Filtering and decoding joint iterative detection method and device for network control system

Abstract

The application discloses a filtering and decoding joint iterative detection method and device for a network control system, and the method comprises the following steps: calculating the prior probability of transmission data according to a control model and a filtering algorithm, and decoding a received signal; updating the prior probability of the transmission data according to a decoding result and the filtering algorithm, and performing iterative detection on the received signal; and updating system output and control input based on an iterative detection result and the filtering algorithm. The filtering and decoding joint iterative detection method and device for the network control system disclosed by the application realize the fusion of communication and control in the bottom layer, can effectively reduce the error probability of transmission by using the prior probability of the control model, and improve the transmission reliability. Moreover, when transmission errors occur, the system output and control can be predicted, retransmission can be avoided, and the transmission delay can be effectively reduced. The application can effectively meet the reliability and delay requirements in an industrial scene.

Description

Technical Field

[0001] This invention relates to the field of control and communication technology, and in particular to a method and apparatus for joint iterative detection of filtering and decoding for network control systems. Background Technology

[0002] In traditional control systems, system components are connected point-to-point. When these components are geographically distributed, this method makes it difficult to build a complete control system, or the cost of building such a system is high. One effective way to solve this problem is to use networked communication, where data exchange between core components such as controlled objects, sensors, and controllers is achieved through network transmission, thereby achieving remote control. Systems based on this connection method are called networked control systems.

[0003] Existing networked control systems primarily focus on research into control systems under conditions of network communication resources, network transmission delays, quantization errors, and network attacks. They neither utilize communication algorithms to improve control performance nor leverage the control system itself to enhance communication performance. Current networked control systems still employ a separate, modular design philosophy, failing to achieve the integration of communication and control. Summary of the Invention

[0004] In view of this, the purpose of this invention is to propose a joint iterative detection method and apparatus for filtering and decoding in network control systems. At the underlying level, it integrates communication and control, effectively reducing the probability of transmission errors by utilizing the prior probabilities of the control model, thus improving transmission reliability. Furthermore, when transmission errors occur, it can predict system output and control, avoiding retransmissions and effectively reducing transmission latency. This invention can effectively meet the reliability and latency requirements in industrial scenarios.

[0005] The technical solution adopted by the present invention to achieve the above objectives is as follows:

[0006] A joint iterative detection method for filtering and decoding in networked control systems includes the following steps:

[0007] 1) Calculate the prior probability of the transmitted data based on the control model and filtering algorithm, and decode the received signal;

[0008] 2) Update the prior probability of the transmitted data based on the decoding result and the filtering algorithm, and perform iterative detection on the received signal;

[0009] 3) Update the system output and control input based on the iterative detection results and filtering algorithm.

[0010] Step 1) includes the following steps:

[0011] 1.1) Determine the state equations of the networked control system;

[0012] 1.2) Based on the state equations, determine the wireless transmission system and channel model;

[0013] 1.3) Calculate the prior probability of the transmitted data, i.e., the maximum likelihood ratio (LLR), based on the control system state equation and filtering algorithm;

[0014] 1.4) Decode the received signal according to LLR.

[0015] Step 1.2) includes the following steps:

[0016] 1.2.1) Determine the quantization method and the transmitted bits of the control system:

[0017] For the system output y k =[y k,1 ,…,y k,d ] T Each element is uniformly quantized using q bits to obtain the quantization result.

[0018]

[0019] Among them, the function v i To quantize the boundary;

[0020] Define quantization index function

[0021]

[0022] Define the binary representation of i as

[0023]

[0024] The control system transmits bit b as

[0025]

[0026] 1.2.2) Encode and modulate the transmitted bits:

[0027] Encode b using the CRC code of (K1, K = dq) to obtain the K1 long sequence s = bG1, where G1 represents the CRC generator matrix and the number of CRC bits is K1-K;

[0028] Encode s using the (N,K1) systematic LDPC code to obtain N-length codeword c = sG2, where G2 represents the generator matrix of the systematic LDPC code, and the overall generator matrix of the encoding scheme is G = G1G2.

[0029] BPSK modulation of c yields the transmitted signal t = 1 - 2c.

[0030] Step 1.3) includes the following steps:

[0031] 1.3.1) Use Kalman filtering on the control system output Prediction is performed to obtain the covariance matrix of the control system output.

[0032] 1.3.2) Calculate the prior probability of the transmitted data based on the predicted control output:

[0033] Let Y k,i The random variable representing the predicted output of the control system. in, express The i-th element on the diagonal, Y k,i The probability density function is p(Y) k,i =y), according to p(Y) k,i =y) estimated Middlebit The LLR is:

[0034]

[0035] Calculate the LLR of the bits in codeword c based on the generator matrix G:

[0036]

[0037] Among them, g i,j This represents the element in the i-th row and j-th column of the generating matrix G.

[0038] Step 1.4) specifically refers to:

[0039] The received signal is r = t + n, n i Let i = 1, 2, ..., N, and n be independent and identically distributed Gaussian white noise, following the rules of... The distribution is obtained by calculating the log-likelihood ratio based on the received signal.

[0040]

[0041] Based on the log-likelihood ratio L(r) i For each i = 1, 2, ..., N, perform BP decoding on the LDPC code to obtain the decoding result. and the corresponding log-likelihood ratio if If the CRC check passes, proceed to step 3; otherwise, proceed to step 2.

[0042] Step 2) includes the following steps:

[0043] 2.1) Initialize the current iteration number i cur =1;

[0044] 2.2) Based on the decoding result Update the prior probability, i.e., the maximum likelihood ratio (LLR);

[0045] 2.3) Iterative detection of the received signal based on the updated LLR.

[0046] Step 2.2) includes the following steps:

[0047] 2.2.1) Calculate the probability of the control system output recovered by the communication system based on the decoding result:

[0048] Based on the decoding results The decoding result of the bits sent by the recovery control system Will Divided into group d The length of each group is q, representing the control system output y. k,i The corresponding decoded bits;

[0049] Based on the log-likelihood ratio Calculate the transmitted bits The probability, i.e. and

[0050] Assume the output of the control system after the communication system is restored is Its probability is and

[0051] 2.2.2) Using Kalman filtering to estimate the system state and output of the control system;

[0052] 2.2.3) Calculate the prior probability based on the output probability and output estimate:

[0053] Based on covariance matrix estimation and control system output use Probability of obtaining and And for all 2 qd kind Perform a traversal to obtain the prior probabilities.

[0054]

[0055] Then, the prior log-likelihood ratio L(c) i )for

[0056]

[0057] Step 2.3) specifically refers to:

[0058] Based on the received signal r = t + n and the prior log-likelihood ratio L(c i ), i = 1, 2, ..., N, update the log-likelihood ratio of the received signal, i.e.

[0059]

[0060] Based on the log-likelihood ratio L(r) i For each i = 1, 2, ..., N, perform BP decoding on the LDPC code to obtain the decoding result. and the corresponding log-likelihood ratio if If the CRC check passes, proceed to step 3); otherwise, determine i. cur Is it less than or equal to the preset number of iterations N? ite If i cur ≤N ite Update i cur =i cur +1, execute 2.2), if i cur >N ite (Execute step 3).

[0061] Step 3) includes the following steps:

[0062] 3.1) If the decoding result If it passes the CRC check, then according to Restore system output The system state and output are estimated using Kalman filtering; otherwise, the system state and output are predicted using system state recursion.

[0063] 3.2) Based on system output For control input u k Update.

[0064] A joint iterative detection system for filtering and decoding of network control systems includes a memory and a processor; the memory is used to store a computer program; the processor is used to implement the joint iterative detection method for filtering and decoding of network control systems when the computer program is executed.

[0065] The present invention has the following beneficial effects and advantages:

[0066] This invention integrates communication and control at the underlying level, effectively reducing the probability of transmission errors and improving transmission reliability by utilizing the prior probabilities of the control model. Furthermore, in the event of a transmission error, it can predict system output and control, avoiding retransmission and effectively reducing transmission latency. This invention effectively meets the reliability and latency requirements of industrial scenarios.

[0067] The method of this invention is particularly suitable for application in control systems in actual industrial scenarios and has good prospects for practical application. Attached Figure Description

[0068] Figure 1 Flowchart of the joint iterative detection method for filtering and decoding in network control systems provided by the present invention;

[0069] Figure 2 This application provides a joint iterative detection device for filtering and decoding in network control systems.

[0070] Figure 3 This is a comparison chart of the bit error rate between the method of the present invention and the traditional communication decoding method;

[0071] Figure 4 This is a comparison diagram of the impact of the method of this invention and the traditional communication decoding method on control trajectory tracking. Detailed Implementation

[0072] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments.

[0073] To make the objectives, technical solutions, and advantages of this disclosure clearer, the following detailed description is provided in conjunction with specific embodiments and the accompanying drawings.

[0074] Unless otherwise defined, the technical or scientific terms used in one or more embodiments of this specification shall have the ordinary meaning as understood by one of ordinary skill in the art to which this disclosure pertains.

[0075] The technical solution of the present invention will now be described in detail with reference to the accompanying drawings.

[0076] like Figure 1 As shown, the joint iterative detection method for filtering and decoding in network control systems includes the following basic steps:

[0077] Step 1: Calculate the prior probability of the transmitted data based on the control model and filtering algorithm, and decode the received signal:

[0078] 1.1) Determine the state equations of the networked control system. A networked control system is described by the following discrete state equations.

[0079]

[0080] Where, x k For the state of an n-dimensional system, u k For m-dimensional control input, y k For the output of a d-dimensional system, and It is independent and identically distributed n-dimensional Gaussian white noise, and W and Z are covariance matrices. A is the n×n system matrix, B is the n×m control input matrix, and C is the d×n output matrix.

[0081] 1.2) Determine the wireless transmission system and channel model.

[0082] 1.2.1) Determine the quantization method and the transmitted bits of the control system. First, determine the system output y. k =[y k,1 ,…,y k,d ] T Each element is uniformly quantized using q bits, with a quantization range of (-V, V) and a step size of [missing value]. The quantization boundary is v0 = -V, v i =v i-1 +Δ, i = 1, 2, ..., 2 q The quantization level is located at the midpoint of the quantization interval, i.e. Quantification results

[0083]

[0084] Among them, the function

[0085] Define quantization index function

[0086]

[0087] Define the binary representation of i as

[0088]

[0089] Therefore, the transmitted bits of the control system are

[0090]

[0091] 1.2.2) Encode and modulate the transmitted bits.

[0092] First, b is encoded using a CRC code of (K1, K = dq) to obtain a K1-length sequence s = bG1, where G1 represents the CRC generator matrix and the number of CRC bits is K1-K. Next, s is encoded using a (N, K1) system LDPC code to obtain an N-length codeword c = sG2, where G2 represents the system LDPC generator matrix. The overall generator matrix of the encoding scheme is G = G1G2. Finally, c is BPSK modulated to obtain the transmitted signal t = 1 - 2c.

[0093] 1.3) Calculate the prior probability of the transmitted data based on the control model and filtering algorithm.

[0094] 1.3.1) Use Kalman filtering to predict the output of the control system.

[0095] Let the initial state of the Kalman filter be... Initial covariance matrix The Kalman filter estimates the system state at time k-1 as follows: The covariance matrix of the system state is estimated as follows: The Kalman filter predicts the state of the control system at time k as follows:

[0096]

[0097] The control system predicts the output as

[0098]

[0099] The covariance matrix of the control system output is

[0100]

[0101] 1.3.2) Calculate the prior probability of the transmitted data based on the predicted control output.

[0102] Let Y k,i The random variable representing the predicted output of the control system. in express The i-th element on the diagonal, Y k,i The probability density function is p(Y) k,i =y). According to p(Y) k,i =y) estimated Middlebit The LLR is

[0103]

[0104] Next, the LLR of the bits in codeword c is calculated based on the generator matrix G, i.e.

[0105]

[0106] Where g represents i,j Generate the element in the i-th row and j-th column of matrix G.

[0107] 1.4) Decode the received signal.

[0108] The received signal is r = t + n, n i Let i = 1, 2, ..., N be independent and identically distributed Gaussian white noise, following the rules of... Distribution. The log-likelihood ratio is calculated based on the received signal, i.e.

[0109]

[0110] Based on the log-likelihood ratio L(r) i For each i = 1, 2, ..., N, perform BP decoding on the LDPC code to obtain the decoding result. and the corresponding log-likelihood ratio if If the CRC check passes, proceed to step 3; otherwise, proceed to step 2.

[0111] Step 2: Update the prior probability of the transmitted data based on the decoding result and filtering algorithm, and perform iterative detection on the received signal:

[0112] 2.1) Initialize the current iteration number i cur =1.

[0113] 2.2) Update the prior probability based on the decoding result.

[0114] 2.2.1) Calculate the restored control system output probability based on the decoding results. The decoding result of the bits sent by the recovery control system Will Divided into group d The length of each group is q, representing the control system output y. k,i The corresponding decoded bits.

[0115] Based on the log-likelihood ratio Calculate the transmitted bits The probability, i.e. and Assume the output of the control system after the communication system is restored is Its probability is and

[0116] 2.2.2) System state and output estimation using Kalman filtering. The Kalman filter estimates the system state at time k-1 as follows: Regarding system status The covariance matrix is ​​estimated as follows The system state recursion based on control input is as follows:

[0117]

[0118] The covariance matrix is ​​recursively derived as follows

[0119]

[0120] The Kalman gain matrix is

[0121]

[0122] The system state estimate based on the system output is:

[0123]

[0124] The covariance matrix is ​​estimated as follows

[0125]

[0126] The control system output obtained by Kalman filtering is

[0127]

[0128] 2.2.3) Calculate the prior probability based on the output probability and the output estimate.

[0129] according to and Obtain using the method in 1.3.2). Further, we can obtain and And for all 2 qd kind Perform a traversal to obtain the prior probabilities.

[0130]

[0131] Then, the prior log-likelihood ratio is

[0132]

[0133] 2.3) Iterative detection of the received signal

[0134] Based on the received signal r = t + n and the prior log-likelihood ratio L(c i ), i = 1, 2, ..., N, update the log-likelihood ratio of the received signal, i.e.

[0135]

[0136] Based on the log-likelihood ratio L(r) i For each i = 1, 2, ..., N, perform BP decoding on the LDPC code to obtain the decoding result. and the corresponding log-likelihood ratio if If the CRC check passes, proceed to step 3; if the CRC check fails, determine i. cur Is it less than or equal to the preset number of iterations N? ite If i cur ≤N ite Update i cur =i cur +1, execute 2.2), if i cur>N ite Proceed to step 3.

[0137] Step 3: Update the system output and control input based on the iterative detection results and filtering algorithm:

[0138] 3.1) If the decoding result It can pass the CRC check, according to Restore system output The system state and output are estimated using Kalman filtering.

[0139] The Kalman filter estimates the system state at time k-1 as follows: Regarding system status The covariance matrix is ​​estimated as follows The system state recursion based on control input is as follows:

[0140]

[0141] The covariance matrix is ​​recursively derived as follows

[0142]

[0143] The Kalman gain matrix is

[0144]

[0145] The system state estimate based on the system output is:

[0146]

[0147] The covariance matrix is ​​estimated as follows

[0148]

[0149] The control system output obtained by Kalman filtering is

[0150]

[0151] 3.2) If the decoding result It is not possible to predict the system state and system output using CRC checksum and system state recursion.

[0152]

[0153] 3.3) Based on system output For control input u k Update.

[0154] The embodiments provided in this application above describe the filtering and decoding joint iterative detection method for network control systems from the perspective of communication and control fusion. It is understood that the above method can be implemented in various network elements. Each network element, such as a terminal, base station, or control node, includes corresponding hardware structures and / or software modules to perform the above functions. Those skilled in the art should readily recognize that, based on the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed by hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0155] Figure 2 The filtering and decoding joint iterative detection device for network control systems provided in this application includes:

[0156] Transceiver S201: Transmitter and receiver, used for sending and receiving signals;

[0157] Controller / processor S202: Used for filtering, decoding, and calculating prior probabilities;

[0158] Memory S203: Used to store calculation results;

[0159] Communication interface S204: Used to support communication between the filtering and decoding joint iterative detection device and other network entities.

[0160] The filtering and decoding joint iterative detection device for network control systems provided in this application can execute the method steps of the previous embodiments, and will not be repeated here.

[0161] The functions of the controller / processor S202 described above can be implemented by circuitry or by executing software code using general-purpose hardware. When the latter is used, the memory S203 is also used to store program code that can be executed by the controller / processor S202. When the controller / processor S202 runs the program code stored in the memory S203, it performs the aforementioned functions.

[0162] Understandable Figure 2 This is just a simplified design of the combined filtering and decoding iterative detection device. In practical applications, the device can include any number of transceivers, controllers / processors, memory, and / or communication interfaces, etc.

[0163] This invention has undergone multiple simulation experiments and applications. The following section details the implementation process and performance analysis of this invention based on the experimental results of the simulation examples:

[0164] Figure 3 This is a comparison chart of the bit error rate between the method of this invention and the traditional communication decoding method. The communication configuration is as follows: code length 64, information bit length 16, CRC length 16, quantization range -10 to 10, and quantization bit number 16.

[0165] The control system model is

[0166]

[0167] The PID control rate is [K] P K I K D ] = [201.5 0.5]. This invention achieves a bit error rate of 10... -3 At that time, there is a performance gain of approximately 1.2 dB.

[0168] Figure 4 This diagram compares the impact of the method of this invention and traditional communication decoding methods on control trajectory tracking. The method of this invention can track the control trajectory better.

[0169] In specific implementations, the aforementioned device can be a terminal or a network-side device. The network-side device can be a base station or a control node.

[0170] The controller / processor of the base station, terminal, or control node described in this application can be a central processing unit (CPU), a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. The processor can also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0171] The steps of the methods or algorithms described in conjunction with the disclosure of this application can be implemented in hardware or by a processor executing software instructions (e.g., program code). The software instructions can consist of corresponding software modules, which can be stored in RAM, flash memory, ROM, EPROM, EEPROM, registers, hard disk, portable hard disk, CD-ROM, or any other form of storage medium known in the art. An exemplary storage medium is coupled to a processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and storage medium can reside in an ASIC. Alternatively, the ASIC can reside in a terminal. Of course, the processor and storage medium can also exist as discrete components in the terminal.

[0172] Those skilled in the art will recognize that, in one or more of the examples above, the functions described in this application can be implemented using hardware, software, firmware, or any combination thereof. When implemented in software, these functions can be stored in a computer-readable medium or transmitted as one or more instructions or code on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein communication media include any medium that facilitates the transfer of a computer program from one place to another. Storage media can be any available medium accessible to a general-purpose or special-purpose computer.

[0173] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of this application. It should be understood that the above description is only a specific embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made on the basis of the technical solution of this application should be included within the scope of protection of this application.

Claims

1. A joint iterative detection method for filtering and decoding in networked control systems, characterized in that, Includes the following steps: 1) Calculate the prior probability of the transmitted data based on the control model and filtering algorithm, and decode the received signal; 2) Update the prior probability of the transmitted data based on the decoding result and the filtering algorithm, and perform iterative detection on the received signal; 3) Update the system output and control input based on the iterative detection results and filtering algorithm; Step 1) includes the following steps: 1.1) Determine the state equations of the networked control system; 1.2) Based on the state equations, determine the wireless transmission system and channel model; 1.3) Calculate the prior probability of the transmitted data, i.e., the maximum likelihood ratio (LLR), based on the control system state equation and filtering algorithm; 1.4) Decode the received signal according to the LLR; Step 2) includes the following steps: 2.1) Initialize the current iteration count ; 2.2) Based on the decoding result Update the prior probability, i.e., the maximum likelihood ratio (LLR); 2.3) Iterative detection of the received signal based on the updated LLR; Step 3) includes the following steps: 3.1) If the decoding result If it passes the CRC check, then according to Restore system output The system state and output are estimated using Kalman filtering; otherwise, the system state and output are predicted using system state recursion. 3.2) Based on system output For control input Update.

2. The joint iterative detection method for filtering and decoding in network control systems according to claim 1, characterized in that, Step 1.2) includes the following steps: 1.2.1) Determine the quantization method and the transmitted bits of the control system: For system output Perform each element Bit uniform quantization yields the quantization result. : Among them, the function , To quantize the boundary; Define quantization index function definition The binary representation of is Control system transmit bits for ； 1.2.2) Encode and modulate the transmitted bits: use CRC code pair Encode to obtain Long sequences ,in, This represents the CRC generation matrix, with the number of CRC bits being... ; use System LDPC code pair Encode to obtain Long code ,in, This represents the generator matrix of the system's LDPC code. The overall generator matrix of the encoding scheme is: ; right BPSK modulation is performed to obtain the transmitted signal. .

3. The joint iterative detection method for filtering and decoding in network control systems according to claim 1, characterized in that, Step 1.3) includes the following steps: 1.3.1) Using Kalman filtering for the control system output Prediction is performed to obtain the covariance matrix of the control system output. ; 1.3.2) Calculate the prior probability of the transmitted data based on the predicted control output: set up The random variable representing the predicted output of the control system. ,in, express The diagonal One element, The probability density function is ,according to estimated Middlebit The LLR is: Based on the generator matrix Calculate codewords Medium-bit LLR: in, Represents the generating matrix The first in Line 1 Column elements.

4. The joint iterative detection method for filtering and decoding in network control systems according to claim 1, characterized in that, Step 1.4) specifically refers to: The received signal is , , It is independent and identically distributed Gaussian white noise, which follows the rules of... The distribution is obtained by calculating the log-likelihood ratio based on the received signal. Based on the log-likelihood ratio BP decoding of the LDPC code is performed to obtain the decoding result. and the corresponding log-likelihood ratio ,if If the CRC check passes, proceed to step 3; otherwise, proceed to step 2.

5. The joint iterative detection method for filtering and decoding in network control systems according to claim 1, characterized in that, Step 2.2) includes the following steps: 2.2.1) Calculate the probability of the control system output that the communication system recovers based on the decoding results: Based on the decoding results The decoding result of the bits sent by the recovery control system ,Will Divided into Group The length of each group is Indicates the output of the control system The corresponding decoded bits; Based on the log-likelihood ratio Calculate the transmitted bits The probability of that, i.e. and ; Assume the output of the control system after the communication system is restored is Its probability is ,and ; 2.2.2) Using Kalman filtering to estimate the system state and output of the control system; 2.2.3) Calculate the prior probability based on the output probability and output estimate: Based on covariance matrix estimation and control system output ,use To obtain the probability and and for all kind Perform a traversal to obtain the prior probabilities. Then, the prior log-likelihood ratio is... for 。 6. The joint iterative detection method for filtering and decoding in network control systems according to claim 1, characterized in that, Step 2.3) specifically refers to: According to the received signal Compared with the prior log-likelihood Update the log-likelihood ratio of the received signal, i.e. Based on the log-likelihood ratio BP decoding of the LDPC code is performed to obtain the decoding result. and the corresponding log-likelihood ratio ,if If the CRC check passes, proceed to step 3); otherwise, determine... Is it less than or equal to the preset number of iterations? ,if ,renew (Execute 2.2), if (Execute step 3).

7. A joint iterative detection system for filtering and decoding in network control systems, characterized in that, It includes a memory and a processor; the memory is used to store a computer program; the processor is used to implement, when executing the computer program, the filtering and decoding joint iterative detection method for network control systems as described in claims 1-6.

Images