A sensor node communication resource reservation method for industrial field status monitoring

Abstract

The present invention discloses a sensor node communication resource reservation method for industrial field status monitoring, which relates to the field of wireless communications. Under the premise of ensuring the observability requirement of each reserved set, an offline sensor node observability diversity algorithm is designed to determine the minimum number of reserved resources, define the priority of each sensor node according to the number of triggering times, use the system perception error as a constraint condition, establish an optimization problem by minimizing the system communication cost, and use a heuristic method to obtain an online resource reservation scheme. The entire technical scheme consists of an offline field subnet sensor node diversity method for system state estimation and an online resource reservation strategy based on node priority. The present invention makes up for the problem of insufficient perception capability of a single sensor node, avoids the problem of excessive power loss caused by frequent triggering of a single sensor node, and leads to an increase in the frequency of manual inspections at the industrial site, thereby ensuring the overall monitoring performance of the system.

Description

Technical Field

[0001] The present invention relates to the field of wireless communications, and in particular to a sensor node communication resource reservation method for industrial field status monitoring. Background Art

[0002] With the rapid development of manufacturing production technology, industrial wireless networks have found widespread applications in industrial automation, such as production process status monitoring, online testing of production equipment, and generation of repair and maintenance tasks. Different applications have varying requirements for data transmission performance. For example, in applications such as equipment failure alarms, alarm signals must be transmitted with the highest priority. In production process status monitoring, in order to estimate the entire system state, the information transmitted to the estimator must meet the system's observability requirements. Existing methods for industrial wireless monitoring applications typically assume observability as a prerequisite for perception and control. Therefore, it is crucial to study communication resource reservation mechanisms for transmission nodes in industrial field-level perception applications.

[0003] At the same time, industrial field environments are complex, with periodic system monitoring tasks coexisting with sudden fault alarm applications. Industrial field applications often require multiple sensor nodes for monitoring, and different sensor nodes have varying impacts on monitoring performance. For example, in production process status monitoring tasks, status information on key indicators / locations is more important than that on non-critical nodes; however, when equipment failure occurs, the alarm signal should be transmitted with the highest priority. However, existing industrial wireless systems are primarily designed for transmission efficiency, which can lead to frequent activation of the same sensor node, shortening the system's lifespan and causing performance degradation. Therefore, how to rationally design sensor node priorities to trigger sensor nodes as evenly as possible while meeting system requirements and extending system lifespan has become a pressing issue.

[0004] Furthermore, a large number of sensing and control nodes exist in industrial sites. In these mission-critical machine-type communications (C-MTC), a large number of uplink-dominated, delay-sensitive transmissions are expected, making uplink scheduling a key issue. However, traditional wireless communication systems employ a shared resource allocation and scheduling approach, which relies on dynamic scheduling strategies to transmit user data. This approach suffers from excessive underlying signaling interactions, and competitive random access mechanisms can cause different nodes to simultaneously request links, leading to network congestion. Therefore, it is necessary to design a wireless resource reservation mechanism for periodic industrial monitoring applications to reduce network congestion.

[0005] To address these characteristics, consider a field subnet where devices access the network via a cellular uplink. Using semi-persistent scheduling, we pre-reserve transmission resources for devices targeting state estimation applications by dividing uplink time-frequency resources into dynamic access resource blocks and semi-persistent scheduling resource blocks. The overall approach is to design an observability diversity algorithm for offline sensor nodes, ensure the joint observability of each reserved set, determine the minimum number of reserved resources, define the priority of each sensor node based on the number of triggers, and formulate an optimization problem to minimize the system communication cost, using system perception error as a constraint, to develop a resource reservation scheme.

[0006] After searching the existing literature, it was found that the most similar implementation scheme is the Chinese patent application number: 202110312461.2, entitled: An on-demand transmission method based on non-orthogonal multiple access for industrial monitoring. Its specific approach is: by determining the performance indicators for evaluating the transmission scheme, constructing the energy efficiency maximization problem under heterogeneous transmission requirements and spectrum constraints, analyzing the requirements of different monitoring applications for transmission performance, designing a channel allocation algorithm based on matching theory, and determining the channel allocation relationship. However, this method may cause nodes with higher sensing accuracy to be frequently triggered, reducing the service life of the system; at the same time, solving the optimization problem requires high computing power of the network controller.

[0007] Existing on-demand transmission resource allocation methods for industrial monitoring applications typically assume that individual sensor nodes have a good understanding of the system status, which does not conform to the current situation where sensor nodes within the field subnet have limited ability to perceive the system status. Existing dynamic scheduling strategies have the disadvantage of excessive underlying signaling interactions. The complex handshake process can also lead to large access delays. At the same time, the competitive random access mechanism may cause network congestion. Existing reservation methods based on periodic scheduling in industrial networks ignore the impact of data packet transmission content on system monitoring performance. This may result in high-priority data not being transmitted in a timely manner in scenarios with limited spectrum, making it difficult to meet system performance requirements. Existing resource scheduling methods for industrial networks require real-time optimization problems and place high demands on the computing power of the equipment.

[0008] Therefore, researchers in this field have dedicated themselves to developing a method for reserving communication resources for sensor nodes for industrial field status monitoring. This method addresses the problem of insufficient sensing capabilities of individual sensor nodes, avoids the problem of excessive power loss caused by frequent triggering of individual sensor nodes, and thus increases the frequency of manual inspections at industrial sites, thereby ensuring the overall monitoring performance of the system. Summary of the Invention

[0009] In view of the above-mentioned defects of the prior art, the technical problem to be solved by the present invention is to find a subset of available sensor nodes that can analyze the system status through data fusion methods in the case that the sensor nodes can only observe part of the system status; dynamically define the transmission priority of different sensor node sets based on the historical records of sensor node activation, reduce the increase in energy consumption caused by the frequent activation of some sensor nodes, and extend the service life of the sensor nodes; when making reservation decisions, judge whether the sensor node set can meet the application requirements by designing judgment conditions to ensure monitoring application needs; combine offline and online calculations to design a priority-based scheduling method to avoid online solution of complex optimization problems.

[0010] To achieve the above objectives, the present invention provides a sensor node communication resource reservation method for industrial field status monitoring. While ensuring the observability requirements of each reserved set, an offline sensor node observability diversity algorithm is designed to determine the minimum number of reserved resources. The priority of each sensor node is defined according to the number of triggering times. With the system perception error as the constraint condition, an optimization problem is established to minimize the system communication cost, and an online resource reservation scheme is obtained using a heuristic method.

[0011] Furthermore, the invention includes an off-line field subnet sensor node diversity method for system state estimation and an on-line resource reservation method based on node priority.

[0012] Furthermore, the field subnet sensor node diversity method for system state estimation has a reliable connection between the edge server and the communication base station, and uses the edge server to offline calculate the set of candidate sensor nodes that meet the system information fusion conditions; wherein, n max The total number of resources that are available for resource reservation.

[0013] Furthermore, the field subnet sensor node diversity method for system state estimation comprises the following steps:

[0014] Step 1.1: Find the elements in the current system that are less than or equal to n by traversing max The set of all sensor nodes;

[0015] Step 1.2: Construct the observation equation of the set according to the perception function of the sensor node;

[0016] Step 1.3: Determine whether each set meets the system information fusion conditions, and store the qualified sets in the edge server as candidate sets for resource reservation;

[0017] Step 1.4: Calculate the noise covariance matrix for each set for subsequent wireless resource allocation decisions.

[0018] Furthermore, in step 1.2, the observation equation of the sensor node set constructed can be used to determine whether the current candidate set contains the information required to estimate the entire state of the system.

[0019] Furthermore, in step 1.3, the system information fusion condition may be determined according to different state estimation algorithms.

[0020] Furthermore, the online resource reservation method based on node priority dynamically updates the priority of the sensor node set and determines the final activated sensor node set and resource allocation result according to system performance requirements.

[0021] Furthermore, the online resource reservation method based on node priority includes the following steps:

[0022] Step 2.1: The base station determines the number of available resources n based on the current channel state, and the remote controller determines the threshold condition tr(P)≤M that the current system state estimate should meet;

[0023] Step 2.2: Sort all sets whose number of elements is less than or equal to n by priority.

[0024] Step 2.3: Determine the resource reservation set according to the system state estimation requirement.

[0025] Furthermore, the online resource reservation method based on node priority is used to determine a set of sensor nodes for which communication resources should be reserved according to current channel status and system requirements.

[0026] Furthermore, in step 2.3, whether the current sensor node set can meet the system performance requirements is judged, and the sensor node set with the highest priority that meets the conditions is selected as the resource reservation set, including the following steps:

[0027] Step 2.3.1, initialize the sensor node set and select variable j = 1;

[0028] Step 2.3.2: Initialize the system's current prior covariance P t|t-1 =AP t-1|t-1 A T +Q, where A is the system matrix and Q is the system process noise covariance matrix.

[0029] Step 2.3.3, select the set with priority j and solve Among them, P j is the information matrix corresponding to the sensor node set j;

[0030] Step 2.3.4: If tr(P)≤M, set the current sensor node set as the resource reservation set;

[0031] Step 2.3.5: If tr(P)>M, set the variable j=j+1 and repeat step 2.3.3 until a set of sensor nodes that meets the conditions is selected;

[0032] Step 2.3.6: Update the posterior covariance of the current system state estimate to P t|t =P.

[0033] In a preferred embodiment of the present invention, the purpose of the present invention is to provide a predictive communication time-frequency resource allocation method for industrial wireless networks with the goal of ensuring the performance of the monitoring system, which can achieve efficient utilization of industrial field spectrum resources.

[0034] To achieve the above object, the present invention adopts the following technical solutions:

[0035] The entire technical solution consists of two parts: offline calculation and online allocation. In the offline design phase, all sensor nodes are grouped by designing a field subnet sensor node diversity method based on system state estimation. In the online calculation phase, a predictive communication time-frequency resource allocation method is designed to make resource reservation decisions based on the priority of each group.

[0036] In the offline computing part, the field subnet sensor node diversity algorithm for system state estimation is executed, and the set of sensor nodes that meet the system conditions is calculated offline on the edge server.

[0037] The first step is to use the enumeration method to construct a set of sensor nodes whose number is less than or equal to the total number of system sensor nodes N.

[0038] The second step is to construct the observation equation y for each set based on the perception function of the sensor node. i =C i x+υ i , where C i and υ i are the combination of the measurement matrix and the observation noise vector of all sensor nodes in set i.

[0039] The third step is to determine whether each set meets the system information fusion conditions. The qualified collection is stored in the edge server as a candidate collection for resource reservation.

[0040] In the online calculation part, a predictive communication time-frequency resource allocation method is executed, the priority of each current set is calculated online, and the resource allocation result is determined.

[0041] The first step is to initialize the priority p of each sensor node k .

[0042] The second step is to select a set of available sensor nodes and prioritize them, which includes the following steps:

[0043] 2.1 In the candidate set, select all sensor node sets whose number of elements is less than or equal to n.

[0044] 2.2 Initialize the priority p of each sensor node k , the priority of each set is composed of the sum of the priorities of each sensor node in it, that is, p i =∑p k .

[0045] The third step is to sort all sensor node sets whose number of elements is less than or equal to n by priority. If two sets have the same priority, the order of the two sets is randomly determined.

[0046] The fourth step is to determine the resource reservation sensor node set according to the system state estimation requirements.

[0047] 4.1 Initialize the sensor node set Select variable j = 1 and initialize the priority p of each sensor node k .

[0048] 4.2 Initialize the system's current prior covariance P t|t-1 =AP t-1|t-1 A T +Q, where A is the system matrix and Q is the system process noise covariance matrix.

[0049] 4.3 Select the set with priority j and calculate the covariance matrix that can be obtained using the information transmitted by the current set of sensor nodes

[0050] 4.4 If tr(P)≤M, let the current sensor node set be the resource reservation set.

[0051] 4.5 If tr(P)>M, set variable j=j+1 and repeat step 3.2 until a set of sensor nodes that meets the conditions is selected.

[0052] 4.6 Update the posterior covariance of the current system state estimate.

[0053] In the fifth step, the devices in the resource reservation set access the channel. Each sensor node selects the transmission power according to the channel status. When the first information transmission fails, it retransmits at the maximum power until the transmission is successful.

[0054] Step 6: Update the priority of each current collection.

[0055] 6.1 The priority of each sensor node can be expressed as p k =p k -t′ k , where t′ kIndicates the number of transmissions of sensor node k at the current transmission moment.

[0056] 6.2 The priority of each set is composed of the sum of the priorities of each sensor node in it, that is, p i =∑p k .

[0057] Compared with the prior art, the present invention has the following obvious substantial features and significant advantages:

[0058] 1. In order to solve the problem that a single sensor node connected to a field subnet has limited perception capability and cannot measure all system states, the present invention designs a subnet sensor node diversity method, which uses the obtained sensor node set as the minimum unit for resource allocation, thereby compensating for the problem of insufficient perception capability of a single sensor node.

[0059] 2. The transmission priority of sensor nodes in the field subnet is defined according to the transmission records of each sensor node, which avoids the problem of excessive power loss caused by frequent triggering of a single sensor node, resulting in an increase in the frequency of manual inspections at the industrial site.

[0060] 3. By designing constraints on the system state estimation covariance matrix at each measurement moment, the overall monitoring performance of the system is guaranteed.

[0061] The concept, specific structure and technical effects of the present invention will be further described below in conjunction with the accompanying drawings to fully understand the purpose, characteristics and effects of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 1 is a schematic diagram of the system structure of a preferred embodiment of the present invention;

[0063] Figure 2 It is an overall flow chart of the algorithm of a preferred embodiment of the present invention;

[0064] Figure 3 This is a flow chart of an offline candidate sensor node diversity method according to a preferred embodiment of the present invention;

[0065] Figure 4 This is a flowchart of online resource reservation decision-making in a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0066] The following describes several preferred embodiments of the present invention with reference to the accompanying drawings to make its technical content clearer and easier to understand. The present invention can be embodied in many different forms of embodiments, and the scope of protection of the present invention is not limited to the embodiments mentioned herein.

[0067] In the drawings, components with identical structures are denoted by the same reference numerals, and components with similar structures or functions are denoted by similar reference numerals. The size and thickness of each component shown in the drawings are arbitrary and are not limited by the present invention. For clarity, the thickness of components in some places in the drawings is appropriately exaggerated.

[0068] A predictive communication time-frequency resource allocation method for industrial wireless networks is designed. While ensuring the observability of each reserved set, an offline sensor node observability diversity algorithm is designed to determine the minimum number of reserved resources. Each sensor node's priority is defined based on the number of triggers. Using system perception error as a constraint, an optimization problem is formulated to minimize the system communication cost. An online resource reservation scheme is derived using a heuristic approach. The overall technical solution consists of an offline field subnet sensor node diversity method for system state estimation and an online resource reservation strategy based on node priority.

[0069] A field subnet sensor node diversity method for system state estimation is proposed. The edge server is reliably connected to the communication base station, and the edge server is used to offline calculate the candidate sensor node set that meets the system information fusion conditions. max The total number of resources that are available for resource reservation.

[0070] The resource reservation strategy based on node priority is online, and the priority of the sensor node set is dynamically updated. The final activated sensor node set and resource allocation results are determined according to the system performance requirements. The specific steps are as follows:

[0071] In the first step, the base station determines the number of available resources n according to the current channel state, and the remote controller determines the threshold condition tr(P)≤M that the current system state estimate should meet.

[0072] The second step is to sort all sets whose number of elements is less than or equal to n by priority.

[0073] The third step is to determine the resource reservation set according to the system status estimation requirements.

[0074] The field subnet sensor node diversity method for system state estimation includes the following steps:

[0075] The first step is to traverse and find the elements in the current system that are less than or equal to n. max The set of all sensor nodes.

[0076] The second step is to construct the observation equation of the set based on the perception function of the sensor node.

[0077] The third step is to determine whether each set meets the system information fusion conditions, and store the qualified sets in the edge server as resource reservation candidate sets.

[0078] The fourth step is to calculate the noise covariance matrix of each set, which is then used for subsequent wireless resource allocation decisions.

[0079] In the second step, the constructed sensor node set can be used to determine whether the current candidate set contains the information required to estimate the entire system state. For the sake of simplicity of analysis, a linear system is used here for description.

[0080] Field subnet sensor node diversity method for system state estimation,In the third step, the system information fusion condition can be determined according to different,state estimation algorithms.

[0081] The online resource reservation method based on node priority is used to determine the set of sensor nodes for which communication resources should be reserved according to the current channel state and system requirements, and includes the following steps:

[0082] The first step is to determine the amount of available resources.

[0083] The second step is to sort all available sensor nodes according to their priorities.

[0084] The third step is to judge whether the current sensor node set can meet the system performance requirements according to the sorting order, and select the sensor node set with the highest priority that meets the conditions as the resource reservation set.

[0085] In the fourth step, the devices in the resource reservation set access the channel. Each sensor node selects the transmission power according to the channel status. When the initial information transmission fails, it retransmits at the maximum power until the transmission is successful.

[0086] The fifth step is to update the priority of each sensor node according to the final communication result, and update the priority of each set according to the priority of each sensor node.

[0087] In the third step of the resource reservation method based on node priority, whether the current sensor node set can meet the system performance requirements is judged, and the sensor node set with the highest priority that meets the conditions is selected as the resource reservation set, including the following steps:

[0088] Step 1: Initialize the sensor node set and select variable j=1.

[0089] Step 2 Initialize the system's current prior covariance P t|t-1 =AP t-1|t-1 A T +Q, where A is the system matrix and Q is the system process noise covariance matrix.

[0090] Step 3: Select the set with priority j and solve

[0091] Step 4: If tr(P)≤M, then set the current sensor node set as the resource reservation set.

[0092] Step 5: If tr(P)>M, set the variable j=j+1 and repeat step 3 until a set of sensor nodes that meets the conditions is selected.

[0093] Step 6 Update the posterior covariance of the current system state estimate to P t|t =P.

[0094] Figure 1 This is a schematic diagram of the edge network system architecture used for steel plate temperature status monitoring applications at a steel hot rolling site.

[0095] On-site sensor nodes measure the temperature of steel plates within a certain range. Each sensor node has wireless communication capabilities and is connected to an edge base station deployed on-site. The edge base station collects temperature information uploaded by the sensor nodes and schedules and allocates wireless communication resources.

[0096] Figure 2 This is the overall flow chart of a resource reservation algorithm for industrial field sensing. The algorithm consists of two parts: offline computation and online scheduling. The offline computation part determines the sensor node diversity to meet the requirements for system state information fusion. The online scheduling part performs real-time scheduling based on the priority of each sensor node set and the system's perception accuracy constraints.

[0097] Figure 3 This is the offline candidate sensor node diversity algorithm process. In the offline calculation part, all frequency bands in the subnet are divided into four channels, where four is the maximum number of resources available for pre-allocation in the subnet. All candidate sensor node sets are calculated offline in the edge base station and the calculation results are saved.

[0098] The first step is to use the exhaustive method to obtain all possible sensor node sets, which are combinations.

[0099] The second step is to construct the observation equation y for each set based on the perception function of the sensor node. i =C i x+υ i ,in and are the combination of the measurement matrix and the observation noise vector of all sensor nodes in set i.

[0100] The third step is to determine whether each set meets the system information fusion conditions. The qualified collection is stored in the edge server as a candidate collection for resource reservation.

[0101] The fourth step is to calculate the information matrix for each set where R i is the observation noise covariance matrix of the current sensor node set.

[0102] Figure 4 This is the online resource reservation decision flow chart. In the online calculation part, the priority of each set is calculated online to determine the resource allocation result.

[0103] In the first step, the base station determines the number of available resources 3 based on the current channel state, and the remote controller determines the threshold condition tr(P)≤M that the current system state estimate should meet.

[0104] The second step is to sort all sensor nodes whose number of elements is less than or equal to n by priority.

[0105] The third step is to determine the set of resource reserved sensor nodes according to the system state estimation requirements.

[0106] 3.1 Initialize the sensor node set and select variable j=1.

[0107] 3.2 Initialize the system's current prior covariance P t|t-1 =AP t-1|t-1 A T +Q, where Q is the system process noise covariance matrix.

[0108] 3.3 Select a set with priority j and find the estimated covariance that can be obtained by transmitting this set

[0109] 3.4 If tr(P)≤M, let the current sensor node set be the resource reservation set.

[0110] 3.5 If tr(P)>M, set variable j=j+1 and repeat step 3.2 until a set of sensor nodes that meets the conditions is selected.

[0111] 3.6 Update the posterior covariance of the current system state estimate to solve the estimated covariance P that can be obtained by transmitting this set t|t =P.

[0112] In the fourth step, the devices in the resource reservation set access the channel. Each sensor node selects the transmission power according to the channel status. When the initial information transmission fails, it retransmits at the maximum power until the transmission is successful.

[0113] Step 5: Update the priority of each current collection.

[0114] 5.1 The priority of each sensor node can be expressed as p k =p k -t′ k, where t′ k Indicates the number of transmissions of sensor node k at the current transmission moment.

[0115] 5.2 The priority of each set i is composed of the sum of the priorities of each sensor node in it, that is, p i =∑p k .

[0116] The preferred embodiments of the present invention have been described in detail above. It should be understood that numerous modifications and variations based on the concepts of the present invention are possible without inventive effort by those skilled in the art. Therefore, any technical solution that can be derived by one skilled in the art through logical analysis, reasoning, or limited experimentation based on the concepts of the present invention and the prior art should be within the scope of protection defined by the claims.

Claims

1. A sensor node communication resource reservation method for industrial field status monitoring, characterized in that: Including offline calculation and online allocation; offline calculation, through the field subnet sensor node diversity method of system state estimation, all sensor nodes are divided; Online Allocation, a predictive communication time-frequency resource allocation method, makes resource reservation decisions based on the priority of each set; Offline computing, performing the field subnet sensor node diversity algorithm for system state estimation, and offline computing the set of sensor nodes that meet the system conditions at the edge server; The first step is to use the enumeration method to construct all the sensor nodes whose number is less than or equal to the total number of system sensor nodes. A collection of sensor nodes ; The second step is to construct the observation equation of each set according to the perception function of the sensor node ,in and Set The combination of the measurement matrix and the observation noise vector of all sensor nodes in; The third step is to determine whether each set meets the system information fusion conditions. ,The qualified set is stored in the edge server as a candidate set for resource reservation; Online allocation, executing the predictive communication time-frequency resource allocation method, online calculating the priority of each current set, and determining the resource allocation result; The first step is to initialize the priority of each sensor node ; The second step is to select a set of available sensor nodes and prioritize them: 2.1 In the candidate set, select the number of elements less than or equal to The set of all sensor nodes; 2.2 Initialize the priority of each sensor node ,The priority of each set is composed of the sum of the priorities of each sensor node, i.e. ; The third step is to In the set of all sensor nodes, they are sorted by priority. If the priorities of two sets are the same, their order is randomly determined. The fourth step is to determine the resource reservation sensor node set according to the system state estimation requirements: 4.1 Initialize sensor node set selection variables , and initialize the priority of each sensor node ; 4.2 Initialize the system covariance at the current moment ,in is the system matrix, is the system process noise covariance matrix; 4.3 Select the priority order The covariance matrix obtained by using the information transmitted by the current set of sensor nodes is calculated. ; 4.4 If , then let the current sensor node set be the resource reservation set; 4.5 If , then let the variable , repeat step 4.3 until a set of sensor nodes that meet the conditions is selected; 4.6 Update the posterior covariance of the current system state estimate; In the fifth step, the devices in the resource reservation set access the channel. Each sensor node selects the transmission power according to the channel status. If the initial information transmission fails, it will retransmit at the maximum power until the transmission is successful. Step 6: Update the priority of each current collection: 6.1 The priority of each sensor node is expressed as ,in Indicates the sensor node at the current transmission moment Number of transmissions; 6.2 The priority of each set is composed of the sum of the priorities of each sensor node in it, that is, .

2. The sensor node communication resource reservation method for industrial field status monitoring according to claim 1, characterized in that: It includes an off-line field subnet sensor node diversity method for system state estimation and an on-line resource reservation method based on node priority.

3. The sensor node communication resource reservation method for industrial field status monitoring according to claim 2, characterized in that: The field subnet sensor node diversity method for system state estimation, the edge server and the communication base station are reliably connected, and the edge server is used to offline calculate the set of candidate sensor nodes that meet the system information fusion conditions; wherein, The total number of resources that are available for resource reservation.

4. The sensor node communication resource reservation method for industrial field status monitoring according to claim 3, characterized in that: The field subnet sensor node diversity method for system state estimation comprises the following steps: Step 1.1, by traversing, get the elements in the current system that are less than or equal to The set of all sensor nodes; Step 1.2: Construct the observation equation of the set according to the perception function of the sensor node; Step 1.3: Determine whether each set meets the system information fusion conditions, and store the qualified sets in the edge server as candidate sets for resource reservation; Step 1.4: Calculate the noise covariance matrix for each set for subsequent wireless resource allocation decisions.

5. The sensor node communication resource reservation method for industrial field status monitoring according to claim 4, characterized in that: In step 1.2, the observation equation of the constructed sensor node set can be used to determine whether the current candidate set contains the information required to estimate the entire state of the system.

6. The sensor node communication resource reservation method for industrial field status monitoring according to claim 4, characterized in that: In step 1.3, the system information fusion condition can be determined according to different state estimation algorithms.

7. The sensor node communication resource reservation method for industrial field status monitoring according to claim 2, characterized in that: The online resource reservation method based on node priority dynamically updates the priority of the sensor node set and determines the final activated sensor node set and resource allocation result according to system performance requirements.

8. The sensor node communication resource reservation method for industrial field status monitoring according to claim 2, characterized in that: The online resource reservation method based on node priority comprises the following steps: Step 2.1: The base station determines the amount of available resources based on the current channel status. , the remote controller determines the threshold condition that the current system state estimate should meet ,in, represents the covariance matrix of the state estimation result, Indicates the threshold that the estimation accuracy should meet; Step 2.2: When the number of sensor nodes is less than or equal to Among all the collections, sort them by priority; Step 2.3: Determine the resource reservation set according to the system state estimation requirement.

9. The sensor node communication resource reservation method for industrial field status monitoring according to claim 2, characterized in that: The online resource reservation method based on node priority is used to determine the set of sensor nodes for which communication resources should be reserved according to the current channel status and system requirements.

10. The sensor node communication resource reservation method for industrial field status monitoring according to claim 8, characterized in that: In step 2.3, whether the current sensor node set can meet the system performance requirements is judged, and the sensor node set with the highest priority that meets the conditions is selected as the resource reservation set, including the following steps: Step 2.3.

1. Initialize sensor node set selection variables ; Step 2.3.2: Initialize the system's current covariance ,in is the system matrix, is the system process noise covariance matrix; Step 2.3.3, select the priority order Set of, calculate the selection set The covariance after ,in A collection of sensor nodes The corresponding information matrix; Step 2.3.4, if , then let the current sensor node set be the resource reservation set; Step 2.3.5, if , then let the selection variable , repeat step 2.3.3 until a set of sensor nodes that meet the conditions is selected; Step 2.3.6: Update the posterior covariance of the current system state estimate to .

Images