A control signal acquisition and transmission system for a host

Abstract

The present invention provides a control signal acquisition and transmission system for a host, which relates to the field of signal acquisition and transmission systems. Real-time monitoring and data processing of multi-dimensional environmental variables are achieved through intelligent sensor modules, high-precision sensors and embedded microprocessors. The data fusion module uses intelligent threshold adjustment and timing prediction models to ensure data integrity, and the edge computing module performs anomaly detection and transmission path optimization through a deep learning algorithm. The adaptive transmission module dynamically selects the best transmission path. Using a machine learning-driven transmission path optimization module, the system can select the best transmission channel in real time, reduce latency and improve data transmission efficiency. This dynamic adjustment capability can maintain efficient data transmission when network conditions change, ensuring the timely arrival of important information. The intelligent acquisition logic module dynamically adjusts the signal acquisition frequency through environmental scene adaptive recognition and predictive data acquisition, and can adapt to changes under different working environments.

Description

Technical Field

[0001] The invention relates to the field of signal acquisition and transmission systems, in particular to a control signal acquisition and transmission system for a host. Background Art

[0002] According to a measurement system and a measurement method for synchronous data collection and comparison at different locations with Chinese patent publication number CN109828144A, the measurement system includes a wireless response timing module, a collection module, a wireless transmission module, and a data processing module. Step 1: the wireless response timing module measures the time of sending and receiving communication response signals, and sets a delay timer according to this time; Step 2: after receiving the collection instruction, the collection module first performs a zero detection on the power frequency of the distribution system detection point; Step 3: the wireless transmission module of the remote slave encodes the collected information in sequence and transmits it to the local host through wireless transmission; Step 4: the local data processing module also encodes the data collected by itself in sequence according to the collection instruction, and performs a matching comparison check on the received synchronous data of the remote slave.

[0003] According to the Chinese patent publication number CN110246317B, the intelligent electricity safety detection device based on EnOcean communication technology and its working method include an intelligent electricity safety detector and a plurality of intelligent sockets; the intelligent electricity safety detector includes a first signal processing module, and a first signal acquisition module, a remote control switch module, a first EnOcean wireless communication module, and a 485 communication module respectively connected to the first signal processing module; the intelligent socket includes a second signal processing module, and a second signal acquisition module and a second EnOcean wireless communication module respectively connected to the second signal processing module.

[0004] The above patents and prior art have the following problems:

[0005] Signal distortion and data inconsistency are common in traditional systems, as well as data loss or delay in complex network environments. Summary of the invention

[0006] 1. Technical issues to be resolved

[0007] In view of the deficiencies in the prior art, the present invention provides a control signal acquisition and transmission system for a host, which solves the problems of signal distortion and data inconsistency commonly found in traditional systems, as well as data loss or delay in a complex network environment.

[0008] (II) Technical solution

[0009] To achieve the above objectives, the present invention is implemented through the following technical solutions: a control signal acquisition and transmission system for a host, the signal acquisition and transmission system comprising:

[0010] An intelligent sensor module, the intelligent sensor module comprising:

[0011] High-precision sensors monitor and collect multi-dimensional environmental variables in real time;

[0012] Embedded microprocessor to execute deep learning-based signal processing and data acquisition algorithms for adaptive regulation;

[0013] Adaptive filter, used to dynamically optimize signal processing parameters and remove noise and outliers in real time;

[0014] Self-diagnostic module for continuous monitoring of sensor status and feedback.

[0015] The data fusion module is used to receive and process data from multiple sensors, including the following parts:

[0016] Intelligent threshold adjustment algorithm automatically optimizes data collection thresholds based on real-time data to reduce interference from invalid data;

[0017] The signal reconstruction algorithm based on the time series prediction model uses historical data to build a mathematical model to infer the signal of the unmonitored area to ensure the continuity and integrity of the data;

[0018] The edge computing module is electrically connected to the smart sensor module and includes the following parts:

[0019] Real-time data analysis and anomaly detection module, with built-in deep learning algorithm to quickly identify abnormal trends and issue early warnings;

[0020] The machine learning-driven transmission path optimization module selects the best transmission channel in real time based on network conditions and data traffic, reducing latency and improving reliability.

[0021] The adaptive transmission module comprises the following steps:

[0022] sp1. Dynamic routing selection: Real-time monitoring of network status, built-in intelligent algorithm to select the best transmission path, and dynamic adjustment to cope with network changes;

[0023] sp2. Priority transmission strategy: hierarchical processing is performed according to the importance of data. Critical data is sent first, and non-critical data is transmitted when the network is idle, ensuring that important information arrives in time;

[0024] sp3. Error correction and data redundancy mechanism: realize data packet redundancy and self-error correction, so that data lost during transmission can be automatically recovered, improving the reliability of data transmission.

[0025] An intelligent acquisition logic module, the intelligent acquisition logic module comprising:

[0026] Environmental scene adaptive recognition unit: built-in deep learning algorithm analyzes the working environment in real time, dynamically adjusts the signal acquisition frequency, and realizes adaptive sampling;

[0027] Predictive data collection unit: Built-in machine learning algorithm predicts future signal trends and collects data based on the results of predicting future signal trends;

[0028] Distributed information fusion unit: used to share information between sensor nodes in real time.

[0029] Preferably, the high-precision sensor of the intelligent sensor module has multiple sensing functions, specifically including:

[0030] Temperature sensor with a measurement range of -40°C to +125°C and a resolution of 0.1°C;

[0031] Humidity sensor with a measurement range of 0% to 100% relative humidity and an accuracy of ±2%;

[0032] Gas concentration sensor, including CO2 and NH3 detection, with a range of 0-5000ppm;

[0033] Light sensor, measuring light intensity from 0 to 100,000lx.

[0034] Preferably, the embedded microprocessor further includes an FPGA architecture to implement parallel processing and signal filtering, and the filtering algorithm is an adaptive filter whose mathematical model is as follows:

[0035]

[0036] is the output signal, is the input signal, is the step size factor, and , is the error signal at the previous moment.

[0037] The dynamic routing selection of the adaptive transmission protocol is specifically: using a weighted average algorithm to select the best transmission path:

[0038]

[0039] B is bandwidth; L is delay; S is signal strength; It is a weight coefficient, which is dynamically adjusted based on historical data to optimize the transmission effect.

[0040] Preferably, the convolutional neural network of the data fusion module includes the following convolution operation:

[0041]

[0042] Among them, X is the input data, K is the convolution kernel, Y is the output feature map, which improves the accuracy of data fusion, and i, j are the coordinates of the output feature map; , is the coordinate of the input data; the Softmax classification of the output layer is:

[0043]

[0044] For input Belongs to category The probability of is the model output, and K is the number of categories.

[0045] Preferably, the real-time data analysis and anomaly detection module further includes a fault prediction module, and the objective function of the built-in support vector machine module in the fault prediction module is:

[0046]

[0047] is the model complexity, C is the penalty factor, is the slack variable, N is The value range of is the number of data collection times of the signal acquisition and transmission system to ensure the accuracy of the prediction.

[0048] Preferably, the scene recognition in the intelligent acquisition logic adopts transfer learning, and the model is:

[0049]

[0050] Dynamically optimize sampling strategies to compensate for differences between source and target domains, Model output for the target domain; is the model output of the source domain.

[0051] Preferably, the error correction and data redundancy mechanism is based on blockchain technology, and the hash function generates data blocks:

[0052]

[0053] To generate data Hash value to ensure data cannot be tampered with; The nth layer of hash functions processes the input data in sequence to ensure that the data cannot be tampered with; the redundant coding algorithm is:

[0054]

[0055] is the redundant encoding function, For the collected data, It is a weighting coefficient with a value of 0-1 to ensure data integrity and recovery capability.

[0056] Preferably, the control signal acquisition and transmission system for a host, the operation process of the signal acquisition and transmission system is as follows:

[0057] SP1: Environmental Monitoring and Data Collection:

[0058] Sp1.1: Smart sensor modules monitor environmental variables in real time, including temperature, humidity, gas concentration, and light intensity;

[0059] Sp1.2: The sensor performs preliminary processing on the collected data through an embedded microprocessor and executes an adaptive filtering algorithm to filter out noise and outliers;

[0060] SP2: Data Fusion and Analysis:

[0061] Sp2.1: The data fusion module receives data from each intelligent sensor module;

[0062] Sp2.2: Use the convolutional neural network algorithm to extract features from the received data and generate an output feature map;

[0063] Sp2.3: Based on the extracted features, the probability of the data belonging to each category is calculated through the Softmax classification algorithm;

[0064] SP3: Forecasting and early warning:

[0065] Sp3.1: Based on historical data and real-time data, use the support vector machine (SVM) algorithm to predict faults and determine whether there are abnormal conditions;

[0066] Sp3.2: If an anomaly is predicted, the edge computing gateway uses machine learning algorithms to quickly identify abnormal trends and issue warning signals;

[0067] SP4: Adaptive Transport:

[0068] Sp4.1: The edge computing gateway monitors network status, including latency, bandwidth, and signal strength;

[0069] Sp4.2: Use dynamic routing selection algorithm to select the best transmission path and adjust the transmission channel according to the real-time feedback mechanism;

[0070] SP5: Data Transfer:

[0071] Sp5.1: Send data packets to the receiving end according to priority, with key data being transmitted first;

[0072] Sp5.2: During the transmission process, error correction and data redundancy mechanism is used to perform data packet redundancy and error correction to ensure data integrity;

[0073] SP6: Feedback and Optimization:

[0074] Sp6.1: The receiving end confirms the data reception and feeds back the result to the edge computing gateway;

[0075] Sp6.2: If transmission delay or data loss is detected, the system will re-evaluate the transmission path and priority;

[0076] SP7: Intelligent collection and information sharing:

[0077] Sp7.1: Share data between sensor nodes to improve overall data quality;

[0078] Sp7.2: When the signal of a certain node is lost, the data of the neighboring nodes is used for compensation to ensure the integrity of the information.

[0079] Preferably, the edge computing gateway further includes a real-time data processing module, and the workflow of the real-time data processing module is as follows:

[0080] Sp1: Receive data streams from multiple smart sensor nodes in real time and use data preprocessing algorithms to denoise and normalize the data;

[0081] Sp2: Through the integrated learning algorithm, the prediction results of multiple models are integrated to improve the accuracy of data analysis;

[0082] Sp3: Automatically adjust the sampling frequency and data transmission strategy of the sensor based on the data analysis results to achieve rapid response to environmental changes;

[0083] Sp4: When an abnormal situation is detected, an alarm is triggered in time, and relevant data is uploaded to the cloud for in-depth analysis to optimize system performance later.

[0084] (III) Beneficial effects

[0085] The present invention provides a control signal acquisition and transmission system for a host, which has the following beneficial effects:

[0086] The present invention adopts high-precision sensors combined with adaptive filters, which can monitor and collect multi-dimensional environmental variables in real time, effectively remove noise and outliers, ensure the reliability and accuracy of data, and adopt the continuous monitoring function of the self-diagnosis module to timely feedback the sensor status, reduce maintenance requirements, and improve the overall reliability of the system.

[0087] The present invention adopts a machine learning-driven transmission path optimization module, and the system can select the best transmission channel in real time, reduce latency and improve the efficiency of data transmission. This dynamic adjustment capability can maintain efficient data transmission when network conditions change, ensuring the timely arrival of important information. The intelligent acquisition logic module dynamically adjusts the signal acquisition frequency through environmental scene adaptive recognition and predictive data acquisition, and can adapt to changes in different working environments. This allows the system to effectively prevent the omission of important information when facing different signal strengths and environmental conditions, and improves the comprehensiveness and efficiency of data acquisition. BRIEF DESCRIPTION OF THE DRAWINGS

[0088] Figure 1 It is a system diagram of the present invention;

[0089] Figure 2 These are experimental diagrams of the present invention, which from top to bottom are the original signal waveform, the signal waveform after adaptive aluminum foil, and the effect index bar graph. DETAILED DESCRIPTION

[0090] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. Specific embodiment (I)

[0092] like Figure 1-Figure 2 As shown, a control signal acquisition and transmission system for a host, the signal acquisition and transmission system comprises:

[0093] Smart sensor module, the smart sensor module includes:

[0094] High-precision sensors monitor and collect multi-dimensional environmental variables in real time;

[0095] Embedded microprocessor to execute deep learning-based signal processing and data acquisition algorithms for adaptive regulation;

[0096] Adaptive filter, used to dynamically optimize signal processing parameters and remove noise and outliers in real time;

[0097] Self-diagnostic module for continuous monitoring of sensor status and feedback, reducing maintenance requirements and improving system reliability.

[0098] The data fusion module is used to receive and process data from multiple sensors, including the following parts:

[0099] Intelligent threshold adjustment algorithm automatically optimizes data collection thresholds based on real-time data to reduce interference from invalid data; Intelligent threshold adjustment algorithm adjustment model: dynamically updates collection thresholds based on the mean and standard deviation of real-time data:

[0100]

[0101] is the new acquisition threshold; is the mean of real-time data; is the standard deviation of the real-time data; is the adjustment parameter that controls the sensitivity of the threshold;

[0102] Collect real-time data and calculate the mean μ and standard deviation σ, dynamically adjust the collection interval value to ensure that important data can always be effectively collected in a changing environment, apply new thresholds, and improve the effectiveness of data collection.

[0103] The signal reconstruction algorithm based on the time series prediction model uses historical data to build a mathematical model to infer the signal in the unmonitored area to ensure the continuity and integrity of the data; the signal reconstruction algorithm uses the ARIMA model:

[0104]

[0105] For time series data, are model parameters, is white noise.

[0106] ARIMA Model:

[0107] Autoregressive part (AR): Use linear combinations of historical data to predict future data.

[0108] Difference (I): Makes a non-stationary time series stationary.

[0109] Moving average (MA): uses a linear combination of historical errors to make predictions.

[0110] Data preparation: Collect historical signal data, perform data cleaning and preprocessing, and perform difference processing on the data to ensure stationarity (for ARIMA).

[0111] Model selection and training: Select ARIMA and set parameters (such as order, number of hidden layers, etc.). Use the training set data to train the model and tune the hyperparameters to improve the prediction performance.

[0112] Signal reconstruction: predict signals in unmonitored areas and use the trained model to generate signal sequences. The reconstructed signals are used to complete the original data to ensure data continuity and integrity.

[0113] Evaluation and optimization: Use indicators such as mean square error (MSE) and mean absolute error (MAE) to evaluate the prediction effect. Adjust the model parameters and structure based on the evaluation results to improve the reconstruction effect.

[0114] The edge computing module is electrically connected to the intelligent sensor module and includes the following parts: a real-time data analysis and anomaly detection module with a built-in deep learning algorithm to quickly identify abnormal trends and issue warnings;

[0115] The transmission path optimization module driven by machine learning selects the best transmission channel in real time based on network conditions and data traffic, reducing latency and improving reliability. The specific algorithm content of machine learning is as follows: Support vector machine (SVM) is used for fault prediction, collecting historical data, and marking normal and abnormal states. The SVM model is trained using the training data set, and the optimal hyperplane is found by optimizing the objective function. New data is classified to determine whether there are abnormal conditions, and an early warning is triggered based on the prediction results. Convolutional neural network is used for data fusion convolution operation. The input data is extracted through the convolution layer to obtain a feature map. The feature map calculates the probability of each category through the fully connected layer and the Softmax classifier. The classification of the data is determined based on the probability results and provided for subsequent decision-making and processing.

[0116] The adaptive transmission module comprises the following steps:

[0117] sp1. Dynamic routing selection: Real-time monitoring of network status, built-in intelligent algorithm to select the best transmission path, dynamic adjustment to cope with network changes; using routing optimization algorithm based on genetic algorithm:

[0118] Fitness function:

[0119]

[0120] is the transmission delay of the path; is the packet loss rate of the path; It is an adjustment parameter to balance the delay and packet loss rate. Explanation-Genetic algorithm is an optimization algorithm based on natural selection, which gradually approaches the optimal solution through selection, crossover and mutation. The fitness function is used to evaluate the quality of the path. The goal is to maximize the fitness and select the optimal transmission path. The usage process is as follows:

[0121] 1. Initialization: Generate an initial population to represent multiple transmission paths;

[0122] 2. Evaluate fitness: Calculate the fitness of each path and evaluate its performance based on delay and packet loss rate;

[0123] 3. Selection operation: select excellent paths for crossover and mutation according to fitness to generate new path combinations;

[0124] 4. Iteration: Repeat the evaluation and selection process until the termination condition is met (reaching the maximum number of iterations or fitness threshold);

[0125] 5. Select the best path: Return the path with the highest fitness as the best transmission path.

[0126] Mathematical model of the shortest path algorithm based on Dijkstra algorithm, path length calculation:

[0127]

[0128] is the shortest path length to node v, for The shortest path length when yes( ), Dijkstra algorithm is used to find the shortest path from the starting point to each node. It is applicable to weighted graphs and gradually expands the shortest path through a greedy strategy to ensure that the global optimal solution is found.

[0129] The usage process is as follows:

[0130] Initialization: Set the distance from the starting point to itself to 0, and other nodes to infinity;

[0131] Priority queue: Use the priority queue to maintain the minimum distance of the current node;

[0132] Traverse nodes: select the node with the current minimum distance and update the distance of its adjacent nodes;

[0133] Repeat: until all nodes are processed, record the shortest path to each node;

[0134] Build Path: Backtrack from the end point to build the shortest path.

[0135] sp2. Priority transmission strategy: hierarchical processing is performed according to the importance of data. Critical data is sent first, and non-critical data is transmitted when the network is idle, ensuring that important information arrives in time;

[0136] sp3. Error correction and data redundancy mechanism: realize data packet redundancy and self-error correction, so that data lost during transmission can be automatically recovered, improving the reliability of data transmission.

[0137] Intelligent acquisition logic module, the intelligent acquisition logic module includes:

[0138] Environmental scene adaptive recognition unit: built-in deep learning algorithm analyzes the working environment in real time, dynamically adjusts the signal acquisition frequency, and realizes adaptive sampling;

[0139] Predictive data collection unit: Built-in machine learning algorithm predicts future signal trends and collects data in advance based on the prediction results to prevent important information from being missed;

[0140] Distributed information fusion unit: used to share information between sensor nodes in real time, enhance the overall data quality through the data of neighboring nodes, and ensure information compensation when the signal is lost.

[0141] The high-precision sensors of the smart sensor module have multiple sensing functions, including:

[0142] Temperature sensor with a measurement range of -40°C to +125°C and a resolution of 0.1°C;

[0143] Humidity sensor with a measurement range of 0% to 100% relative humidity and an accuracy of ±2%;

[0144] Gas concentration sensor, including CO2 and NH3 detection, with a range of 0-5000ppm;

[0145] Light sensor, measuring light intensity from 0 to 100,000lx.

[0146] The embedded microprocessor also includes an FPGA architecture to implement parallel processing and signal filtering. The filtering algorithm is an adaptive filter with the following mathematical model:

[0147]

[0148] is the output signal, is the input signal, is the step size factor, and , is the error signal at the previous moment.

[0149] The dynamic routing selection of the adaptive transmission protocol is as follows: Use the weighted average algorithm to select the best transmission path:

[0150]

[0151] B is bandwidth; L is delay; S is signal strength; It is a weight coefficient, which is dynamically adjusted based on historical data to optimize the transmission effect.

[0152] The convolutional neural network of the data fusion module includes the following convolution operations:

[0153]

[0154] Among them, X is the input data, K is the convolution kernel, Y is the output feature map, which improves the accuracy of data fusion, and i, j are the coordinates of the output feature map; , is the coordinate of the input data; the Softmax classification of the output layer is:

[0155]

[0156] For input Belongs to category The probability of is the model output, and K is the number of categories. The real-time data analysis and anomaly detection module also includes a fault prediction module. The objective function of the built-in support vector machine module in the fault prediction module is:

[0157]

[0158] is the model complexity, C is the penalty factor, is a slack variable. The actual value range of the slack variable is between -1 and 1. A positive number represents a positive trend, and a negative number represents a non-optimal trend. N is The value range of is the number of data collection times of the signal acquisition and transmission system to ensure the accuracy of the prediction.

[0159] The scene recognition in the intelligent collection logic uses transfer learning, and the model is:

[0160]

[0161] Dynamically optimize sampling strategies to compensate for differences between source and target domains, Model output for the target domain; is the model output of the source domain.

[0162] The error correction and data redundancy mechanism is based on blockchain technology, and the hash function generates data blocks:

[0163]

[0164] To generate data Hash value to ensure data cannot be tampered with; The nth layer of hash functions processes the input data in sequence to ensure that the data cannot be tampered with; the redundant coding algorithm is:

[0165]

[0166] is the redundant encoding function, For the collected data, It is a weighting coefficient with a value of 0-1 to ensure data integrity and recovery capability.

[0167] The control signal acquisition and transmission system aims to achieve real-time monitoring and data transmission of multi-dimensional environmental variables through efficient intelligent sensors and data processing modules. The system consists of multiple modules, including intelligent sensor module, data fusion module, edge computing module, adaptive transmission module and intelligent acquisition logic module, which work together through clear data input and output interfaces to ensure high reliability and real-time performance. First, the intelligent sensor module monitors environmental variables in real time through high-precision sensors (such as temperature, humidity, gas concentration and light sensors), and the embedded microprocessor executes signal processing and data acquisition algorithms based on deep learning to ensure data quality; the adaptive filter is responsible for removing noise and outliers, and the self-diagnosis module continuously monitors the sensor status to improve system reliability. The processed multi-dimensional data stream is transmitted to the data fusion module, where the data is fused and optimized through the intelligent threshold adjustment algorithm and the signal reconstruction algorithm based on the time series prediction model to fill the signal in the unmonitored area and ensure the continuity and integrity of the data. The edge computing module analyzes the information from the data fusion module, identifies abnormal trends through the real-time data analysis and anomaly detection module, and uses the machine learning-driven transmission path optimization module to select the best data transmission path to reduce latency and improve reliability. The adaptive transmission module implements dynamic routing selection, priority transmission strategy, error correction and data redundancy mechanism according to the network status information provided by the edge computing module to ensure that important data is sent first, and at the same time improve the reliability of data transmission through redundancy and self-correction mechanism. The intelligent acquisition logic module is responsible for dynamically analyzing the working environment and adjusting the signal acquisition frequency in real time, while predicting future signal trends to ensure that important information is not missed. This module realizes real-time information sharing between sensor nodes through distributed information fusion units, thereby improving the overall data quality. When the system is running, it first starts and self-checks to ensure that each module is working properly, then the intelligent sensor collects environmental variables in real time, and outputs standardized data after processing by the microprocessor. Subsequently, the data fusion module receives and optimizes this data, the edge computing module performs anomaly detection and selects the best transmission path, and finally the adaptive transmission module implements dynamic routing and priority management of data. The entire system runs on an industrial-grade embedded device, using a platform that supports deep learning and machine learning, and relies on a stable network environment to ensure real-time data transmission and storage.

[0168] High-precision sensors: Each sensor is connected to the embedded microprocessor via an I2C or SPI interface to ensure low-latency data transmission. The sensor's pins are connected in a standard configuration to support serial data transmission. The analog signals (such as temperature, voltage, etc.) measured by the sensor are converted into digital signals by the built-in ADC and transmitted to the microprocessor for preliminary processing.

[0169] Embedded microprocessor: Choose ARMCortex-M series or FPGA architecture with high computing power and multiple built-in GPIO ports for connecting peripherals. The microprocessor executes signal processing algorithms, analyzes sensor data in real time, removes noise, and sends the processed data to the data fusion module.

[0170] Adaptive filter: Connected to the microprocessor via DMA (direct memory access) to ensure efficient signal processing. The microprocessor sends the original signal to the adaptive filter, which returns the processed signal in real time for subsequent modules to use.

[0171] Data processing unit: Uses a multi-core processor (such as ARMCortex-A series) to support high concurrent processing capabilities. The processed data received from multiple sensors is transmitted to the data processing unit via a high-speed bus (such as USB3.0 or PCIe) for fusion and analysis. Storage: Uses high-speed SSD or NVMe storage to support fast read and write operations. Connects to the data processing unit via a SATA or PCIe interface to provide stable data storage support.

[0172] Edge computing device: Use NVIDIA Jetson or other edge computing devices, equipped with GPU for deep learning algorithm processing. Connect to the data fusion module through the Internet, receive processed data in real time, and perform anomaly detection and analysis. Use dual Gigabit network ports, support load balancing and redundancy, and ensure the stability and reliability of data transmission.

[0173] Network router: Use high-performance routers that support multiple network protocols (including Wi-Fi6, Zigbee). Connect to edge computing modules and other nodes through wireless signals or wired Internet to achieve dynamic data routing. Contains fiber optic interfaces and wireless modules for data transmission in different network environments. According to the real-time network status and traffic, the router dynamically selects the best transmission path to reduce latency.

[0174] The intelligent acquisition logic module includes:

[0175] Machine Learning Unit: A computing unit based on GPU or TPU to support fast execution of complex algorithms. It shares information with sensor nodes in real time through low-latency communication interfaces (such as UART or CAN) to improve overall data quality.

[0176] Information fusion unit: Bluetooth or Zigbee protocol is used to ensure wireless data transmission with low power consumption. By sharing information between neighboring nodes, the continuity and accuracy of data are enhanced, and information compensation is ensured when the signal is lost.

[0177] The data collected from the sensor is converted by ADC and sent to the microprocessor for processing. The microprocessor transmits the processed signal to the data fusion module for further data analysis and fusion. The fused data is detected for anomalies through the edge computing module and then transmitted to the target host through the adaptive transmission module. Each module exchanges data through standardized interfaces (I2C, SPI, USB, Internet, etc.) to ensure interoperability and compatibility between modules. The real-time bus architecture is adopted to support efficient data transmission between modules, reduce latency and improve the overall response speed of the system. The system design includes a monitoring mechanism that can monitor the status of each module in real time, identify and warn faults through the self-diagnosis module, and reduce maintenance requirements. See the attached manual for details. Figure 2 From top to bottom, the original signal waveform, the signal waveform after adaptive aluminum foil, and the effect index bar graph are shown. The original signal waveform graph clearly shows that when the original data is collected, there will be a large number of interference peaks and the waveform curve is not smooth. In the subsequent further processing, a large number of interference values ​​will appear. Therefore, adaptive optimization is performed to obtain the signal waveform after adaptive aluminum foil. While retaining the overall operating trend and without additional reduction of the peak value, the data is optimized and integrated, and the curve is smoother. The effect index bar graph is an authenticity score for the data after adaptive processing. It can be seen that its index is close to 900 points, which is enough to prove the overall accuracy of the data and the authenticity of the data after adaptive data processing, and realizes the function of data processing while ensuring the completeness of the data results as much as possible. Specific embodiment (II)

[0179] SP1: Environmental Monitoring and Data Collection:

[0180] Sp1.1: Intelligent sensor module monitors environmental variables in real time:

[0181] Various sensors (including but not limited to temperature sensors, humidity sensors, gas sensors, and light sensors) periodically collect environmental data. Real-time monitoring of environmental variables ensures that the system responds quickly to environmental changes, and sensor data provides a basis for subsequent analysis and decision-making.

[0182] Sp1.2: Preliminary data processing:

[0183] The sensor sends the collected data to the embedded microprocessor, which applies adaptive filtering algorithms (including but not limited to Kalman filtering) to remove noise and outliers and output processed clean data. Preliminary processing can improve data quality, ensure the accuracy of subsequent analysis, and reduce errors caused by environmental interference.

[0184] SP2: Data Fusion and Analysis:

[0185] Sp2.1: Data fusion module receives data:

[0186] The fusion module receives data from each sensor through a high-speed bus (including but not limited to USB or the Internet) and uses a cache mechanism to ensure continuous data reception. Data fusion can integrate multi-source information and improve the representativeness and accuracy of the overall data.

[0187] Sp2.2: Feature extraction:

[0188] The received data is processed using a convolutional neural network (CNN), key features are extracted, and output feature maps are generated to identify important patterns and trends in the data. Feature extraction helps to efficiently identify and distinguish different data categories, improving the efficiency of subsequent analysis.

[0189] Sp2.3: Probability calculation:

[0190] Based on the extracted features, the Softmax classification algorithm is used to calculate the probability of each data point belonging to each category and output the classification result. Probability calculation provides a quantitative basis for decision-making, making it easier for the system to automatically judge the environmental status.

[0191] SP3: Forecasting and early warning:

[0192] Sp3.1: Fault prediction:

[0193] Combining historical data with real-time data, the support vector machine (SVM) algorithm analyzes data trends to determine whether there are potential abnormalities in the system. The fault prediction function enables the system to foresee problems, so that measures can be taken in advance to reduce risks.

[0194] Sp3.2: Abnormal trend identification:

[0195] The edge computing gateway uses machine learning algorithms to identify abnormal trends in real time. When an abnormal situation is found, it immediately sends out an early warning signal to notify relevant personnel or systems. Timely warnings can help respond quickly and reduce losses and risks.

[0196] SP4: Adaptive Transport:

[0197] Sp4.1: Network status monitoring:

[0198] The edge computing gateway regularly monitors network latency, bandwidth, and signal strength and generates network status reports. Real-time monitoring of network status provides a basis for adaptive transmission and ensures the reliability of data transmission.

[0199] Sp4.2: Dynamic routing selection:

[0200] According to the network status report, the data transmission path is adjusted in real time using a dynamic routing selection algorithm, giving priority to channels with low latency and high bandwidth. The optimized transmission path ensures the efficiency and stability of data transmission.

[0201] SP5: Data Transfer:

[0202] Sp5.1: Priority data packet sending:

[0203] According to the importance and urgency of the data, data packets are assigned priorities, and critical data is transmitted first. The priority sending mechanism ensures that important information is not delayed when the network is congested.

[0204] Sp5.2: Error correction and data redundancy mechanism:

[0205] During data transmission, redundancy and error correction mechanisms (including but not limited to FEC forward error correction) are used to ensure data integrity and correctness. Error correction and data redundancy mechanisms can improve the reliability of data transmission and avoid data loss or damage.

[0206] SP6: Feedback and Optimization:

[0207] Sp6.1: Receiver confirmation:

[0208] The receiving end confirms the data reception and feeds back the result to the edge computing gateway, including detailed information on whether the reception is successful or not. The feedback mechanism helps the system understand the actual effect of data transmission so that adjustments can be made.

[0209] Sp6.2: Path and priority re-evaluation:

[0210] Including but not limited to discovering transmission delays or data loss, the system will automatically re-evaluate the transmission path and data packet priority and adjust the strategy to improve transmission quality. Dynamic optimization ensures that the system maintains efficient operation in a changing network environment.

[0211] SP7: Intelligent collection and information sharing:

[0212] Sp7.1: Data sharing between sensor nodes:

[0213] Each sensor node shares data in real time through wireless or wired networks, using a unified protocol to ensure consistent data formats. Data sharing between nodes enhances the overall data quality of the system and improves the accuracy and reliability of information.

[0214] Sp7.2: Neighboring node data compensation:

[0215] When a node signal is lost, the data of the neighboring nodes is used to compensate, ensuring the integrity and coherence of the information. This mechanism ensures that the system can still operate stably in the case of local failures and enhances the system's fault tolerance.

[0216] Real-time data processing module workflow:

[0217] Sp1: Real-time data reception and preprocessing:

[0218] The real-time data processing module continuously receives data streams from multiple sensors and processes the data using denoising and normalization algorithms to ensure data quality. Data preprocessing improves the accuracy of subsequent analysis and provides a good data foundation for decision-making.

[0219] Sp2: Ensemble learning algorithm:

[0220] Through ensemble learning algorithms (including but not limited to Bagging or Boosting), the prediction results of multiple models are integrated to output the final analysis results. Ensemble learning improves the accuracy of data analysis and reduces the errors caused by a single model.

[0221] Sp3: Automatically adjust sensor strategy:

[0222] Based on the data analysis results, the system automatically adjusts the sampling frequency and data transmission strategy of the sensor to quickly respond to environmental changes. This dynamic adjustment mechanism ensures that the system can adapt to the environment in real time and improve monitoring efficiency.

[0223] Sp4: Abnormal situation handling:

[0224] When an abnormal situation is detected, the real-time data processing module will trigger an alarm in time and upload the relevant data to the cloud for further analysis. In-depth analysis in the cloud helps optimize system performance and ensure the effectiveness and accuracy of subsequent operations.

[0225] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or equipment. In the absence of further restrictions, the elements defined by the statement "comprising a reference structure" do not exclude the existence of other identical elements in the process, method, article or equipment including the elements.

[0226] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A control signal acquisition and transmission system for a host, characterized in that: The signal acquisition and transmission system comprises: An intelligent sensor module, the intelligent sensor module comprising: High-precision sensors monitor and collect multi-dimensional environmental variables in real time; Embedded microprocessor to execute deep learning-based signal processing and data acquisition algorithms for adaptive regulation; Adaptive filter, used to dynamically optimize signal processing parameters and remove noise and outliers in real time; Self-diagnosis module for continuous monitoring of sensor status and feedback; The data fusion module is used to receive and process data from multiple sensors, including the following parts: Intelligent threshold adjustment algorithm, automatically optimizes data collection thresholds based on real-time data; The signal reconstruction algorithm based on the time series prediction model uses historical data to build a mathematical model to infer the signal of the unmonitored area; The edge computing module is electrically connected to the smart sensor module and includes the following parts: Real-time data analysis and anomaly detection module, with built-in deep learning algorithm to identify abnormal trends and issue early warnings; The machine learning-driven transmission path optimization module selects the best transmission channel in real time based on network conditions and data traffic; An adaptive transmission module, wherein the adaptive transmission module has a built-in adaptive transmission protocol, and the adaptive transmission module comprises the following steps: sp1. Dynamic routing selection: Real-time monitoring of network status, built-in intelligent algorithm to select the best transmission path, and dynamic adjustment to cope with network changes; sp2. Priority transmission strategy: hierarchical processing is performed according to the importance of data. Critical data is sent first, and non-critical data is transmitted when the network is idle; sp3, Error correction and data redundancy mechanism: realize data packet redundancy and self-error correction, and automatically recover the data lost during transmission; An intelligent acquisition logic module, the intelligent acquisition logic module comprising: Environmental scene adaptive recognition unit: built-in deep learning algorithm analyzes the working environment in real time and dynamically adjusts the signal acquisition frequency; Predictive data collection unit: Built-in machine learning algorithm predicts future signal trends and collects data based on the results of predicting future signal trends; Distributed information fusion unit: used to share information between sensor nodes in real time and enhance the overall data quality through the data of neighboring nodes.

2. A control signal acquisition and transmission system for a host according to claim 1, characterized in that: The high-precision sensor of the intelligent sensor module has multiple sensing functions, including: Temperature sensor with a measurement range of -40°C to +125°C and a resolution of 0.1°C; Humidity sensor with a measurement range of 0% to 100% relative humidity and an accuracy of ±2%; Gas concentration sensor, including CO2 and NH3 detection, with a range of 0-5000ppm; Light sensor, measuring light intensity from 0 to 100,000lx.

3. The control signal acquisition and transmission system for a host according to claim 1, characterized in that: The embedded microprocessor also includes an FPGA architecture to implement parallel processing and signal filtering. The filtering algorithm is an adaptive filter whose mathematical model is as follows: ， is the output signal, is the input signal, is the step size factor, and , is the error signal at the previous moment.

4. A control signal acquisition and transmission system for a host according to claim 1, characterized in that: The dynamic routing selection of the adaptive transmission protocol is specifically: using a weighted average algorithm to select the best transmission path: ， B is bandwidth; L is delay; S is signal strength; It is a weight coefficient, which is dynamically adjusted based on historical data to optimize the transmission effect.

5. The control signal acquisition and transmission system for a host according to claim 1, characterized in that: The convolutional neural network of the data fusion module includes the following convolution operations: ， Among them, X is the input data, K is the convolution kernel, Y is the output feature map, which improves the accuracy of data fusion, and i, j are the coordinates of the output feature map; , is the coordinate of the input data; the Softmax classification of the output layer is: ， For input Belongs to category probability; is the model output, and K is the number of categories.

6. A control signal acquisition and transmission system for a host according to claim 1, characterized in that: The real-time data analysis and anomaly detection module also includes a fault prediction module. The objective function of the built-in support vector machine module in the fault prediction module is: ， is the model complexity, C is the penalty factor, and its value is between 0 and 1. is the slack variable, N is The value range of is the number of data collection times of the signal acquisition and transmission system.

7. A control signal acquisition and transmission system for a host according to claim 1, characterized in that: The scene recognition in the intelligent acquisition logic adopts transfer learning, and the model is: ， Dynamically optimize sampling strategies to compensate for differences between source and target domains, Model output for the target domain; is the model output of the source domain.

8. The control signal acquisition and transmission system for a host according to claim 1, characterized in that: The error correction and data redundancy mechanism is based on blockchain technology, and the hash function generates data blocks: ， To generate data The hash value of The nth layer of hash functions processes the input data in sequence; The redundant coding algorithm is: ， is the redundant encoding function, For the collected data, is the weighting coefficient, and its value ranges from 0 to 1.

9. A control signal acquisition and transmission system for a host computer according to any one of claims 1 to 8, characterized in that: The operation process of the signal acquisition and transmission system is as follows: SP1: Environmental Monitoring and Data Collection: Sp1.1: Smart sensor modules monitor environmental variables in real time, including temperature, humidity, gas concentration, and light intensity; Sp1.2: The sensor performs preliminary processing on the collected data through an embedded microprocessor, executing an adaptive filtering algorithm to filter out noise and outliers; SP2: Data Fusion and Analysis: Sp2.1: The data fusion module receives data from each intelligent sensor module; Sp2.2: Use the convolutional neural network algorithm to extract features from the received data and generate an output feature map; Sp2.3: Based on the extracted features, the probability of the data belonging to each category is calculated through the Softmax classification algorithm; SP3: Forecasting and early warning: Sp3.1: Based on historical data and real-time data, use the support vector machine (SVM) algorithm to predict faults and determine whether there are abnormal conditions; Sp3.2: If an anomaly is predicted, the edge computing gateway uses machine learning algorithms to quickly identify abnormal trends and issue warning signals; SP4: Adaptive Transport: Sp4.1: The edge computing gateway monitors network status, including latency, bandwidth, and signal strength; Sp4.2: Use dynamic routing selection algorithm to select the best transmission path and adjust the transmission channel according to the real-time feedback mechanism; SP5: Data Transfer: Sp5.1: Send data packets to the receiving end according to priority, with key data being transmitted first; Sp5.2: During the transmission process, error correction and data redundancy mechanism is used for data packet redundancy and error correction; SP6: Feedback and Optimization: Sp6.1: The receiving end confirms the data reception and feeds back the result to the edge computing gateway; Sp6.2: If transmission delay or data loss is found, the system re-evaluates the transmission path and priority; SP7: Intelligent collection and information sharing: Sp7.1: Sharing data between sensor nodes; Sp7.2: When the signal of a certain node is lost, the data of the neighboring nodes is used for compensation.

10. A control signal acquisition and transmission system for a host according to claim 9, characterized in that: The edge computing gateway also includes a real-time data processing module, and the workflow of the real-time data processing module is as follows: Sp1: Receive data streams from multiple smart sensor nodes in real time and use data preprocessing algorithms to denoise and normalize the data; Sp2: Fusion of prediction results from multiple models through ensemble learning algorithms; Sp3: Automatically adjust the sampling frequency and data transmission strategy of the sensor based on the data analysis results; Sp4: When an abnormal situation is detected, an alarm is triggered in time and relevant data is uploaded to the cloud for in-depth analysis.

Images