Elevator Internet of Things data high-concurrency hierarchical compression transmission method based on edge computing

Abstract

The invention discloses an elevator Internet of Things data high-concurrency hierarchical compression transmission method and system based on edge computing, and the method comprises the steps: an elevator Internet of Things edge computing node collects elevator operation full-amount Internet of Things data in real time, and the data is graded into key data and conventional data through a multi-dimensional data value quantitative evaluation method. The node carries out lossless packaging on key data, carries out high-power compression on conventional data, initializes adaptive transmission strategies for the two types of data and then transmits the two types of data to the cloud platform. The edge computing node senses the network transmission state in real time, dynamically adjusts the transmission strategy according to the network state, improves the conventional data compression ratio and reduces the transmission rate when the network is congested, and allocates higher bandwidth for key data. Meanwhile, a dynamic adaptive mathematical model is constructed based on network state parameters and transmission performance parameters, optimal parameters are solved, and self-adaptive adjustment and optimization of a transmission strategy are completed. And the real-time performance and effectiveness of elevator emergency disposal are guaranteed.

Description

A High-Concurrency Hierarchical Compression Transmission Method for Elevator IoT Data Based on Edge Computing Technical Field

[0001] This invention belongs to the field of elevator Internet of Things (IoT) technology, and more specifically, relates to a method and system for high-concurrency hierarchical compression and transmission of elevator IoT data based on edge computing. Background Technology

[0002] With the continuous increase in the number of high-rise buildings in cities, elevators, as core equipment in vertical transportation, have become an important component of urban public safety in terms of operational safety and emergency response capabilities. Currently, elevator IoT systems generally adopt a centralized cloud-based data processing model, requiring a large amount of elevator operation monitoring data, fault alarm data, and emergency rescue data to be uploaded to the cloud platform in real time for analysis and decision-making. In high-concurrency scenarios, such as when multiple elevators simultaneously trigger fault alarms or emergency rescue events, network bandwidth resources between edge nodes and the cloud platform are prone to bottlenecks, leading to delays and increased packet loss rates in critical emergency data transmission, severely impacting the efficiency of elevator fault diagnosis and entrapment rescue responses.

[0003] Most existing elevator IoT data transmission solutions employ a uniform transmission and compression strategy, failing to differentiate data value. This results in a large amount of non-critical routine operation monitoring data consuming limited network bandwidth resources, while critical data crucial for fault location and rescue route optimization cannot be prioritized. Furthermore, traditional transmission strategies lack dynamic awareness and adaptive adjustment capabilities regarding network status. When network congestion occurs, they cannot promptly adjust data compression ratios and transmission rates, further exacerbating transmission delays for critical data and failing to meet the millisecond-level response requirements of elevator emergency handling.

[0004] Furthermore, existing data value assessment methods largely rely on manually set weights or coefficients, which are highly subjective and easily influenced by human factors, failing to accurately identify data that truly has core value for elevator emergency response. This grading approach not only struggles to adapt to the complex and ever-changing elevator operating scenarios but also leads to a lack of scientific and rational allocation of transmission resources, hindering the overall performance and reliability of elevator IoT systems under high-concurrency scenarios. Therefore, how to achieve objective data value grading at the edge computing node level and construct a tiered compression transmission strategy that dynamically adapts to network conditions has become a key technical bottleneck in improving the emergency response capabilities and operational efficiency of elevator IoT systems. Summary of the Invention

[0005] This invention aims to solve problems such as data transmission congestion, insufficient protection of critical data, and strong subjectivity in value classification in high-concurrency transmission scenarios of elevator Internet of Things (IoT). By using edge computing, it achieves objective data classification and adaptive optimization of transmission strategies, thereby improving data transmission efficiency and reliability and ensuring the real-time and effective handling of elevator emergencies.

[0006] To address the aforementioned deficiencies or improvement needs of existing technologies, as a first aspect of this invention, the present invention provides a high-concurrency hierarchical compression and transmission method for elevator IoT data based on edge computing, comprising: S1. An elevator IoT edge computing node collects all IoT data generated during elevator operation in real time, and simultaneously evaluates the value of the all IoT data using a preset multi-dimensional data value quantification evaluation method, and then classifies the all IoT data into key data and regular data based on the evaluation results; S2. The edge computing node performs lossless processing and encapsulation on the key data, performs high-compression processing on the regular data, and then initializes corresponding transmission strategies for the two types of hierarchical data, namely key data and regular data; subsequently, the edge computing node initiates data transmission to the cloud platform based on the transmission strategies, completing the data transmission. S3. Edge computing nodes perceive the network transmission status between themselves and the cloud platform in real time, acquire network status parameters, and quantitatively compare them with preset threshold ranges. Based on the comparison results, they autonomously and dynamically adjust the transmission strategy locally. When the network is congested, the compression ratio of regular data is increased and its transmission rate is reduced. At the same time, a higher proportion of network bandwidth is quantitatively allocated to key data to ensure transmission priority. When the network status is good, the initial transmission strategy is maintained. S4. The edge computing nodes construct a dynamic adaptation mathematical model based on the quantitative change value of the network status parameters and the data transmission performance parameters. Through the dynamic adaptation mathematical model, the compression ratio adjustment range of regular data, the reduction ratio of transmission rate, and the bandwidth allocation ratio of key data are quantitatively solved and the optimal value is output. The transmission strategy is optimized based on the output results.

[0007] Furthermore, the full amount of IoT data in S1 includes elevator safety operation status data, fault warning related data, emergency response related data, and elevator routine operation monitoring data.

[0008] Furthermore, the key data in S1 includes elevator fault triggering characteristic data, elevator entrapment alarm data, elevator safety protection device action data, and positioning and status data required for elevator emergency rescue; the routine data includes floor feedback data, operating speed data, and routine status monitoring data during normal elevator operation.

[0009] Furthermore, the multi-dimensional data value quantification and evaluation method in S1 specifically involves: first constructing an anomaly correlation feature vector of elevator IoT data. Timeliness requirement feature vector , rescue support feature vector ; then calculate Elevator Abnormal Event Baseline Feature Vector cosine similarity , This characterizes the degree of linear correlation between a single data point and abnormal events, including elevator malfunctions and people entrapment. The standardized baseline feature vector is trained based on historical elevator anomaly data; calculation vector magnitude This characterizes the urgency of the timeliness requirement for transmitting a single piece of data. The larger the value, the higher the demand for real-time data transmission; calculation Reference feature vector of core information for elevator emergency rescue The reciprocal of the distance to Manhattan , , For the feature vector dimension, This is a standardized benchmark feature vector constructed based on the core information requirements of the elevator emergency rescue process. The reciprocal of this vector represents the degree to which a single data point supports elevator emergency rescue information; the smaller the distance, the higher the degree of support. Based on these three quantitative values, the comprehensive value assessment value of a single IoT data point is calculated. , This allows for the quantitative assessment of data value without manually setting weights or coefficients; it also establishes data grading thresholds. , The critical value obtained through statistical analysis of the value distribution characteristics of historical elevator IoT data is used to determine the comprehensive value assessment value. The IoT data was identified as key data, and its comprehensive value was assessed. The IoT data was judged to be regular data.

[0010] Furthermore, the network status parameters in S3 include the real-time available bandwidth between the edge computing node and the cloud platform, the end-to-end transmission latency, and the data transmission packet loss rate; the real-time available bandwidth is the real-time remaining network bandwidth when the edge computing node transmits data to the cloud platform, the end-to-end transmission latency is the real-time one-way transmission time from the edge computing node to the cloud platform, and the data transmission packet loss rate is the ratio of the number of data packets lost by the edge computing node to the total number of packets sent to the cloud platform within a preset statistical period.

[0011] Furthermore, the dynamic adaptation mathematical model in S4 specifically involves: first defining the network state vector. ,in This refers to the real-time available bandwidth between edge computing nodes and the cloud platform. For real-time end-to-end transmission latency, Define the packet loss rate for real-time data transmission; define the historical steady-state network state vector. ,in This is the baseline available bandwidth when the network is in steady state. The reference transmission delay in steady state of the network. Calculate the network state offset based on the baseline packet loss rate in steady state. This is used to characterize the degree of deviation of the current network transmission state from the steady-state state; the optimization vector is then redefined. ,in This refers to the adjustment range of the compression ratio for regular data. The percentage reduction in the transmission rate for regular data. Allocate bandwidth proportions for key data; construct a multi-objective optimization function. ;in This represents the overall data transmission efficiency of the elevator IoT system; the larger the value, the higher the transmission efficiency. This value represents the latency of key data transmission; the smaller the value, the lower the transmission latency. This characterizes the compression distortion of conventional data; the smaller the value, the lower the compression distortion. The multi-objective optimization objective is... Simultaneously, constraints are constructed, and the improved Lagrange multiplier method is used to solve the multi-objective optimization model.

[0012] Furthermore, the constraints include: key data transmission delay constraints. Bandwidth resource conservation constraint Compression ratio constraint , Minimum allowable compression ratio adjustment range and transmission rate constraints for regular data. , The percentage reduction is the maximum allowed transmission rate for regular data.

[0013] Furthermore, the improved Lagrange multiplier method specifically involves introducing Lagrange multipliers. , , , These correspond to four constraints, among which Corresponding to the critical data transmission latency constraints, This corresponds to the bandwidth resource conservation constraint. Corresponding to compression ratio constraints, To address the transmission rate constraint, construct the augmented Lagrangian function: The optimization vector is iteratively updated using the gradient descent method. With Lagrange multipliers until the augmented Lagrange function When it converges to a minimum, the output optimized vector is... This is the optimal solution; edge computing nodes are based on the optimal solution. , , The value is determined to achieve adaptive optimization of the transmission strategy.

[0014] As a second aspect of the present invention, a high-concurrency hierarchical compression and transmission system for elevator IoT data based on edge computing is also provided, comprising: a data acquisition and classification unit, used for real-time acquisition of all IoT data generated by elevator operation by the elevator IoT edge computing node, and simultaneously using a preset multi-dimensional data value quantification evaluation method to evaluate the value of the all IoT data, and then classifying the all IoT data into key data and regular data according to the evaluation results; a hierarchical processing and transmission unit, used for the edge computing node to perform lossless processing and encapsulation on the key data, perform high-compression processing on the regular data, and then initialize corresponding transmission strategies for the two types of hierarchical data, respectively, and then the edge computing node initiates data transmission to the cloud platform based on the transmission strategy to complete the hierarchical compression and transmission of the data; and a network sensing unit. The knowledge and parameter tuning unit is used by edge computing nodes to perceive the network transmission status between themselves and the cloud platform in real time, obtain network status parameters and compare them with preset threshold ranges. Based on the comparison results, the transmission strategy is dynamically adjusted locally. When the network is congested, the compression ratio of regular data is increased and its transmission rate is reduced. At the same time, a higher proportion of network bandwidth is allocated to key data to ensure transmission priority. When the network status is good, the initial transmission strategy is maintained. The model building and solution unit is used by edge computing nodes to build a dynamic adaptation mathematical model based on the quantitative change value of network status parameters and data transmission performance parameters. Through the dynamic adaptation mathematical model, the compression ratio adjustment range of regular data, the reduction ratio of transmission rate, and the bandwidth allocation ratio of key data are quantitatively solved and the optimal value is output. The transmission strategy is optimized based on the output results.

[0015] As a third aspect of the invention, a computer-readable storage medium is also provided, on which a computer program is stored, which is executed by a processor as described in any one of the claims: a method for high-concurrency hierarchical compression transmission of elevator Internet of Things data based on edge computing.

[0016] Overall, compared with the prior art, the above-described technical solutions conceived in this invention can achieve the following beneficial effects: 1. The high-concurrency hierarchical compression and transmission method for elevator IoT data based on edge computing of this invention constructs a multi-dimensional data value quantification and evaluation model on the edge computing node side. First, it constructs the abnormal correlation feature vector, timeliness demand feature vector, and rescue support feature vector of elevator IoT data. Then, it calculates the comprehensive value evaluation value of a single data point through mathematical measurement methods such as cosine similarity, vector magnitude, and reciprocal of Manhattan distance. This achieves automatic classification of key data and regular data without the need for manual setting of weights or coefficients. It ensures the objectivity and scientific nature of data classification and can accurately identify data with core value for elevator emergency response, providing a reliable basis for the formulation of subsequent differentiated transmission strategies. It also avoids classification deviations caused by manual intervention.

[0017] 2. The high-concurrency hierarchical compression and transmission method for elevator IoT data based on edge computing of the present invention performs differentiated processing and transmission strategy initialization on the hierarchical data, adopts lossless encapsulation processing for key data, and high-compression processing for regular data, and initializes corresponding transmission strategies respectively. While ensuring the integrity and real-time performance of key data, it effectively reduces the transmission bandwidth occupation of regular data, improves the overall transmission efficiency, and enables edge computing nodes to prioritize the transmission needs of core data related to emergency response in high-concurrency scenarios, while taking into account the transmission efficiency of regular operation monitoring data, thus realizing the rational allocation of data transmission resources.

[0018] 3. The edge computing-based high-concurrency hierarchical compression and transmission method for elevator IoT data of the present invention, by real-time sensing of network transmission status and construction of a dynamic adaptation mathematical model, constructs a multi-objective optimization function based on the quantified change value of network status parameters and data transmission performance parameters, and uses the improved Lagrange multiplier method to solve for the optimal solution, realizes adaptive optimization of the adjustment range of conventional data compression ratio, the reduction ratio of transmission rate, and the proportion of bandwidth allocation for key data, so that the transmission strategy can accurately adapt to changes in network status, prioritize the transmission of key data when the network is congested, and minimize the bandwidth occupation of conventional data, thereby improving the transmission reliability and efficiency of elevator IoT data in high-concurrency scenarios. Attached Figure Description

[0019] Figure 1 is a flowchart of the high-concurrency hierarchical compression and transmission method for elevator IoT data based on edge computing according to an embodiment of the present invention; Figure 2 is a general flowchart of the high-concurrency hierarchical compression and transmission method according to an embodiment of the present invention; Figure 3 is a schematic diagram of multi-dimensional data value quantification assessment and hierarchical classification according to an embodiment of the present invention; Figure 4 is a schematic diagram of dynamic adaptation mathematical model solving and transmission strategy optimization according to an embodiment of the present invention; Figure 5 is a schematic diagram of system units according to an embodiment of the present invention. Detailed Implementation

[0020] 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.

[0021] Example 1 (Referring to Figure 1) provides a high-concurrency hierarchical compression and transmission method for elevator IoT data based on edge computing, including: S1. The elevator IoT edge computing node collects all IoT data generated by elevator operation in real time, and simultaneously uses a preset multi-dimensional data value quantification evaluation method to evaluate the value of the all IoT data. Based on the evaluation results, the all IoT data is classified into key data and regular data; S2. The edge computing node performs lossless processing and encapsulation on the key data, performs high-compression processing on the regular data, and initializes corresponding transmission strategies for the two types of hierarchical data, namely key data and regular data. Subsequently, the edge computing node initiates data transmission to the cloud platform based on the transmission strategy, completing the hierarchical compression and transmission of the data; 3. Edge computing nodes perceive the network transmission status between themselves and the cloud platform in real time, acquire network status parameters, and quantitatively compare them with preset threshold ranges. Based on the comparison results, they autonomously and dynamically adjust the transmission strategy locally. When the network is congested, they increase the compression ratio of regular data and reduce its transmission rate, while allocating a higher proportion of network bandwidth to key data to ensure transmission priority. When the network status is good, they maintain the initial transmission strategy. S4. Edge computing nodes construct a dynamic adaptation mathematical model based on the quantitative change value of network status parameters and data transmission performance parameters. Through the dynamic adaptation mathematical model, they quantitatively solve and output the optimal value of the compression ratio adjustment of regular data, the reduction ratio of transmission rate, and the bandwidth allocation ratio of key data. Based on the output results, they complete the optimization of the transmission strategy.

[0022] Please refer to Figure 2. This embodiment 1 will further elaborate on the above steps.

[0023] (1) Data Acquisition and Classification In this embodiment, the elevator IoT edge computing node collects all IoT data generated during elevator operation in real time. The full IoT data specifically includes elevator safety operation status data, fault warning related data, emergency response related data, and elevator routine operation monitoring data, which can comprehensively reflect the real-time operating conditions and safety status of the elevator. In order to achieve reasonable allocation of transmission resources in high-concurrency scenarios, after completing data acquisition, the edge computing node adopts a pre-constructed multi-dimensional data value quantification evaluation method to uniformly quantify and evaluate the collected IoT data of various types, and classifies the data into key data and routine data according to the evaluation results.

[0024] Please refer to Figure 3. The multi-dimensional data value quantification assessment process first constructs an anomaly correlation feature vector of elevator IoT data. Timeliness requirement feature vector , rescue support feature vector ; then calculate Elevator Abnormal Event Baseline Feature Vector cosine similarity , This characterizes the degree of linear correlation between a single data point and abnormal events, including elevator malfunctions and people entrapment. The standardized baseline feature vector is trained based on historical elevator anomaly data; calculation vector magnitude This characterizes the urgency of the timeliness requirement for transmitting a single piece of data. The larger the value, the higher the demand for real-time data transmission; calculation Reference feature vector of core information for elevator emergency rescue The reciprocal of the distance to Manhattan , , For the feature vector dimension, This is a standardized benchmark feature vector constructed based on the core information requirements of the elevator emergency rescue process. The reciprocal of this vector represents the degree to which a single data point supports elevator emergency rescue information; the smaller the distance, the higher the degree of support. Based on these three quantitative values, the comprehensive value assessment value of a single IoT data point is calculated. , This allows for the quantitative assessment of data value without manually setting weights or coefficients; it also establishes data grading thresholds. , The critical value obtained through statistical analysis of the value distribution characteristics of historical elevator IoT data is used to determine the comprehensive value assessment value. The IoT data is identified as critical data, specifically including elevator malfunction triggering characteristic data, elevator entrapment alarm data, elevator safety protection device action data, and location and status data required for elevator emergency rescue; the comprehensive value assessment value will be... The IoT data is classified as routine data, specifically including floor feedback data, operating speed data, and routine status monitoring data during normal elevator operation.

[0025] (2) Hierarchical Processing and Transmission After data hierarchical processing is completed, edge computing nodes adopt corresponding processing methods based on the value differences and transmission requirements of critical data and regular data to adapt to the network transmission environment under high concurrency scenarios. For information determined to be critical data, to ensure data integrity and availability, edge computing nodes perform lossless processing and encapsulation, completing data regularization without changing the data content and structure, ensuring that fault, alarm, and rescue-related information is not distorted or lost during transmission and parsing. For information determined to be regular data, under the premise of meeting normal monitoring needs, edge computing nodes perform high-compression processing to reduce data volume and bandwidth usage, alleviating network pressure during high-concurrency transmission.

[0026] After completing differentiated data processing, edge computing nodes initialize appropriate transmission strategies based on the importance, real-time requirements, and network resource usage of the two types of data. The transmission strategy for critical data aims for high priority, low latency, and high reliability, while the strategy for regular data aims for efficient bandwidth utilization and reduced transmission overhead. Following the set transmission strategies, the edge computing nodes simultaneously transmit the processed critical data and regular data to the cloud platform, achieving hierarchical, orderly, and efficient data transmission. This ensures priority transmission of critical emergency data while improving overall data transmission efficiency, thereby completing the hierarchical and compressed transmission of elevator IoT data.

[0027] (3) Network perception and parameter tuning During the data hierarchical compression and transmission process, the stability of the network transmission status directly affects the reliability and efficiency of data transmission. Especially in the high-concurrency scenario of elevator IoT, the network status is prone to fluctuation, which may lead to problems such as delay in key data transmission and excessive bandwidth consumption of regular data. Therefore, edge computing nodes need to perceive the network transmission status between themselves and the cloud platform in real time.

[0028] Edge computing nodes, through their built-in network awareness modules, acquire core parameters characterizing network transmission status in real time, including real-time available bandwidth, end-to-end transmission latency, and data transmission packet loss rate. Real-time available bandwidth refers to the available remaining network bandwidth that can be utilized when the edge computing node transmits data to the cloud platform; end-to-end transmission latency refers to the real-time one-way transmission time from when data is sent from the edge computing node to when it is successfully received by the cloud platform; and data transmission packet loss rate refers to the ratio of the number of lost data packets to the total number of packets sent by the edge computing node to the cloud platform within a preset statistical period.

[0029] Edge computing nodes acquire various network status parameters and compare them with preset threshold ranges. Based on the comparison results, they dynamically adjust their transmission strategies locally. When the comparison results indicate network congestion, the edge computing nodes increase the compression ratio of regular data, reduce the transmission rate of regular data, and decrease the bandwidth usage of regular data. At the same time, they allocate a higher proportion of network bandwidth to critical data to ensure that the transmission priority of critical data is not affected. When the comparison results indicate good network status and all parameters are within the preset threshold range, the edge computing nodes maintain the initially set transmission strategy to ensure the stability and efficiency of data transmission, achieving dynamic adaptation between network status and transmission strategy.

[0030] (4) Model construction and solution: Please refer to Figure 4. Based on the dynamic adjustment of network transmission status, in order to further improve the adaptation accuracy of transmission strategy and solve the problem that the transmission parameter adjustment is not accurate enough and the optimal resource allocation cannot be achieved when the network status fluctuates, the edge computing node constructs a dynamic adaptation mathematical model based on the quantized change value of network status parameters and data transmission performance parameters. Through this model, the transmission strategy is optimized to ensure the transmission efficiency and reliability of elevator IoT data in high-concurrency scenarios.

[0031] When constructing this dynamic adaptation mathematical model, the network state vector is first defined. ,in This refers to the real-time available bandwidth between edge computing nodes and the cloud platform. For real-time end-to-end transmission latency, Define the packet loss rate for real-time data transmission; define the historical steady-state network state vector. ,in This is the baseline available bandwidth when the network is in steady state. The reference transmission delay in steady state of the network. Calculate the network state offset based on the baseline packet loss rate in steady state. This is used to characterize the degree of deviation of the current network transmission state from the steady-state state; the optimization vector is then redefined. ,in This refers to the adjustment range of the compression ratio for regular data. The percentage reduction in the transmission rate for regular data. Allocate bandwidth proportions for key data; construct a multi-objective optimization function. ;in This represents the overall data transmission efficiency of the elevator IoT system; the larger the value, the higher the transmission efficiency. This value represents the latency of key data transmission; the smaller the value, the lower the transmission latency. This characterizes the compression distortion of conventional data; the smaller the value, the lower the compression distortion. The multi-objective optimization objective is... Simultaneously, constraints are constructed, including: key data transmission delay constraints. Bandwidth resource conservation constraint Compression ratio constraint , Minimum allowable compression ratio adjustment range and transmission rate constraints for regular data. , The percentage reduction is the maximum allowed transmission rate for regular data.

[0032] The improved Lagrange multiplier method is then used to solve this multi-objective optimization model. Specifically, this involves introducing Lagrange multipliers. , , , These correspond to four constraints, among which Corresponding to the critical data transmission latency constraints, This corresponds to the bandwidth resource conservation constraint. Corresponding to compression ratio constraints, To address the transmission rate constraint, construct the augmented Lagrangian function: The optimization vector is iteratively updated using the gradient descent method. With Lagrange multipliers until the augmented Lagrange function When it converges to a minimum, the output optimized vector is... This is the optimal solution; edge computing nodes are based on the optimal solution. , , The values ​​are determined to achieve adaptive optimization of the transmission strategy, enabling accurate and dynamic adaptation between network conditions and the transmission strategy.

[0033] Example 2, referring to Figure 5, provides a high-concurrency hierarchical compression and transmission system for elevator IoT data based on edge computing. It includes: a data acquisition and classification unit, used by the elevator IoT edge computing node to collect all IoT data generated during elevator operation in real time, and simultaneously using a preset multi-dimensional data value quantification evaluation method to assess the value of the all IoT data, then classifying the all IoT data into key data and regular data based on the evaluation results; and a hierarchical processing and transmission unit, used by the edge computing node to perform lossless processing and encapsulation on key data, perform high-compression processing on regular data, and initialize corresponding transmission strategies for the two types of hierarchical data (key data and regular data). Subsequently, the edge computing node initiates data transmission to the cloud platform based on the transmission strategy, completing the hierarchical compression and transmission of the data. The network sensing and parameter tuning unit is used by edge computing nodes to perceive the network transmission status between themselves and the cloud platform in real time, acquire network status parameters and compare them with preset threshold ranges. Based on the comparison results, the transmission strategy is dynamically adjusted locally. When the network is congested, the compression ratio of regular data is increased and its transmission rate is reduced. At the same time, a higher proportion of network bandwidth is allocated to key data to ensure transmission priority. When the network status is good, the initial transmission strategy is maintained. The model building and solution unit is used by edge computing nodes to build a dynamic adaptation mathematical model based on the quantitative change value of network status parameters and data transmission performance parameters. Through the dynamic adaptation mathematical model, the compression ratio adjustment range of regular data, the reduction ratio of transmission rate, and the bandwidth allocation ratio of key data are quantitatively solved and the optimal value is output. The transmission strategy is optimized based on the output results.

[0034] Example 3 This example 3 also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, can implement any step of a high-concurrency hierarchical compression transmission method for elevator Internet of Things data based on edge computing.

[0035] The computer-readable storage medium may include various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0036] For a description of the computer-readable storage medium provided in this application, please refer to the above method embodiments; further details will not be repeated here.

[0037] 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 high-concurrency hierarchical compression and transmission of elevator IoT data based on edge computing, characterized in that, include: S1. The elevator IoT edge computing node collects all IoT data generated by elevator operation in real time. At the same time, it uses a preset multi-dimensional data value quantification evaluation method to evaluate the value of all IoT data. Based on the evaluation results, the all IoT data is classified into key data and regular data. S2. Edge computing nodes perform lossless processing and encapsulation on critical data and high-compression processing on regular data. Then, they initialize corresponding transmission strategies for the two types of hierarchical data: critical data and regular data. Subsequently, the edge computing nodes initiate data transmission to the cloud platform based on the transmission strategies to complete the hierarchical compression transmission of data. S3. Edge computing nodes perceive the network transmission status between themselves and the cloud platform in real time, obtain network status parameters and make quantitative comparisons with preset threshold ranges. Based on the comparison results, they autonomously and dynamically adjust the transmission strategy locally. When the network is congested, they increase the compression ratio of regular data and reduce its transmission rate. At the same time, they quantitatively allocate a higher proportion of network bandwidth to key data to ensure transmission priority. When the network status is good, they maintain the initial transmission strategy. S4. Edge computing nodes construct a dynamic adaptation mathematical model based on the quantified change values ​​of network state parameters and data transmission performance parameters. Through the dynamic adaptation mathematical model, the compression ratio adjustment range of regular data, the reduction ratio of transmission rate, and the bandwidth allocation ratio of key data are quantified and the optimal value is output. Based on the output results, the transmission strategy is optimized.

2. The method for high-concurrency hierarchical compression and transmission of elevator IoT data based on edge computing according to claim 1, characterized in that, The full amount of IoT data in S1 includes elevator safety operation status data, fault early warning related data, emergency response related data, and elevator routine operation monitoring data.

3. The method for high-concurrency hierarchical compression and transmission of elevator IoT data based on edge computing according to claim 1, characterized in that, The key data in S1 includes elevator fault triggering characteristic data, elevator entrapment alarm data, elevator safety protection device action data, and positioning and status data required for elevator emergency rescue; the routine data includes floor feedback data, operating speed data, and routine status monitoring data during normal elevator operation.

4. The method for high-concurrency hierarchical compression and transmission of elevator IoT data based on edge computing according to claim 1, characterized in that, The multi-dimensional data value quantification and evaluation method in S1 is as follows: First, construct the abnormal correlation feature vector of elevator Internet of Things data. Timeliness requirement feature vector , rescue support feature vector ; then calculate Elevator Abnormal Event Baseline Feature Vector cosine similarity , This characterizes the degree of linear correlation between a single data point and abnormal events, including elevator malfunctions and people entrapment. This is a standardized baseline feature vector trained based on historical elevator anomaly event data; calculate vector magnitude This characterizes the urgency of the timeliness requirement for transmitting a single piece of data. The larger the value, the higher the demand for real-time data transmission; calculate Reference feature vector of core information for elevator emergency rescue The reciprocal of the distance to Manhattan ， ， For the feature vector dimension, This is a standardized benchmark feature vector constructed based on the core information requirements of the elevator emergency rescue process. The reciprocal of this vector represents the degree to which a single data point supports elevator emergency rescue information; the smaller the distance, the higher the degree of support. Based on these three quantitative values, the comprehensive value assessment value of a single IoT data point is calculated. , This allows for the quantitative assessment of data value without manually setting weights or coefficients; it also establishes data grading thresholds. , The critical value obtained through statistical analysis of the value distribution characteristics of historical elevator IoT data is used to determine the comprehensive value assessment value. The IoT data was identified as key data, and its comprehensive value was assessed. The IoT data was judged to be regular data.

5. The method for high-concurrency hierarchical compression and transmission of elevator IoT data based on edge computing according to claim 1, characterized in that, The network status parameters in S3 include the real-time available bandwidth between the edge computing node and the cloud platform, the end-to-end transmission latency, and the data transmission packet loss rate. The real-time available bandwidth is the real-time remaining network bandwidth when the edge computing node transmits data to the cloud platform. The end-to-end transmission latency is the real-time one-way transmission time from the edge computing node to the cloud platform. The data transmission packet loss rate is the ratio of the number of data packets lost by the edge computing node to the total number of packets sent to the cloud platform within a preset statistical period.

6. The method for high-concurrency hierarchical compression and transmission of elevator IoT data based on edge computing according to claim 1, characterized in that, The dynamic adaptation mathematical model in S4 is specifically defined as follows: First, the network state vector is defined. ,in This refers to the real-time available bandwidth between edge computing nodes and the cloud platform. For real-time end-to-end transmission latency, Define the packet loss rate for real-time data transmission; define the historical steady-state network state vector. ,in This is the baseline available bandwidth when the network is in steady state. The reference transmission delay in steady state of the network. Calculate the network state offset based on the baseline packet loss rate in steady state. It is used to characterize the degree of deviation of the current network transmission state from the steady state; Redefining the optimization vector ,in This refers to the adjustment range of the compression ratio for regular data. The percentage reduction in the transmission rate for regular data. The bandwidth allocation ratio for critical data; Constructing a multi-objective optimization function ;in This represents the overall data transmission efficiency of the elevator IoT system; the larger the value, the higher the transmission efficiency. This value represents the latency of key data transmission; the smaller the value, the lower the transmission latency. This characterizes the compression distortion of conventional data; the smaller the value, the lower the compression distortion. The multi-objective optimization objective is... Simultaneously, constraints are constructed, and the improved Lagrange multiplier method is used to solve the multi-objective optimization model.

7. The method for high-concurrency hierarchical compression and transmission of elevator IoT data based on edge computing according to claim 6, characterized in that, The constraints include: key data transmission delay constraints. Bandwidth resource conservation constraint Compression ratio constraint ， Minimum allowable compression ratio adjustment range and transmission rate constraints for regular data. ， The percentage reduction is the maximum allowed transmission rate for regular data.

8. The method for high-concurrency hierarchical compression and transmission of elevator IoT data based on edge computing according to claim 7, characterized in that, The improved Lagrange multiplier method specifically involves introducing Lagrange multipliers. 、 、 、 These correspond to four constraints, among which Corresponding to the critical data transmission latency constraints, This corresponds to the bandwidth resource conservation constraint. Corresponding to compression ratio constraints, To address the transmission rate constraint, construct the augmented Lagrangian function: The optimization vector is iteratively updated using the gradient descent method. With Lagrange multipliers until the augmented Lagrange function When it converges to a minimum, the output optimized vector is... This is the optimal solution; edge computing nodes are based on the optimal solution. 、 、 The value is determined to achieve adaptive optimization of the transmission strategy.

9. A high-concurrency hierarchical compression and transmission system for elevator IoT data based on edge computing, characterized in that, include: The data acquisition and classification unit is used by the elevator IoT edge computing node to collect all IoT data generated by elevator operation in real time. It also uses a preset multi-dimensional data value quantification assessment method to evaluate the value of all IoT data, and then classifies the data into key data and regular data based on the assessment results. The classification processing and transmission unit is used by the edge computing node to perform lossless processing and encapsulation on key data and high-compression processing on regular data. It then initializes appropriate transmission strategies for each of the two types of classified data. Subsequently, the edge computing node initiates data transmission to the cloud platform based on the transmission strategy, completing the classified and compressed data transmission. The network perception and parameter tuning unit is used by the edge computing node to perceive the network transmission status between itself and the cloud platform in real time, obtain network status parameters and quantify them against preset threshold ranges. Based on the comparison results, it autonomously and dynamically adjusts the transmission strategy locally. When the network is congested, it increases the compression ratio of regular data and reduces its transmission rate, while allocating a higher proportion of network bandwidth to key data to ensure transmission priority. When the network status is good, it maintains the initial transmission strategy. The model building and solution unit is used by edge computing nodes to build a dynamic adaptation mathematical model based on the quantified change values ​​of network state parameters and data transmission performance parameters. Through the dynamic adaptation mathematical model, the compression ratio adjustment range of regular data, the reduction ratio of transmission rate, and the bandwidth allocation ratio of key data are quantified and the optimal value is output. Based on the output results, the transmission strategy is optimized.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, The computer program is executed by a processor as described in any one of claims 1-8: a method for high-concurrency hierarchical compression transmission of elevator IoT data based on edge computing.

Images