CN117411463A - Edge computing gateway data acquisition-oriented adaptive filtering method - Google Patents
Edge computing gateway data acquisition-oriented adaptive filtering method Download PDFInfo
- Publication number
- CN117411463A CN117411463A CN202311724481.6A CN202311724481A CN117411463A CN 117411463 A CN117411463 A CN 117411463A CN 202311724481 A CN202311724481 A CN 202311724481A CN 117411463 A CN117411463 A CN 117411463A
- Authority
- CN
- China
- Prior art keywords
- data
- board card
- host
- model
- state
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 82
- 238000001914 filtration Methods 0.000 title claims abstract description 58
- 230000003044 adaptive effect Effects 0.000 title claims abstract description 36
- 230000008569 process Effects 0.000 claims abstract description 42
- 239000011159 matrix material Substances 0.000 claims description 27
- 238000012546 transfer Methods 0.000 claims description 9
- 230000005540 biological transmission Effects 0.000 claims description 7
- 230000004044 response Effects 0.000 claims description 7
- 238000001514 detection method Methods 0.000 claims description 5
- 238000004458 analytical method Methods 0.000 claims description 3
- 238000012544 monitoring process Methods 0.000 claims description 3
- 230000007704 transition Effects 0.000 claims description 3
- 238000007639 printing Methods 0.000 claims description 2
- 238000013480 data collection Methods 0.000 claims 2
- 230000006854 communication Effects 0.000 description 14
- 238000004891 communication Methods 0.000 description 13
- 238000004088 simulation Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 238000007405 data analysis Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000001934 delay Effects 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000011664 signaling Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03H—IMPEDANCE NETWORKS, e.g. RESONANT CIRCUITS; RESONATORS
- H03H21/00—Adaptive networks
- H03H21/0012—Digital adaptive filters
- H03H21/0043—Adaptive algorithms
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/05—Digital input using the sampling of an analogue quantity at regular intervals of time, input from a/d converter or output to d/a converter
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5061—Partitioning or combining of resources
- G06F9/5072—Grid computing
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03H—IMPEDANCE NETWORKS, e.g. RESONANT CIRCUITS; RESONATORS
- H03H21/00—Adaptive networks
- H03H21/0012—Digital adaptive filters
- H03H21/0025—Particular filtering methods
- H03H21/0029—Particular filtering methods based on statistics
- H03H21/003—KALMAN filters
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Human Computer Interaction (AREA)
- Mathematical Physics (AREA)
- Feedback Control In General (AREA)
Abstract
The invention discloses an edge computing gateway data acquisition self-adaptive filtering method, which comprises the following steps: step 1, a host communicates with an IO board card to acquire an IO board card equipment address; step 2, the host reads the information of the registration table of the IO board card and obtains the model of the IO board card; step 3, judging that the type of the IO board card corresponds to a channel containing analog data so as to acquire analog data to be filtered; and 4, calling an adaptive filter to filter the analog data. The invention avoids complex equipment inquiry and matching processes, simplifies the data acquisition process, and improves the quality and reliability of data; the adaptive Kalman filter is adopted to carry out adaptive filtering on the nonlinear system effectively, so that the adaptive Kalman filter is more flexible to use, has lower computational complexity, is more suitable for environments with limited resources, and realizes the full-automatic process of collecting and filtering data of the edge computing gateway.
Description
Technical Field
The invention relates to the technical field of self-adaptive filtering, in particular to an edge computing gateway data acquisition self-adaptive filtering method.
Background
Edge computing is a distributed computing paradigm that places computing resources closer to data sources and end devices to reduce latency and improve performance. Understanding the principles and architecture of edge computing is critical to developing edge computing oriented data acquisition and processing methods. In the embedded gateway technology, communication between a host and an IO board is of great importance. In order to ensure the stability and efficiency of the system, the host needs to be able to identify and control the connected IO board card timely and accurately. Conventional approaches may involve cumbersome manual configuration or rely on external hardware identification, which are not only complex to operate, but also prone to error.
In order to effectively implement the above communication, a mechanism is required to identify and control various devices connected to the host, that is, to enable the host to automatically detect the connected IO board, read the specific model of the IO board, and determine the registration information thereof, thereby obtaining the analog data. Current adaptive filtering techniques are mostly reference adaptive, considering that there is no suitable reference signal and the model is not complicated.
Therefore, we have designed an edge-oriented computing gateway data acquisition adaptive filtering method to solve the above problems.
Disclosure of Invention
The invention provides an edge-calculation-oriented gateway data acquisition self-adaptive filtering method aiming at the problems existing in the prior art of gateway data acquisition and filteringThe technical scheme provides a method for acquiring gateway analog data and performing self-adaptive filtering, which utilizes serial port communication, GPIO pin state detection and RS485 control to automatically detect and read registration information of an IO board card through communication between a host and the IO board card, further determines the model of the IO board card to acquire analog data, and initializes state quantityCovariance matrix->And predicting the state quantity by combining the real observed value, updating the initialization variable, carrying out new iteration, and finally obtaining a filtering result Z, wherein the filtering effect is more ideal along with the continuous progress of the iteration.
In order to achieve the above purpose, the present invention adopts the following technical scheme:
an edge computing gateway data acquisition-oriented adaptive filtering method mainly comprises the following steps:
step 1, a host communicates with an IO board card to acquire an IO board device address;
step 2, the host reads the information of the registration table of the IO board card and obtains the model of the IO board card;
step 3, judging that the type of the IO board card corresponds to a channel containing analog data so as to acquire analog data to be filtered;
and 4, calling an adaptive filter to filter the analog data.
Specifically, as a preferred scheme of the present invention, in step 1, the host communicates with the IO board card, and the step of obtaining the address of the IO board device includes the following steps:
step 1.1, initializing UART serial equipment and RS-485 control equipment, and opening the UART serial equipment and the RS-485 control equipment;
step 1.2, configuring UART serial equipment, and initializing baud rate parameters, data bit parameters, stop bit parameters and check parameters after the UART serial equipment is opened;
step 1.3, initializing and configuring GPIO pins as input modes, and configuring 8 cs pins as input modes;
and step 1.4, the host communicates with the IO board card and acquires the address of the IO board device.
Specifically, as a preferred solution of the present invention, in step 2, the step of reading the information of the IO board registry by the host, and obtaining the IO board model includes:
step 2.1, monitoring the state of the GPIO pin, continuously checking the state of the GPIO pin, and reading the registry information of the IO board card when the state of the GPIO pin is 0;
step 2.2, obtaining the model of the IO board card by reading registry information of the IO board card comprises the following steps:
step 2.2.1, the host configures UART serial equipment into a sending mode, and sends a request to the IO board through the UART serial equipment, and the IO board is requested to return registration information, and waits for data transmission to be completed;
step 2.2.2, the host configures UART serial equipment as a receiving mode, and reads the response of the IO board card from the UART serial equipment;
2.2.3, analyzing the obtained data and judging whether the data is valid, and for the successfully received data, printing the value of the data by the host computer in a circulating way;
step 2.2.4, verifying whether the received data accords with the expected format, and confirming the validity of the data by checking start and end marks, position information and other specific values of the frame;
and 2.2.5, after the host confirms that the response is valid, acquiring the model of the IO board card, and updating a registry of the model of the IO board card in the host.
Specifically, as a preferred solution of the present invention, in step 3, the step of determining that the IO board card type corresponds to the channel containing the analog data to obtain the analog data to be filtered includes the following steps:
step 3.1, the host reads the model of the IO board card from the registry of the updated IO board card model;
step 3.2, checking a model database or a predefined model list according to the model of the IO board card, determining whether the IO board card and the pins of the model support an analog data input channel, and skipping if not;
step 3.3, configuring channels and acquiring analog data, if the IO board card supports analog data input, configuring a corresponding channel by a host to prepare data for receiving, and sending a data reading request to the IO board card by the host through a specific command or request;
step 3.4, the host waits for and reads the analog data returned from the IO board card;
step 3.5, after the analog data is successfully read by the host, the host calls a predefined filter type judgment model to analyze the analog data;
and 3.6, according to the analysis result of the filter type judgment model, the host gives a suggestion or automatically selects the most suitable filter to process the analog data.
Specifically, in step 4, the step of invoking the adaptive filter to filter the analog data includes:
step 4.1, acquiring data acquired by an edge computing gateway, and presenting the data in the form of m paths of gateway data Z acquired n times;
step 4.2, kth column data of data ZRepresents->Data collected at the moment;
step 4.3, modeling a detection model of the m-path sensor, and performing taylor expansion on the nonlinear system to linearize to obtain a state equation and an observation equation or perform feature engineering direct linear modeling:
;
;
in the method, in the process of the invention,and->System->Time of day,/->State variable of time->And->System->Time of day,Inputting time; />Is process noise subject to mean 0, covariance +.>Is a normal distribution of (2);for observing noise, obey the mean value 0 and covariance +.>Is a normal distribution of (2); />For a nonlinear state transfer function, the system is described as being at time +.>To->Evolution between (I) and (II)>And->Respectively indicate->Time and->Estimating a posterior state of the moment; />For non-linear observation transfer function, < >>Jacobian, which is a state transition matrix, < >>A jacobian matrix for observing the transfer matrix;
step 4.4, in the prediction stage of the filtering process, predicting the state estimation, involving the estimated value of the predicted state and the covariance matrix:
in the method, in the process of the invention,representing predicted state estimate->Representing a priori covariance matrix,/->Representation->Time-of-day posterior estimation covariance, +.>Representing the transpose of the matrix>Representing a process noise covariance matrix;
step 4.5, in the updating stage of filtering, updating the value of the state quantity and the value of the covariance matrix according to the difference between the estimated value of the predicted state and the observed value of the actual sensor, and continuously correcting and optimizing the estimation of the system state:
wherein the method comprises the steps of
;
In the method, in the process of the invention,for Kalman gain, ++>Is a unitary matrix->Is->Time posterior covariance matrix->Representation->Time observation noise covariance matrix->Is->Sensor actual observation of time, +.>Is->Predicting the observation of the sensor at the moment;
step 4.6: for the process noise matrix in the steps 4.4 and 4.5And observation noise matrix->Updating:
;
;
in the method, in the process of the invention,a superparameter updated for control Q, R;
and 4.7, obtaining a final filtering result based on the updated state quantity:
。
compared with the prior art, the invention has the beneficial effects that: according to the invention, a high-efficiency and stable communication mode is selected for the host and the IO board card, so that real-time transmission and high accuracy of data are ensured, and the device address of the IO board card is successfully obtained through code design, thereby avoiding complex device inquiry and matching processes, simplifying the data acquisition process, and improving the quality and reliability of the data; in the process of reading the information of the registry of the IO board card by the host, the model of the IO board card can be rapidly obtained, and a set of efficient data analysis algorithm is provided, so that the required information can be accurately extracted from the registry; the system has a self-adaptive model matching function, can be compatible with various IO board cards, and further greatly improves the expandability and flexibility of the system; the invention can adaptively judge the analog data channel corresponding to the IO board card model, and ensure that the system can be accurately matched with the correct data channel, thereby obtaining the analog data to be filtered; the self-adaptive Kalman filtering mode adopted for filtering the gateway data can process the filtering of any common gateway equipment data, so that the human intervention of the traditional gateway data filtering and the limitation of a filtering method are avoided; the adaptive Kalman filter is adopted to carry out adaptive filtering on the nonlinear system effectively, so that the adaptive Kalman filter is more flexible to use, has lower computational complexity, is more suitable for environments with limited resources, and realizes the full-automatic process of collecting and filtering data of the edge computing gateway.
Drawings
FIG. 1 is a flow chart of the invention in which a host and an IO board card acquire analog channel data;
FIG. 2 is a flow chart of the adaptive Kalman filtering of analog data according to the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments.
According to the edge computing gateway data acquisition self-adaptive filtering method provided by the embodiment, the nonlinear model is linearized, the linear model is directly built for the characteristic engineering, and then the self-adaptive Kalman filtering technology is applied. Compared with other filtering methods, the method has the advantages of minimum mean square error, effective elimination of noise and jitter, certain robustness to system uncertainty, high response speed and the like.
As shown in fig. 1, the related edge-oriented computing gateway data acquisition adaptive filtering method mainly comprises the following steps:
step 1, a host communicates with an IO board card to acquire an IO board device address;
step 2, the host reads the information of the registration table of the IO board card and obtains the model of the IO board card;
step 3, judging that the type of the IO board card corresponds to a channel containing analog data so as to acquire analog data to be filtered;
and 4, calling an adaptive filter to filter the analog data.
First, the names involved in the present embodiment will be explained, and UART (Universal Asynchronous Receiver-Transmitter), which is an abbreviation for universal asynchronous receiver-Transmitter, is a part of computer hardware for asynchronous serial communication. It converts the bytes into a serial bit stream for transmission and converts the received serial bit stream back into bytes. Including baud rate (symbols or modulation order transmitted per second), data bits (number of bits per byte in UART communication), stop bits (identifying the end of a data byte), check (checking whether a data byte is in error during transmission), etc. RS-485 is a differential signaling standard for multipoint communications, often used for communications over long distances or in environments with electrical noise. GPIO (General-Purpose Input/Output) is an abbreviation for General Purpose Input/Output, a type of General Purpose pin on a computer, microcontroller or other device that can be customized by a user as an Input or Output. In the present invention, the value of GPIOX is used to determine whether the corresponding board card is registered.
Next, a detailed description will be developed for main steps of the edge computing gateway data acquisition adaptive filtering method according to this embodiment.
Step 1, a host communicates with an IO board card, and the step of obtaining the address of the IO board device comprises the following detailed steps:
step 1.1, initializing serial port and 485 control equipment:
at this step, the program tries to open two devices, namely UART serial equipment and RS-485 control equipment;
UART serial equipment: opening UART serial equipment through a path of the UART serial equipment, and returning error information by a program if the equipment fails to be opened;
RS-485 control device: and opening the RS-485 control device through the path of the RS-485 control device, and returning error information by the program if the device fails to be opened.
Step 1.2, configuring UART serial equipment:
once the UART serial device is successfully turned on, UART serial device parameters such as baud rate, data bit, stop bit, check, etc. need to be initialized.
Step 1.3, configuring GPIO:
the GPIO is initialized and configured by the following method:
step 1.3.1, initializing and configuring a GPIO pin as an input mode;
step 1.3.2, 8 cs (chip select) pins are configured as an input mode.
And step 1.4, the host communicates with the IO board card and acquires the address of the IO board card equipment.
In this embodiment, in order to obtain the address of the IO board card device, the host needs to be able to identify which IO board card has been inserted in the corresponding card handling position (or is currently activated), in the code, the host uses 8 cs (chip select) pins to identify and select a specific IO board card, in the main cycle, the program will continuously check the status of each cs pin, and if the status of a certain cs pin is 0, it indicates that the IO board card has been inserted in the corresponding position, so as to obtain the position of the IO board card.
In the embodiment, the technical means of communication between the host and the IO board card is subjected to intensive research and optimization. And the high-efficiency stable communication protocol is selected, so that the real-time transmission and high accuracy of data are ensured. Through code design, the IO board device address is successfully obtained, so that complex device query and matching processes are avoided. In addition, multiple error detection and correction mechanisms are added in the communication process in consideration of various interference possibly existing in the industrial field, so that the stability of communication is greatly improved. Compared with the traditional communication method, the invention not only simplifies the data acquisition process, but also improves the quality and reliability of the data.
Step 2, reading the information of the register table of the IO board card for the host computer, and acquiring the model of the IO board card comprises the following detailed steps:
step 2.1, monitoring the state of the GPIO pin,
the program continuously checks the state of the GPIO pin, and reads the register information of the IO board card when the GPIO pin is 0.
Step 2.2, reading the registry information to determine the model of the IO board card, and further obtaining the model of the IO board card by communicating with the IO board card so as to read the registry information of the IO board card:
step 2.2.1, the host firstly configures UART serial equipment into a sending mode, then sends a request to the IO board through the UART serial equipment, requests the IO board to return registration information, and delays for 1 millisecond to wait for data transmission to finish.
And 2.2.2, the host configures UART serial equipment into a receiving mode, and reads the response of the IO board card from the UART serial equipment.
And 2.2.3, the host analyzes the obtained data and judges whether the data is valid, and the host prints the value of the data through circulation for the successfully received data.
Step 2.2.4, verifying whether the received data conforms to the expected format, and confirming the validity of the data by checking start and end flags, position information and other specific values of the frame.
And 2.2.5, after the host confirms that the response is valid, acquiring the model of the IO board card, and updating a registry of the model of the IO board card in the host.
In this embodiment, the invention shows innovation and practicability in the process of reading the information of the IO board card registry by the host. The method not only can rapidly acquire the type of the IO board card, but also has a set of efficient data analysis algorithm, and ensures that the required information can be accurately extracted from the registry. In addition, considering that IO boards of different models possibly have different registry structures, the system has a self-adaptive model matching function, can be compatible with various IO boards, and further greatly improves the expandability and flexibility of the system.
In step 3, judging that the model of the IO board corresponds to the channel containing the analog data, so as to obtain the analog data to be filtered, including the following detailed steps:
and 3.1, reading the model of the IO board card, and reading the model of the IO board card from a registry of the updated model of the IO board card by the host.
And 3.2, judging whether analog data input is supported, checking a model database or a predefined model list according to the model of the IO board card, determining whether the model of the IO board card and the pins support analog data input channels, if yes, entering the next step, if not, skipping, and returning to reading the registry of the updated model of the IO board card to determine the model of the IO board card.
And 3.3, configuring a channel and acquiring analog data, wherein if the IO board card supports analog data input, the host configures the corresponding channel to prepare for data receiving, and the host sends a data reading request to the IO board card through a specific command or request.
And 3.4, reading the analog data, and waiting and reading the analog data returned from the IO board card by the host.
And 3.5, calling a filter type judgment model, and once the simulation data is successfully read by the host, calling a predefined filter type judgment model by the host to analyze the simulation data.
And 3.6, outputting a filter suggestion, and according to the analysis result of the filter type judgment model, giving the suggestion or automatically selecting the most suitable filter to process the simulation data Z by the host.
The implementation adopts an advanced data processing technology, and can adaptively judge the analog data channel corresponding to the IO board card model. By establishing a complete database of the types of the IO board card and the data channels or a predefined type list, the system is ensured to be accurately matched with the correct data channels, so that the analog data to be filtered is obtained.
As shown in fig. 2, in step 4, the state quantity is initializedCovariance matrix->Process noise matrix->And observation noise matrix->Predicting state quantity, correcting state quantity by combining real observed value, and updating initialization changeAnd continuously carrying out new iterative updating on x, P, Q, R, and calling the adaptive filter to filter the analog data, wherein the implementation is as follows:
step 4.1, in which we first obtain the data collected by the edge computation gateway, typically in the form of m-way n-time collected gateway data Z.
Step 4.2, kth column data of data ZRepresents->And data acquired at the moment.
Step 4.3, in which we model the detection model of the m-way sensor, which typically involves taylor expansion of the nonlinear system to linearize to get a state equation and observation equation or to perform direct linear modeling of the feature engineering:
;
;
in the method, in the process of the invention,and->System->Time of day,/->State variable of time->And->System->Time of day,/->Inputting time; />Is process noise subject to mean 0, covariance +.>Is a normal distribution of (2); />For observing noise, obey the mean value 0 and covariance +.>Is a normal distribution of (2); />For a nonlinear state transfer function, the system is described as being at time +.>To->Evolution between (I) and (II)>And->Respectively indicate->Time and->Estimating a posterior state of the moment; />For non-linear observation transfer function, < >>Jacobian, which is a state transition matrix, < >>A jacobian matrix for observing the transfer matrix; these equations will assist us in the next step in state estimation.
Step 4.4, in the prediction stage of the filtering process, predicting the state estimation, involving the estimated value of the predicted state and the covariance matrix:
in the method, in the process of the invention,representing predicted state estimate->Representing a priori covariance matrix,/->Representation->Time-of-day posterior estimation covariance, +.>Representing the transpose of the matrix>Representing the process noise covariance matrix.
Step 4.5, in the updating stage of filtering, updating the value of the state quantity and the value of the covariance matrix according to the difference between the estimated value of the predicted state and the observed value of the actual sensor, and continuously correcting and optimizing the estimation of the system state:
wherein the method comprises the steps of
In the method, in the process of the invention,for Kalman gain, ++>Is a unitary matrix->Is->Time posterior covariance matrix->Representation->Time observation noise covariance matrix->Is->Sensor actual observation of time, +.>Is->Prediction of sensor observations at time.
Step 4.6: for the process noise matrix in the steps 4.4 and 4.5And observation noise matrix->Updating:
;
;
in the method, in the process of the invention,for controlling->、/>An updated hyper-parameter may be selected from the empirical values.
And 4.7, obtaining a final filtering result based on the updated state quantity:
,
thereby obtaining virtual data Z, which represents the best estimate of the sensor simulation data for the present implementation, taking into account the information of previous predictions and observations, thus achieving the goal of adaptive filtering.
The adaptive Kalman filtering mode adopted for filtering the gateway data in the embodiment can almost process the filtering of any common gateway equipment data, and avoids the human intervention of the traditional gateway data filtering and the limitation of a filtering method. The adaptive kalman filter is an optimal estimator which provides optimal estimation by minimizing the covariance of the estimation error based on a bayesian framework, i.e. provides state estimation of minimum mean square error in the case of linear system and gaussian noise; the adaptive kalman filter performs well in data filtering in high-dimensional systems, which means that real-time filtering of data can be achieved even for systems with complex modeling; the adaptive Kalman filtering has strong robustness, and adapts to different measurement and system dynamic conditions by automatically adjusting the weight in the estimation process, so that the adaptive Kalman filtering can cope with the problems of noise level change, system parameter drift and the like, thereby being suitable for various complex application occasions; the adaptive Kalman filter adopted by the embodiment can also effectively carry out adaptive filtering on a nonlinear system, and particularly after carrying out characteristic engineering pretreatment on the system in advance, the adaptive Kalman filter can be used more flexibly in practical application; the adaptive kalman filter also typically has better smoothness, it estimates the current state based on past state estimates and measurements, without introducing abrupt large changes, which is also characteristic of most gateway analog data volumes. In addition, the self-adaptive Kalman filtering has lower computational complexity, and is more suitable for the environment with limited resources; the unique gateway analog data acquisition and self-adaptive filtering mode of the embodiment realizes the full-automatic process of collecting and filtering the edge computing gateway data. In the whole, the method has excellent application potential in the field of edge computing gateway data filtering.
The steps together form an adaptive filtering method for the data acquisition of the edge computing gateway, and accurate information is extracted from the original data through continuous iterative prediction and observation, so that along with the filtering, an ideal filtering treatment effect can be achieved more and more.
The foregoing is only a preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art, who is within the scope of the present invention, should make equivalent substitutions or modifications according to the technical scheme of the present invention and the inventive concept thereof, and should be covered by the scope of the present invention.
Claims (5)
1. An edge-oriented computing gateway data acquisition adaptive filtering method is characterized in that,
step 1, a host communicates with an IO board card to acquire an IO board card equipment address;
step 2, the host reads the information of the registration table of the IO board card and obtains the model of the IO board card;
step 3, judging that the model of the IO board card corresponds to a channel containing analog data so as to acquire analog data to be filtered;
and 4, calling an adaptive filter to filter the analog data.
2. The method for adaptive filtering data collection of an edge-oriented computing gateway according to claim 1, wherein in step 1, the host communicates with the IO board card, and the step of obtaining the address of the IO board card device comprises the following steps:
step 1.1, initializing UART serial equipment and RS-485 control equipment, and opening the UART serial equipment and the RS-485 control equipment;
step 1.2, configuring UART serial equipment, and initializing baud rate parameters, data bit parameters, stop bit parameters and check parameters after the UART serial equipment is opened;
step 1.3, initializing and configuring GPIO pins as input modes, and configuring 8 cs pins as input modes;
and step 1.4, the host communicates with the IO board card and acquires the address of the IO board card equipment.
3. The method for adaptively filtering data collected by an edge-oriented computing gateway according to claim 1, wherein in step 2, the host reads information of an IO board registration table, and the obtaining of the model of the IO board comprises:
step 2.1, monitoring the state of the GPIO pin, continuously checking the state of the GPIO pin, and reading the registry information of the IO board card when the state of the GPIO pin is 0;
step 2.2, obtaining the model of the IO board card by reading registry information of the IO board card comprises the following steps:
step 2.2.1, the host configures UART serial equipment into a sending mode, and sends a request to the IO board through the UART serial equipment, and the IO board is requested to return registration information, and waits for data transmission to be completed;
step 2.2.2, the host configures UART serial equipment as a receiving mode, and reads the response of the IO board card from the UART serial equipment;
2.2.3, analyzing the obtained data and judging whether the data is valid, and for the successfully received data, printing the value of the data by the host computer in a circulating way;
step 2.2.4, verifying whether the received data accords with the expected format, and confirming the validity of the data by checking the start and end marks of the frame and the position information;
and 2.2.5, after the host confirms that the response is valid, acquiring the model of the IO board card, and updating a registry of the model of the IO board card in the host.
4. The method for adaptive filtering of edge-oriented computing gateway data collection according to claim 1, wherein in step 3, the step of determining that the IO board card model corresponds to a channel containing analog data to obtain analog data to be filtered comprises the following steps:
step 3.1, the host reads the model of the IO board card from the registry of the updated IO board card model;
step 3.2, checking a model database or a predefined model list according to the model of the IO board card, determining whether the IO board card and the pins of the model support an analog data input channel, and skipping if not;
step 3.3, configuring channels and acquiring analog data, if the IO board card supports analog data input, configuring a corresponding channel by a host to prepare data for receiving, and sending a data reading request to the IO board card by the host through a specific command or request;
step 3.4, the host waits for and reads the analog data returned from the IO board card;
step 3.5, after the analog data is successfully read by the host, the host calls a predefined filter type judgment model to analyze the analog data;
and 3.6, according to the analysis result of the filter type judgment model, the host gives a suggestion or automatically selects the most suitable filter to process the analog data.
5. The method of claim 1, wherein in step 4, the invoking the adaptive filter to filter the analog data comprises:
step 4.1, acquiring data acquired by an edge computing gateway, and presenting the data in the form of m paths of gateway data Z acquired n times;
step 4.2, kth column data of data ZRepresents->Data collected at the moment;
step 4.3, modeling a detection model of the m-path sensor, and performing taylor expansion on the nonlinear system to linearize to obtain a state equation and an observation equation or perform feature engineering direct linear modeling:
;
;
in the method, in the process of the invention,and->System->Time of day,/->State variable of time->And->System->Time of day,Inputting time; />Is process noise subject to mean 0, covariance +.>Is a normal distribution of (2);for observing noise, obey the mean value 0 and covariance +.>Is a normal distribution of (2); />For a nonlinear state transfer function, the system is described as being at time +.>To->Evolution between (I) and (II)>And->Respectively indicate->Time and->Estimating a posterior state of the moment; />For non-linear observation transfer function, < >>Jacobian, which is a state transition matrix, < >>A jacobian matrix for observing the transfer matrix;
step 4.4, in the prediction stage of the filtering process, predicting the state estimation, involving the estimated value of the predicted state and the covariance matrix:
;
in the method, in the process of the invention,representing predicted state estimate->Representing a priori covariance matrix,/->Representation->Time-of-day posterior estimation covariance, +.>Representing the transpose of the matrix>Representing a process noise covariance matrix;
step 4.5, in the updating stage of filtering, updating the value of the state quantity and the value of the covariance matrix according to the difference between the estimated value of the predicted state and the observed value of the actual sensor, and continuously correcting and optimizing the estimation of the system state:
;
in the method, in the process of the invention,for Kalman gain, ++>Is a unitary matrix->Is->Time posterior covariance matrix->Representation->Time observation noise covariance matrix->Is->Sensor actual observation of time, +.>Is->Predicting the observation of the sensor at the moment;
step 4.6: for the process noise matrix in the steps 4.4 and 4.5And observation noise matrix->Updating:
;
;
in the method, in the process of the invention,to control an updated one of the super parameters;
and 4.7, obtaining a final filtering result based on the updated state quantity:
。
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311724481.6A CN117411463B (en) | 2023-12-15 | 2023-12-15 | Edge computing gateway data acquisition-oriented adaptive filtering method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311724481.6A CN117411463B (en) | 2023-12-15 | 2023-12-15 | Edge computing gateway data acquisition-oriented adaptive filtering method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117411463A true CN117411463A (en) | 2024-01-16 |
CN117411463B CN117411463B (en) | 2024-02-20 |
Family
ID=89492998
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311724481.6A Active CN117411463B (en) | 2023-12-15 | 2023-12-15 | Edge computing gateway data acquisition-oriented adaptive filtering method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117411463B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117827467A (en) * | 2024-03-05 | 2024-04-05 | 南京群顶科技股份有限公司 | Dynamic portrait-based virtual machine resource allocation method |
CN117992389A (en) * | 2024-04-03 | 2024-05-07 | 南京群顶科技股份有限公司 | Power management device and method for edge computing gateway |
Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2003250043A (en) * | 2002-02-26 | 2003-09-05 | Ricoh Co Ltd | Image forming device |
CN106054856A (en) * | 2016-05-26 | 2016-10-26 | 中广核研究院有限公司北京分公司 | Nuclear power plant proportional integral controller-based noise analysis method and device |
CN109541994A (en) * | 2018-10-29 | 2019-03-29 | 浙江求是科教设备有限公司 | A kind of real-time acquisition system and its acquisition method of high anti-interference ability |
CN111740723A (en) * | 2020-06-12 | 2020-10-02 | 武汉三相电力科技有限公司 | Method, equipment and device for filtering monitoring data |
CN112367276A (en) * | 2020-12-31 | 2021-02-12 | 南京群顶科技有限公司 | Network resource dynamic self-adaption method and system based on network flow priority |
CN113613287A (en) * | 2021-06-21 | 2021-11-05 | 工业云制造(四川)创新中心有限公司 | Automatic data acquisition system based on edge calculation |
CN113672608A (en) * | 2021-08-25 | 2021-11-19 | 东北大学 | Internet of things perception data reduction system and method based on self-adaptive reduction threshold |
CN114124033A (en) * | 2021-10-11 | 2022-03-01 | 北京川速微波科技有限公司 | Kalman filter implementation method, device, storage medium and equipment |
US20230052553A1 (en) * | 2021-08-16 | 2023-02-16 | Ambarella International Lp | Adding an adaptive offset term using convolution techniques to a local adaptive binarization expression |
CN116008688A (en) * | 2022-12-02 | 2023-04-25 | 上海宝康电子控制工程有限公司 | Portable fault detector |
CN117095314A (en) * | 2023-08-22 | 2023-11-21 | 中国电子科技集团公司第五十四研究所 | Target detection and re-identification method under cross-domain multi-dimensional air-space environment |
-
2023
- 2023-12-15 CN CN202311724481.6A patent/CN117411463B/en active Active
Patent Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2003250043A (en) * | 2002-02-26 | 2003-09-05 | Ricoh Co Ltd | Image forming device |
CN106054856A (en) * | 2016-05-26 | 2016-10-26 | 中广核研究院有限公司北京分公司 | Nuclear power plant proportional integral controller-based noise analysis method and device |
CN109541994A (en) * | 2018-10-29 | 2019-03-29 | 浙江求是科教设备有限公司 | A kind of real-time acquisition system and its acquisition method of high anti-interference ability |
CN111740723A (en) * | 2020-06-12 | 2020-10-02 | 武汉三相电力科技有限公司 | Method, equipment and device for filtering monitoring data |
CN112367276A (en) * | 2020-12-31 | 2021-02-12 | 南京群顶科技有限公司 | Network resource dynamic self-adaption method and system based on network flow priority |
CN113613287A (en) * | 2021-06-21 | 2021-11-05 | 工业云制造(四川)创新中心有限公司 | Automatic data acquisition system based on edge calculation |
US20230052553A1 (en) * | 2021-08-16 | 2023-02-16 | Ambarella International Lp | Adding an adaptive offset term using convolution techniques to a local adaptive binarization expression |
CN113672608A (en) * | 2021-08-25 | 2021-11-19 | 东北大学 | Internet of things perception data reduction system and method based on self-adaptive reduction threshold |
CN114124033A (en) * | 2021-10-11 | 2022-03-01 | 北京川速微波科技有限公司 | Kalman filter implementation method, device, storage medium and equipment |
CN116008688A (en) * | 2022-12-02 | 2023-04-25 | 上海宝康电子控制工程有限公司 | Portable fault detector |
CN117095314A (en) * | 2023-08-22 | 2023-11-21 | 中国电子科技集团公司第五十四研究所 | Target detection and re-identification method under cross-domain multi-dimensional air-space environment |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117827467A (en) * | 2024-03-05 | 2024-04-05 | 南京群顶科技股份有限公司 | Dynamic portrait-based virtual machine resource allocation method |
CN117827467B (en) * | 2024-03-05 | 2024-05-10 | 南京群顶科技股份有限公司 | Dynamic portrait-based virtual machine resource allocation method |
CN117992389A (en) * | 2024-04-03 | 2024-05-07 | 南京群顶科技股份有限公司 | Power management device and method for edge computing gateway |
Also Published As
Publication number | Publication date |
---|---|
CN117411463B (en) | 2024-02-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN117411463B (en) | Edge computing gateway data acquisition-oriented adaptive filtering method | |
Geng et al. | Bias estimation for asynchronous multi-rate multi-sensor fusion with unknown inputs | |
Yu et al. | Practical implementation of an efficient forward-backward algorithm for an explicit-duration hidden Markov model | |
Strobach | Linear prediction theory: a mathematical basis for adaptive systems | |
WO2018076571A1 (en) | Method and system for detecting abnormal value in lte network | |
CN110458187B (en) | Malicious code family clustering method and system | |
Rojas et al. | A note on the SPICE method | |
CN110503071A (en) | Multi-object tracking method based on the more Bernoulli Jacob's Additive Models of variation Bayes's label | |
Han et al. | Parameter estimation for multirate multi-input systems using auxiliary model and multi-innovation | |
CN108153259B (en) | Multi-controller optimal state estimation control strategy design method based on Kalman filtering | |
WO2020253039A1 (en) | Road section characteristic model training method and apparatus, and computer device and storage medium | |
CN113315781B (en) | HMM model-based abnormal data detection method | |
Wang et al. | Measurement random latency probability identification | |
CN106909656A (en) | Obtain the method and device of Text Feature Extraction model | |
CN101223727A (en) | Data - dependent noise predictor in data-aided timing recovery | |
WO2004090782A1 (en) | Accurate linear parameter estimation with noisy inputs | |
CN116743468A (en) | Dynamic attack path generation method based on reinforcement learning | |
CN111754589A (en) | Color matching method and device, computer equipment and storage medium | |
CN110049442A (en) | Indoor WiFi fingerprint location automatic calibration method and system based on smart phone | |
CN114708307A (en) | Target tracking method, system, storage medium and device based on correlation filter | |
CN114900835A (en) | Malicious traffic intelligent detection method and device and storage medium | |
CN112989340A (en) | Model back door detection method, device, medium and computing equipment | |
CN108470016B (en) | System state prediction method of industrial dryer | |
CN115988100B (en) | Gateway management method for intelligent perception of Internet of things of equipment based on multi-protocol self-adaption | |
CN117473396B (en) | New energy automobile intelligent battery management system based on deep learning |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |