Controller communication management system and method based on star flash

Abstract

The invention relates to the technical field of communication, in particular to a controller communication management system and method based on star flash, and the system comprises a protocol adaptation module, a node dynamic networking module, a data flow optimization module, a behavior pattern matching module and a risk early warning output module. And arranging communication frame timestamps through a time axis to extract a high-risk area, identifying a signal intensity reversal point and a networking abnormal area, analyzing a data flow disturbance inflection point sequence, matching a behavior pattern, and finally generating a communication risk signal set. According to the method, quantitative analysis can be carried out from a multi-dimensional structure, signal intensity, a dynamic path and a time sequence behavior, full-process closed-loop tracking and node hierarchical positioning are realized, the accuracy and timeliness of communication risk prediction are improved, and accurate logic support is provided for high-risk evolution signals.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a controller communication management system and method based on Starflash. Background Art

[0002] With the rapid development of industrial automation and intelligence, controller communication management systems are becoming increasingly important in modern industrial equipment. Achieving efficient and reliable communication management is a key technical challenge, particularly in multi-node, distributed control systems. Existing controller communication management systems mostly rely on traditional protocols or hardware architectures, which present limitations in flexibility, scalability, and real-time performance. A search revealed a patent application for an AC drive system management and communication controller with publication number CN101290513B. This patent proposes a microprocessor-based communication controller equipped with multiple communication resources, including Ethernet and RS485 bus interfaces, and implements address decoding and data flow control through programmable logic devices. However, this technical solution is primarily designed for train control scenarios. While it offers a wide range of communication interfaces, it lacks support for emerging communication protocols (such as the StarFlash protocol) and has limited performance in scenarios with dynamic multi-node networking and high real-time requirements. Furthermore, the system's complex hardware architecture can increase costs and maintenance difficulties.

[0003] On the other hand, patent CN111966617B discloses a sensor information communication method, central processing unit, and baseboard management controller. By integrating sensor information into the heartbeat signal, this patent reduces the number of communication packets between the central processing unit and the baseboard management controller, thereby improving communication efficiency and reliability. However, this technical solution is mainly applicable to sensor information management within server systems and lacks support for efficient communication between multiple controller nodes in distributed systems. In addition, its reliance on heartbeat signals may lead to latency issues in application scenarios with high real-time requirements and fails to fully utilize the advantages of new communication protocols.

[0004] The above issues demonstrate that existing controller communication management systems still have shortcomings in supporting new communication protocols, adapting to multi-node dynamic networking requirements, improving real-time performance, and reducing system complexity. Therefore, the present invention provides a controller communication management system and method based on Starflash. By introducing the Starflash protocol, the system aims to optimize communication efficiency, simplify system architecture, enhance real-time performance and scalability, and thus meet the modern industrial demand for efficient and intelligent communication management systems. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a controller communication management system and method based on Star Flash.

[0006] In order to achieve the above objectives, the present invention adopts the following technical solution: a controller communication management system based on Star Flash, the system comprising:

[0007] The protocol adaptation module initializes and configures the data exchange between the controller and the front-end sensor based on the Star Flash protocol. It parses the communication frame format and extracts the signaling and data fields. It arranges the timestamps of the communication frames in chronological order, analyzes the time difference distribution between signaling and data transmission, filters communication frames whose time difference exceeds the real-time benchmark threshold, and generates a high-risk frame set for communication delay.

[0008] The node dynamic networking module identifies the signal strength fluctuation range within the communication frame node in the communication delay high-risk frame concentration and the signal strength difference between the previous and next time points, selects the node area where the signal strength direction is reversed and the fluctuation amplitude exceeds the network stability threshold, and generates a dynamic networking abnormal area map;

[0009] The data flow optimization module extracts the data flow paths at multiple time points according to the node numbers corresponding to the dynamic networking abnormal area map, performs sequence comparison on the path rate change values, locates the data flow rate mutation points, and generates a data flow disturbance inflection point sequence;

[0010] The behavior pattern matching module extracts the high-frequency disturbance segments and corresponding disturbance time periods in the data flow disturbance inflection point sequence, clusters the disturbance trajectory nodes, analyzes the trajectory turning sequence and distribution consistency, screens the disturbance behavior concentration area, and obtains the data flow behavior clustering map.

[0011] As a further solution of the present invention, the communication delay high-risk frame set includes a signaling delay aggregation area, a data frame loss high-frequency area, and a timestamp deviation abnormal area; the dynamic networking abnormal area diagram includes a signal strength attenuation point, a connection stability low-frequency area, and an inter-node interference edge area; the data flow disturbance inflection point sequence includes a rate mutation time point, a transmission strength mark point, and a data flow path inflection point; the data flow behavior clustering map includes an inflection point clustering sequence, a trajectory morphology category, and a disturbance segment frequency characteristic.

[0012] As a further solution of the present invention, the protocol adaptation module includes:

[0013] The communication frame parsing submodule initializes and configures the data exchange between the controller and the front-end sensor or device based on the Star Flash protocol. It parses the communication frame format and extracts the signaling field and data field. It arranges the timestamps of the communication frames in time sequence, identifies the time differences between adjacent time series and forms a time difference sequence. It compares the time differences with the real-time benchmark threshold item by item, locates the time difference frame position, and generates the time difference distribution value.

[0014] The delay high-risk screening submodule screens the communication frame sets whose time differences exceed the real-time benchmark threshold based on the time difference distribution value, extracts the timestamps of the communication frame sets and matches them with the node numbers, extracts the areas where the timestamps overlap and defines the connectivity range to obtain the delay high-risk area interval;

[0015] The high-risk frame positioning submodule spatially overlaps the delay high-risk area interval with the node topology map, selects the area with the highest signal strength and connection stability in the overlapping nodes, calibrates the corresponding position in the edge area of ​​the topology map, and generates a communication delay high-risk frame set.

[0016] As a further solution of the present invention, the node dynamic networking module includes:

[0017] The signal strength extraction submodule calls the communication delay high-risk frame set, divides the node area, extracts the local signal strength extreme value and average value, records the signal strength distribution at each time point, and generates a signal strength distribution data set;

[0018] The signal trend judgment submodule compares the mean signal strength of the node before and after the signal strength distribution data set, identifies the reversal trend of the signal strength direction, records the corresponding node index, and generates a signal strength change type set;

[0019] The network stability screening submodule extracts the signal strength variation value and neighborhood signal strength variance of the direction-reversed node based on the signal strength variation type set, calculates the network variation value by combining the direction angle difference and the signal strength mean difference, screens the node positions whose variation exceeds the network stability threshold, identifies the corresponding coordinates in the topology map, and generates a dynamic network anomaly area map.

[0020] As a further solution of the present invention, the data flow optimization module includes:

[0021] The node number extraction submodule matches the node topology map with the node area according to the dynamic networking abnormal area map, extracts and integrates the node numbers, performs node and path mapping, and obtains the node identification code value;

[0022] The data flow path calculation submodule calls the node identification code value, collects the data flow rate of the node at the time point, analyzes the time series and calculates the unit time rate, identifies the data flow rate sequence, and fuses it to obtain the data flow rate sequence;

[0023] The disturbance inflection point positioning submodule identifies the absolute peak value of the node data flow rate difference based on the data flow rate sequence, extracts the disturbance sensitivity array, analyzes the mutation interval, calculates the rate disturbance coefficient within the monitoring period, selects the extreme value corresponding time point and the node to generate the data flow disturbance inflection point sequence.

[0024] As a further solution of the present invention, the behavior pattern matching module includes:

[0025] The disturbance segment extraction submodule calls the data stream disturbance inflection point sequence, filters the segments whose disturbance amplitude exceeds the disturbance amplitude reference value, identifies the disturbance frequency and time interval, clusters the segments whose time interval is lower than the time interval threshold, and obtains a high-frequency disturbance segment set;

[0026] The inflection point trajectory analysis submodule extracts the inflection points of the cluster center trajectory based on the high-frequency disturbance segment set, collects the inflection point time series and spatial coordinates, identifies the time series variance and distribution density, calculates the inflection point trajectory dispersion value, determines the trajectory inflection point distribution characteristics based on the dispersion, and obtains the inflection point trajectory distribution characteristic set;

[0027] The path alignment consistency comparison submodule extracts the target path inflection point data based on the inflection point trajectory distribution characteristic set, analyzes the matching inflection point number ratio and trajectory shape consistency ratio, screens paths that meet the alignment rate and consistency requirements, and obtains a data flow behavior clustering map.

[0028] As a further embodiment of the present invention, the system further comprises:

[0029] The risk warning output module selects nodes and time periods with overlapping trajectories based on the matching degree between the disturbance trajectory path and the risk center trajectory in the data flow behavior clustering map, marks the disturbance and risk synchronization intervals, classifies and locates the corresponding time points and node numbers, and obtains a communication risk signal set;

[0030] The communication risk signal set includes a risk node number, a potential risk period, and a trend synchronization feature identifier.

[0031] As a further solution of the present invention, the risk warning output module includes:

[0032] The path screening submodule selects paths that meet the disturbance proximity threshold based on the matching degree between the disturbance trajectory path in the data flow behavior clustering map and the controller risk center trajectory, extracts the node number and corresponding time period of the path, and generates a communication risk path interval value set;

[0033] The synchronization marking submodule calls the node number and time period in the communication risk path interval value set, marks the time segment of disturbance synchronization according to the consistency of the node disturbance trend change in the path time sequence, and obtains the node disturbance trend synchronization interval value;

[0034] The key event positioning submodule extracts the ratio of the disturbance amplitude change rate to the duration based on the node number and time period in the node disturbance trend synchronization interval value, and screens the node sequences that exceed the evolution mutation threshold, locates the corresponding time point and node number, and obtains the communication risk signal set.

[0035] The controller communication management method based on Star Flash is executed based on the above-mentioned controller communication management system based on Star Flash, and includes the following steps:

[0036] S1: Initializes and configures data exchange between the controller and front-end sensors based on the Star Flash protocol. It parses the communication frame format and extracts the signaling and data fields. It then arranges the timestamps of the communication frames in chronological order. It analyzes the time difference distribution between signaling and data transmission, filters communication frames whose time difference exceeds the real-time benchmark threshold, and generates a high-risk frame set for communication delay.

[0037] S2: Identify the signal strength fluctuation range within the communication frame nodes in the high-risk communication delay frame set and the signal strength difference between the previous and next time points, screen out node areas where the signal strength direction is reversed and the fluctuation amplitude exceeds the network stability threshold, and generate a dynamic network abnormal area map;

[0038] S3: extracting data flow paths at multiple time points according to the node numbers corresponding to the dynamic networking abnormal area graph, performing sequence comparison on the path rate change values, locating the data flow rate mutation points, and generating a data flow disturbance inflection point sequence;

[0039] S4: extracting high-frequency disturbance segments and corresponding disturbance time periods in the data stream disturbance inflection point sequence, clustering disturbance trajectory nodes, analyzing trajectory turning sequence and distribution consistency, screening disturbance behavior concentration areas, and obtaining a data stream behavior clustering map;

[0040] S5: Based on the matching degree between the disturbance trajectory path and the risk center trajectory in the data flow behavior clustering map, the nodes and time periods where the trajectories overlap are screened, the disturbance and risk synchronization intervals are marked, the corresponding time points and node numbers are classified and located, and a communication risk signal set is obtained.

[0041] Compared with the prior art, the advantages and positive effects of the present invention are:

[0042] In the present invention, by continuously arranging the communication frame timestamps on the time axis and extracting the areas that overlap with the node topology map, it is possible to focus on the location of high-risk areas for communication delays, achieve highly sensitive identification of the evolution trend of communication functions, and enhance the timing analysis capability of communication variation detection. By comparing the local signal strength range and extracting the direction reversal point, a fine network anomaly map is constructed at the network level, so that the identification of abnormal areas has stronger quantitative accuracy and structural recognition clarity. With the help of the rate comparison of the data flow path under the unified time axis, the continuity and mutation of the node data flow disturbance can be grasped from the dynamic dimension, the key turning point state in the evolution process can be captured, and a stable foundation can be laid for subsequent path analysis. The path behavior is compared by the inflection point alignment rate and trajectory consistency index, and the data flow disturbance law is clustered and integrated, which improves the classification ability of the communication evolution logic in the node area, so that the potential evolution trend has the characteristics of traceability, classification, and evaluation. Based on the comparison of the path and the risk trajectory, the disturbance trend synchronization interval and evolution period are extracted, which effectively improves the forward-looking identification ability of key warning events and establishes a logical support system for the accurate output of high-risk evolution signals. By conducting detailed quantitative analysis from dimensions such as multi-dimensional structure, signal strength, dynamic path and timing behavior, the integrated prediction basis has the ability to track the entire process in a closed loop and locate node levels, making the expression of risk trends more accurate, differentiated and timely. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 is a system flow chart of the present invention;

[0044] Figure 2 This is a flowchart of obtaining the protocol adapter module of the present invention;

[0045] Figure 3 This is a flowchart for obtaining the node dynamic networking module of the present invention;

[0046] Figure 4 This is a flowchart of obtaining the data flow optimization module of the present invention;

[0047] Figure 5 This is a flowchart of obtaining the behavior pattern matching module of the present invention;

[0048] Figure 6 This is the acquisition flow chart of the risk warning output module of the present invention. DETAILED DESCRIPTION

[0049] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0050] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as an "exemplary" in the present invention should not be interpreted as being preferred or advantageous over other embodiments or designs. Rather, the use of the word "exemplary" is intended to present concepts in a concrete manner. Furthermore, in the embodiments of the present invention, "and / or" can mean both or either of the two.

[0051] In the embodiments of the present invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, when the distinction is not emphasized, the meanings they convey are the same. The terms "of," "corresponding," and "corresponding" may sometimes be used interchangeably. It should be noted that, when the distinction is not emphasized, the meanings they convey are the same.

[0052] In the embodiments of the present invention, sometimes a subscript such as W1 may be written as a non-subscript such as W1. When the difference is not emphasized, the meanings to be expressed are the same.

[0053] In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0054] See also Figure 1 The present invention provides a technical solution: a controller communication management system based on Star Flash, the system includes:

[0055] The protocol adaptation module initializes and configures the data exchange between the controller and the front-end sensor based on the Star Flash protocol. It parses the communication frame format and extracts the signaling and data fields. It arranges the timestamps of the communication frames in chronological order, analyzes the time difference distribution between signaling and data transmission, filters communication frames whose time difference exceeds the real-time benchmark threshold, and generates a high-risk frame set for communication delay.

[0056] The node dynamic networking module identifies the signal strength fluctuation range within the communication frame node in the high-risk communication delay frame concentration and the signal strength difference between the previous and next time points. It selects the node areas where the signal strength direction is reversed and the fluctuation amplitude exceeds the network stability threshold, and generates a dynamic networking abnormal area map.

[0057] The data flow optimization module extracts data flow paths at multiple time points based on the node numbers corresponding to the dynamic networking anomaly area graph, performs sequence comparison on the path rate change values, locates the data flow rate mutation points, and generates a data flow disturbance inflection point sequence;

[0058] The behavior pattern matching module extracts high-frequency disturbance segments and corresponding disturbance time periods in the data stream disturbance inflection point sequence, clusters disturbance trajectory nodes, analyzes the trajectory turning sequence and distribution consistency, screens the disturbance behavior concentration area, and obtains the data stream behavior clustering map;

[0059] The risk warning output module selects the nodes and time periods with overlapping trajectories based on the matching degree between the disturbance trajectory and the risk center trajectory in the data flow behavior clustering map, marks the disturbance and risk synchronization intervals, classifies and locates the corresponding time points and node numbers, and obtains the communication risk signal set.

[0060] The high-risk frame set for communication delay includes signaling delay aggregation areas, high-frequency areas for data frame loss, and abnormal timestamp deviation areas. The dynamic networking abnormal area map includes signal strength attenuation points, low-frequency areas for connection stability, and edge areas for interference between nodes. The data flow disturbance inflection point sequence includes rate mutation time points, transmission strength landmark points, and data flow path inflection points. The data flow behavior clustering map includes inflection point clustering sequences, trajectory morphology categories, and disturbance segment frequency characteristics. The communication risk signal set includes risk node numbers, potential risk periods, and trend synchronization feature identifiers.

[0061] See also Figure 2 , the protocol adaptation module includes:

[0062] The communication frame parsing submodule initializes and configures the data exchange between the controller and the front-end sensor or device based on the Star Flash protocol. It parses the communication frame format and extracts the signaling field and data field. It arranges the timestamps of the communication frames in time sequence, identifies the time differences between adjacent time series and forms a time difference sequence. It compares the time differences with the real-time benchmark threshold item by item, locates the time difference frame position, and generates the time difference distribution value.

[0063] Initialize the data exchange between the controller SN004 and the front-end lidar sensor SN001 based on the Star Flash protocol. This configuration sets the communication to Star Flash G mode, uses time division duplex (TDD), and allocates 40MHz of communication bandwidth. After the configuration is complete, the controller continuously monitors and collects the data frames exchanged between the two. When a binary sequence "01111110..." is captured, it is parsed as "0x7E 01 0A F1 C3 B2 99 45 10 01 04 D2 ... 0x7E", recognize that the flag bits at the beginning and end of the frame are both 0x7E, the address field is 0x01 (pointing to SN001), and the control field is 0x0A, thus determining that this is an information frame carrying data. Then extract the signaling field "F1C3B299" and the data field "45100104D2..." defined in the data link layer, parse the signaling field, treat it as a 32-bit unsigned integer, and obtain the hexadecimal timestamp value 0xF1C3B299, which is then converted to a decimal timestamp. milliseconds, process the subsequent ten communication frames in the same way, obtain their respective timestamps, and arrange these timestamp values ​​strictly in the order of acquisition to form a timestamp sequence containing eleven elements, specifically { }={4056127129,4056127131,4056127133,4056127136,4056127138,4056127140,4056127142,4056127145,4056127147,4056127149,4056127151}, the unit is milliseconds. Then, by subtracting two adjacent timestamps in the sequence, , to calculate the time difference between adjacent frames and obtain a time difference sequence consisting of ten elements {2,2,3,2,2,2,3,2,2,2}, in milliseconds, the real-time benchmark threshold The setting is based on the maximum end-to-end delay required by the automatic emergency braking (AEB) service that carries the communication, which is 10 milliseconds. After budgeting and deducting the delay of the non-physical layer link (such as protocol stack processing and software scheduling) in the system (a total of 7.5 milliseconds), the inter-frame time difference threshold of the physical layer transmission is set to 2.5 milliseconds. To verify the rationality of this value, 1000 communication tests were conducted in a standard urban road simulated driving environment. The statistical results show that the inter-frame time difference of 99.7% of the normal communication samples is less than 2.5 milliseconds, so Set to 2.5 milliseconds, and compare each element in the time difference sequence, such as the first element 2, the second element 2, and the third element 3, with The value 2.5 is compared one by one. When the third element 3 in the sequence is found to be greater than 2.5, the time difference is located to occur between the third and fourth timestamps of the original timestamp sequence, that is, between the third and fourth frames. Similarly, when the seventh element 3 is found to be greater than 2.5, it is located to occur between the seventh and eighth frames. These time differences greater than the threshold and their position information are integrated to generate the time difference distribution value.

[0064] The high-risk delay screening submodule, based on the time difference distribution value, screens the communication frame set whose time difference exceeds the real-time benchmark threshold, extracts the timestamps of the communication frame set and matches them with the node number, extracts the timestamp overlapping area and delineates the connectivity range to obtain the high-risk delay area interval;

[0065] Based on the generated time difference distribution value, that is, the array {2,2,3,2,2,2,3,2,2,2} and its corresponding frame position information, high-risk frames are screened. First, each element in the array is compared with the set real-time benchmark threshold. The time difference is compared with milliseconds, and all values greater than 2.5 milliseconds are screened out to form a time difference over-limit set {3, 3}, and the index positions of these over-limit values in the original time difference sequence are recorded as index 2 and index 6 respectively. These two indexes correspond to the time intervals between the 3rd frame and the 4th frame and between the 7th frame and the 8th frame. Therefore, the 4th frame and the 8th frame are preliminarily identified as the starting frames of the delay event, and the complete information of these two frames is extracted, including their respective time stamps and the source node number contained in the signaling field. According to the frame data record, the time stamp of the 4th frame is 4056127136 milliseconds, and the source node is SN001. The time stamp of the 8th frame is 4056127145 milliseconds, and the source node is also SN001. Matching these two sets of information generates a matching pair set: {(4056127136, SN001), (4056127145, SN001)}. Next, the relevance of these delay events is evaluated. A delay event correlation time window is set. The value of this window is set by referring to the vehicle speed (120 km / h) when the vehicle is running at high speed. In order to ensure the continuity of safety decision-making, the reaction time of related events should not exceed 50 milliseconds. Here, it is conservatively set to 20 milliseconds. Since the time stamp difference of the two delay events is 9 milliseconds, which is less than the correlation window of 20 milliseconds, and the source nodes of the events are both SN001, these two events are determined to be associated events. The area covered by these two time stamps is extracted and divided into a continuous connected range. The starting point of this range is the time stamp of the first event, 4056127136, and the ending point is the time stamp of the second event, 4056127145, forming the time interval [4056127136, 4056127145]. This interval is bound with the node number SN001 to obtain the delay high-risk area interval.

[0066] The high-risk frame positioning sub-module performs spatial overlap between the delay high-risk area interval and the node topology graph, selects the area where the signal strength and connection stability are in the top ranks of the sorting, and marks the corresponding position on the edge area of the topology graph to generate a set of communication delay high-risk frames.

[0067] ​According to the output delay high-risk area interval, that is, the time period [4056127136, 4056127145] and the associated node SN001, this information is spatially overlapped with the node topology map pre-stored in the system that describes the physical layout of the vehicle network. The topology map is a three-dimensional coordinate system that accurately calibrates the physical locations of all Star Flash nodes such as SN001 (left side of the front bumper), SN002 (right side of the front bumper), SN003 (roof) and controller SN004 (under the main driver's seat). The overlay operation is specifically manifested as highlighting the icon of SN001 on the graphical interface and marking it as being in an active risk analysis state. Subsequently, all links in the network log that have interacted with node SN001 during the precise time interval [4056127136, 4056127145] are queried, and the bidirectional signal strength and connection stability data on these links are extracted. The log shows that during this time period, SN001 only communicated with the controller SN004. During this time period, the average received signal strength indication (RSSI) of the SN001 to SN004 link was -58dBm, and the connection stability (obtained by calculating the proportion of data packets that were not requested for retransmission in 100 data packets during this time period) was 99.0%. At the same time, the average RSSI of the SN004 to SN001 link was -57dBm, and the connection stability was 99.5%. The two indicators of these link areas are sorted, and the sorting rule is: the larger the signal strength value (the closer to 0dBm), the higher the ranking, and the higher the connection stability percentage. The closer it is to the front, through weighted comprehensive sorting, both indicators of the link from SN004 to SN001 are in a better position, so this unidirectional link is selected as the analysis object. According to the topology map, the physical position of SN001 is at the forefront of the entire vehicle network, which meets the definition of the edge area of ​​the topology map. Finally, all communication frames sent by SN001 to SN004 with timestamps falling within the interval of [4056127136,4056127145], a total of 5 frames, are finally calibrated to generate a high-risk frame set for communication delay.

[0068] See also Figure 3 , the node dynamic networking module includes:

[0069] The signal strength extraction submodule calls the high-risk frame set for communication delay, divides the node area, extracts the local signal strength extreme value and average value, records the signal strength distribution at each time point, and generates a signal strength distribution data set;

[0070] The high-risk frame set for communication delay is a set of 5 frames of data. These frames are all sent by SN001 and received by SN004 in the time interval [4056127136, 4056127145]. First, based on the source node SN001 and the destination node SN004 recorded in these 5 frames of data, the analysis range is defined as the unidirectional link from SN001 to SN004 as an independent node area. Then, the received signal strength indication (RSSI) value recorded by the physical layer chip of the receiving end SN004 is extracted frame by frame. The precise timestamp recorded when these 5 frames of data are received is consistent with the corresponding RSSI value. The SI values ​​are (4056127136, -54dBm), (4056127138, -58dBm), (4056127140, -61dBm), (4056127142, -57dBm), (4056127145, -55dBm), and a basic statistical analysis is performed on the RSSI value set {-54, -58, -61, -57, -55} of the node area. By traversing the set, the maximum value of -54dBm is found as the upper limit of the local signal strength extreme value, and the minimum value of -61dBm is found as the lower limit, and then the summation is performed. , and divide the sum by the number of elements, 5, to calculate the average signal strength during this time period as -57 dBm. Record each time point and its corresponding signal strength distribution, and finally organize it into a structured data set, as shown in the following table, to generate a signal strength distribution data set.

[0071] Table 1: Signal strength distribution dataset

[0072] Timestamp (milliseconds) Node Link Signal strength (dBm) 4056127136 SN001-SN004 -54 4056127138 SN001-SN004 -58 4056127140 SN001-SN004 -61 4056127142 SN001-SN004 -57 4056127145 SN001-SN004 -55

[0073] As shown in Table 1, the dataset accurately records the continuous changes in signal strength on the target node link during the identified high-risk delay time period.

[0074] The signal trend judgment submodule compares the mean signal strength of the node before and after the signal strength distribution data set, identifies the reversal trend of the signal strength direction, records the corresponding node index, and generates a signal strength change type set;

[0075] Based on the signal strength distribution data set shown in Table 1, we perform trend judgment and compare the signal strength means of two adjacent time points in the data set sorted by time. The signal strength of the first time point 4056127136 is -54dBm, and the signal strength of the second time point 4056127138 is -58dBm. The calculated change is: dBm, the change trend is decreasing, and the change between the second and third time points (4056127140) is dBm, the change trend is still downward, and the change between the third and fourth time points (4056127142) is dBm. Since the sign of the change changes from negative to positive, it indicates that the direction of the signal strength change has reversed from decreasing to increasing. Immediately record the time point 4056127140 (as the last point before the reversal) and the index information of its associated node links SN001-SN004. Then, continue the backward analysis. The change between the fourth and fifth time points (4056127145) is dBm, the change trend is rising, which is the same as the trend in the previous stage, and no reversal has occurred. After completing the traversal of the entire data set, it is found that on the SN001-SN004 link, only one clear signal strength direction reversal occurred between time points 4056127140 and 4056127142, and the type of this event is marked as "falling followed by rising", generating a signal strength change type set.

[0076] The network stability screening submodule extracts the signal strength variation amplitude and neighborhood signal strength variance of direction-reversing nodes based on the signal strength variation type set. It then calculates the network variation amplitude value by combining the direction angle difference with the signal strength mean difference. It then screens the locations of nodes whose variation amplitude exceeds the network stability threshold, identifies the corresponding coordinates in the topology map, and generates a dynamic network anomaly area map.

[0077] According to the signal strength change type set, the node links SN001-SN004 where the direction reversal occurs are extracted, and the data near the reversal point is retrieved. First, the signal strength change amplitude is calculated. , that is, the absolute value of the difference between the signal strength of -61dBm at the reversal point 4056127140 and the signal strength of -57dBm at the first point after reversal 4056127142, is calculated as dBm. At the same time, extract the signal strength data of the neighboring node of node SN001 (the node with the closest physical distance in the topology diagram and communicating with the same controller, here SN002) to SN004 during the same period, and obtain the sequence {-50,-51,-52,-51,-50}dBm, and calculate its variance , with a mean of -50.8 and a variance of Since this embodiment is a vehicle-mounted fixed scenario, the relative angle between nodes remains unchanged, and the direction angle difference is set to 0. Then calculate the signal strength mean difference , that is, the absolute value of the difference between the average strength of the SN001-SN004 link -57dBm and the average strength of the neighboring SN002-SN004 link -50.8dBm, is dBm, then calculate the network variable amplitude value , and its calculation formula is , wherein represents a quantitative index of network stability, is a signal strength amplitude, is a neighborhood signal variance, is a mean difference from the neighborhood, and the weight is set based on regression analysis of 1000 groups of historical data, including 500 groups of stable data and 500 groups of known interference data, and the analysis result shows that the direct contribution of the amplitude value is the highest, the neighborhood deviation degree is the second, and the neighborhood stability has the least influence, so the weight is set to 0.6, the numerical value is substituted into the formula for calculation: , and the network stability threshold is set by statistically analyzing the values of the above 1000 groups of historical data, it is found that the values of 99% unstable scenarios are all greater than 3.5, and the values of 99% stable scenarios are all less than 3.2, so the intermediate value 3.5 is taken as the threshold, and the selection of this value has undergone ROC curve analysis to determine the best discrimination point, since the calculated value 3.972 is greater than the threshold 3.5, the result shows that the stability state of the node is determined to be abnormal, the node SN001 is screened out, and its corresponding physical coordinates in the node topology graph are highlighted to generate a dynamic network anomaly area graph.

[0078] Please refer to Figure 4 , the data stream optimization module includes:

[0079] a node number extraction submodule, according to the dynamic network anomaly area graph, matching the node area of the node topology graph, extracting the node number and integrating, mapping the node and the path, and obtaining the node identification code value;

[0080] The dynamic networking abnormal area map is a data structure that contains the physical coordinates of node SN001 and the abnormal status label at the data level. By comparing the coordinates with the node topology map database, the node number corresponding to the abnormal area is accurately matched to SN001. Since the analysis of the communication data flow requires complete path information, the network routing table is further queried to determine that the data flow of SN001 is directly sent to the main controller SN004 under the current configuration. Therefore, the nodes included in the complete data flow path related to the abnormal event are determined to be SN001 and SN004. The two node numbers {SN001, SN004} are added. 4} are integrated into an ordered set, and this frequently used path is mapped and assigned a unique path identifier, such as path P01. Then, in order to facilitate subsequent calculations and indexing, each node on path P01 is assigned an internally used integer identification code value. SN001 is assigned the code 1001 according to its device type (lidar) and serial number, and SN004 is assigned the code 4001 according to its device type (controller) and serial number. Finally, a structured data containing path identification and node coding is obtained, that is, the node identification code value set corresponding to path P01 is {1001, 4001}.

[0081] The data flow path calculation submodule calls the node identification code value, collects the data flow rate of the node at the time point, analyzes the time series and calculates the unit time rate, identifies the data flow rate sequence, and fuses it to obtain the data flow rate sequence;

[0082] The path P01 represented by the node identification code value {1001, 4001} is called, and based on this path information, the network monitoring system is requested to obtain the data flow rate information on path P01 within a wider time window before and after the exception occurs, that is, the interval [4056127130, 4056127155] milliseconds. The monitoring system returns the instantaneous data throughput statistics of the network interface controller (NIC) at the receiving end node SN004 of the path with an interval of 2 milliseconds, forming an original time series, of which some data are: at time point 4056127134, the rate is 12.5Mbps, and at time point 4056127136, it is 12.6 Mbps, 4056127138 is 12.4Mbps, 4056127140 is 8.2Mbps, 4056127142 is 12.5Mbps, 4056127144 is 9.1Mbps, and 4056127146 is 12.4Mbps. The submodule analyzes the time series. Since the acquisition time interval is fixed, the unit time rate is the instantaneous rate acquired. The rate sequence is identified as the data flow rate sequence of path P01. In order to eliminate sampling glitches and occasional small fluctuations, the submodule performs data fusion processing on the original sequence. The fusion here uses a moving average filter algorithm with a window size of 3 to perform data fusion on the time points. The rate value , by calculating For example, the new rate value for time point 4056127138 is Mbps, the new rate value for time point 4056127140 is Mbps, after processing the entire sequence, a smoother data flow rate sequence is obtained.

[0083] The disturbance inflection point location submodule identifies the absolute peak of the node data flow rate difference based on the data flow rate sequence, extracts the disturbance sensitivity array, analyzes the mutation interval, calculates the rate disturbance coefficient within the monitoring period, selects the time point corresponding to the extreme value and corresponds to the node, and generates the data flow disturbance inflection point sequence;

[0084] Based on the generated smoothed data flow rate sequence, for example, {…, 12.52, 12.50, 11.07, 11.03, 10.93, 11.33, …} Mbps, the disturbance inflection point is located. First, the rate difference between each two adjacent data points in the sequence is calculated to obtain a difference sequence: {…, -0.02, -1.43, -0.04, -0.10, +0.40, …}. Then, the absolute value of each element in the difference sequence is taken to obtain a disturbance sensitivity value that reflects the severity of the disturbance. The perceptual array is: {…, 0.02, 1.43, 0.04, 0.10, 0.40, …}. We traverse this array and identify the local maximum points, that is, the value of this point is greater than the previous and next values. In this array, 1.43 is a significant local maximum, and its corresponding raw rate change is from 12.50 Mbps to 11.07 Mbps (in the raw data before smoothing, it corresponds to a sharp drop from 12.4 Mbps to 8.2 Mbps). We then calculate the rate disturbance coefficient within the monitoring period. , which is calculated as the maximum value in the disturbance sensitivity array (1.43) divided by the average rate during the time period (assuming it is 11.5Mbps). , set a disturbance inflection point screening threshold, select all extreme points in the disturbance sensitivity array that exceed the preset threshold (for example, 0.5, which is set based on the 99.9% percentile of normal rate fluctuations in historical data), and match the timestamps of these points with the node paths to which they belong. Finally, record the time point 4056127140 (the time point when the sharp drop occurred), path P01, and the specific rate change value -4.2Mbps (the change in the original data) as a disturbance inflection point event to generate a data stream disturbance inflection point sequence.

[0085] Table 2: Example of data flow perturbation inflection point sequence table

[0086] Timestamp (milliseconds) Path identification Raw rate change (Mbps) 4056127140 P01 -4.20 4056127144 P01 -3.40

[0087] As shown in Table 2, the sequence records the specific time point, path, and change amount when the data flow rate changes dramatically.

[0088] See also Figure 5 , the behavioral pattern matching module includes:

[0089] The disturbance segment extraction submodule calls the data stream disturbance inflection point sequence, filters the segments whose disturbance amplitude exceeds the disturbance amplitude baseline value, identifies the disturbance frequency and time interval, clusters the segments whose time interval is lower than the time interval threshold, and obtains the high-frequency disturbance segment set;

[0090] Call the data stream perturbation inflection point sequence shown in Table 2. First, set a perturbation amplitude reference value The setting of this benchmark value refers to a large amount of data collected under various normal working conditions (such as different weather conditions and different road conditions). Statistics show that 99.9% of the absolute value of the fluctuation of normal data flow rate is less than 2.0Mbps. Set to 2.0Mbps, traverse the disturbance inflection point sequence, and filter out the inflection points whose absolute value of the disturbance amplitude exceeds 2.0Mbps. The absolute values ​​of the disturbance amplitudes of the two inflection points in Table 2 are 4.20 and 3.40 respectively, both exceeding the threshold, so all are selected. Next, identify the time intervals between these selected inflection points. The timestamps of the two inflection points are 4056127140 and 4056127144 respectively, and the time interval is milliseconds, then set a time interval threshold The threshold is set based on the assumption that multiple data stream disturbances caused by the same physical cause (such as a short occlusion or interference) usually occur at very short intervals. Based on the frame structure and scheduling period of star flashes, the threshold is set to 10 milliseconds. Since the calculated time interval of 4 milliseconds is lower than the time interval threshold of 10 milliseconds, the two inflection point events are determined to belong to the same disturbance event cluster and are clustered into a high-frequency disturbance segment. The time range of this segment covers the time from the first inflection point to the last inflection point, that is, [4056127140,4056127144], and the high-frequency disturbance segment set is obtained.

[0091] The inflection point trajectory analysis submodule extracts the inflection points of the cluster center trajectory based on the high-frequency disturbance segment set, collects the inflection point time series and spatial coordinates, identifies the time series variance and distribution density, calculates the inflection point trajectory dispersion value, determines the trajectory inflection point distribution characteristics based on the dispersion, and obtains the inflection point trajectory distribution characteristic set;

[0092] Extract all cluster inflection points of the high-frequency disturbance segment set in the time period [4056127140, 4056127144], that is, the two inflection points in Table 2, and collect the time series {4056127140, 4056127144} of these two inflection points and their spatial coordinates. Since both inflection points belong to path P01, their spatial coordinates are logically the same (both on P01), so the spatial coordinates can be regarded as a constant value. Then analyze the distribution characteristics of the inflection points in the time dimension and the spatial dimension, and calculate the variance of the inflection point time series. The mean is 4056127142 and the variance is , the distribution density is the number of inflection points divided by the segment length, that is, times / millisecond, and then calculate the inflection point trajectory dispersion value , which is defined as the weighted sum of the spatial coordinate variance and the time coordinate variance. Since the spatial coordinate is constant, its variance is 0, so The value of is determined only by the time variance, where , according to the preset discreteness judgment rule ( The distribution characteristics of the trajectory are judged to be "time-concentrated" and the spatial variance is 0, which is judged to be "space-concentrated". The distribution characteristics set of the inflection point trajectory is obtained.

[0093] The path alignment consistency comparison submodule extracts the target path inflection point data based on the inflection point trajectory distribution characteristic set, analyzes the matching inflection point number ratio and trajectory shape consistency ratio, selects paths that meet the alignment rate and consistency requirements, and obtains the data flow behavior clustering map;

[0094] Based on the inflection point trajectory distribution characteristic set, the "time-space dual concentration" type disturbance trajectory data of the target path P01 is extracted, and the trajectory is aligned and compared with a typical behavior pattern stored in the risk pattern knowledge base, which is known to be caused by "close overtaking by large vehicles". The characteristics of this typical pattern are also "time-space dual concentration", and its disturbance sequence is characterized by "a rapid and large rate drop, followed by a small rate recovery, and then a second large drop". The matching degree calculation is performed. The first item is the number ratio of matching inflection points. The target path has two main inflection points. The typical pattern is defined as having 2 to 3 main inflection points. 2 falls within this interval, and the number matching degree is 100%. The second item is the trajectory shape consistency ratio. The distance between the perturbation value sequence {-4.20, -3.40} of the target path and the template sequence {-4.5, -3.8} of the typical pattern is calculated by the dynamic time warping (DTW) algorithm. The calculated distance is 0.44, which is less than the preset shape consistency distance threshold of 0.5. Therefore, the shapes are judged to be consistent. The overall alignment rate threshold of the path is set to 90%, and the consistency requirement threshold is set to 90%. Since the quantity matching and shape consistency of the current path P01 meet the requirements, the behavior pattern of the path is judged to be a successful match, and it is classified into the "vehicle occlusion risk" category in the system, and a data flow behavior clustering map is obtained.

[0095] See also Figure 6 , the risk warning output module includes:

[0096] The path screening submodule selects paths that meet the disturbance proximity threshold based on the matching degree between the disturbance trajectory path in the data flow behavior clustering graph and the controller risk center trajectory, extracts the node number and corresponding time period of the path, and generates a communication risk path interval value set;

[0097] The perturbation trajectory of path P01 in the data flow behavior clustering map is screened based on the degree of match between it and the controller risk center trajectory. The controller risk center trajectory here is defined as the set of all paths that directly input key perception data (such as lidar, camera) to the main controller SN004. Path P01 (from lidar SN001 to controller SN004) is a member of this set, so its matching degree with the risk center trajectory is 100%, and the set perturbation is close to the threshold. It is required that the path must be a direct component of the risk center trajectory in topology, that is, the topological distance is 0 hops. Path P01 obviously meets this condition. Therefore, path P01 is screened out, and the numbers of all nodes on the path, namely {SN001, SN004}, and the specific time period determined in the previous module that matches the risk behavior pattern, namely [4056127140, 4056127144], are extracted. This information is combined to generate a communication risk path interval value set.

[0098] The synchronization marking submodule calls the node number and time period in the communication risk path interval value set, marks the time segment of disturbance synchronization according to the consistency of the node disturbance trend change in the path time sequence, and obtains the node disturbance trend synchronization interval value;

[0099] The node numbers {SN001, SN004} and time periods [4056127140, 4056127144] in the communication risk path interval value set are called, and based on this information, the disturbance event sequence of each node on the path (sender SN001 and receiver SN004) is retrieved from the lower-level log. The log shows that the data sending queue of SN001 began to be congested at time 4056127139 (manifested by an instantaneous drop in the sending rate), while the receiving end of SN004 detected a sudden drop in the rate at 4056127140. The disturbance trend changes of these two events (both rate drops) are consistent, and the time difference between their occurrences is 2. milliseconds, which is consistent with the agreed physical propagation delay of the signal between the two (for example, 0.8 milliseconds) plus the processing delay of a single node. Therefore, it is determined that the two disturbance events occurring on different nodes are synchronized, and the time segment [4056127139, 4056127140] showing this synchronization is marked to obtain the node disturbance trend synchronization interval value.

[0100] The key event location submodule extracts the ratio of the disturbance amplitude change rate to the duration based on the node number and time period in the node disturbance trend synchronization interval value, and screens the node sequences that exceed the evolution mutation threshold, locates the corresponding time point and node number, and obtains the communication risk signal set;

[0101] According to the node {SN001, SN004} marked in the node disturbance trend synchronization interval value and the synchronization time period [4056127139, 4056127140], the most severe disturbance data in the interval is extracted for final evaluation, that is, the event that the rate of SN004 at the receiving end suddenly dropped from 12.4Mbps to 8.2Mbps. The disturbance amplitude change rate is calculated, that is, the rate change divided by the time it takes for the change to occur, which is , and then calculate the duration ratio, where the ratio is the absolute value of the amplitude change rate, i.e. 4.2, to set the evolution mutation threshold of a key event This threshold is obtained by analyzing 50 communication failure cases that have historically caused the degradation or failure of key functions such as AEB. The analysis found that the absolute value of the disturbance amplitude change rate in these cases exceeded 3.0Mbps / ms, while 99.5% of non-failure disturbances had this value below 2.5Mbps / ms. Therefore, the threshold is set Since the currently calculated value of 4.2 exceeds the threshold of 3.0, this event is judged as a high-risk critical communication event, and the node sequence {SN001, SN004} directly related to the event is screened out. Finally, the risk is located at the node and time point where the disturbance is most severe, that is, node SN004 and time point 4056127140 milliseconds, and the communication risk signal set is obtained.

[0102] The controller communication management method based on Star Flash includes the following steps:

[0103] S1: Initializes and configures data exchange between the controller and front-end sensors based on the Star Flash protocol. It parses the communication frame format and extracts the signaling and data fields. It then arranges the timestamps of the communication frames in chronological order. It analyzes the time difference distribution between signaling and data transmission, filters communication frames whose time difference exceeds the real-time benchmark threshold, and generates a high-risk frame set for communication delay.

[0104] S2: Identify the signal strength fluctuation range within the communication frame nodes in the high-risk communication delay frame cluster and the signal strength difference between the previous and next time points. Filter out the node areas where the signal strength direction reverses and the fluctuation amplitude exceeds the network stability threshold, and generate a dynamic network abnormal area map.

[0105] S3: Extract the data flow paths at multiple time points based on the node numbers corresponding to the dynamic networking anomaly area graph, perform sequence comparison on the path rate change values, locate the data flow rate mutation points, and generate a data flow disturbance inflection point sequence;

[0106] S4: Extract high-frequency disturbance segments and corresponding disturbance time periods in the data flow disturbance inflection point sequence, cluster disturbance trajectory nodes, analyze trajectory turning sequence and distribution consistency, screen disturbance behavior concentration areas, and obtain data flow behavior clustering map;

[0107] S5: Based on the matching degree between the disturbance trajectory path and the risk center trajectory in the data flow behavior clustering map, the nodes and time periods with overlapping trajectories are screened, the disturbance and risk synchronization intervals are marked, the corresponding time points and node numbers are classified and located, and the communication risk signal set is obtained.

[0108] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. The controller communication management system based on Star Flash is characterized by: The system comprises: The protocol adaptation module initializes and configures the data exchange between the controller and the front-end sensor based on the Star Flash protocol. It parses the communication frame format and extracts the signaling and data fields. It arranges the timestamps of the communication frames in chronological order, analyzes the time difference distribution between signaling and data transmission, filters communication frames whose time difference exceeds the real-time benchmark threshold, and generates a high-risk frame set for communication delay. The node dynamic networking module identifies the signal strength fluctuation range within the communication frame node in the communication delay high-risk frame concentration and the signal strength difference between the previous and next time points, selects the node area where the signal strength direction is reversed and the fluctuation amplitude exceeds the network stability threshold, and generates a dynamic networking abnormal area map; The data flow optimization module extracts the data flow paths at multiple time points according to the node numbers corresponding to the dynamic networking abnormal area map, performs sequence comparison on the path rate change values, locates the data flow rate mutation points, and generates a data flow disturbance inflection point sequence; The behavior pattern matching module extracts the high-frequency disturbance segments and corresponding disturbance time periods in the data flow disturbance inflection point sequence, clusters the disturbance trajectory nodes, analyzes the trajectory turning sequence and distribution consistency, screens the disturbance behavior concentration area, and obtains the data flow behavior clustering map.

2. The controller communication management system based on Star Flash according to claim 1 is characterized in that: The communication delay high-risk frame set includes the signaling delay aggregation area, the data frame loss high-frequency area, and the timestamp deviation abnormal area. The dynamic networking abnormal area map includes the signal strength attenuation point, the connection stability low-frequency area, and the node interference edge area. The data flow disturbance inflection point sequence includes the rate mutation time point, the transmission strength mark point, and the data flow path inflection point. The data flow behavior clustering map includes the inflection point clustering sequence, the trajectory morphology category, and the disturbance segment frequency characteristics.

3. The controller communication management system based on Star Flash according to claim 1 is characterized in that: The protocol adaptation module includes: The communication frame parsing submodule initializes and configures the data exchange between the controller and the front-end sensor or device based on the Star Flash protocol. It parses the communication frame format and extracts the signaling field and data field. It arranges the timestamps of the communication frames in time sequence, identifies the time differences between adjacent time series and forms a time difference sequence. It compares the time differences with the real-time benchmark threshold item by item, locates the time difference frame position, and generates the time difference distribution value. The delay high-risk screening submodule screens the communication frame sets whose time differences exceed the real-time benchmark threshold based on the time difference distribution value, extracts the timestamps of the communication frame sets and matches them with the node numbers, extracts the areas where the timestamps overlap and defines the connectivity range to obtain the delay high-risk area interval; The high-risk frame positioning submodule spatially overlaps the delay high-risk area interval with the node topology map, selects the area with the highest signal strength and connection stability in the overlapping nodes, calibrates the corresponding position in the edge area of ​​the topology map, and generates a communication delay high-risk frame set.

4. The controller communication management system based on Star Flash according to claim 1, characterized in that: The node dynamic networking module includes: The signal strength extraction submodule calls the communication delay high-risk frame set, divides the node area, extracts the local signal strength extreme value and average value, records the signal strength distribution at each time point, and generates a signal strength distribution data set; The signal trend judgment submodule compares the mean signal strength of the node before and after the signal strength distribution data set, identifies the reversal trend of the signal strength direction, records the corresponding node index, and generates a signal strength change type set; The network stability screening submodule extracts the signal strength variation value and neighborhood signal strength variance of the direction-reversed node based on the signal strength variation type set, calculates the network variation value by combining the direction angle difference and the signal strength mean difference, screens the node positions whose variation exceeds the network stability threshold, identifies the corresponding coordinates in the topology map, and generates a dynamic network anomaly area map.

5. The controller communication management system based on Star Flash according to claim 1, characterized in that: The data flow optimization module includes: The node number extraction submodule matches the node topology map with the node area according to the dynamic networking abnormal area map, extracts and integrates the node numbers, performs node and path mapping, and obtains the node identification code value; The data flow path calculation submodule calls the node identification code value, collects the data flow rate of the node at the time point, analyzes the time series and calculates the unit time rate, identifies the data flow rate sequence, and fuses it to obtain the data flow rate sequence; The disturbance inflection point positioning submodule identifies the absolute peak value of the node data flow rate difference based on the data flow rate sequence, extracts the disturbance sensitivity array, analyzes the mutation interval, calculates the rate disturbance coefficient within the monitoring period, selects the extreme value corresponding time point and the node to generate the data flow disturbance inflection point sequence.

6. The controller communication management system based on Star Flash according to claim 1, characterized in that: The behavior pattern matching module includes: The disturbance segment extraction submodule calls the data stream disturbance inflection point sequence, filters the segments whose disturbance amplitude exceeds the disturbance amplitude reference value, identifies the disturbance frequency and time interval, clusters the segments whose time interval is lower than the time interval threshold, and obtains a high-frequency disturbance segment set; The inflection point trajectory analysis submodule extracts the inflection points of the cluster center trajectory based on the high-frequency disturbance segment set, collects the inflection point time series and spatial coordinates, identifies the time series variance and distribution density, calculates the inflection point trajectory dispersion value, determines the trajectory inflection point distribution characteristics based on the dispersion, and obtains the inflection point trajectory distribution characteristic set; The path alignment consistency comparison submodule extracts the target path inflection point data based on the inflection point trajectory distribution characteristic set, analyzes the matching inflection point number ratio and trajectory shape consistency ratio, screens paths that meet the alignment rate and consistency requirements, and obtains a data flow behavior clustering map.

7. The controller communication management system based on Star Flash according to claim 1 is characterized in that: The system further comprises: The risk warning output module selects nodes and time periods with overlapping trajectories based on the matching degree between the disturbance trajectory path and the risk center trajectory in the data flow behavior clustering map, marks the disturbance and risk synchronization intervals, classifies and locates the corresponding time points and node numbers, and obtains a communication risk signal set; The communication risk signal set includes a risk node number, a potential risk period, and a trend synchronization feature identifier.

8. The controller communication management system based on Star Flash according to claim 7, characterized in that: The risk warning output module includes: The path screening submodule selects paths that meet the disturbance proximity threshold based on the matching degree between the disturbance trajectory path in the data flow behavior clustering map and the controller risk center trajectory, extracts the node number and corresponding time period of the path, and generates a communication risk path interval value set; The synchronization marking submodule calls the node number and time period in the communication risk path interval value set, marks the time segment of disturbance synchronization according to the consistency of the node disturbance trend change in the path time sequence, and obtains the node disturbance trend synchronization interval value; The key event positioning submodule extracts the ratio of the disturbance amplitude change rate to the duration based on the node number and time period in the node disturbance trend synchronization interval value, and screens the node sequences that exceed the evolution mutation threshold, locates the corresponding time point and node number, and obtains the communication risk signal set.

9. The controller communication management method based on Star Flash is characterized by: The method is used for the Starflash-based controller communication management system according to any one of claims 1 to 8, comprising the following steps: S1: Initializes and configures data exchange between the controller and front-end sensors based on the Star Flash protocol. It parses the communication frame format and extracts the signaling and data fields. It then arranges the timestamps of the communication frames in chronological order. It analyzes the time difference distribution between signaling and data transmission, filters communication frames whose time difference exceeds the real-time benchmark threshold, and generates a high-risk frame set for communication delay. S2: Identify the signal strength fluctuation range within the communication frame nodes in the high-risk communication delay frame set and the signal strength difference between the previous and next time points, screen out node areas where the signal strength direction is reversed and the fluctuation amplitude exceeds the network stability threshold, and generate a dynamic network abnormal area map; S3: extracting data flow paths at multiple time points according to the node numbers corresponding to the dynamic networking abnormal area graph, performing sequence comparison on the path rate change values, locating the data flow rate mutation points, and generating a data flow disturbance inflection point sequence; S4: extracting high-frequency disturbance segments and corresponding disturbance time periods in the data stream disturbance inflection point sequence, clustering disturbance trajectory nodes, analyzing trajectory turning sequence and distribution consistency, screening disturbance behavior concentration areas, and obtaining a data stream behavior clustering map; S5: Based on the matching degree between the disturbance trajectory path and the risk center trajectory in the data flow behavior clustering map, the nodes and time periods where the trajectories overlap are screened, the disturbance and risk synchronization intervals are marked, the corresponding time points and node numbers are classified and located, and a communication risk signal set is obtained.

Images