Dual-module networking method and system

Abstract

The invention relates to the technical field of dual-mode communication, in particular to a dual-mode networking method and system. The method comprises the steps that a networking anomaly recognition model is constructed, the input of the model is networking monitoring data, the output of the model is a networking state prediction result of each station, and the networking anomaly recognition model is trained based on historical networking monitoring data; when the preset network access time of the station arrives, the station preferentially applies for network access in a carrier mode; inputting networking monitoring data acquired in real time into the trained networking anomaly identification model to obtain a networking state prediction result of each site output by the model, and when the networking state prediction result of the site is abnormal, selecting an optimal proxy node from the rest sites based on the networking monitoring data, and performing networking anomaly identification on the optimal proxy node. And the site is bridged to the optimal proxy node in a wireless communication mode to apply for network access. The data transmission path can be adaptively adjusted according to the current networking state, the flexibility and adaptability of the network are improved, and the overall efficiency of dual-module networking communication is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of dual-mode communications, and in particular to a dual-mode networking method and system. Background Art

[0002] In recent years, both at home and abroad, combining the advantages of wireless communication, a dual-mode communication solution of high-speed power line carrier (HPLC) and high-speed wireless communication (HRF) has been formed in the power consumption information collection system to increase the coverage of communication signals, ensure the basic functions of the monitoring system, and achieve effective management of low-voltage distribution substations. Among them, high-speed wireless communication refers to wireless communication technology that supports high-speed data transmission, which is usually used in Internet of Things (IoT) devices, sensor networks, etc.

[0003] At present, the existing dual-mode communication solutions are mainly as follows: the carrier and the wireless channel have the ability to receive and send synchronously, that is, when the carrier channel receives data, the wireless channel receives data at the same time. According to the load capacity of the communication module, you can choose to send in time or at the same time. Considering the problems of channel collision and information overlap, time slot management and scheduling solutions must be designed. For example, in Chinese patent CN116318259A, a method, apparatus, device and storage medium for optimizing contention time slots in a dual-mode system, a master node determines the proxy node of a site that needs to participate in time slot optimization according to the network status of the dual-mode system, recorded as a target PCO, and obtains the time slot optimization fragment length according to the number of target PCOs and the number of sites loaded by the target PCO on the HRF link, and updates the wireless beacon flag value corresponding to the target PCO, and configures a standard beacon frame according to the wireless beacon flag value and the time slot optimization fragment length to obtain a first beacon frame; the proxy node updates the first beacon frame using the wireless beacon flag value in the first beacon frame to obtain a second beacon frame; the site determines the target time slot segment from the contention time slot according to the second beacon frame, thereby realizing a reasonable dynamic allocation of contention time slots, making the competition within the time slot orderly, and reducing the occurrence of conflicts.

[0004] Although the existing HPLC-HRF dual-mode communication scheme utilizes the design of dual-channel simultaneous information transmission and reception as well as time slot management and scheduling scheme, which improves the communication stability and system robustness to a certain extent, the existing HPLC-HRF dual-mode communication scheme requires a large amount of control signaling in order to maintain time slot synchronization and management. The signaling overhead will reduce the efficiency of effective data transmission, resulting in low communication and networking efficiency. At the same time, in a high-density network environment and with dynamic changes in traffic, it has poor adaptability to changes in network topology and load, which can easily lead to uneven resource allocation and reduce overall efficiency. Summary of the invention

[0005] In response to the above technical problems, the present invention proposes a dual-mode networking method and system, which aims to adaptively adjust the data transmission path according to the current networking status, improve the flexibility and adaptability of the network, and improve the overall efficiency of dual-mode networking communications.

[0006] In a first aspect, the present application provides a dual-mode networking method for realizing a dual-mode networking system of high-speed power line carrier and high-speed wireless communication, wherein the dual-mode networking system includes a master node and a plurality of sites, wherein the sites are connected to the master node by a carrier, and the sites are connected by wireless communication, and the method includes the following steps:

[0007] S1, obtain historical network monitoring data;

[0008] S2, constructing a networking anomaly identification model, wherein the input of the networking anomaly identification model is the networking monitoring data, and the output of the networking anomaly identification model is the networking status prediction result of each site;

[0009] S3, training the network anomaly recognition model based on historical network monitoring data to obtain a trained network anomaly recognition model;

[0010] S4, when the preset network access time of the site is reached, the site preferentially applies for network access by carrier mode and obtains network monitoring data in real time;

[0011] S5, input the real-time acquired networking monitoring data into the trained networking anomaly recognition model to obtain the networking status prediction result of each site output by the model. If the networking status prediction result of the site is normal, the carrier transmission between the site and the main node is maintained. If the networking status prediction result of the site is abnormal, the optimal proxy node is selected from the remaining sites based on the networking monitoring data, and the site uses wireless communication to bridge to the optimal proxy node to apply for network access.

[0012] In some embodiments, the historical networking monitoring data includes a list of sites that have been connected to the network in the historical period and the actual networking status of each network in the historical period. In S3, the network anomaly recognition model is trained based on the historical networking monitoring data to obtain a trained network anomaly recognition model, including:

[0013] S301, based on the actual networking status of each network in the historical period, annotate the historical networking monitoring data in the normal networking status and the abnormal networking status, and divide the annotated historical networking monitoring data into a training set and a verification set;

[0014] S302, setting an initial value of an abnormal time threshold;

[0015] S303, training the networking anomaly recognition model based on the training set, using the initial value of the abnormal time threshold to make a prediction, if the site is in the list of sites that have been connected to the network when the preset network access time of the site is reached and the abnormal time threshold is exceeded, then the prediction result of the networking state of the site is judged to be normal, if the site is not in the list of sites that have been connected to the network when the preset network access time of the site is reached and the abnormal time threshold is exceeded, then the prediction result of the networking state of the site is abnormal;

[0016] S304, after the training is completed, using the validation set to evaluate the error between the prediction result of the model and the actual networking state;

[0017] S305, adjusting the abnormal time threshold based on the error;

[0018] S306, retraining the networking anomaly recognition model based on the training set, and using the adjusted anomaly time threshold for prediction;

[0019] S307, repeatedly execute S304 to S306 until the error is lower than the preset threshold, and obtain the optimal abnormal time threshold, which is used as the judgment standard for the trained networking abnormality recognition model to predict the networking status.

[0020] In some embodiments, the networking monitoring data includes a list of sites that have been connected to the network in the current cycle and data packet transmission information of each site. In S5, selecting the optimal proxy node from the remaining sites based on the networking monitoring data includes:

[0021] S501, selecting a site within a preset distance range from a list of sites that have been connected to the network in the current cycle as a candidate proxy node;

[0022] S502, based on the data packet information of the candidate proxy site, calculate the wireless channel bandwidth occupancy rate of each candidate proxy site in the current cycle;

[0023] S503: Select a candidate proxy site with the lowest wireless channel bandwidth occupancy rate in the current cycle as the optimal proxy node.

[0024] In some embodiments, the networking monitoring data includes signal strength information of each site, then in S503, selecting a candidate proxy site with the lowest wireless channel bandwidth occupancy rate in the current cycle as the optimal proxy node also includes:

[0025] Based on the signal strength information of the candidate proxy sites, the signal strength of each candidate proxy site in the current cycle is calculated using the first-order lag filtering method;

[0026] Based on the wireless channel bandwidth occupancy rate and signal strength of each candidate proxy site in the current cycle, a first evaluation value of each candidate proxy site is calculated;

[0027] The candidate proxy site with the smallest first evaluation value is selected as the optimal proxy node.

[0028] In some embodiments, based on the wireless channel bandwidth occupancy rate and signal strength of each candidate proxy site in the current cycle, the expression for calculating the first evaluation value of each candidate proxy site is:

[0029]

[0030] Where i represents the serial number of the candidate proxy site, V 1i represents the first evaluation value of the i-th candidate proxy site, U i represents the wireless channel bandwidth occupancy rate of the i-th candidate proxy site, R i Represents the signal strength of the i-th candidate proxy site.

[0031] In some embodiments, the networking monitoring data includes the total number of communications between each site and the master node and the number of successful communications in the current cycle, and the candidate proxy site with the smallest first evaluation value is selected as the optimal proxy node, and further includes:

[0032] The quotient of the number of successful communications between each candidate proxy site and the master node and the total number of communications between the corresponding candidate proxy site and the master node in the current cycle is calculated to obtain the first communication success rate of each candidate proxy point;

[0033] Based on the first communication success rate of each candidate proxy point, a first-order lag filtering method is used to calculate the second communication success rate of each candidate proxy site in the current cycle;

[0034] Based on the wireless channel bandwidth occupancy rate, signal strength and second communication success rate of each candidate proxy site in the current cycle, a second evaluation value of each candidate proxy site is calculated;

[0035] The candidate proxy site with the smallest second evaluation value is selected as the optimal proxy node.

[0036] In some embodiments, based on the wireless channel bandwidth occupancy rate, signal strength and second communication success rate of each candidate proxy site in the current cycle, the expression for calculating the second evaluation value of each candidate proxy site is:

[0037]

[0038] Where i represents the serial number of the candidate proxy site, V 2i represents the second evaluation value of the i-th candidate proxy site, P 2i represents the second communication success rate of the i-th candidate proxy site, U i represents the wireless channel bandwidth occupancy rate of the i-th candidate proxy site, R iRepresents the signal strength of the i-th candidate proxy site.

[0039] In some embodiments, the networking monitoring data includes the networking level of each site, and selecting the optimal proxy node from the remaining sites based on the networking monitoring data also includes:

[0040] When there are at least two optimal proxy nodes selected, the optimal proxy node with the smallest networking level is selected as the final optimal proxy node;

[0041] When at least two optimal proxy nodes are located at the same networking level, the optimal proxy node with the smallest wireless channel bandwidth occupancy rate among the optimal proxy nodes at the same level is selected as the final optimal proxy node.

[0042] In some embodiments, the following steps are also included:

[0043] S6, in the process of the current period site using wireless communication to bridge to the optimal proxy node to apply for network access, the wireless channel bandwidth occupancy rate of the optimal proxy node is calculated in real time;

[0044] S7, when the wireless channel bandwidth occupancy rate of the optimal proxy node reaches a preset upper limit threshold, calculate the second evaluation values ​​of the remaining candidate proxy sites except the current optimal proxy node, and select the candidate proxy site with the smallest second evaluation value from the remaining candidate proxy sites as an auxiliary proxy node, the auxiliary proxy node is used to share the transmission data volume of the optimal proxy node;

[0045] S8, when the wireless channel bandwidth occupancy rate of the optimal proxy node and the wireless channel bandwidth occupancy rate of the auxiliary proxy node drop to a preset lower limit threshold and the sum of the wireless channel bandwidth occupancy rates of the optimal proxy node and the auxiliary proxy node does not exceed the wireless channel bandwidth occupancy rate of the optimal proxy node or the wireless channel bandwidth occupancy rate of the auxiliary proxy node, disconnect the bridge between the node with a lower second communication success rate among the optimal proxy node and the auxiliary proxy node and the site.

[0046] In a second aspect, the present application provides a dual-mode networking system, including a master node, a plurality of sites, a data acquisition module, a model building module, a model training module, and a dual-mode communication scheduling module, wherein the sites are connected to the master node by a carrier wave, and the sites are connected by wireless communication, wherein:

[0047] Data acquisition module, used to obtain historical network monitoring data and real-time network monitoring data;

[0048] A model building module is used to build a network anomaly identification model, the input of the network anomaly identification model is the network monitoring data, and the output of the network anomaly identification model is the network status prediction result of each site;

[0049] A model training module is used to train a network anomaly recognition model based on historical network monitoring data to obtain a trained network anomaly recognition model;

[0050] The dual-mode communication scheduling module is used for, when the preset network access time of the site is reached, the site preferentially applies for network access by carrier mode, and inputs the real-time acquired networking monitoring data into the trained networking anomaly recognition model to obtain the network status prediction result of each site output by the model. If the networking status prediction result of the site is normal, the carrier transmission between the site and the main node is maintained. If the networking status prediction result of the site is abnormal, the optimal proxy node is selected from the remaining sites based on the networking monitoring data, and the site uses wireless communication to bridge to the optimal proxy node to apply for network access.

[0051] The beneficial technical effects of the present invention include at least:

[0052] 1. A dual-mode networking method and system is adopted. First, a networking anomaly recognition model is constructed, and the model is trained by a dynamic adjustment mechanism of the anomaly time threshold based on historical networking monitoring data, so that the model can predict the networking status of the site based on real-time monitoring data, thereby being able to quickly respond to network topology changes and load fluctuations and improve network adaptability; then, by preliminarily selecting sites within a preset distance range from the list of sites that have been connected to the network in the current cycle as candidate proxy nodes, the communication stability between the candidate proxy nodes and the main node can be guaranteed, while reducing the signal attenuation and carrier communication interruption problems caused by the long distance, and then the wireless channel bandwidth occupancy rate is calculated based on the data packet information of the candidate proxy site, and the candidate proxy site with the highest wireless channel bandwidth occupancy rate is selected as the optimal proxy node, which can Quickly and effectively identify idle resources in the network to avoid further increasing traffic on high-load sites, thereby more effectively utilizing network bandwidth and helping to quickly realize adaptive switching of subsequent abnormal site communication channels, thereby improving the transmission efficiency of the entire network; finally, using the wireless communication connection between sites in the dual-mode networking architecture, the data that the abnormal site needs to transmit is first transmitted to the best proxy node through wireless communication, and then the best proxy node forwards the data that the abnormal site needs to transmit to the main node through carrier communication. Therefore, the best proxy node updates the dynamic routing table according to the new network topology, ensuring that the data packets of each site can be correctly transmitted to the main node through the new path, realizing the network adaptive switching of the communication channel of the site with abnormal operating environment status, thereby effectively improving the dual-mode communication and networking efficiency;

[0053] 2. Since signal strength is easily affected by factors such as multipath effect, shadow fading, and interference, which causes large fluctuations in signal strength in a short period of time, the present application adopts a first-order lag filtering method to filter the signal strength information of candidate proxy sites, providing a smoother and more stable signal strength estimation, and combining the two key indicators of filtered signal strength and wireless channel bandwidth occupancy rate as the basis for subsequent optimal proxy node selection, which can reduce the adverse effects of instantaneous fluctuations on the optimal proxy node decision, thereby avoiding frequent network adjustments, making the decision based on signal strength and wireless channel bandwidth occupancy rate more stable and reliable, so that the selected optimal proxy node can ensure the subsequent high efficiency of data transmission to abnormal sites and network load balancing;

[0054] 3. In an actual dual-mode networking environment, the communication success rate of a site may be affected by various noises and instantaneous interferences, resulting in data fluctuations. Therefore, this application uses a first-order lag filtering method to filter the communication success rate of the candidate proxy site, and selects the optimal proxy node by comprehensively considering the wireless channel bandwidth occupancy rate, the signal strength after filtering, and the communication success rate after filtering. Among the candidate proxy sites that have been searched and have been connected to the network, the candidate proxy site with the smallest wireless channel bandwidth occupancy rate, the strongest signal strength, and the highest communication success rate is adaptively selected as the optimal proxy node to apply for network access, ensuring that the abnormal site has a more stable communication connection with the main node through the optimal proxy node, reducing the risk of communication failure and reducing the delay in data transmission, thereby optimizing the communication path of the entire network, enhancing resistance to interference and abnormal situations, and thereby improving the stability, reliability and intelligence level of the entire dual-mode network system;

[0055] 4. Further, by calculating the wireless channel bandwidth occupancy rate of the optimal proxy node in real time and introducing auxiliary proxy nodes to share the data transmission load when the preset upper limit threshold is reached, it is helpful to achieve dynamic load balancing, avoid overload of a single node, and effectively improve the stability of the network. When the wireless channel bandwidth occupancy rate of the optimal proxy node and the auxiliary proxy node decreases and reaches the preset conditions, the bridge between the node with the second lower communication success rate among the optimal proxy node and the auxiliary proxy node and the site is disconnected, which helps to simplify the network structure and avoid wasting network resources. This solution can adaptively adjust the bridging relationship between the proxy node and the abnormal site according to the actual operating environment of the dual-mode network, so that the dual-mode networking system can better adapt to fluctuations in network traffic.

[0056] Other features and advantages of the present invention will be disclosed in detail in the following specific embodiments and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] The present invention will be further described below in conjunction with the accompanying drawings:

[0058] Figure 1 The figure is a schematic diagram of the dual-mode networking method flow in an embodiment of the present invention.

[0059] Figure 2 It is a schematic diagram of the structure of the dual-mode networking system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0060] The technical solutions of the embodiments of the present invention are explained and described below in conjunction with the drawings of the embodiments of the present invention, but the following embodiments are only preferred embodiments of the present invention, not all. Based on the embodiments in the implementation mode, other embodiments obtained by those skilled in the art without creative work are all within the protection scope of the present invention.

[0061] In the following description, terms such as "inside", "outside", "up", "down", "left", "right", etc. that indicate directions or positional relationships are only used to facilitate the description of the embodiments and simplify the description, and do not indicate or imply that the referred device or element must have a specific direction, be constructed and operate in a specific direction. Therefore, it should not be understood as a limitation of the present invention.

[0062] Embodiment 1:

[0063] Please refer to the attached Figure 1 , Figure 1 A schematic diagram of the process of a dual-mode networking method provided in one embodiment of this specification is shown.

[0064] Before describing the dual-mode networking method provided by this embodiment, the dual-mode networking architecture of this embodiment is described first. This embodiment is used to implement a dual-mode networking system of high-speed power line carrier and high-speed wireless communication. The dual-mode networking system includes a master node and several sites. Among them, the dual-mode networking system uses the local communication unit of the low-voltage distribution station area fusion terminal as the master node, also known as the central coordinator (CCO), which has the functions of networking control, network maintenance management and data collection. The station (STA) in the dual-mode networking system is also called a slave node, which mainly includes a collector and a dual-mode communication module on the meter. The station and the master node are connected through a high-speed power line carrier. Therefore, if the site is an intelligent meter, a dual-mode communication module can be installed to communicate and transmit directly with the master node. If the site is a traditional 485 meter, a collector can be installed to communicate and transmit with the master node. The master node of the station area fusion terminal can aggregate site data through the carrier communication channel, and the aggregated site data can also be uploaded to the remote power system cloud platform through a wireless network for storage or further analysis. High-speed wireless communication is used to connect the sites. For example, the collector can collect data information from other meters with wireless functions, such as electricity meters, water meters, gas meters, and data information from wireless sensors through high-speed wireless communication.

[0065] like Figure 1 As shown, the dual-mode networking method may at least include the following steps:

[0066] S1, obtain historical network monitoring data.

[0067] Among them, since the master node can aggregate site data through the carrier communication channel, the historical networking monitoring data can be obtained from the master node or the remote power system cloud platform. The historical networking monitoring data in this embodiment includes but is not limited to the list of sites that have been connected to the network during the historical period and the actual networking status of each network during the historical period.

[0068] S2, building a networking anomaly identification model, the input of the networking anomaly identification model is the networking monitoring data, and the output of the networking anomaly identification model is the networking status prediction result of each site.

[0069] It can be understood that the network anomaly recognition model in this embodiment is a typical supervised learning problem, especially a classification problem (network status normal or abnormal). Considering the efficiency of model training and prediction and the prediction accuracy of the model on unknown data, this embodiment can select one of the following machine learning algorithms to construct a network anomaly recognition model, which is not limited in this embodiment:

[0070] 1. Logistic Regression: Applicable to binary classification problems, easy to understand and implement, and highly efficient. It can provide probability scores and help adjust the abnormal time threshold.

[0071] 2. Support Vector Machine (SVM): It performs well in high-dimensional space and is suitable for small and medium-sized data sets. It can effectively handle nonlinear problems and is implemented through kernel functions.

[0072] 3. Random Forest: Performs well for large datasets and data with many features. Can handle classification and regression problems and is insensitive to outliers.

[0073] 4. Gradient Boosting Machine (GBM): More accurate than random forest, but may take longer to train. It iteratively trains weak learners to minimize the loss function.

[0074] S3: training a network anomaly recognition model based on historical network monitoring data to obtain a trained network anomaly recognition model.

[0075] Specifically, in this embodiment, the network anomaly recognition model is trained based on historical network monitoring data to obtain a trained network anomaly recognition model, including the following steps:

[0076] S301, based on the actual networking status of each network in the historical period, the historical networking monitoring data under the normal networking status and the abnormal networking status are respectively labeled to facilitate model learning, and the labeled historical networking monitoring data are divided into a training set and a verification set.

[0077] S302, setting an initial value of the abnormal time threshold.

[0078] Before starting to train the model, an initial abnormal time threshold is set based on expert experience or preliminary data analysis. The initial value of this abnormal time threshold will be used by the network anomaly recognition model to initially determine whether the site access time is abnormal, and will be continuously adjusted and optimized during the subsequent training process.

[0079] S303, training the networking anomaly recognition model based on the training set, and using the initial value of the abnormal time threshold for prediction. If the site is in the list of sites that have been connected to the network when the preset network access time of the site is reached and exceeds the abnormal time threshold, then the network status prediction result of the site is judged to be normal. If the site is not in the list of sites that have been connected to the network when the preset network access time of the site is reached and exceeds the abnormal time threshold, then the network status prediction result of the site is abnormal.

[0080] S304: After the training is completed, the validation set is used to evaluate the error between the prediction result of the model and the actual networking status.

[0081] Furthermore, the validation set can also be used to evaluate the accuracy, recall rate, F1 score and other indicators of the model through cross-validation and other methods, focusing on the model's ability to identify abnormal states. The operation method of using the validation set to evaluate the network anomaly recognition model in this embodiment is similar to the operation method of using the validation set to evaluate the classification model in the prior art, and this embodiment will not be repeated.

[0082] S305: Adjust the abnormal time threshold based on the error.

[0083] Specifically, the adjustment strategy for the abnormal time threshold is:

[0084] If the network anomaly recognition model does not recognize the abnormal state sufficiently (i.e., the false negative rate is high), then the abnormal time threshold is lowered so that the network anomaly recognition model can more easily identify edge cases as abnormalities;

[0085] If the networking anomaly recognition model over-recognizes abnormal states (i.e., the false positive rate is high), the abnormal time threshold is increased to reduce the sensitivity of the networking anomaly recognition model.

[0086] S306, retraining the networking anomaly recognition model based on the training set, and using the adjusted anomaly time threshold for prediction;

[0087] S307, repeat S304 to S306, and retrain and evaluate the model after each threshold adjustment until the error between the prediction result and the actual networking status is lower than the preset threshold or a high accuracy and appropriate recall rate are achieved on the validation set, indicating that the prediction performance of the model has reached a satisfactory level. Then, the threshold adjustment can be stopped, and the abnormal time threshold used at this time is determined to be the optimal abnormal time threshold, which is used as the judgment criterion for the trained networking anomaly recognition model to predict the networking status.

[0088] This embodiment constructs a networking anomaly identification model and trains the model using a dynamic adjustment mechanism of the anomaly time threshold based on historical networking monitoring data, so that the model can predict the networking status of the site based on real-time monitoring data, thereby being able to quickly respond to network topology changes and load fluctuations and improve network adaptability.

[0089] S4, when the preset network access time of the site is reached, the site applies for network access in a carrier mode first and obtains network monitoring data in real time.

[0090] It is understandable that in order to synchronize data collection and transmission of each site and ensure that the data in the entire network is consistent, the master node usually sends the corresponding preset network access time to each site. The time period between the previous preset network access time and the current preset network access time is a cycle.

[0091] The networking monitoring data acquired in real time in this embodiment includes, but is not limited to, a list of sites that have accessed the network in the current cycle and data packet transmission information of each site.

[0092] S5, input the real-time acquired networking monitoring data into the trained networking anomaly recognition model to obtain the networking status prediction result of each site output by the model. If the networking status prediction result of the site is normal, the carrier transmission between the site and the main node is maintained. If the networking status prediction result of the site is abnormal, the optimal proxy node is selected from the remaining sites based on the networking monitoring data, and the site uses wireless communication to bridge to the optimal proxy node to apply for network access.

[0093] Among them, when the site is currently unable to access the network in carrier mode, that is, the carrier does not communicate, the power line carrier signal is interfered, the hardware fails, or the network is congested, the networking status prediction result of the site is abnormal.

[0094] It can be understood that in this embodiment, the site whose networking status is predicted to be abnormal is bridged to the optimal proxy node by wireless communication to apply for network access, that is, the wireless communication connection between the sites in the dual-mode networking architecture is utilized to first transmit the data that the abnormal site needs to transmit to the optimal proxy node through wireless communication, and then the optimal proxy node forwards the data that the abnormal site needs to transmit to the main node through carrier communication. Therefore, the optimal proxy node updates the dynamic routing table entries according to the new network topology, ensuring that the data packets of each site can be correctly transmitted to the main node through the new path, and realizing the network adaptive switching of the communication channel of the site with abnormal operating environment status, thereby effectively improving the dual-mode communication and networking efficiency.

[0095] By repeatedly executing S4-S5, it is possible to predict the networking status of each site according to the current networking monitoring data in each cycle, and select the optimal proxy node from the remaining sites based on the networking monitoring data when the networking status prediction result of the site is abnormal, thereby adaptively switching the communication channel of the site with abnormal operating environment status, forming a unified dynamic routing table entry in the entire network, and greatly improving the dual-mode networking efficiency and networking effect.

[0096] Furthermore, in S5 of this embodiment, selecting the optimal proxy node from the remaining sites based on the network monitoring data includes:

[0097] S501, selecting a site within a preset distance range from a list of sites that have joined the network in the current cycle as a candidate proxy node.

[0098] Among them, in this embodiment, sites within a preset distance range are selected, that is, surrounding sites of sites with abnormal networking status prediction results are selected. Through the preliminary screening condition of distance, the communication stability between the candidate proxy node and the main node can be guaranteed to a certain extent.

[0099] S502: Based on the data packet information of the candidate proxy sites, calculate and obtain the wireless channel bandwidth occupancy rate of each candidate proxy site in the current cycle.

[0100] Specifically, the wireless channel bandwidth occupancy rate of each candidate proxy site can be expressed as:

[0101]

[0102] Among them, U represents the wireless channel bandwidth occupancy rate of the candidate proxy site, n represents the number of data packets required to be transmitted by the candidate proxy site, s represents the size of the data packet required to be transmitted by the candidate proxy site, and B represents the channel bandwidth.

[0103] S503: Select a candidate proxy site with the lowest wireless channel bandwidth occupancy rate in the current cycle as the optimal proxy node.

[0104] This embodiment preliminarily selects sites within a preset distance range from the list of sites that have been connected to the network in the current cycle as candidate proxy nodes, which can ensure the communication stability between the candidate proxy nodes and the main node, while reducing the signal attenuation and carrier communication interruption problems caused by excessive distance, and then calculates the wireless channel bandwidth occupancy rate based on the data packet information of the candidate proxy site, and selects the candidate proxy site with the wireless channel bandwidth occupancy rate as the optimal proxy node, which can quickly and effectively identify idle resources in the network and avoid further increasing traffic on high-load sites, thereby more effectively utilizing network bandwidth, helping to quickly realize the adaptive switching of subsequent abnormal site communication channels, and thus improving the transmission efficiency of the entire network.

[0105] Embodiment 2:

[0106] Compared with the dual-mode networking method provided in Example 1, this embodiment further considers the influence of signal strength on the optimal proxy node forwarding data of abnormal sites. This embodiment only describes the added content, and does not repeat the parts similar to Example 1 (such as S511-S512). Specifically, in this embodiment, the networking monitoring data also includes the signal strength information of each site. Then, in S5, the optimal proxy node is selected from the remaining sites based on the networking monitoring data, and it also includes:

[0107] S511, selecting a site within a preset distance range from the list of sites that have joined the network in the current cycle as a candidate proxy node.

[0108] S512: Based on the data packet information of the candidate proxy sites, calculate and obtain the wireless channel bandwidth occupancy rate of each candidate proxy site in the current cycle.

[0109] S513: Based on the signal strength information of the candidate proxy sites, a first-order lag filtering method is used to calculate the signal strength of each candidate proxy site in the current cycle.

[0110] Specifically, the signal strength R of the candidate proxy site can be expressed as:

[0111]

[0112] in, It represents the average signal strength of the candidate proxy site in history, α represents the filter coefficient, and R0 represents the signal strength value obtained by the candidate proxy site in the current period.

[0113] S514: Calculate and obtain a first evaluation value of each candidate proxy site based on the wireless channel bandwidth occupancy rate and signal strength of each candidate proxy site in the current cycle.

[0114] Specifically, in this embodiment, based on the wireless channel bandwidth occupancy rate and signal strength of each candidate proxy site in the current cycle, the expression for calculating the first evaluation value of each candidate proxy site is:

[0115]

[0116] Where i represents the serial number of the candidate proxy site, V 1i represents the first evaluation value of the i-th candidate proxy site, U i represents the wireless channel bandwidth occupancy rate of the i-th candidate proxy site, R i Represents the signal strength of the i-th candidate proxy site.

[0117] It can be understood that, in this embodiment, the candidate proxy station having the smallest wireless channel bandwidth occupancy rate and the strongest signal strength in the current cycle has the smallest first evaluation value.

[0118] S515: Select the candidate proxy site with the smallest first evaluation value as the optimal proxy node.

[0119] It is understandable that signal strength is easily affected by factors such as multipath effects, shadow fading, and interference, resulting in large fluctuations in signal strength in a short period of time. Therefore, this embodiment uses a first-order lag filtering method to filter the signal strength information of the candidate proxy site, providing a smoother and more stable signal strength estimation, and combining the two key indicators of the filtered signal strength and the wireless channel bandwidth occupancy rate as the basis for the subsequent optimal proxy node selection, which can reduce the adverse effects of instantaneous fluctuations on the optimal proxy node decision, thereby avoiding frequent network adjustments, making the decision based on signal strength and wireless channel bandwidth occupancy rate more stable and reliable, so that the selected optimal proxy node can ensure the subsequent high efficiency of data transmission to abnormal sites and the load balancing of the network.

[0120] Embodiment three:

[0121] Compared with the dual-mode networking method provided in Example 2, this embodiment further considers the influence of the communication success rate on the optimal proxy node forwarding the data of the abnormal site. This embodiment only describes the added content, and does not repeat the parts similar to Example 2 (such as S521-S523). Specifically, in this embodiment, the networking monitoring data also includes the total number of communications and successful communications between each site and the master node in the current cycle. In S5, the optimal proxy node is selected from the remaining sites based on the networking monitoring data, and it also includes:

[0122] S521, selecting a site within a preset distance range from the list of sites that have joined the network in the current cycle as a candidate proxy node.

[0123] S522: Based on the data packet information of the candidate proxy sites, calculate and obtain the wireless channel bandwidth occupancy rate of each candidate proxy site in the current cycle.

[0124] S523: Based on the signal strength information of the candidate proxy sites, a first-order lag filtering method is used to calculate the signal strength of each candidate proxy site in the current cycle.

[0125] S524, calculating the quotient of the number of successful communications between each candidate proxy site and the master node and the total number of communications between the corresponding candidate proxy site and the master node in the current cycle to obtain a first communication success rate of each candidate proxy site.

[0126] S525: Based on the first communication success rate of each candidate proxy point, a first-order lag filtering method is used to calculate a second communication success rate of each candidate proxy site in the current cycle.

[0127] Specifically, the second communication success rate P2 of the candidate proxy site can be expressed as:

[0128]

[0129] in, represents the communication success rate of the candidate proxy site in history, α represents the filter coefficient, and P1 represents the communication success rate of the candidate proxy site in the current period.

[0130] S526: Calculate and obtain a second evaluation value of each candidate proxy site based on the wireless channel bandwidth occupancy rate, signal strength, and second communication success rate of each candidate proxy site in the current cycle.

[0131] Specifically, in this embodiment, based on the wireless channel bandwidth occupancy rate, signal strength and second communication success rate of each candidate proxy site in the current cycle, the expression for calculating the second evaluation value of each candidate proxy site is:

[0132]

[0133] Where i represents the serial number of the candidate proxy site, V 2i represents the second evaluation value of the i-th candidate proxy site, P 2i represents the second communication success rate of the i-th candidate proxy site, U i represents the wireless channel bandwidth occupancy rate of the i-th candidate proxy site, R i Represents the signal strength of the i-th candidate proxy site.

[0134] It can be understood that, in this embodiment, the candidate proxy station with the smallest wireless channel bandwidth occupancy rate, the strongest signal strength and the highest communication success rate in the current cycle has the smallest second evaluation value.

[0135] S527: Select the candidate proxy site with the smallest second evaluation value as the optimal proxy node.

[0136] It is understandable that in an actual dual-mode networking environment, the communication success rate of a site may be affected by various noises and instantaneous interferences, resulting in data fluctuations. To this end, this embodiment uses a first-order lag filtering method to filter the communication success rate of the candidate proxy site, and selects the optimal proxy node by comprehensively considering the wireless channel bandwidth occupancy rate, the signal strength after filtering, and the communication success rate after filtering. Among the candidate proxy sites that have been searched and have been connected to the network, the candidate proxy site with the smallest wireless channel bandwidth occupancy rate, the strongest signal strength and the highest communication success rate is adaptively selected as the optimal proxy node to apply for network access, ensuring that the abnormal site has a more stable communication connection with the main node through the optimal proxy node, reducing the risk of communication failure and reducing the delay in data transmission, thereby optimizing the communication path of the entire network, enhancing resistance to interference and abnormal situations, and thereby improving the stability, reliability and intelligence level of the entire dual-mode network system.

[0137] Furthermore, in this embodiment, the networking monitoring data includes the networking level at which each site is located, and selecting the optimal proxy node from the remaining sites based on the networking monitoring data also includes:

[0138] When there are at least two optimal proxy nodes selected, the optimal proxy node with the smallest networking level is selected as the final optimal proxy node;

[0139] When at least two optimal proxy nodes are located at the same networking level, the optimal proxy node with the smallest wireless channel bandwidth occupancy rate among the optimal proxy nodes at the same level is selected as the final optimal proxy node.

[0140] This embodiment helps to maintain the flatness of the network structure, reduce the number of communication hops, and reduce the transmission delay of data in the network by selecting the optimal proxy node with the smallest networking level from multiple optimal proxy nodes. It also helps to balance the network load and prevent certain nodes from becoming performance bottlenecks due to taking on too many forwarding tasks, which is particularly important for power system applications with high real-time requirements.

[0141] On the other hand, in this embodiment, the following steps may be further included after step S5:

[0142] S6, in the process of the current period site using wireless communication to bridge to the optimal proxy node to apply for network access, the wireless channel bandwidth occupancy rate of the optimal proxy node is calculated in real time;

[0143] S7, when the wireless channel bandwidth occupancy rate of the optimal proxy node reaches a preset upper limit threshold, calculate the second evaluation values ​​of the remaining candidate proxy sites except the current optimal proxy node, and select the candidate proxy site with the smallest second evaluation value from the remaining candidate proxy sites as an auxiliary proxy node, and the auxiliary proxy node is used to share the transmission data volume of the optimal proxy node;

[0144] S8, when the wireless channel bandwidth occupancy rate Wt1 of the optimal proxy node and the wireless channel bandwidth occupancy rate Wt2 of the auxiliary proxy node drop to the preset lower limit threshold and the sum of the wireless channel bandwidth occupancy rates of the optimal proxy node and the auxiliary proxy node does not exceed the wireless channel bandwidth occupancy rate of the optimal proxy node or the wireless channel bandwidth occupancy rate of the auxiliary proxy node, that is, Wr1+Wr2≤(Wt1 or Wt2), disconnect the bridge between the node with the second lower communication success rate among the optimal proxy node and the auxiliary proxy node and the site.

[0145] This embodiment further calculates the wireless channel bandwidth occupancy rate of the optimal proxy node in real time, and introduces an auxiliary proxy node to share the data transmission load when the preset upper limit threshold is reached, which helps to achieve dynamic load balancing, avoid overload of a single node, and effectively improve the stability of the network. When the wireless channel bandwidth occupancy rate of the optimal proxy node and the auxiliary proxy node decreases and reaches the preset condition, the bridge between the node with the second lower communication success rate among the optimal proxy node and the auxiliary proxy node and the site is disconnected, which helps to simplify the network structure and avoid wasting network resources. This solution can adaptively adjust the bridging relationship between the proxy node and the abnormal site according to the actual operating environment of the dual-mode network, so that the dual-mode networking system can better adapt to fluctuations in network traffic.

[0146] Embodiment 4:

[0147] See also Figure 2 , Figure 2 A schematic diagram of the structure of a dual-mode networking system provided for one embodiment of this specification.

[0148] As shown in the figure, the dual-mode networking system may at least include a master node, several sites, a data acquisition module 1, a model building module 2, a model training module 3, and a dual-mode communication scheduling module 4. The sites are connected to the master node by carriers, and the sites are connected by wireless communication.

[0149] Data acquisition module 1, used to acquire historical network monitoring data and real-time network monitoring data;

[0150] Model building module 2, used to build a network anomaly identification model, the input of the network anomaly identification model is the network monitoring data, and the output of the network anomaly identification model is the network status prediction result of each site;

[0151] Model training module 3, used to train the network anomaly recognition model based on historical network monitoring data to obtain a trained network anomaly recognition model;

[0152] The dual-mode communication scheduling module 4 is used for, when the preset network access time of the site is reached, the site preferentially applies for network access by carrier mode, and inputs the real-time acquired networking monitoring data into the trained networking anomaly recognition model to obtain the network status prediction result of each site output by the model. If the networking status prediction result of the site is normal, the carrier transmission between the site and the main node is maintained. If the networking status prediction result of the site is abnormal, the optimal proxy node is selected from the remaining sites based on the networking monitoring data, and the site uses wireless communication to bridge to the optimal proxy node to apply for network access.

[0153] It can be understood that the technical concept of a dual-mode networking system provided in this embodiment is similar to the technical concept of a dual-mode networking method provided in the aforementioned multiple embodiments, and this embodiment will not be repeated here.

[0154] The above descriptions are only preferred embodiments disclosed in this application and descriptions of the technical principles used. Those skilled in the art should understand that the scope of protection involved in this disclosure is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in this disclosure (but not limited to) to form a technical solution.

[0155] In addition, although each operation is described in a specific order, this should not be understood as requiring these operations to be performed in the specific order shown or in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

Claims

1. A dual-mode networking method for realizing a dual-mode networking system of high-speed power line carrier and high-speed wireless communication, characterized in that: The dual-mode networking system includes a master node and a plurality of sites, the sites are connected to the master node via a carrier wave, and the sites are connected to each other via wireless communication. The method includes the following steps: S1, obtain historical network monitoring data; S2, constructing a networking anomaly identification model, wherein the input of the networking anomaly identification model is the networking monitoring data, and the output of the networking anomaly identification model is the networking status prediction result of each site; S3, training the network anomaly recognition model based on historical network monitoring data to obtain a trained network anomaly recognition model; S4, when the preset network access time of the site is reached, the site preferentially applies for network access by carrier mode and obtains network monitoring data in real time; S5, input the real-time acquired networking monitoring data into the trained networking anomaly recognition model to obtain the networking status prediction result of each site output by the model. If the networking status prediction result of the site is normal, the carrier transmission between the site and the main node is maintained. If the networking status prediction result of the site is abnormal, the optimal proxy node is selected from the remaining sites based on the networking monitoring data, and the site uses wireless communication to bridge to the optimal proxy node to apply for network access.

2. A dual-mode networking method as claimed in claim 1, characterized in that: The historical networking monitoring data includes a list of sites that have been connected to the network in the historical period and the actual networking status of each network in the historical period. In S3, the network anomaly recognition model is trained based on the historical networking monitoring data to obtain a trained network anomaly recognition model, including: S301, based on the actual networking status of each network in the historical period, annotate the historical networking monitoring data in the normal networking status and the abnormal networking status, and divide the annotated historical networking monitoring data into a training set and a verification set; S302, setting an initial value of an abnormal time threshold; S303, training the networking anomaly recognition model based on the training set, using the initial value of the abnormal time threshold to make a prediction, if the site is in the list of sites that have been connected to the network when the preset network access time of the site is reached and the abnormal time threshold is exceeded, then the prediction result of the networking state of the site is judged to be normal, if the site is not in the list of sites that have been connected to the network when the preset network access time of the site is reached and the abnormal time threshold is exceeded, then the prediction result of the networking state of the site is abnormal; S304, after the training is completed, using the validation set to evaluate the error between the prediction result of the model and the actual networking state; S305, adjusting the abnormal time threshold based on the error; S306, retraining the networking anomaly recognition model based on the training set, and using the adjusted anomaly time threshold for prediction; S307, repeatedly execute S304 to S306 until the error is lower than the preset threshold, and obtain the optimal abnormal time threshold, which is used as the judgment standard for the trained networking abnormality recognition model to predict the networking status.

3. A dual-mode networking method as claimed in claim 1, characterized in that: The network monitoring data includes a list of sites that have been connected to the network in the current cycle and data packet transmission information of each site. In S5, the optimal proxy node is selected from the remaining sites based on the network monitoring data, including: S501, selecting a site within a preset distance range from a list of sites that have been connected to the network in the current cycle as a candidate proxy node; S502, based on the data packet information of the candidate proxy site, calculate the wireless channel bandwidth occupancy rate of each candidate proxy site in the current cycle; S503: Select a candidate proxy site with the lowest wireless channel bandwidth occupancy rate in the current cycle as the optimal proxy node.

4. A dual-mode networking method as claimed in claim 3, characterized in that: The networking monitoring data includes the signal strength information of each site. Then, in S503, the candidate proxy site with the lowest wireless channel bandwidth occupancy rate in the current cycle is selected as the optimal proxy node, and the following further includes: Based on the signal strength information of the candidate proxy sites, the signal strength of each candidate proxy site in the current cycle is calculated using the first-order lag filtering method; Based on the wireless channel bandwidth occupancy rate and signal strength of each candidate proxy site in the current cycle, a first evaluation value of each candidate proxy site is calculated; The candidate proxy site with the smallest first evaluation value is selected as the optimal proxy node.

5. A dual-mode networking method as claimed in claim 4, characterized in that: Based on the wireless channel bandwidth occupancy rate and signal strength of each candidate proxy site in the current cycle, the expression for calculating the first evaluation value of each candidate proxy site is: Where i represents the serial number of the candidate proxy site, V 1i represents the first evaluation value of the i-th candidate proxy site, U i represents the wireless channel bandwidth occupancy rate of the i-th candidate proxy site, R i Represents the signal strength of the i-th candidate proxy site.

6. A dual-mode networking method as claimed in claim 4, characterized in that: The network monitoring data includes the total number of communications between each site and the master node and the number of successful communications in the current cycle, and the candidate proxy site with the smallest first evaluation value is selected as the optimal proxy node, and also includes: The quotient of the number of successful communications between each candidate proxy site and the master node and the total number of communications between the corresponding candidate proxy site and the master node in the current cycle is calculated to obtain the first communication success rate of each candidate proxy point; Based on the first communication success rate of each candidate proxy point, a first-order lag filtering method is used to calculate the second communication success rate of each candidate proxy site in the current cycle; Based on the wireless channel bandwidth occupancy rate, signal strength and second communication success rate of each candidate proxy site in the current cycle, a second evaluation value of each candidate proxy site is calculated; The candidate proxy site with the smallest second evaluation value is selected as the optimal proxy node.

7. A dual-mode networking method as claimed in claim 6, characterized in that: Based on the wireless channel bandwidth occupancy rate, signal strength and second communication success rate of each candidate proxy site in the current cycle, the expression for calculating the second evaluation value of each candidate proxy site is: Where i represents the serial number of the candidate proxy site, V 2i represents the second evaluation value of the i-th candidate proxy site, P 2i represents the second communication success rate of the i-th candidate proxy site, U i represents the wireless channel bandwidth occupancy rate of the i-th candidate proxy site, R i Represents the signal strength of the i-th candidate proxy site.

8. A dual-mode networking method according to any one of claims 4 to 6, characterized in that: The network monitoring data includes the network level of each site, and selecting the best proxy node from the remaining sites based on the network monitoring data also includes: When there are at least two optimal proxy nodes selected, the optimal proxy node with the smallest networking level is selected as the final optimal proxy node; When at least two optimal proxy nodes are located at the same networking level, the optimal proxy node with the smallest wireless channel bandwidth occupancy rate among the optimal proxy nodes at the same level is selected as the final optimal proxy node.

9. A dual-mode networking method as claimed in claim 6, characterized in that: The following steps are also included: S6, in the process of the current period site using wireless communication to bridge to the optimal proxy node to apply for network access, the wireless channel bandwidth occupancy rate of the optimal proxy node is calculated in real time; S7, when the wireless channel bandwidth occupancy rate of the optimal proxy node reaches a preset upper limit threshold, calculate the second evaluation values ​​of the remaining candidate proxy sites except the current optimal proxy node, and select the candidate proxy site with the smallest second evaluation value from the remaining candidate proxy sites as an auxiliary proxy node, the auxiliary proxy node is used to share the transmission data volume of the optimal proxy node; S8, when the wireless channel bandwidth occupancy rate of the optimal proxy node and the wireless channel bandwidth occupancy rate of the auxiliary proxy node drop to a preset lower limit threshold and the sum of the wireless channel bandwidth occupancy rates of the optimal proxy node and the auxiliary proxy node does not exceed the wireless channel bandwidth occupancy rate of the optimal proxy node or the wireless channel bandwidth occupancy rate of the auxiliary proxy node, disconnect the bridge between the node with a lower second communication success rate among the optimal proxy node and the auxiliary proxy node and the site.

10. A dual-mode networking system, characterized in that: It includes a master node, several sites, a data acquisition module, a model building module, a model training module, and a dual-mode communication scheduling module. The sites are connected to the master node by a carrier wave, and the sites are connected by wireless communication. Data acquisition module, used to obtain historical network monitoring data and real-time network monitoring data; A model building module is used to build a network anomaly identification model, the input of the network anomaly identification model is the network monitoring data, and the output of the network anomaly identification model is the network status prediction result of each site; A model training module is used to train a network anomaly recognition model based on historical network monitoring data to obtain a trained network anomaly recognition model; The dual-mode communication scheduling module is used for, when the preset network access time of the site is reached, the site preferentially applies for network access by carrier mode, and inputs the real-time acquired networking monitoring data into the trained networking anomaly recognition model to obtain the network status prediction result of each site output by the model. If the networking status prediction result of the site is normal, the carrier transmission between the site and the main node is maintained. If the networking status prediction result of the site is abnormal, the optimal proxy node is selected from the remaining sites based on the networking monitoring data, and the site uses wireless communication to bridge to the optimal proxy node to apply for network access.

Images