Industrial Internet of Things bidirectional information transmission and full-loop control collaborative optimization method

Abstract

The invention provides an industrial Internet of Things bidirectional information transmission and full loop control collaborative optimization method, which belongs to the technical field of wireless communication, and comprises the following steps: constructing a bidirectional information transmission architecture based on a sensor, a controller and an actuator, and introducing a multicast transmission mode into a controller-actuator downlink, the utilization efficiency of downlink transmission resources is improved; performing joint optimization on time-frequency resources of uplink transmission and downlink multicast packet transmission, and constructing a problem of total cost minimization of the industrial Internet of Things system; and solving the total cost minimization problem of the industrial Internet of Things system based on a transmission scheduling joint optimization algorithm to obtain a time-frequency resource allocation scheme and the minimum total cost of the industrial Internet of Things system. According to the invention, the resource allocation can be adaptively adjusted according to the channel condition of the industrial environment and the scale of the sensor network, and the overhead of the control system is effectively reduced.

Description

Technical Field

[0001] The present invention relates to the field of wireless communication technology, and in particular to a method for collaborative optimization of bidirectional information transmission and full-loop control of an industrial Internet of Things. Background Art

[0002] Industrial network systems are two-way communication systems deployed in industrial production environments to support real-time monitoring and control of production processes. Typically, industrial network systems consist of spatially distributed sensors, controllers, and actuators paired with sensors. The more state information delivered to the controller and the more control instructions delivered to the actuator, the better the estimation and control performance. However, there are two major challenges in industrial environments. First, with limited transmission resources, concurrent transmission of multiple subsystems can lead to network congestion; second, complex electromagnetic environments and interference lead to poor channel conditions. Both of these problems will increase information transmission delays, thereby reducing the reliability of data transmission, which in turn reduces the estimation and control performance of the system and wastes communication resources.

[0003] In the field of control, many existing studies are devoted to designing estimation or control algorithms to mitigate the impact of wireless transmission on data transmission reliability, such as proposing an energy-aware sensor scheduling and opportunistic transmission scheme to balance estimation performance and resource consumption; proposing a sensor scheduler for state estimation of heterogeneous systems over constrained and lossy wireless channels, which reduces the computation time while achieving near-optimal estimation performance.

[0004] However, state estimation algorithms or control strategies can only passively mitigate the impact of complex wireless channels on system performance, and cannot guarantee the convergence of estimation and the stability of control. Therefore, it is crucial to find an effective information transmission scheme to improve transmission reliability.

[0005] To sum up, the problems with the existing technology are: due to the complex and changeable channel conditions in industrial environments, only designing state estimation algorithms or control strategies without considering the time-varying nature of the channel is not always effective in improving system performance; at the same time, a more complete information transmission scheme needs to be formulated to actively improve the transmission reliability of the network, and whether the use of multicast for downlink transmission can increase the information transmission rate to reduce transmission energy consumption and control instruction transmission errors, thereby improving the overall performance of the industrial network system, remains to be verified. Summary of the invention

[0006] According to the technical problems raised above, a method for collaborative optimization of bidirectional information transmission and full-loop control of industrial Internet of Things is provided, comprising the following steps:

[0007] S1. Build a two-way information transmission architecture based on sensor-controller-actuator, and introduce multicast transmission in the controller-actuator downlink to improve the utilization efficiency of downlink transmission resources;

[0008] S2, jointly optimize the time and frequency resources of uplink transmission and downlink multicast packet transmission to construct the total cost minimization problem of the industrial Internet of Things system;

[0009] S3. Based on the transmission scheduling joint optimization algorithm, the total cost minimization problem of the industrial Internet of Things system is solved to obtain the allocation plan of time and frequency resources and the minimum total cost of the industrial Internet of Things system.

[0010] Furthermore, the specific process of constructing the bidirectional information transmission architecture based on sensor-controller-actuator is as follows:

[0011] S11. In the uplink, orthogonal frequency division multiple access technology is used for information transmission, and a binary variable θ is used to represent the subcarrier allocation. n，s (k) = 1 means that subcarrier s is assigned to sensor n in control cycle k, otherwise, θ n，s (k) = 0. In the absence of co-channel interference, the signal-to-noise ratio of sensor n on subcarrier s in control period k is:

[0012]

[0013] Among them, p n (k) is the transmission power of sensor n, h n，s (k) is the channel gain of sensor n when transmitting on subcarrier s, σ 2 is the noise power, is a collection of sensors, is a set of subcarriers;

[0014] S12, the achievable transmission rate of sensor n on subcarrier s is expressed as:

[0015]

[0016] in, is the bandwidth of each subcarrier, B is the transmission bandwidth of the entire channel, and S is the number of subcarriers;

[0017] S13, the uplink transmission delay from sensor n to the wireless access point is expressed as:

[0018]

[0019] in, represents the sensing data packet size of sensor n,

[0020] When the transmission delay is greater than the delay constraint of the uplink transmission, sensor n does not transmit data in control period k.

[0021] Furthermore, the specific process of introducing a multicast transmission mode in the controller-actuator downlink to improve the utilization efficiency of downlink transmission resources is as follows:

[0022] In the downlink, multicast transmission is adopted; N executors are divided into M groups. Indicates that executor n is assigned to multicast group m, otherwise, The set of executors assigned to group m is denoted by The signal-to-noise ratio of actuator n in group m is expressed as:

[0023]

[0024] Among them, h n (k) is the channel gain of actuator n, q m (k) is the power allocated by the wireless access point to group m, is a set of multicast groups;

[0025] The achievable transmission rate of multicast group m is:

[0026]

[0027] in, b m (k) is the transmission bandwidth of multicast group m, is the set of executors in group m;

[0028] The transmission delay of multicast group m is expressed as:

[0029]

[0030] in, d n (k) represents the size of the data packet received by executor n.

[0031] The transmission rate of executor n in multicast group m is:

[0032]

[0033] The achievable transmission rate from the wireless access point to actuator n is:

[0034]

[0035] Therefore, the downlink transmission delay from the wireless access point to the executor n is:

[0036]

[0037] Furthermore, the time-frequency resources of the uplink transmission and the downlink multicast packet transmission are optimized, and the specific process of constructing the total cost minimization problem of the industrial Internet of Things system is as follows:

[0038] The state estimation error of the uplink at control cycle k quantifies the state value x of the industrial Internet of Things system n (k) and state estimates The difference between is expressed as:

[0039]

[0040] The control command error of the downlink at control cycle k quantifies the control command calculated by the controller The control command received by actuator n The difference between is expressed as:

[0041]

[0042] The system control cost of control period k is expressed as:

[0043]

[0044] in, Expressing hope;

[0045] The system communication cost of control cycle k is expressed as:

[0046]

[0047] Where E1(k) is the energy consumed by uplink transmission, and E2(k) is the energy consumed by downlink transmission;

[0048] The division of delay constraints in the two stages of uplink transmission and downlink transmission, subcarrier allocation, sensor transmit power, multicast group division, downlink bandwidth and power allocation of wireless access points are jointly optimized to obtain the minimization problem of the total cost of the industrial Internet of Things system. The expression of the minimization problem of the total cost of the industrial Internet of Things system is as follows;

[0049]

[0050] Where K is the number of control cycles and β is the weighting factor. max and q max Represent the maximum transmission power of the sensor and the wireless access point respectively, C1 and C2 represent the matching relationship between the subcarrier and the sensor; C3, C4 and C5 are the transmission bandwidth and power constraints; C6 means that each actuator can only be assigned to one multicast group; C7 limits the number of multicast groups; C8, C9 and C10 describe the feasible area of ​​the variable.

[0051] Furthermore, the downlink multicast packet transmission sub-problem is as follows;

[0052] The downlink transmission sub-problem includes multicast group division, bandwidth allocation and power allocation, which can be expressed as:

[0053]

[0054] Among them, β2 is a positive weighting factor;

[0055] The first half of the objective function of subproblem SP2 is the mean square value of the control instruction error, and the calculation is simplified as follows:

[0056]

[0057] in, is the probability that the control command transmitted to actuator n is successfully transmitted.

[0058] Subproblem SP2 is rewritten as follows:

[0059]

[0060] Further: the downlink multicast packet transmission problem is solved by using a multicast resource allocation algorithm, which is as follows:

[0061] In the first step, the multicast grouping problem of the actuators is solved by fixing the communication resource allocation;

[0062] In the second step, bandwidth and power allocation are obtained by solving the resource allocation problem for the given grouping results.

[0063] Further: after the uplink transmission sub-problem and the downlink multicast packet transmission sub-problem are solved, the value of η is searched and updated in the interval 0 to 1 so that the objective function value of the original problem P0 is minimized.

[0064] Further: the specific process of solving the multicast grouping problem of the actuator by allocating fixed communication resources is as follows:

[0065] First, consider the multicast packet problem. Assuming that the downlink transmission bandwidth and the wireless access point transmission power are evenly distributed to each actuator, the objective function of problem SP2.1 is analyzed and it is known that the maximum This is equivalent to minimizing the following objective:

[0066]

[0067] in, is a constant, and Can be rewritten as an actuator point and the center point of group m The distance between

[0068] The multicast grouping problem is equivalent to a clustering problem. In this problem, given N actuator points ε1, ε2, ..., ε N , the goal is to divide these points into M clusters The multicast packet problem is expressed as:

[0069]

[0070] First, determine the value of M. The silhouette coefficient is introduced as an internal indicator to evaluate the clustering effect, which aims to describe the similarity between the target and its group and other groups. The silhouette coefficient calculation formula of sample i is:

[0071]

[0072] Among them, F(i) and R(i) represent cohesion and separation respectively, which reflect the closeness of the sample point to the elements within the group and the elements outside the group. The calculation formula is:

[0073]

[0074] Among them, m′ is the group other than the group where sample i belongs, N m and N m′ denote the number of executors in groups m and m′ respectively, is the set of actuators in group m′, and dist(i, j) represents the distance between sample i and sample j.

[0075] The average silhouette coefficient of the clustering results is expressed as:

[0076]

[0077] Where N is the number of actuators;

[0078] Statistically speaking, there is an optimal value M for the number of groups. * , which is less than Therefore, the initial value of M is set to in, Indicates the rounding up operation. First, by looping through the number of groups, the group with the largest average silhouette coefficient value is selected as M * .

[0079] Further: the process of obtaining bandwidth and power allocation by solving the resource allocation problem of a given grouping result is as follows:

[0080] The improved K-means++ algorithm is used to group the executors. After completing the multicast grouping, first, the downlink bandwidth allocation is optimized. Assuming that the entire bandwidth B is allocated to group m, the transmission delay of group m is expressed as:

[0081]

[0082] Among them, r m，B (k) is the transmission rate when the entire bandwidth is allocated to group m;

[0083] When the bandwidth ρ m When B is allocated to group m, the bandwidth allocation problem is simplified to:

[0084]

[0085] Among them, ρ m (k) is the proportional coefficient of the bandwidth allocated to group m;

[0086] Given the bandwidth allocation result, the allocation problem of wireless access point transmit power is expressed as:

[0087]

[0088] From the analysis, we can see that the above two problems are convex optimization problems, which are solved by convex optimization solver, and then the solution of problem SP2 is obtained by alternately optimizing the bandwidth and the transmission power allocation of the wireless access point.

[0089] The present invention includes a method for collaborative optimization of bidirectional information transmission and full-loop control of the industrial Internet of Things: constructing a bidirectional information transmission architecture based on sensors, controllers, and actuators, and introducing multicast transmission technology in the controller-actuator downlink; jointly optimizing the time and frequency resources of uplink transmission and downlink multicast packet transmission, and constructing a system total cost minimization problem; proposing a transmission scheduling joint optimization algorithm to solve the optimization problem, wherein the multicast resource allocation algorithm for downlink transmission problems can effectively reduce the system cost of downlink transmission. Compared with the prior art, the present invention has the following advantages:

[0090] The method for collaborative optimization of bidirectional information transmission and full-loop control of the industrial Internet of Things provided by the present invention can adaptively adjust the time allocation of uplink transmission and downlink transmission according to the dynamic requirements of the industrial network system for the reliability of transmission of perception data and control instructions; and can further improve the reliability of downlink control instruction transmission and reduce energy consumption by using multicast transmission technology. Through the joint optimization of communication and control, the reliability of data transmission in the industrial network system is effectively improved, and the overall performance of the industrial network system under resource constraints is improved, including control cost and communication cost.

[0091] Based on the above reasons, the present invention can be widely promoted in fields such as wireless communication. BRIEF DESCRIPTION OF THE DRAWINGS

[0092] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0093] Figure 1 is a flow chart of the method of the present invention;

[0094] Figure 2 A network scene graph used in an embodiment of the present invention;

[0095] Figure 3 A comparison chart of total system costs of different methods when the number of random simulations provided by an embodiment of the present invention changes;

[0096] Figure 4 A comparison chart of the total system cost of different methods when the number of sensors provided by the embodiment of the present invention changes;

[0097] Figure 5 A comparison chart of total system costs of different methods when the uplink transmission bandwidth changes according to an embodiment of the present invention. DETAILED DESCRIPTION

[0098] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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 should fall within the scope of protection of the present invention.

[0099] like Figure 1 As shown, an embodiment of the present invention provides an industrial Internet of Things two-way information transmission and full-loop control collaborative optimization method, including the following steps:

[0100] S1. Build a two-way information transmission architecture based on sensor-controller-actuator, and introduce multicast transmission in the controller-actuator downlink to improve the utilization efficiency of downlink transmission resources, that is, reduce resource waste, have strong flexibility, can adapt to the ever-changing network environment and user needs, and have relatively high security;

[0101] S2, jointly optimize the time and frequency resources of uplink transmission and downlink multicast packet transmission to construct the total cost minimization problem of the industrial Internet of Things system;

[0102] S3. Based on the transmission scheduling joint optimization algorithm, the total cost minimization problem of the industrial Internet of Things system is solved to obtain the allocation plan of time and frequency resources and the minimum total cost of the industrial Internet of Things system.

[0103] Steps S1 / S2 / S3 are executed sequentially;

[0104] Figure 2 A network scenario diagram used in an embodiment of the present invention.

[0105] In this industrial network system, N sensors are deployed to monitor the operating status of the equipment, and the sensors transmit the collected status information to the controller. Then, the controller performs state estimation based on the received information to infer the operating status of the system, and passes the calculated control instructions to the actuator. Finally, the actuator performs the corresponding operation according to the received control instructions. In the scenario considered, the uplink adopts orthogonal frequency division multiple access technology for transmission; the downlink adopts multicast packet transmission, which can further reduce the system cost compared with broadcast and unicast.

[0106] Furthermore, the construction is based on a two-way information transmission architecture of sensor-controller-actuator, and the multicast transmission technology is introduced in the controller-actuator downlink. The specific process is as follows:

[0107] S11. In the uplink, orthogonal frequency division multiple access technology is used for information transmission. The binary variable θ is used to represent the subcarrier allocation. θ n，s (k) = 1 means that subcarrier s is assigned to sensor n in control cycle k, otherwise, θ n，s (k) = 0. In the absence of co-channel interference, the signal-to-noise ratio of sensor n on subcarrier s in control period k is:

[0108]

[0109] Among them, p n (k) is the transmission power of sensor n, h n，s (k) is the channel gain of sensor n when transmitting on subcarrier s, σ 2 is the noise power, is a collection of sensors, is a set of subcarriers;

[0110] S12, the achievable transmission rate of sensor n on subcarrier s can be expressed as:

[0111]

[0112] in, is the bandwidth of each subcarrier, B is the transmission bandwidth of the entire channel, and S is the number of subcarriers;

[0113] S13, the uplink transmission delay from sensor n to the wireless access point is expressed as:

[0114]

[0115] in, represents the sensing packet size of sensor n.

[0116] When the transmission delay is greater than the delay constraint of the uplink transmission, sensor n does not perform data transmission in the control period k, further reducing the transmission energy consumption;

[0117] When the transmission delay is less than or equal to the delay constraint of the uplink transmission, the sensor transmits the perception data to the controller. The controller receives the data and estimates the state of the industrial Internet of Things system and calculates the control instructions.

[0118] The specific process of introducing a multicast transmission mode in the controller-actuator downlink to improve the utilization efficiency of downlink transmission resources is as follows:

[0119] S14. In the downlink, multicast transmission is adopted. N actuators are divided into M groups. Indicates that executor n is assigned to multicast group m, otherwise, The set of executors assigned to group m is denoted by The signal-to-noise ratio of actuator n in group m is expressed as:

[0120]

[0121] Among them, h n (k) is the channel gain of actuator n, q m (k) is the power allocated by the wireless access point to group m, is a set of multicast groups;

[0122] S15. The achievable transmission rate of multicast group m is:

[0123]

[0124] in, b m (k) is the transmission bandwidth of multicast group m, is the set of executors in group m;

[0125] S16. The transmission delay of multicast group m is expressed as:

[0126]

[0127] in, dn (k) represents the size of the data packet received by executor n.

[0128] S17. The transmission rate of executor n in multicast group m is:

[0129]

[0130] S18, the achievable transmission rate from the wireless access point to actuator n is:

[0131]

[0132] Therefore, the downlink transmission delay from the wireless access point to the executor n is:

[0133]

[0134] Furthermore, the time-frequency resources of the uplink transmission and the downlink multicast packet transmission are optimized, and the specific process of constructing the total cost minimization problem of the industrial Internet of Things system is as follows:

[0135] S21, the state estimation error of the uplink at control period k, quantifies the system state value x n (k) and state estimates The difference between is expressed as:

[0136]

[0137] S22, the control command error of the downlink at control cycle k, quantifies the control command calculated by the controller The control command received by actuator n The difference between is expressed as:

[0138]

[0139] S23, the system control cost of control cycle k is expressed as:

[0140]

[0141] in, Expressing hope;

[0142] S24, the system communication cost of control cycle k is expressed as:

[0143]

[0144] Wherein, E1(k) is the energy consumed by uplink transmission, and E2(k) is the energy consumed by downlink transmission.

[0145] Through reasonable resource allocation and optimization strategies, the total cost can be reduced while ensuring system performance. That is, the goal of this minimization problem is the total cost of the system. What needs to be done is to jointly optimize the time-frequency resources to achieve the purpose of minimizing the total cost of the system. The total cost of the minimized system is obtained by jointly optimizing the division of two-stage delay constraints, subcarrier allocation, sensor transmission power, multicast group division, and downlink bandwidth and power allocation of wireless access points. The expression of the total cost of the minimized system is as follows:

[0146]

[0147] Where K is the number of control cycles and β is the weighting factor. max and q max Represent the maximum transmission power of the sensor and the wireless access point respectively. C1 and C2 represent the matching relationship between the subcarrier and the sensor; C3, C4 and C5 are the transmission bandwidth and power constraints; C6 indicates that each actuator can only be assigned to one multicast group; C7 limits the number of multicast groups; C8, C9 and C10 describe the feasible region of the variable.

[0148] The transmission scheduling joint optimization algorithm includes:

[0149] Furthermore, the transmission scheduling joint optimization algorithm is used to solve the total cost minimization problem of the industrial Internet of Things system, and the allocation scheme of time-frequency resources and the minimum total cost of the industrial Internet of Things system are obtained. The specific process is as follows:

[0150] S31, fix the value of the two-stage delay constraint partition point η(k) for uplink transmission and downlink multicast packet transmission, and express the uplink transmission subproblem in control period k as follows:

[0151]

[0152] Among them, β1 is a positive weighting factor.

[0153] S32, sub-problem SP1 can be solved by alternately optimizing subcarrier allocation θ and sensor transmit power p. Among them, the subcarrier optimization problem is a binary linear programming problem, which is solved using the KM algorithm; the objective function of the sensor transmit power optimization problem is a unimodal function about power first decreasing and then increasing, so it can be solved using the dichotomy method;

[0154] S33, after sub-problem SP1 is solved, the transmission status of uplink sensing data can be obtained, the sensors scheduled in control period k can be obtained, and the subcarriers can be redistributed to the scheduled sensors;

[0155] S34, the downlink multicast packet transmission sub-problem includes multicast group division, bandwidth allocation and power allocation, which is expressed as:

[0156]

[0157] Among them, β2 is a positive weighting factor.

[0158] S35. The first half of the objective function of problem SP2 is the mean square value of the control instruction error, which can be calculated and simplified as:

[0159]

[0160] in, is the probability that the control command transmitted to actuator n is successfully transmitted.

[0161] S36. Question SP2 can be rewritten as:

[0162]

[0163] This problem is a mixed integer nonlinear programming problem and is difficult to solve. A multicast resource allocation algorithm is proposed to solve it. In the first step, the multicast grouping problem of the actuators is solved by fixing the communication resource allocation. In the second step, the bandwidth and power allocation are obtained by solving the resource allocation problem for the given grouping results.

[0164] S37. After the uplink transmission subproblem SP1 and the downlink multicast packet transmission subproblem SP2 are solved, search and update the value of η in the interval 0 to 1 so that the objective function value of the original problem P0 is minimized.

[0165] The downlink transmission problem is solved by using a multicast resource allocation algorithm, which can effectively reduce the system cost of downlink transmission. The specific process is as follows:

[0166] S341. First, consider the multicast packet problem. Assume that the downlink transmission bandwidth and the wireless access point transmission power are evenly distributed to each actuator. Analyzing the objective function of problem SP2.1, we know that we need to maximize This is equivalent to minimizing the following objective:

[0167]

[0168] in, is a constant, and Can be rewritten as an actuator point and the center point of group m The distance between.

[0169] S342, the multicast grouping problem can be equivalent to a clustering problem. In this problem, given N actuator points ε1, ε2, ..., εN , the goal is to divide these points into M clusters The multicast packet problem can be expressed as:

[0170]

[0171] S343. First, determine the value of M. The silhouette coefficient is introduced as an internal indicator for evaluating clustering effect, which aims to describe the similarity between the target and its group and other groups. The silhouette coefficient calculation formula of sample i is:

[0172]

[0173] Among them, F(i) and R(i) represent cohesion and separation respectively, which reflect the closeness of the sample point to the elements within the group and the elements outside the group. The calculation formula is:

[0174]

[0175] Among them, m′ is the group other than the group where sample i belongs, N m and N m′ denote the number of executors in groups m and m′ respectively, is the set of actuators in group m′, and dist(i, j) represents the distance between sample i and sample j.

[0176] S344. The average silhouette coefficient of the clustering result is expressed as:

[0177]

[0178] Where N is the number of actuators.

[0179] S345. From a statistical point of view, there is an optimal value M for the number of groups. * , which is less than Therefore, the initial value of M is set to in, Indicates the rounding up operation. First, by looping through the number of groups, the group with the largest average silhouette coefficient value is selected as M * .

[0180] S346. Then, determine the grouping of the executors. Since the K-means algorithm has the advantages of fast convergence and good clustering results, it can be used to solve clustering problems. K-means++ improves the method of selecting the initial cluster center of K-means, that is, points farther away from the current cluster center have a higher probability of being selected as the next cluster center, rather than random selection. Therefore, the improved K-means++ algorithm is used to group the executors. After completing the multicast group division, first, optimize the downlink bandwidth allocation. Assuming that the entire bandwidth B is allocated to group m, the transmission delay of group m is expressed as:

[0181]

[0182] Among them, r m，B (k) is the transmission rate when the entire bandwidth is allocated to group m;

[0183] S347, downlink control error is only affected by the probability of successful transmission of downlink control command. Therefore, when bandwidth ρ m When B is allocated to group m, the bandwidth allocation problem can be simplified to:

[0184]

[0185] Among them, ρ m (k) is the proportional coefficient of the bandwidth allocated to group m.

[0186] S348. Given the bandwidth allocation result, the allocation problem of the wireless access point transmission power can be expressed as:

[0187]

[0188] Analysis shows that the above two problems are convex optimization problems, which can be solved by convex optimization solvers. Then, by alternately optimizing the bandwidth and the transmission power allocation of the wireless access point, the solution to problem SP2 is obtained.

[0189] In order to verify the effectiveness of the method of the present invention, the application effect of the present invention is described in detail below in combination with simulation.

[0190] Simulation conditions

[0191] In the simulation scenario, an industrial production workshop with a coverage range of 50m*200m rectangular area is considered, and 10 pairs of sensors and actuators are randomly arranged within the range. The channel transmission bandwidth is 1.4MHz, the maximum transmission power of the sensor is 16mW, and the maximum transmission power of the wireless access point is 64mW. The Gaussian white noise power in the industrial environment is -42dBm, the average size of the perception data packet of each sensor is 6Kbits, the control instruction data packet of each actuator ranges from 450 to 1100bits, and the length of a control cycle is 100ms.

[0192] Simulation content and result analysis

[0193] The effectiveness of the method of the present invention is verified by comparing it with the following three information transmission methods.

[0194] Comparative method 1: uplink bandwidth allocation transmission method, in which frequency domain resource allocation for uplink transmission adopts bandwidth allocation instead of subcarrier allocation.

[0195] Comparative method 2: Downlink broadcast transmission method, this method uses broadcast instead of multicast to transmit downlink control instructions.

[0196] Comparative method 3: Downlink unicast transmission method, this method uses unicast instead of multicast to transmit downlink control instructions.

[0197] Simulation 1: Performance comparison of different approaches in terms of total system cost.

[0198] from Figure 3 It can be seen that the proposed method shows the best performance in terms of total system cost, which is 15% and 90% lower than that of comparison method 3 and comparison method 2, respectively. This advantage is mainly attributed to the multicast transmission mechanism of the downlink. In addition, compared with comparison method 1, the total system cost of the proposed method is reduced by 68%, indicating the effectiveness of the uplink transmission subcarrier allocation scheme. Compared with the three comparison methods, the performance improvement of the proposed method is attributed to its adaptability to the dynamic characteristics of the system. In addition, the proposed method also realizes efficient control scheduling and coordinated transmission of uplink and downlink information, effectively alleviating the impact of wireless resource scarcity on the performance of the control system.

[0199] Simulation 2: Comparative analysis of system total cost curves based on different methods when the number of sensors changes.

[0200] from Figure 4It can be seen that as the number of sensors increases, the total system cost of the proposed method is always lower than that of the three comparison methods, and the increase is significantly smaller than that of comparison method 3 and comparison method 1. This is because if the system consists of more sensors and controllers, more state information needs to be transmitted to the controller and actuator during the information transmission process, resulting in higher communication costs. In addition, the control cost also increases with the increase in the number of sensors and actuators, because it is the sum of the state estimation errors of all sensors and the control command errors of all actuators due to unsuccessful control command transmission.

[0201] Simulation 3: Comparative analysis of system total cost curves based on different methods when the transmission bandwidth changes.

[0202] from Figure 5 It can be seen that when the number of sensors remains unchanged, as the transmission bandwidth increases, each sensor can be allocated more frequency band resources, thereby improving the information transmission rate, reducing transmission delay, and increasing the probability of successful information transmission. The total system cost of the four methods decreases with the increase of transmission bandwidth, but the total system cost of the proposed method is always less than that of the other three comparison methods.

[0203] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for collaborative optimization of bidirectional information transmission and full-loop control of industrial Internet of Things, characterized in that: The steps include: S1. Build a two-way information transmission architecture based on sensor-controller-actuator, and introduce multicast transmission in the controller-actuator downlink to improve the utilization efficiency of downlink transmission resources; S2, jointly optimize the time and frequency resources of uplink transmission and downlink multicast packet transmission to construct the total cost minimization problem of the industrial Internet of Things system; S3. Based on the transmission scheduling joint optimization algorithm, the total cost minimization problem of the industrial Internet of Things system is solved to obtain the allocation plan of time and frequency resources and the minimum total cost of the industrial Internet of Things system.

2. The method for collaborative optimization of bidirectional information transmission and full-loop control of industrial Internet of Things according to claim 1 is characterized in that: The specific process of constructing a two-way information transmission architecture based on sensor-controller-actuator is as follows: S11. In the uplink, orthogonal frequency division multiple access technology is used for information transmission, and a binary variable θ is used to represent the subcarrier allocation. n，s (k) = 1 means that subcarrier s is assigned to sensor n in control cycle k, otherwise, θ n，s (k) = 0. In the absence of co-channel interference, the signal-to-noise ratio of sensor n on subcarrier s in control period k is: Among them, p n (k) is the transmission power of sensor n, h n，s (k) is the channel gain of sensor n when transmitting on subcarrier s, σ 2 is the noise power, is a collection of sensors, is a set of subcarriers; S12, the achievable transmission rate of sensor n on subcarrier s is expressed as: in, is the bandwidth of each subcarrier, B is the transmission bandwidth of the entire channel, and S is the number of subcarriers; S13, the uplink transmission delay from sensor n to the wireless access point is expressed as: in, represents the sensing data packet size of sensor n, When the transmission delay is greater than the delay constraint of the uplink transmission, sensor n does not transmit data in control period k.

3. The method for collaborative optimization of bidirectional information transmission and full-loop control of industrial Internet of Things according to claim 1 is characterized in that: The specific process of introducing a multicast transmission mode in the controller-actuator downlink to improve the utilization efficiency of downlink transmission resources is as follows: In the downlink, multicast transmission is adopted; N executors are divided into M groups. Indicates that executor n is assigned to multicast group m, otherwise, The set of executors assigned to group m is denoted by The signal-to-noise ratio of actuator n in group m is expressed as: Among them, h n (k) is the channel gain of actuator n, q m (k) is the power allocated by the wireless access point to group m, is a set of multicast groups; The achievable transmission rate of multicast group m is: in, b m (k) is the transmission bandwidth of multicast group m, is the set of executors in group m; The transmission delay of multicast group m is expressed as: in, d n (k) represents the size of the data packet received by executor n; The transmission rate of executor n in multicast group m is: The achievable transmission rate from the wireless access point to actuator n is: Therefore, the downlink transmission delay from the wireless access point to the executor n is:

4. The method for collaborative optimization of bidirectional information transmission and full-loop control of industrial Internet of Things according to claim 1 is characterized in that: The specific process of jointly optimizing the time and frequency resources of the uplink transmission and the downlink multicast packet transmission and constructing the total cost minimization problem of the industrial Internet of Things system is as follows: The state estimation error of the uplink at control cycle k quantifies the state value x of the industrial Internet of Things system n (k) and state estimates The difference between is expressed as: The control command error of the downlink at control cycle k quantifies the control command calculated by the controller The control command received by actuator n The difference between is expressed as: The system control cost of control period k is expressed as: in, Expressing hope; The system communication cost of control cycle k is expressed as: Where E1(k) is the energy consumed by uplink transmission, and E2(k) is the energy consumed by downlink transmission; The division of delay constraints in the two stages of uplink transmission and downlink transmission, subcarrier allocation, sensor transmit power, multicast group division, downlink bandwidth and power allocation of wireless access points are jointly optimized to obtain the minimization problem of the total cost of the industrial Internet of Things system. The expression of the minimization problem of the total cost of the industrial Internet of Things system is as follows: s.t.C1: C2: C3: C4: C5: C6: C7: C8: C9: C10:η(k)∈(0,1) Among them, K is the number of control cycles, β is the weighting factor, and p max and q max Represent the maximum transmission power of the sensor and the wireless access point respectively, C1 and C2 represent the matching relationship between the subcarrier and the sensor; C3, C4 and C5 are the transmission bandwidth and power constraints; C6 means that each actuator can only be assigned to one multicast group; C7 limits the number of multicast groups; C8, C9 and C10 describe the feasible area of ​​the variable.

5. The method for collaborative optimization of bidirectional information transmission and full-loop control of industrial Internet of Things according to claim 1 is characterized in that: The downlink multicast packet transmission sub-problem is as follows; The downlink transmission sub-problem includes multicast group division, bandwidth allocation and power allocation, which can be expressed as: stC4-C7,C9 Among them, β2 is a positive weighting factor; The first half of the objective function of subproblem SP2 is the mean square value of the control instruction error, and the calculation is simplified as follows: in, is the probability that the control command transmitted to actuator n is successfully transmitted; Subproblem SP2 is rewritten as follows: stC4-C7,C9.

6. The method for collaborative optimization of bidirectional information transmission and full-loop control of industrial Internet of Things according to claim 1 is characterized in that: The downlink multicast packet transmission problem is solved by using a multicast resource allocation algorithm, which is as follows: In the first step, the multicast grouping problem of the actuators is solved by fixing the communication resource allocation; In the second step, bandwidth and power allocation are obtained by solving the resource allocation problem for the given grouping results.

7. The method for collaborative optimization of bidirectional information transmission and full-loop control of industrial Internet of Things according to claim 1 is characterized in that: After the uplink transmission subproblem and the downlink multicast packet transmission subproblem are solved, the value of η is searched and updated in the interval 0 to 1 so that the objective function value of the original problem P0 is minimized.

8. The method for collaborative optimization of bidirectional information transmission and full-loop control of industrial Internet of Things according to claim 6 is characterized in that: The specific process of solving the multicast grouping problem of the actuator by fixed communication resource allocation is as follows: First, consider the multicast packet problem. Assuming that the downlink transmission bandwidth and the wireless access point transmission power are evenly distributed to each actuator, the objective function of problem SP2.1 is analyzed and it is known that the maximum This is equivalent to minimizing the following objective: in, is a constant, and Can be rewritten as an actuator point and the center point of group m The distance between The multicast grouping problem is equivalent to a clustering problem. In this problem, given N actuator points ε1, ε2, ..., ε N , the goal is to divide these points into M clusters The multicast packet problem is expressed as: stC6,C7,C9 First, determine the value of M and introduce the silhouette coefficient as an internal indicator for evaluating clustering effects. It aims to describe the similarity between the target and its group and other groups. The silhouette coefficient calculation formula for sample i is: Among them, F(i) and R(i) represent cohesion and separation respectively, which reflect the closeness of the sample point to the elements within the group and the elements outside the group. The calculation formula is: Among them, m′ is the group other than the group where sample i belongs, N m and N m′ denote the number of executors in groups m and m′ respectively, is the set of actuators in group m′, dist(i, j) represents the distance between sample i and sample j; The average silhouette coefficient of the clustering results is expressed as: Where N is the number of actuators; Statistically speaking, there is an optimal value M for the number of groups. * , which is less than Therefore, the initial value of M is set to in, It represents the rounding up operation. By looping through the number of groups, the number of groups with the largest average silhouette coefficient value is selected as M * .

9. The method for collaborative optimization of bidirectional information transmission and full-loop control of industrial Internet of Things according to claim 6 is characterized in that: The process of obtaining bandwidth and power allocation by solving the resource allocation problem for a given grouping result is as follows: The improved K-means++ algorithm is used to group the executors. After completing the multicast grouping, first, the downlink bandwidth allocation is optimized. Assuming that the entire bandwidth B is allocated to group m, the transmission delay of group m is expressed as: Among them, r m，B (k) is the transmission rate when the entire bandwidth is allocated to group m; When the bandwidth ρ m When B is allocated to group m, the bandwidth allocation problem is simplified to: Among them, ρ m (k) is the proportional coefficient of the bandwidth allocated to group m; Given the bandwidth allocation result, the allocation problem of wireless access point transmit power is expressed as: stC5. From the analysis, we can see that the above two problems are convex optimization problems, which are solved by convex optimization solver, and then the solution of problem SP2 is obtained by alternately optimizing the bandwidth and the transmission power allocation of the wireless access point.

Images