An industrial wireless signal transmission system

Abstract

This invention relates to the field of industrial communication technology, and more particularly to an industrial wireless signal transmission system, comprising: a signal acquisition and access module that accesses a network via a wired communication link and multiple wireless communication links; a link status perception module that acquires the operating status of each wireless link and forms status feature information; a multi-layer operation prediction module that predicts the future status of the links based on equipment-level, production line-level, and factory-level data and generates risk assessment information; a constraint decision module that, based on the status features and risk assessment, determines the target link or link combination and transmission control parameters through an optimizable decision model; an adaptive coding module that performs predictive difference coding based on the control parameters and dynamically adjusts the coding threshold and synchronization interval; and a transmission execution module that completes data transmission and performs link switching and resource adjustment when the risk exceeds the threshold. This achieves proactive identification and adaptive scheduling control of link risks.

Description

Technical Field

[0001] This invention relates to the field of industrial communication technology, and in particular to an industrial wireless signal transmission system. Background Technology

[0002] In the field of industrial automation and intelligent manufacturing, the large amount of operational data generated by industrial equipment usually needs to be transmitted to upper-layer application systems for monitoring and analysis via industrial Ethernet or wireless communication links. With the application of communication technologies such as 4G, 5G and wireless LAN in industrial scenarios, the combination of wired and multiple wireless links for data transmission has become a common technical solution. Existing systems usually obtain parameters such as latency, packet loss rate and bandwidth through link quality detection, and select or switch links based on the current link status to ensure the continuity of data transmission.

[0003] Most existing technologies only perform real-time scheduling based on the current link operating status, lacking the ability to predict future link load changes and operating trends. This makes it difficult to respond in a timely manner to the risk of link congestion caused by fluctuations in industrial field services or changes in network resources, resulting in a lag in link switching. Summary of the Invention

[0004] To overcome the above shortcomings, this invention provides an industrial wireless signal transmission system, which aims to improve the problem of link congestion risk caused by fluctuations in industrial field services or changes in network resources, resulting in delayed link switching.

[0005] This invention provides the following technical solution: an industrial wireless signal transmission system, comprising the following modules:

[0006] The signal acquisition and access module is used to acquire operating signals of industrial equipment and access the network through wired communication links and various wireless communication links;

[0007] The link status awareness module is used to acquire the operating status parameters of each wireless communication link in real time and form the status feature information of the corresponding link.

[0008] The multi-layer operation prediction module is used to predict the future operation status of each wireless communication link within a preset time window based on equipment-level operation data, production line-level load data, and factory-level network operation data, and generate risk assessment information for each wireless communication link.

[0009] The constraint decision module is used to determine the transmission control strategy through a decision model based on state characteristic information and risk assessment information. The decision model is optimized based on historical transmission results and determines the target wireless communication link or link combination and the corresponding transmission control parameters under the condition of meeting preset service performance constraints.

[0010] The adaptive coding module is used to perform predictive difference coding on the data to be transmitted based on the transmission control parameters, and dynamically adjust the coding threshold and synchronization interval based on risk assessment information, so as to realize data reconstruction and correction through the synchronization mechanism when data loss occurs.

[0011] The transmission execution module is used to perform data transmission according to the target wireless communication link or link combination and transmission control parameters, and to perform link switching control and wireless resource adjustment when the risk assessment information exceeds a preset threshold.

[0012] By adopting the above technical solution, a multi-layer operation prediction mechanism based on equipment-level, production line-level, and factory-level data was constructed. This mechanism predicts the future operating status of wireless communication links within a preset time window and generates risk assessment information. It enables early identification and forward-looking scheduling control of link congestion risks, and improves the link switching lag problem caused by real-time scheduling based solely on the current link status.

[0013] Preferably, in the signal acquisition and access module, the network access via wired communication links and multiple wireless communication links includes:

[0014] Collect operating signals from industrial equipment and establish a wired data channel via industrial Ethernet;

[0015] Perform data encapsulation processing on the operating signals;

[0016] At the same time, at least two different wireless communication links are established, including at least one of 4G and 5G links;

[0017] According to the preset transmission strategy, the encapsulated operating signal is distributed to the wired data channel and wireless communication link for parallel or redundant transmission.

[0018] Preferably, in the link state awareness module, the real-time acquisition of the operating state parameters of each wireless communication link includes:

[0019] Perform periodic link quality checks on each wireless communication link;

[0020] Collect latency, packet loss rate, available bandwidth, and link load parameters for each wireless communication link;

[0021] The collected operating status parameters are statistically processed over a time window and then normalized. The processing results are used as the status feature information of the corresponding link.

[0022] Preferably, in the multi-layer operation prediction module, the equipment-level operation data, production line-level load data, and factory-level network operation data include:

[0023] Real-time sampling of the operating status of industrial equipment, recording the frequency of equipment data generation and the magnitude of data changes, forming equipment-level operating data;

[0024] The number of online devices and the total amount of data transmitted per unit time within the production line area are counted, the regional link load value is calculated, and production line-level load data is generated.

[0025] Monitor the bandwidth occupancy and link utilization of each communication link throughout the plant, calculate the overall network load level, and generate plant-level network operation data.

[0026] Preferably, in the multi-layer operation prediction module, the prediction of the future operating status of each wireless communication link within a preset time window includes:

[0027] Collect historical link status data within a preset time window;

[0028] Construct a time series-based prediction model to calculate the trends of link latency, packet loss rate, and load changes;

[0029] The future fluctuation range and reliability change trend of the link are calculated based on the prediction results, and the risk assessment information is generated accordingly.

[0030] Preferably, in the constraint decision module, determining the transmission control strategy through the decision model includes:

[0031] The state characteristics information and risk assessment information are used as input states;

[0032] Based on the input status, select the corresponding link selection strategy and coding control strategy from the preset action set as the decision action;

[0033] After the data transmission is completed, the actual transmitted data is obtained, and the model parameters are updated based on the actual transmitted data to form an iteratively optimized decision-making mechanism.

[0034] Preferably, in the constraint decision module, determining the target wireless communication link or link combination and its corresponding transmission control parameters under the condition of satisfying preset service performance constraints includes:

[0035] Set business performance constraint thresholds, including the maximum allowable latency threshold and the maximum allowable packet loss rate threshold;

[0036] The performance of candidate wireless communication links or link combinations is evaluated, and the predicted delay and predicted packet loss rate are calculated within a preset time window.

[0037] When the predicted latency and predicted packet loss rate meet the service performance constraint thresholds, the corresponding wireless communication link or link combination and transmission control parameters are determined.

[0038] Preferably, in the adaptive coding module, the predictive difference coding process for the data to be transmitted based on the transmission control parameters includes:

[0039] Build a data change prediction model based on historical data collection;

[0040] The predicted value is obtained by using a prediction model to predict the data to be sent.

[0041] Calculate the difference between the current data to be sent and the predicted value;

[0042] The difference is encoded and used as the actual data to be sent.

[0043] Preferably, in the adaptive coding module, the step of dynamically adjusting the coding threshold and synchronization interval based on risk assessment information includes:

[0044] Calculate the risk level based on the link risk assessment information;

[0045] When the risk level increases, the trigger threshold for prediction difference coding is increased and the synchronization interval is shortened;

[0046] When the risk level decreases, the trigger threshold for prediction difference coding is lowered and the synchronization interval is extended;

[0047] Synchronization data is periodically sent to correct for data reconstruction errors at the receiving end.

[0048] Preferably, in the transmission execution module, the step of performing link switching control and radio resource adjustment when the risk assessment information exceeds a preset threshold includes:

[0049] The performance of the current target wireless communication link is retested and confirmed;

[0050] Select links from the candidate wireless communication link set that meet the service performance constraints in terms of predicted latency and predicted packet loss rate as the handover target links;

[0051] Before performing link switching, establish a communication connection to the target link and perform short-term parallel transmission;

[0052] After the link switch is completed, the allocation ratio of wireless communication resources is adjusted according to the link load.

[0053] The present invention has the following beneficial effects:

[0054] 1. In this invention, by constructing a multi-layered operation prediction mechanism at the equipment level, production line level, and factory level, a comprehensive assessment of the future operating status of wireless communication links is achieved, and link risk assessment information is generated. This solves the problem that traditional scheduling based solely on the current link status makes it difficult to anticipate link congestion and fluctuations.

[0055] 2. In this invention, by introducing an iteratively optimized decision model, link state feature information and risk assessment information are used as inputs and combined with historical transmission results to update parameters, a closed-loop decision mechanism of state-action-feedback is realized, which solves the problem that fixed rule scheduling is difficult to adapt to changes in complex wireless environments.

[0056] 3. In this invention, by constructing a risk-driven adaptive coding and link switching control mechanism, dynamic adjustment of coding threshold and synchronization interval and smooth link switching control are realized, solving the problems of data reconstruction error accumulation and service interruption caused by sudden switching under link fluctuation conditions. Attached Figure Description

[0057] Figure 1 This is an architectural diagram of an industrial wireless signal transmission system proposed in this invention. Detailed Implementation

[0058] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0059] In a first embodiment of the present invention, the present invention provides an industrial wireless signal transmission system, such as... Figure 1 As shown, it includes the following modules:

[0060] The signal acquisition and access module is used to acquire operating signals of industrial equipment and access the network through wired communication links and various wireless communication links;

[0061] Furthermore, in the signal acquisition and access module, network access is achieved through wired communication links and various wireless communication links, including:

[0062] Collect operating signals from industrial equipment and establish a wired data channel via industrial Ethernet;

[0063] Perform data encapsulation processing on the operating signals;

[0064] At the same time, at least two different wireless communication links are established, including at least one of 4G and 5G links;

[0065] According to the preset transmission strategy, the encapsulated operating signals are distributed to wired data channels and wireless communication links for parallel or redundant transmission.

[0066] Specifically, the signal acquisition and access module is deployed on the edge gateway side of the industrial site to complete the acquisition, encapsulation, and multi-link collaborative access processing of industrial equipment operating signals. The industrial equipment can be sensors, PLC controllers, or actuators in the production line, and its operating signals include process monitoring data and control feedback data. The equipment output signals first enter the edge gateway through a local interface and are periodically acquired according to a preset sampling frequency; the sampling frequency is denoted as... The sampling period is Satisfying the relationship By controlling the sampling frequency, a balance can be achieved between data real-time performance and network load.

[0067] Before the signal enters the transmission stage, the collected data undergoes unified encapsulation processing. This encapsulation process includes adding device identifiers, timestamps, and verification fields to form a standard data frame structure. A data frame can be represented as... ,in Indicates equipment identification. Represents a timestamp. Represents load data, This represents the verification field; the verification field is generated using a cyclic redundancy check function to ensure data consistency in multi-link transmission scenarios.

[0068] When establishing a wired communication link, a stable data channel is established through industrial Ethernet; the edge gateway detects link reachability and obtains the currently available bandwidth during the link establishment phase; the rated bandwidth of the wired link is... The current bandwidth used is Link utilization is defined as When the link utilization rate is lower than a preset threshold, it is included in the set of schedulable links; this mechanism ensures that wired links participate in data transmission within their carrying capacity.

[0069] Simultaneously establish at least two different wireless communication links, such as 4G and 5G links; after each wireless link is established, record its current available bandwidth. Link load rate and instantaneous latency To uniformly assess the schedulable capabilities of different links, a comprehensive link availability index is introduced. The calculation method is as follows:

[0070] ;

[0071] in The reference latency constant is used for normalization; this indicator comprehensively reflects bandwidth resources, load level, and latency status, and is used for subsequent data allocation decisions.

[0072] After establishing wired and wireless links, the encapsulated data frames are allocated according to a preset transmission strategy. The preset transmission strategy determines the data flow direction based on service priority and overall link availability. Link allocation weights are defined as follows: ,in This is the business priority coefficient; the allocation ratio is calculated based on the weights.

[0073] ;

[0074] in The number of candidate links; the data to be sent is divided into sub-data streams according to the proportion, and sent in parallel through different links; this parallel mode is used to improve throughput stability.

[0075] In redundant transmission scenarios, when the service involves critical control data, the data frame is copied to multiple links that meet the minimum availability threshold and sent simultaneously. The receiving end sorts the data according to the timestamp and removes duplicate data. This mechanism ensures that data access can still be completed even when some links are fluctuating.

[0076] During continuous system operation, the edge gateway updates the overall availability of the links in real time and dynamically adjusts the allocation ratio according to changes in link status. When the availability of wireless links decreases, the data share of wired links is automatically increased. When the load on wired links increases, the data sharing ratio of wireless links is increased. Through the above operation process, the collaborative access and dynamic scheduling of industrial equipment operation signals between wired communication links and multiple wireless communication links are realized.

[0077] Through the synergistic effect of sampling control, data encapsulation, link status assessment, and weight ratio allocation mechanism, a complete multimodal network access structure is formed. This structure provides a unified data entry point for subsequent link status awareness, multi-layer operation prediction, and adaptive decision-making, and supports stable data access and transmission scheduling in industrial environments.

[0078] The link status awareness module is used to acquire the operating status parameters of each wireless communication link in real time and form the status feature information of the corresponding link.

[0079] Furthermore, in the link status awareness module, the real-time acquisition of operational status parameters for each wireless communication link includes:

[0080] Perform periodic link quality checks on each wireless communication link;

[0081] Collect latency, packet loss rate, available bandwidth, and link load parameters for each wireless communication link;

[0082] The collected operating status parameters are statistically processed over a time window and then normalized. The processing results are used as the status feature information of the corresponding link.

[0083] Specifically, the link status awareness module is deployed inside the edge gateway or communication control unit to continuously monitor and model the status of multiple established wireless communication links. These wireless communication links can include 4G, 5G, or wireless LAN links. To ensure the timeliness and comparability of the status data, the module performs periodic link quality checks on each link at a fixed detection cycle; this detection cycle is denoted as... , which represents the time interval between two adjacent detections; a link quality detection process is triggered once in each detection cycle, and the link operating parameters are obtained by sending test data packets and receiving confirmation responses;

[0084] During the link quality detection process, for the first The first wireless link collects instantaneous latency, packet loss rate, available bandwidth, and link load parameters; the second... The link is in the 1st The instantaneous delay during the second detection is denoted as Packet loss rate is denoted as Available bandwidth is denoted as The link load rate is denoted as ;in Indicates the link number. The number of tests is indicated; the instantaneous latency can be calculated as the difference between the sending time and the acknowledgment time of the test data packet; the packet loss rate can be calculated as the ratio of the total number of test packets sent to the number of successful responses; the link load rate can be calculated as the ratio of the current bandwidth occupied by the link to the rated bandwidth of the link.

[0085] To eliminate the impact of instantaneous fluctuations on the decision-making system, the collected operating status parameters are subjected to time window statistical processing; let the length of the statistical time window be... , representing the number of continuously collected detection samples; the average delay is calculated within the window range to obtain the window average delay:

[0086] ;

[0087] in Indicates the first The average latency of each link within the current statistical window; similarly, the packet loss rate is calculated by averaging within a window.

[0088] ;

[0089] The available bandwidth was calculated using a window average.

[0090] ;

[0091] The above statistical processing can smooth out short-term jitter and make the link status reflect the overall operating trend;

[0092] After statistical processing, parameters of different dimensions are normalized to form state characteristic information of a uniform scale; let the maximum reference time delay be... The maximum packet loss rate is The maximum available bandwidth is The normalized delay indicator is defined as follows:

[0093] ;

[0094] The normalized packet loss rate is denoted as:

[0095] ;

[0096] The normalized bandwidth indicator is defined as follows:

[0097] ;

[0098] The meanings of all symbols remain consistent; the purpose of normalization is to transform different physical quantities into a unified proportional range so that subsequent models can perform comprehensive analysis.

[0099] After generating normalized parameters, the latency index will be... Packet loss rate indicator Bandwidth indicators and load factor metrics The data are combined to form link status feature information; this status feature information can be stored in structured data form and continuously updated with the detection cycle; as the link environment changes, the data in the statistical window is updated synchronously, thereby realizing real-time mapping of the link status;

[0100] The link status feature information formed by periodic detection, window statistics and normalization processing provides a stable input basis for subsequent multi-layer operation prediction and decision-making models. This mechanism ensures that the link status information is both real-time and avoids judgment deviations caused by instantaneous fluctuations, thereby supporting dynamic transmission scheduling and control in multimodal network environments.

[0101] The multi-layer operation prediction module is used to predict the future operation status of each wireless communication link within a preset time window based on equipment-level operation data, production line-level load data, and factory-level network operation data, and generate risk assessment information for each wireless communication link.

[0102] Furthermore, in the multi-layer operation prediction module, equipment-level operation data, production line-level load data, and factory-level network operation data include:

[0103] Real-time sampling of the operating status of industrial equipment, recording the frequency of equipment data generation and the magnitude of data changes, forming equipment-level operating data;

[0104] The number of online devices and the total amount of data transmitted per unit time within the production line area are counted, the regional link load value is calculated, and production line-level load data is generated.

[0105] Monitor the bandwidth occupancy and link utilization of each communication link throughout the plant, calculate the overall network load level, and generate plant-level network operation data.

[0106] Furthermore, in the multi-layer operation prediction module, the prediction of the future operating status of each wireless communication link within a preset time window includes:

[0107] Collect historical link status data within a preset time window;

[0108] Construct a time series-based prediction model to calculate the trends of link latency, packet loss rate, and load changes;

[0109] Based on the prediction results, the future fluctuation range and reliability change trend of the link are calculated, and risk assessment information is generated accordingly.

[0110] Specifically, the multi-layer operation prediction module is deployed on edge gateways or factory-side computing nodes to predict the future operating status of wireless communication links in a multi-modal network environment and generate risk assessment information that can be used for transmission control. The multi-layer operation prediction module uses equipment-level operation data, production line-level load data, and factory-level network operation data as its basic data sources. Through hierarchical modeling, it unifies the behavior of field equipment data, changes in regional business load, and global network resource occupancy, thereby providing engineering-meaning input conditions for subsequent prediction calculations.

[0111] When generating device-level operational data, the operating status of each industrial device is sampled in real time, and the frequency of data generation and the magnitude of data changes are recorded. The device data generation frequency is used to characterize the number of data frames generated by the device per unit time, denoted as [missing information]. The magnitude of data change is used to characterize the intensity of the change in sampled values ​​over time, denoted as . In one possible approach, the frequency of device data generation is calculated by the number of samples within a statistical time window, and the magnitude of data variation is obtained by the difference or root mean square error between the maximum and minimum values ​​within the window; device-level operational data is stored in a structured format. ;in Indicates the frequency of device data generation. This indicates the magnitude of changes in device data; this device-level operational data is used to characterize the strength of future data injection into the network from the device side, thereby providing a basis for link load prediction.

[0112] When generating production line-level load data, the number of online devices within the production line and the total amount of data transmitted per unit time are statistically analyzed; the number of online devices is denoted as... The total amount of data transmitted per unit time is denoted as Further calculate the regional link load value. To characterize the extent to which business operations within the production line consume link resources; in one possible manner, the regional link load value is calculated according to the following relationship:

[0113] ;

[0114] in Indicates the length of the statistical time window. This indicates the total amount of data transferred within the statistics window. This represents the regional data injection intensity per unit time; the production line-level load data is stored in a structured format. ;in Indicates the number of online devices. This represents the regional link load value; production line-level load data is used to characterize the potential impact of regional business peaks or sudden data reporting on wireless links.

[0115] When generating factory-level network operation data, monitor the bandwidth occupancy ratio and link utilization of each communication link across the entire factory, and calculate the global network load level; let the first... The rated bandwidth of each communication link is The bandwidth already used is Then the bandwidth usage ratio Defined as:

[0116] ;

[0117] in Indicates the proportion of link bandwidth utilization; in one possible way, it represents the global network load level. It is obtained by weighted average of the bandwidth occupancy ratios of multiple links:

[0118] ;

[0119] in This indicates the number of communication links included in the statistics. Indicates the first The weight coefficients of each link satisfy the following conditions: This method allows for the quantification of factory-level network resource scarcity into a single indicator; factory-level network operation data is stored in a structured format. Or further include each link Sequence; This factory-level network operation data is used to reflect the global network occupancy trends at different time periods, thereby predicting the congestion risk of wireless links in future windows;

[0120] Complete device-level operational data Production line level load data and factory-level network operation data After data collection, the multi-layer operation prediction module uses a hierarchical weighted fusion method to construct a comprehensive prediction input vector V. pred V pred =[G,α·P,β·F]; where α and β are the hierarchical fusion weight coefficients, determined through training with historical data, used to balance the influence of different levels of data on link status prediction. This fusion vector serves as the input to the subsequent time series prediction model, enabling the prediction results to simultaneously reflect the multiple influences of device behavior, regional load, and global network status.

[0121] After completing the three-layer data acquisition, historical link status data is collected within a preset time window to predict future operational status; historical link status data includes link latency sequences, packet loss rate sequences, and link load sequences; [The text abruptly ends here, likely due to an incomplete sentence or missing information.] Taking a wireless link as an example, the delay sequence is defined as follows: The packet loss rate sequence is The load sequence is ;in The time index is represented; the multi-level running prediction module performs time alignment and missing data compensation on the above sequences to ensure the continuity of the time series input; missing data compensation can be achieved by linear interpolation, forward filling, or moving average filling, etc.

[0122] When constructing a time-series-based prediction model, a recursive time-series prediction structure can be optionally used to calculate the trend of link status changes; in one possible approach, the prediction model is an autoregressive model, utilizing the most recent... The observations at one time point predict the link status at a future time point; taking latency prediction as an example, the future latency prediction value is denoted as... ,satisfy:

[0123] ;

[0124] in Indicates the backtracking order. Represents the autoregressive coefficient; predicted packet loss rate. Compared with load forecast The same calculation method is used; the autoregressive coefficients can be obtained through least squares fitting, which is used to minimize the sum of squared prediction errors in order to obtain the set of coefficients that minimizes the error; in this way, the predicted trends of link latency, packet loss rate and load within the future time window can be obtained;

[0125] When calculating the future fluctuation amplitude and reliability trend of the link based on the prediction results, fluctuation amplitude and reliability indicators are introduced; fluctuation amplitude is used to characterize the intensity of state changes within the prediction window; taking latency as an example, the fluctuation amplitude is denoted as... It can be calculated as the difference between the maximum and minimum predicted values ​​within the prediction window:

[0126] ;

[0127] in and These represent the maximum and minimum values ​​of the latency prediction sequence within the prediction window, respectively; the reliability trend is used to characterize the combined change direction of packet loss rate and load within the prediction window; in one possible approach, a link reliability index is defined. for:

[0128] ;

[0129] in This represents the average value of the predicted packet loss rate within the prediction window. This represents the average value of the load forecast within the forecast window. and The weighting coefficients are and satisfy the following conditions: This reliability metric allows for a unified expression of the impact of packet loss and load on link stability.

[0130] When generating risk assessment information, a risk score is formed by combining volatility and reliability indicators; the risk score is denoted as... In one possible manner, the risk score is calculated according to the following relationship:

[0131] ;

[0132] in This represents the reference value for the maximum time delay fluctuation used for normalization. , , The weighting coefficients are and satisfy the following conditions: , Indicates the overall network load level; risk score A higher risk score indicates a higher likelihood of congestion, jitter, or reliability degradation on the link within the prediction window; risk assessment information may include a risk score. And its corresponding risk level range; the risk level range can be optionally divided into low risk, medium risk and high risk, for use in subsequent transmission control strategy selection;

[0133] Through the aforementioned data formation and time series prediction processes at the equipment, production line, and factory levels, the multi-layer operation prediction module can quantitatively predict the future operational status of wireless links within a preset time window and generate risk assessment information. This risk assessment information is used to characterize the future availability and fluctuation trends of the links, providing a basis for subsequent link selection, resource adjustment, and coding control. This multi-layer operation prediction process can incorporate changes in industrial field service load and global network resource occupancy into the same prediction framework, thereby forming a risk assessment foundation that matches multimodal collaborative transmission.

[0134] The constraint decision module is used to determine the transmission control strategy through a decision model based on state characteristic information and risk assessment information. The decision model is optimized based on historical transmission results and determines the target wireless communication link or link combination and the corresponding transmission control parameters under the condition of meeting the preset service performance constraints.

[0135] Furthermore, in the constraint decision module, the transmission control strategy is determined through the decision model, including:

[0136] The state characteristics information and risk assessment information are used as input states;

[0137] Based on the input status, select the corresponding link selection strategy and coding control strategy from the preset action set as the decision action;

[0138] After the data transmission is completed, the actual transmitted data is obtained, and the model parameters are updated based on the actual transmitted data to form an iteratively optimized decision-making mechanism.

[0139] Furthermore, in the constraint decision module, determining the target wireless communication link or link combination and its corresponding transmission control parameters under the condition of satisfying preset service performance constraints includes:

[0140] Set business performance constraint thresholds, including the maximum allowable latency threshold and the maximum allowable packet loss rate threshold;

[0141] The performance of candidate wireless communication links or link combinations is evaluated, and the predicted delay and predicted packet loss rate are calculated within a preset time window.

[0142] When the predicted latency and predicted packet loss rate meet the service performance constraint thresholds, the corresponding wireless communication link or link combination and transmission control parameters are determined.

[0143] Specifically, the constraint decision module is deployed in the control unit or independent computing node of the edge gateway to generate executable transmission control strategies based on link state feature information and risk assessment information in a multimodal communication environment. The link state feature information comes from the link state perception module, and the risk assessment information comes from the multi-layer operation prediction module. Together, they constitute the decision input state at the current moment. To uniformly express the input state, a state vector is defined. ,in This represents the current decision-making moment; the state vector can be represented as:

[0144] ;

[0145] in Indicates the first Normalized latency metrics for each link This represents the normalized packet loss rate index. This represents the normalized bandwidth metric. This indicates the risk score of the corresponding link. The state vector represents the number of candidate links; it comprehensively reflects the current link operation status and future risk trends.

[0146] The decision model is used to select a transmission control strategy based on a state vector; in one possible approach, the decision model is a decision model built on a reinforcement learning framework; reinforcement learning is an algorithmic structure that continuously optimizes decision strategies through interaction with the environment, and its basic elements include state, action, and reward; in this embodiment, the state is... The set of actions is denoted as Each action corresponds to a combination of link selection and coding control strategies; for example, an action may include single-link transmission, multi-link parallel transmission, or multi-link redundant transmission, as well as the corresponding coding threshold level selection; the number of actions is denoted as... ;

[0147] To quantify the merits of decisions, a reward function is introduced. , used to represent the transmission effect after the current action is executed; in one possible way, the reward function is defined as:

[0148] ;

[0149] in This represents the actual average delay within the current transmission cycle. This represents the actual packet loss rate. and The weighting coefficients are and satisfy the following conditions: The reward function reflects the overall transmission quality level by weighting latency and packet loss rate; the smaller the latency and packet loss rate values, the higher the reward value.

[0150] The decision model achieves iterative optimization by updating parameters; one possible approach is to use a value function-based update mechanism to adjust the model parameters; defining state-action value functions. Indicates the state Next action The expected cumulative return; the parameter update relationship is:

[0151] ;

[0152] in Indicates the learning rate. Indicates the discount factor. This indicates the possible action to be taken in the next state; the update method is based on the idea of ​​temporal difference, and optimizes parameters by comparing the difference between the current estimate and the target value; learning rate. Control the update range, discount factor Control the weight of future returns in current decisions; through continuous iteration and updates, the decision model can gradually optimize its strategy selection capabilities based on historical transmission results;

[0153] When determining the target wireless communication link or link combination and transmission control parameters under the condition of meeting service performance constraints, a constraint screening mechanism is introduced; let the maximum allowable delay threshold be... The maximum allowable packet loss rate threshold is ;Perform performance evaluation on candidate links or link combinations, and denot the predicted latency as . The predicted packet loss rate is denoted as When satisfied When the time comes, the corresponding link or link combination is included in the feasible solution set; in the feasible solution set, the final target link or link combination is selected according to the action priority output by the decision model.

[0154] Transmission control parameters include link allocation ratio, coding threshold level, and synchronization interval level. After determining the target link or link combination, the coding control strategy output by the decision model and the link selection strategy together constitute the final transmission control strategy. If the prediction results indicate that the future risk will increase, the decision model may tend to choose redundant transmission or increase the coding threshold level. If the risk is low and the bandwidth is sufficient, parallel transmission or reduced coding redundancy can be selected.

[0155] Through the synergistic effect of the aforementioned state input construction, reinforcement learning decision update, and service constraint screening mechanism, the constraint decision module can form an iteratively optimized transmission control strategy in a dynamic network environment. This mechanism integrates the current link state, future risk prediction, and service performance requirements into the decision framework, giving link selection and coding control a clear logical basis and an updateable mechanism, thereby supporting the stable operation of industrial wireless signal transmission systems in multimodal network environments.

[0156] The adaptive coding module is used to perform predictive difference coding on the data to be transmitted based on the transmission control parameters, and dynamically adjust the coding threshold and synchronization interval based on risk assessment information, so as to realize data reconstruction and correction through the synchronization mechanism when data loss occurs.

[0157] Furthermore, in the adaptive coding module, the predictive difference coding process for the data to be transmitted based on the transmission control parameters includes:

[0158] Build a data change prediction model based on historical data collection;

[0159] The predicted value is obtained by using a prediction model to predict the data to be sent.

[0160] Calculate the difference between the current data to be sent and the predicted value;

[0161] The difference is encoded and used as the actual data to be sent.

[0162] Furthermore, in the adaptive coding module, the coding threshold and synchronization interval are dynamically adjusted based on risk assessment information, including:

[0163] Calculate the risk level based on the link risk assessment information;

[0164] When the risk level increases, the trigger threshold for prediction difference coding is increased and the synchronization interval is shortened;

[0165] When the risk level decreases, the trigger threshold for prediction difference coding is lowered and the synchronization interval is extended;

[0166] Synchronization data is periodically sent to correct for data reconstruction errors at the receiving end.

[0167] Specifically, the adaptive coding module is deployed in the data processing unit of the edge gateway. It is used to perform predictive differential coding on the data to be transmitted under conditions of limited link resources or changing risks, and dynamically adjusts the coding threshold and synchronization interval based on the risk assessment results, thereby achieving a balance between bandwidth utilization and data reliability. The data to be transmitted can be continuous sampled values ​​or periodically reported data from the device's operating data stream, denoted as a time series. ,in Indicates a discrete-time index;

[0168] When constructing a data change prediction model, a time series prediction structure is established based on historically collected data; in one possible approach, an autoregressive prediction model is used to estimate the data change trend; the autoregressive model utilizes previous data... The prediction expression for using historical data from a given point in time to predict current data is as follows:

[0169] ;

[0170] in Indicates time The predicted value of the data, Represents the autoregressive coefficient. This indicates the order of regression; the autoregressive coefficients can be solved using the least squares method, which determines the parameter values ​​by minimizing the sum of squared prediction errors. The sum of squared errors is defined as:

[0171] ;

[0172] Through the The optimal parameters are obtained by taking the partial derivatives and setting them to zero; this predictive model can estimate the future trend of data changes.

[0173] After obtaining the predicted value, calculate the difference between the current data to be sent and the predicted value, and denote it as:

[0174] ;

[0175] in This represents the prediction error or difference; a small difference indicates that the current data trend is consistent with the prediction result; a large difference indicates a sudden change or abnormal fluctuation in the data; when encoding the difference, fixed-point quantization or variable-length encoding can be used; for example, the difference can be encoded according to the quantization step size. Quantization is performed to obtain the quantized value:

[0176] ;

[0177] in This represents the result of the quantized difference encoding. This represents the rounding function. This indicates the quantization step size; the encoded difference data is used as the actual transmitted data, replacing the complete original data, thereby reducing the amount of data transmitted.

[0178] When making dynamic adjustments based on risk assessment information, the risk level is first calculated according to the link risk assessment results; the risk score is recorded as follows: Risks are classified into low, medium, and high risk levels by setting a preset risk classification threshold; the risk level function is defined as follows:

[0179] ;

[0180] in and This is the risk classification threshold. This indicates the risk level; when the risk level increases, to reduce the impact of data loss on reconstruction accuracy, the prediction difference coding trigger threshold is increased; the coding trigger threshold is denoted as... Under the high-risk level, it is set as Set as a low-risk level ,satisfy By increasing the trigger threshold, more data is sent in its complete or synchronous form, thereby enhancing reconstruction reliability.

[0181] The synchronization interval is used to control the time interval for periodically sending complete data frames, denoted as . When the risk level increases, the synchronization interval will be adjusted to a shorter period. When the risk level decreases, the synchronization interval will be adjusted to a longer period. ,satisfy Synchronization data frames contain complete raw data values, which are used to correct the prediction model at the receiving end.

[0182] When reconstructing data at the receiving end, the reconstructed value is calculated based on the received difference data and the local prediction model, and denoted as:

[0183] ;

[0184] in This represents the reconstructed data value. When data loss occurs, the receiving end can use a prediction model to estimate the lost data and then correct it after the next synchronized data frame arrives. The periodic synchronization mechanism can limit the accumulation of prediction errors and ensure the stability of data reconstruction.

[0185] Through the synergistic effect of the aforementioned predictive differential coding process and risk-driven threshold adjustment mechanism, the adaptive coding module can dynamically adjust the data transmission granularity according to the link risk status. When the link risk is low, differential coding is given priority to reduce the transmission load, and when the link risk increases, the proportion of complete data transmission is increased to enhance reconstruction reliability. This coding control logic forms a closed-loop structure with the aforementioned link risk assessment and decision-making model, enabling the industrial wireless signal transmission system to achieve adaptive data compression and reliable transmission coordinated control in a multimodal network environment.

[0186] The transmission execution module is used to perform data transmission according to the target wireless communication link or link combination and transmission control parameters, and to perform link switching control and wireless resource adjustment when the risk assessment information exceeds a preset threshold.

[0187] Furthermore, in the transmission execution module, when the risk assessment information exceeds a preset threshold, link switching control and radio resource adjustment are performed, including:

[0188] The performance of the current target wireless communication link is retested and confirmed;

[0189] Select links from the candidate wireless communication link set that meet the service performance constraints in terms of predicted latency and predicted packet loss rate as the handover target links;

[0190] Before performing link switching, establish a communication connection to the target link and perform short-term parallel transmission;

[0191] After the link switch is completed, the allocation ratio of wireless communication resources is adjusted according to the link load.

[0192] Specifically, the transmission execution module is deployed in the communication control unit of the edge gateway. After the target wireless communication link or link combination is determined, it combines transmission control parameters to complete data transmission, acknowledgment, and retransmission control, and performs link switching control and wireless resource adjustment when the link risk reaches the trigger condition. The target wireless communication link or link combination is given by the constraint decision module. The transmission control parameters may include link selection parameters, link allocation ratio parameters, parallel or redundant transmission mode parameters, and resource allocation related parameters. In each transmission cycle, the transmission execution module selects the transmission path according to the transmission control parameters and records the status of the transmission process so that it can quickly switch to the alternative path when the link risk changes.

[0193] During the normal transmission phase, the transmission execution module performs data transmission according to the target wireless communication link or link combination. In single-link mode, data frames enter the target link's transmission queue sequentially and are acknowledged upon transmission. In link combination mode, the transmission execution module splits or duplicates the data stream according to the link allocation ratio; splitting is used for parallel transmission, and duplication is used for redundant transmission. In one possible approach, let the amount of data to be transmitted per unit transmission cycle be... , for the The target link is set with an allocation ratio of 100%. The amount of data sent to this link is:

[0194] ;

[0195] in Indicates the first The amount of data transmitted per link, Indicates the link allocation ratio and satisfies The transmission execution module determines the transmission path based on the data of each link. Complete scheduling and transmission;

[0196] During the risk trigger determination phase, the transmission execution module receives link risk assessment information generated by the multi-layer operation prediction module and compares the risk score with a preset risk threshold; let the risk score of the target link be... The preset risk threshold is When the following conditions are met:

[0197] ;

[0198] The switching control process is initiated at this time; among which... This indicates the risk level of the current target link within the prediction time window. This indicates the threshold for triggering link switching; this determination is used to map future risks in advance into executable link scheduling behaviors.

[0199] Before performing a link switch, the performance of the current target wireless communication link is first re-tested and confirmed. This re-testing and confirmation is used to eliminate false triggers caused by short-term statistical errors or transient jitter. The testing process can obtain the instantaneous latency and instantaneous packet loss rate by sending probe data packets and statistically analyzing the response results. The instantaneous latency is denoted as... Instantaneous packet loss rate is denoted as When the test results still show that the link performance is degraded, the candidate link screening process begins.

[0200] When selecting a target link for switching from the set of candidate wireless communication links, the transmission execution module performs constraint filtering on the candidate links based on predicted latency and predicted packet loss rate; let the first... The prediction delay of each candidate link within the prediction time window is The predicted packet loss rate is The business performance constraint threshold is the maximum allowable latency threshold. With the maximum allowable packet loss rate threshold When satisfied When multiple switchable links exist, the transmission execution module can determine the priority by combining the candidate link risk score, available bandwidth, or load level to select the final target link for switching.

[0201] During the link switching process, the transmission execution module establishes a communication connection to the target link before the switch and performs short-term parallel transmission; the short-term parallel transmission is used to ensure data continuity during the switch; the duration of the parallel transmission is denoted as... This indicates the duration of data transmission on both the original link and the target link simultaneously. During the parallel phase, critical business data can be transmitted to both links simultaneously using a replication method, and the target link is judged to have stable transmission capability by receiving confirmation results. When the target link continuously meets the preset confirmation conditions, the transmission on the original link is terminated and the switchover is completed.

[0202] After link handover is completed, the wireless communication resource allocation ratio is adjusted according to the link load. The resource allocation ratio describes the transmission share or wireless side resource occupancy ratio of each link in the link combination mode. Let the handover result be the first... The load rate of the target link The load rate can be obtained by the ratio of the bandwidth occupied by the link to the rated bandwidth of the link; in one possible approach, the resource allocation ratio is adjusted in reverse based on the load rate, and a new allocation ratio is defined as follows:

[0203] ;

[0204] in This indicates the adjusted allocation ratio. This represents the set of link numbers participating in the allocation; this method allows for a reduction in the proportion of links that bear the load when the link load increases, and an increase in the proportion of links that bear the load when the link load decreases; when the wireless standard supports resource unit scheduling, the resource allocation ratio can also be mapped to bandwidth resource block allocation, time slot occupancy ratio, or subcarrier occupancy ratio; this mapping method can be implemented in conjunction with a specific wireless standard and is an optional implementation method.

[0205] Through a collaborative process involving risk threshold triggering, re-detection and confirmation, candidate link constraint screening, short-term parallel transmission before handover, and resource ratio adjustment after handover, the transmission execution module can achieve controllable link handover and adaptive resource scheduling when the link risk increases in the future. This process forms a closed-loop control relationship with the aforementioned link status perception, multi-layer operation prediction, and constraint decision-making, enabling industrial wireless signal transmission to have an implementable handover and scheduling mechanism under multi-modal network conditions.

[0206] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An industrial wireless signal transmission system, characterized by Includes the following modules: The signal acquisition and access module is used to acquire operating signals of industrial equipment and access the network through wired communication links and various wireless communication links; The link status awareness module is used to acquire the operating status parameters of each wireless communication link in real time and form the status feature information of the corresponding link. The multi-layer operation prediction module is used to predict the future operation status of each wireless communication link within a preset time window based on equipment-level operation data, production line-level load data, and factory-level network operation data, and generate risk assessment information for each wireless communication link. The constraint decision module is used to determine the transmission control strategy through a decision model based on state characteristic information and risk assessment information. The decision model is optimized based on historical transmission results and determines the target wireless communication link or link combination and the corresponding transmission control parameters under the condition of meeting preset service performance constraints. The adaptive coding module is used to perform predictive difference coding on the data to be transmitted based on the transmission control parameters, and dynamically adjust the coding threshold and synchronization interval based on risk assessment information, so as to realize data reconstruction and correction through the synchronization mechanism when data loss occurs. The transmission execution module is used to execute data transmission according to the target wireless communication link or link combination and transmission control parameters, and to perform link switching control and wireless resource adjustment when the risk assessment information exceeds a preset threshold.

2. An industrial wireless signal transmission system according to claim 1, wherein, In the signal acquisition and access module, the network access via wired communication links and multiple wireless communication links includes: Collect operating signals from industrial equipment and establish a wired data channel via industrial Ethernet; Perform data encapsulation processing on the operating signals; At the same time, at least two different wireless communication links are established, including at least one of 4G and 5G links; According to the preset transmission strategy, the encapsulated operating signal is distributed to the wired data channel and wireless communication link for parallel or redundant transmission.

3. An industrial wireless signal transmission system according to claim 1, wherein, In the link status awareness module, the real-time acquisition of the operating status parameters of each wireless communication link includes: Perform periodic link quality checks on each wireless communication link; Collect latency, packet loss rate, available bandwidth, and link load parameters for each wireless communication link; The collected operating status parameters are statistically processed over a time window and then normalized. The processing results are used as the status feature information of the corresponding link.

4. The industrial wireless signal transmission system according to claim 1, characterized in that, In the multi-layer operation prediction module, the equipment-level operation data, production line-level load data, and factory-level network operation data include: Real-time sampling of the operating status of industrial equipment, recording the frequency of equipment data generation and the magnitude of data changes, forming equipment-level operating data; The number of online devices and the total amount of data transmitted per unit time within the production line area are counted, the regional link load value is calculated, and production line-level load data is generated. Monitor the bandwidth occupancy and link utilization of each communication link throughout the plant, calculate the overall network load level, and generate plant-level network operation data.

5. An industrial wireless signal transmission system according to claim 1, characterized in that, In the multi-layer operation prediction module, the prediction of the future operating status of each wireless communication link within a preset time window includes: Collect historical link status data within a preset time window; Construct a time series-based prediction model to calculate the trends of link latency, packet loss rate, and load changes; The future fluctuation range and reliability change trend of the link are calculated based on the prediction results, and the risk assessment information is generated accordingly.

6. An industrial wireless signal transmission system according to claim 1, characterized in that, In the constraint decision module, determining the transmission control strategy through the decision model includes: The state characteristics information and risk assessment information are used as input states; Based on the input status, select the corresponding link selection strategy and coding control strategy from the preset action set as the decision action; After the data transmission is completed, the actual transmitted data is obtained, and the model parameters are updated based on the actual transmitted data to form an iteratively optimized decision-making mechanism.

7. An industrial wireless signal transmission system according to claim 1, characterized in that, In the constraint decision module, determining the target wireless communication link or link combination and its corresponding transmission control parameters under the condition of satisfying preset service performance constraints includes: Set business performance constraint thresholds, including the maximum allowable latency threshold and the maximum allowable packet loss rate threshold; The performance of candidate wireless communication links or link combinations is evaluated, and the predicted delay and predicted packet loss rate are calculated within a preset time window. When the predicted latency and predicted packet loss rate meet the service performance constraint thresholds, the corresponding wireless communication link or link combination and transmission control parameters are determined.

8. An industrial wireless signal transmission system according to claim 1, characterized in that, In the adaptive coding module, the predictive difference coding process for the data to be transmitted based on the transmission control parameters includes: Build a data change prediction model based on historical data collection; The predicted value is obtained by using a prediction model to predict the data to be sent. Calculate the difference between the current data to be sent and the predicted value; The difference is encoded and used as the actual data to be sent.

9. An industrial wireless signal transmission system according to claim 1, characterized in that, In the adaptive coding module, the dynamic adjustment of the coding threshold and synchronization interval based on risk assessment information includes: Calculate the risk level based on the link risk assessment information; When the risk level increases, the trigger threshold for prediction difference coding is increased and the synchronization interval is shortened; When the risk level decreases, the trigger threshold for prediction difference coding is lowered and the synchronization interval is extended; Synchronization data is periodically sent to correct for data reconstruction errors at the receiving end.

10. An industrial wireless signal transmission system according to any one of claims 1 to 9, characterized in that, In the transmission execution module, the step of performing link switching control and radio resource adjustment when the risk assessment information exceeds a preset threshold includes: The performance of the current target wireless communication link is retested and confirmed; Select links from the candidate wireless communication link set that meet the service performance constraints in terms of predicted latency and predicted packet loss rate as the handover target links; Before performing link switching, establish a communication connection to the target link and perform short-term parallel transmission; After the link switch is completed, the allocation ratio of wireless communication resources is adjusted according to the link load.

Images