the target neighborhood node list based on the dynamic threshold, thereby avoiding unnecessary energy consumption and improving the intelligence and effectiveness of the linkage response. This will enable the entire courtyard light linkage control system to be more responsive, make smarter decisions, and achieve better energy-saving effects when realizing the intelligent lighting of "light follows people". It will be able to better adapt to the complex environmental changes and dynamic needs of users in the courtyard, thereby improving the user experience and the overall performance of the system.

[0084] Specifically, in step S400, based on the dynamic RSSI screening threshold, a target neighboring node list is extracted from the RSSI-containing neighboring light node list. It should be understood that although the system has already obtained the wireless signal strength (RSSI) of surrounding neighboring light fixtures and combined it with spatial distribution information to generate a dynamic threshold, this threshold itself is merely a standard and has not yet been directly converted into specific linkage instructions. A clear screening process is required to apply this dynamically and intelligently calculated "qualified threshold" to the actual RSSI values ​​of each neighboring node to identify which light fixtures, in the current environment, both meet communication reliability requirements (sufficient signal strength) and are determined by the system to be closely spatially associated with the trigger point and should participate in this linkage. Therefore, based on the dynamic RSSI screening threshold, a target neighboring node list is extracted from the RSSI-containing neighboring light node list. In particular, in one specific example of this application, the target neighboring node list can be obtained by traversing the neighboring light node list and screening for neighboring light nodes with RSSI values ​​greater than or equal to the dynamic RSSI screening threshold. This screening process directly determines which lights will be turned on, thereby achieving precise "light follows people" and efficient energy utilization, overcoming the problems of over-lighting or under-lighting caused by improper thresholds in traditional solutions.

[0085] Specifically, in step S500, the high priority, the ID of the current garden light node, the unique event ID, and the target neighbor node list are encapsulated to produce the linkage trigger message to be sent. It should be understood that after the aforementioned complex processing, the system has determined the trigger source of this linkage event (the current garden light node), the time of its occurrence (represented by the unique event ID), its importance (the high priority marker), and most importantly, which specific neighbor light nodes should participate in this linkage (the target neighbor node list). To effectively communicate this linkage decision to the relevant neighbor light nodes, this fragmented information must be integrated into a structured message carrier. Therefore, the high priority, the ID of the current garden light node, the unique event ID, and the target neighbor node list are further encapsulated to produce the linkage trigger message to be sent. This encapsulation ensures that the receiving light node obtains all the necessary context and instructions required to execute the linkage operation in one go, avoiding information omissions or mismatches. This is the foundation for efficient and reliable communication in distributed systems.

[0086] Generally speaking, the implementation process involves filling the corresponding fields with prepared information within the trigger light node or centralized control unit that issues the linkage command according to a pre-defined communication protocol or message structure. This structured data is then converted into a byte sequence suitable for transmission over a wireless network. A specific implementation example is as follows: First, during the system design phase, a standard data format or structure for linkage trigger messages is defined. This format specifies the fields contained in the message, as well as the order, size, and data type of each field. For example, there may be a field to indicate the message type (linkage trigger), a field for a high-priority flag (such as a Boolean value or a specific flag bit), a field to store the trigger light node ID (typically a fixed-length integer or byte array), a field to store a unique event ID (a composite value as described above), and a field or area to store a list of target neighboring nodes. This list typically begins with a field indicating the number of nodes in the list, followed by a list of the IDs of each target node.

[0087] Next, once the intelligent algorithm has derived the list of target neighboring nodes, the system processing unit allocates a memory buffer to construct the message to be sent. The system then sequentially fills each information item into the corresponding position in the memory buffer according to the predetermined message format. For example, the high priority flag is set to true; the ID of the currently triggering light node is copied into a specified field; the unique event ID generated for this event is copied into its corresponding field; the number of nodes in the target neighboring node list is written into the list length field, and the ID of each target node in the list is written one by one into the subsequent list data area. Finally, after all information fields are filled, depending on the communication protocol, a message header (containing source / destination addresses, checksum, etc.) may need to be added or data serialization (for example, encoding an integer or list into a byte stream) may be performed. This ultimately forms a complete byte sequence for the linkage trigger message to be sent that meets wireless transmission requirements. This byte sequence is the linkage instruction that will be broadcast or unicast to the target neighboring light fixtures via the wireless communication module.

[0088] In summary, the wireless networking-based intelligent garden light linkage control method according to the embodiments of this application is described. When any garden light senses a person passing by, it not only activates and illuminates itself, but also proactively communicates with neighboring lights via wireless networks, collecting their real-time signal strengths (RSSIs). The system then cleverly combines this dynamically changing wireless signal information with the pre-defined physical spatial layout of the garden lights for in-depth analysis. This analysis determines which neighboring lights are not only reachable but also closely spatially associated with the current trigger point, making them ideal for simultaneous illumination. Based on this intelligent analysis, which integrates signal quality and spatial relationships, the system dynamically calculates an optimal signal strength screening threshold and, based on this threshold, accurately selects the "target neighbor lights" that should participate in the linkage. Ultimately, only these intelligently selected target neighbor lights receive the command and are linked to illuminate. This entire process dynamically adjusts in real time as the user moves, ensuring that the lighting area accurately follows the pedestrian's trajectory. This significantly reduces unnecessary long-term light on time while meeting lighting needs, maximizing energy efficiency and efficiency.

[0089] Furthermore, an intelligent garden light linkage control system based on wireless networking is also provided.

[0090] Figure 5 FIG is a block diagram of an intelligent garden light linkage control system based on wireless networking according to an embodiment of the present application. Figure 5As shown, according to the embodiment of the present application, the intelligent garden light linkage control system 600 based on wireless networking includes: a light node RSSI data acquisition module 610, which is used to obtain the RSSI value of each light node; a light node spatial distribution topology construction module 620, which is used to construct the spatial distribution topology matrix between the light nodes; a light node RSSI distribution feature extraction module 630, which is used to arrange the RSSI values ​​of the light nodes into a light node RSSI distribution sequence, and then perform sequence encoding on it to obtain a light node RSSI distribution feature coding vector; a light node spatial distribution activation module 640, which is used to input the spatial distribution topology matrix into a nonlinear activation function to obtain a spatial distribution modulation activation matrix; a light node RSSI distribution feature space mapping module 650, which is used to map the light node RSSI distribution feature coding vector to the feature space defined by the spatial distribution modulation activation matrix and perform spatial fine-grained feature modulation enhancement processing to obtain an enhanced light node RSSI distribution spatial modulation feature coding vector; a dynamic RSSI screening threshold decoding module 660, which is used to feature decode the enhanced light node RSSI distribution spatial modulation feature coding vector to obtain the dynamic RSSI screening threshold.

[0091] As described above, the wireless networking-based intelligent garden light linkage control system 600 according to the embodiment of the present application can be implemented in various wireless terminals, such as a server equipped with a wireless networking-based intelligent garden light linkage control algorithm. In one possible implementation, the wireless networking-based intelligent garden light linkage control system 600 according to the embodiment of the present application can be integrated into the wireless terminal as a software module and / or hardware module. For example, the wireless networking-based intelligent garden light linkage control system 600 can be a software module in the operating system of the wireless terminal, or it can be an application developed for the wireless terminal; of course, the wireless networking-based intelligent garden light linkage control system 600 can also be one of the many hardware modules of the wireless terminal.

[0092] Alternatively, in another example, the smart garden light linkage control system 600 based on wireless networking and the wireless terminal may also be separate devices, and the smart garden light linkage control system 600 based on wireless networking may be connected to the wireless terminal through a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

[0093] While various embodiments of the present disclosure have been described above, the above descriptions are illustrative, non-exhaustive, and not intended to be limiting of the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is selected to best explain the principles of the embodiments, their practical applications, or improvements to existing technologies, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A method for controlling the linkage of intelligent garden lights based on wireless networking, characterized in that: include: In response to the MCU control unit detecting an event that the PIR sensor is activated, marking the event that the PIR sensor is activated as a high priority, and extracting the ID of the current garden light node; Generate a unique event ID based on the current timestamp; Obtain a list of neighboring light nodes containing RSSIs, and generate a dynamic RSSI screening threshold based on the list of neighboring light nodes containing RSSIs; Extracting a target neighborhood node list from the neighbor light node list containing RSSI based on the dynamic RSSI screening threshold; Encapsulating the high priority, the ID of the current garden light node, the unique event ID, and the target neighboring node list to obtain a linkage trigger message to be sent; The process of obtaining a list of neighboring light nodes including RSSIs and generating a dynamic RSSI screening threshold based on the list of neighboring light nodes including RSSIs includes: Get the RSSI value of each light node; Constructing a spatial distribution topology matrix between the light nodes; After arranging the RSSI values ​​of the light nodes into a light node RSSI distribution sequence, performing sequence encoding on the sequence to obtain a light node RSSI distribution feature encoding vector; Inputting the spatial distribution topology matrix into a nonlinear activation function to obtain a spatial distribution modulation activation matrix; Mapping the light node RSSI distribution feature coding vector to the feature space defined by the spatial distribution modulation activation matrix and performing spatial fine-grained feature modulation enhancement processing to obtain an enhanced light node RSSI distribution spatial modulation feature coding vector; The enhanced light node RSSI distribution spatial modulation feature coding vector is feature decoded to obtain the dynamic RSSI screening threshold.

2. The method for controlling the intelligent garden light linkage based on wireless networking according to claim 1, characterized in that: The value of each non-diagonal position in the spatial distribution topology matrix is ​​the spatial distance between the corresponding two light nodes.

3. The method for controlling the intelligent garden light linkage based on wireless networking according to claim 2, characterized in that: After arranging the RSSI values ​​of the light nodes into a light node RSSI distribution sequence, sequence encoding is performed to obtain a light node RSSI distribution feature encoding vector, including: Arranging the RSSI values ​​of the light nodes into a light node RSSI distribution sequence; The light node RSSI distribution sequence is passed through an LSTM-based sequence encoder to obtain the light node RSSI distribution feature encoding vector.

4. The method for controlling the intelligent garden light linkage based on wireless networking according to claim 3, characterized in that: Mapping the light node RSSI distribution feature coding vector to the feature space defined by the spatial distribution modulation activation matrix and performing spatial fine-grained feature modulation enhancement processing to obtain an enhanced light node RSSI distribution spatial modulation feature coding vector, including: Performing matrix multiplication on the RSSI distribution feature coding vector of the light node and the spatial distribution modulation activation matrix to map the RSSI distribution feature coding vector of the light node to the feature space defined by the spatial distribution modulation activation matrix to obtain the RSSI distribution spatial modulation feature coding vector of the light node; The spatial fine-grained feature modulation enhancement is performed on the light node RSSI distribution spatial modulation feature coding vector to obtain the enhanced light node RSSI distribution spatial modulation feature coding vector.

5. The method for controlling the intelligent garden light linkage based on wireless networking according to claim 4, characterized in that: Performing spatial fine-grained feature modulation enhancement on the light node RSSI distribution spatial modulation feature coding vector to obtain the enhanced light node RSSI distribution spatial modulation feature coding vector includes: Performing feature decomposition based on one-dimensional convolution coding on the modulation feature coding vector of the light node RSSI distribution space to obtain a set of initial local implicit feature vectors of the light node RSSI distribution space; Generate a fine-grained correlation mask topology matrix of the manifold structure of the light node RSSI distribution space based on the manifold structure correlation coefficient between any two initial local implicit eigenvectors of the light node RSSI distribution space in the set of initial local implicit eigenvectors of the light node RSSI distribution space; Based on the fine-grained associative mask topology matrix of the light node RSSI distribution space manifold structure, feedback distillation is performed on each light node RSSI distribution space initial local implicit feature vector in the set of light node RSSI distribution space initial local implicit feature vectors to obtain a set of light node RSSI distribution space distillation initial local implicit feature vectors; The set of initial local implicit feature vectors of the spatial distillation of the light node RSSI distribution is subjected to feature reconstruction based on the self-attention mechanism to obtain the enhanced light node RSSI distribution spatial modulation feature coding vector.

6. The method for controlling the intelligent garden light linkage based on wireless networking according to claim 5, characterized in that: Based on the manifold structure correlation coefficient between any two initial local implicit feature vectors of the light node RSSI distribution space in the set of the initial local implicit feature vectors of the light node RSSI distribution space, a fine-grained correlation mask topology matrix of the manifold structure of the light node RSSI distribution space is generated, including: Calculating the manifold structure correlation coefficient between any two initial local implicit eigenvectors of the light node RSSI distribution space in the set of the initial local implicit eigenvectors of the light node RSSI distribution space to obtain a light node RSSI distribution space manifold structure correlation topology matrix composed of multiple light node RSSI distribution space manifold structure correlation coefficients; The light node RSSI distribution space manifold structure association topology matrix is ​​input into a gated mask function to obtain the light node RSSI distribution space manifold structure fine-grained association mask topology matrix.

7. The method for controlling the intelligent garden light linkage based on wireless networking according to claim 6, characterized in that: The enhanced light node RSSI distribution spatial modulation feature coding vector is feature decoded to obtain the dynamic RSSI screening threshold, including: the enhanced light node RSSI distribution spatial modulation feature coding vector is passed through a decoder-based dynamic threshold generator to obtain the dynamic RSSI screening threshold.

8. The method for controlling the intelligent garden lights based on wireless networking according to claim 7, characterized in that: Based on the dynamic RSSI screening threshold, extracting a target neighborhood node list from the neighbor light node list containing RSSI, including: traversing the neighbor light node list, screening neighbor light nodes with RSSI values ​​greater than or equal to the dynamic RSSI screening threshold to obtain the target neighborhood node list.

9. An intelligent garden light linkage control system based on wireless networking, used to execute the intelligent garden light linkage control method based on wireless networking according to any one of claims 1 to 8, characterized in that: include: Light node RSSI data acquisition module, used to obtain the RSSI value of each light node; A light node spatial distribution topology construction module, used to construct a spatial distribution topology matrix between the light nodes; a light node RSSI distribution feature extraction module, configured to arrange the RSSI values ​​of the light nodes into a light node RSSI distribution sequence, and then perform sequence encoding on the sequence to obtain a light node RSSI distribution feature encoding vector; a light node spatial distribution activation module, configured to input the spatial distribution topology matrix into a nonlinear activation function to obtain a spatial distribution modulation activation matrix; A light node RSSI distribution feature space mapping module is used to map the light node RSSI distribution feature coding vector to the feature space defined by the spatial distribution modulation activation matrix and perform spatial fine-grained feature modulation enhancement processing to obtain an enhanced light node RSSI distribution spatial modulation feature coding vector; The dynamic RSSI screening threshold decoding module is used to perform feature decoding on the spatial modulation feature coding vector of the enhanced light node RSSI distribution to obtain the dynamic RSSI screening threshold.

Images