Multi-channel switching method for intelligent water meter communication

Abstract

The invention relates to the technical field of digital information transmission, in particular to a multi-channel switching method for intelligent water meter communication, which comprises the following steps of: acquiring round-trip time observation data of a water meter node in a communication process; if the deviation accumulation rate corresponding to the round-trip time observation data is greater than a preset communication anomaly threshold, requesting and acquiring a historical deviation accumulation rate sequence returned by the plurality of neighbor nodes; constructing a channel switching probability of the water meter node according to the change trend consistency and trend intensity of the historical deviation accumulation rate sequence corresponding to the water meter node and the plurality of neighbor nodes; and if the channel switching probability is greater than a preset switching threshold, performing channel switching on the water meter node. According to the scheme, the regional congestion condition can be dynamically evaluated by utilizing the current network state and the collaborative sensing result of the neighbor nodes, and part of the affected nodes are intelligently selected to be switched to the alternative communication channels, so that the real-time performance and continuity of key meter reading data and the transmission reliability of the whole communication network are effectively guaranteed.

Description

Technical Field

[0001] This application relates to the field of digital information transmission technology, and particularly to a multi-channel switching method for intelligent water meter communication. Background Art

[0002] In high-density residential areas, with the advancement of smart water service construction, intelligent water meters are widely deployed to achieve remote meter reading and centralized management of water consumption data. LoRa (Long Range) technology is widely used in the wireless communication of intelligent water meters due to its characteristics of low power consumption, wide coverage, and easy deployment. The LoRa network adopts an ALOHA-like random access mechanism and lacks a strict collision avoidance strategy. Especially in high-density deployment scenarios, when a large number of nodes report data or perform periodic meter reading within a short period of time, regional communication congestion is likely to occur.

[0003] The regional congestion caused by LoRa technology will result in an increase in data reporting delay, an increase in packet loss rate, and energy consumption waste caused by frequent retries. Existing technologies mainly adopt communication quality optimization strategies focused on individual nodes, such as increasing the number of data retransmissions, dynamically adjusting the transmission power, or spreading factor, etc. to recover communication. These methods have the following deficiencies: Single-node optimization cannot identify the spatial aggregation characteristics of communication anomalies and it is difficult to timely judge whether systematic congestion occurs in a local area. At the same time, these methods rely on frequent retransmissions, lack dynamic adaptive adjustment and fast self-healing capabilities, and it is difficult to achieve a quick response and effective solution to regional congestion. Therefore, existing technologies lack an effective identification and response mechanism for regional communication anomalies, cannot achieve precise diversion of some key nodes, and limit the scalability and operational stability of the network in high-density deployment scenarios. Summary of the Invention

[0004] To ensure the stable transmission of remote meter reading data and the overall performance of the network, this application provides a multi-channel switching method for intelligent water meter communication.

[0005] In a first aspect, this application provides a multi-channel switching method for intelligent water meter communication, adopting the following technical solution: A multi-channel switching method for intelligent water meter communication includes the following steps: Obtain the round-trip time observation data of the water meter node during the communication process; If the deviation accumulation rate corresponding to the round-trip time observation data is greater than a preset communication anomaly threshold, request and obtain the historical deviation accumulation rate sequences returned by multiple neighbor nodes; Construct the channel switching probability of the water meter node according to the change trend consistency and trend intensity of the historical deviation accumulation rate sequences corresponding to the water meter node and the multiple neighbor nodes; If the channel switching probability is greater than a preset switching threshold, perform channel switching on the water meter node; Among them, constructing the channel switching probability of the water meter node includes: performing Pearson correlation calculation on the historical deviation accumulation rate sequences of the water meter node and the multiple neighbor nodes to obtain n Pearson correlation coefficients, calculating the average correlation coefficient of the n Pearson correlation coefficients, performing linear fitting on the historical deviation accumulation rate sequence of each neighbor node and calculating the trend slope, and calculating the average trend intensity of the multiple neighbor nodes according to the trend slope. The channel switching probability of the water meter node is equal to the product of the average correlation coefficient and the average trend intensity.

[0006] Optionally, calculating the deviation accumulation rate corresponding to the round-trip time observation data includes: Obtain the communication round-trip time when the water meter node sends data for the i-th time in the round-trip time observation data; Select the communication round-trip time values of the M consecutive successful communications in the history of the water meter node, and calculate the mean value of the communication round-trip time values as the reference communication round-trip time; Calculate the corresponding deviation accumulation rate by continuously monitoring the relative deviation between the communication round-trip time of the water meter node and the reference communication round-trip time.

[0007] Optionally, the calculating the corresponding deviation accumulation rate by continuously monitoring the relative deviation between the communication round-trip time of the water meter node and the reference communication round-trip time includes: Obtain the first N successful communication round-trip time values before the water meter node sends data for the i-th time as a sliding observation window, and calculate the deviation accumulation rate between the communication round-trip time corresponding to the sliding observation window and the reference communication round-trip time, where the window length is N.

[0008] Optionally, obtaining the round-trip time observation data during the communication process of the water meter node includes: Each water meter node records its own sending time and the time when the gateway returns the Ack message indicating receipt during communication; Calculate the round-trip time observation data of the water meter node a when it sends data for the i-th time .

[0009] Optionally, requesting and obtaining the historical deviation accumulation rate sequences returned by multiple neighbor nodes includes: Set a communication quality abnormal judgment threshold according to the deviation accumulation rate of the current water meter node during the communication process ; If the deviation accumulation rate when the current water meter node sends data satisfies , then judge that the communication of the current water meter node is in an abnormal state; Based on the abnormal state, start a broadcast request, send a collection request to surrounding neighbor nodes, and obtain the historical deviation accumulation rate sequences of the neighbor nodes.

[0010] Optionally, the slope of the linear fitting of the historical deviation accumulation rate sequence is the degree of change of the corresponding values of the historical deviation accumulation rate sequence, and the degree of change of the corresponding values of the historical deviation accumulation rate sequence reflects the degree of deterioration of the communication quality of each neighbor node.

[0011] Optionally, after calculating the average trend intensity of the multiple neighbor nodes, it further includes: Normalize the average trend intensity to the range of (0, 1) through the arctangent normalization method to obtain the normalized average trend intensity.

[0012] Optionally, if the channel switching probability is greater than a preset switching threshold, the channel switching of the water meter node includes: When the channel switching probability is greater than the preset switching threshold, change the communication mode of the water meter node from LoRa to NB-IoT.

[0013] The present application has the following technical effects: It can utilize the current network state and the collaborative perception results of neighbor nodes to dynamically evaluate the regional congestion situation, and intelligently select some affected nodes to switch to the alternative NB-IoT communication channel, effectively ensuring the real-time performance, continuity of key meter reading data, and the transmission reliability of the overall communication network. Description of the Drawings

[0014] Figure 1 is a schematic flowchart of steps S1 - S4 in a multi-channel switching method for intelligent water meter communication in the present application. Detailed Embodiments

[0015] The embodiments of the present application disclose a multi-channel switching method for intelligent water meter communication. Referring to Figure 1 , it includes the following steps: S1: Obtain the round-trip time observation data of the water meter node during the communication process.

[0016] It should be noted that the specific scenario targeted by the present application is: an intelligent water meter remote meter reading system in a high-density residential area. In such an environment, due to the extensive deployment of intelligent water meters and the real-time data upload requirements of a large amount of data, the communication network will face congestion problems. Especially in the case of the application of LoRa technology, when multiple intelligent water meter nodes frequently report data, it is easy to form network resource competition within a local area, resulting in communication bottlenecks, data loss, and transmission delays.

[0017] Thus, in the wireless communication of intelligent water meters, the communication quality of nodes is affected by factors such as channel interference and load changes. The round-trip time observation data (round-trip time ), as a real-time and continuous communication status feedback signal, will show an upward trend when communication is congested or the communication quality deteriorates. However, a single fluctuation is vulnerable to instantaneous interference (such as temporary occlusion, environmental noise, electromagnetic interference), and directly using a single value to judge communication anomalies is prone to false alarms. Therefore, by introducing an observation window for continuous observation, the relative deviation between the actual communication for multiple consecutive times and the reference is calculated, and the average degree of the increase in the communication time for multiple consecutive times is defined as the deviation accumulation rate to determine whether the node is in an abnormal situation.

[0018] S2: If the deviation accumulation rate corresponding to the round-trip time observation data is greater than the preset communication anomaly threshold, request and obtain the historical deviation accumulation rate sequences returned by multiple neighbor nodes.

[0019] In an embodiment of the present application, the analysis process of the round-trip time of a single water meter node is as follows: 1. Obtain the round-trip time for each communication of the node; Each node records its own sending time and the time when the gateway returns the Ack message indicating receipt during communication, and calculates the round-trip time of node a when sending data for the i-th time.

[0020] 2. Set the average value of the historical M times of the node as the reference ; 3. Calculate the deviation accumulation rate through the values of the previous N times before each communication; Due to differences in the installation environment, the during normal communication of different nodes itself varies. For each node, during the initial stage of node deployment or when the communication status is good, the values of multiple communications are selected to calculate the average value as the reference, which can more stably and objectively reflect the communication level of the node in the non-abnormal state and effectively reduce the misjudgment risk caused by a single deviation. Select the values of M consecutive successful communications, calculate the average value as the reference , denoted as , and M takes the empirical value of 5.

[0021] 3. Calculate the deviation accumulation rate through the values of the previous N times before each communication ; Among them, , in the actual communication of remote meter reading for intelligent water meters, the round-trip time of nodes ( ) will fluctuate due to the transmission conditions in the communication link. If relying solely on a single communication to determine whether the communication is abnormal, it is extremely vulnerable to instantaneous interference and environmental noise, resulting in false alarms. Therefore, an anomaly metric based on an observation window is introduced to more stably reflect the changing trend of the communication quality of nodes. By continuously monitoring the relative deviation between the actual communication of nodes and a benchmark , using the recent N communication data as the observation data for the sliding window, with the window length being N, and N taking the empirical value of 10, the average degree of increase in consecutive communication times is calculated and defined as the deviation accumulation rate, which measures whether there is an increase in the recent communication time of nodes and reflects whether there is an anomaly in the communication quality of the link. Obtain the values of the previous N successful communications when the node sends data for the i-th time as the sliding observation window , and calculate the deviation accumulation rate of this sliding observation window as an indicator of whether there is an anomaly in the communication quality of this node. Composition: max(0, the relative growth of the difference between the single communication

[0022] and the benchmark relative to the benchmark value). The meaning of max(0,) is to focus on the increase in the communication round-trip time and ignore short-term decreases. Furthermore, when the communication quality of a certain node deteriorates and causes a communication anomaly (the deviation accumulation rate exceeds the threshold), if directly switching channels based on the deviation accumulation rate

[0023] of a single communication directly, it may cause multiple neighbor nodes to switch channels simultaneously in a local area, resulting in a sudden drop in the usage rate of a certain channel and affecting the usage stability of the transmission system. To avoid the instability impact caused by multiple neighbor nodes switching simultaneously in a local area, in this step, when detecting a node communication anomaly, the historical deviation accumulation rate sequence of neighbor nodes is obtained through broadcasting to determine whether the anomaly has "regionality and consistency", so as to adopt a more appropriate switching strategy.

[0024] In an embodiment of the present application, the specific implementation method is as follows: 1. Determine whether a node has a communication anomaly by setting a communication anomaly threshold (preset communication anomaly threshold).

[0025] 2. When a communication anomaly occurs, a broadcast request is initiated to request multiple neighbor nodes to return the historical deviation accumulation rate sequence R.

[0026] The specific implementation is as follows: (1) By setting the communication abnormality threshold Determine whether the node has communication abnormality; According to the current node Cumulative deviation rate during communication , set the judgment threshold of abnormal communication quality , when sending data satisfy , it is determined that the communication of the current node is abnormal.

[0027] (2) When a communication anomaly occurs, a broadcast request is initiated to request multiple neighbor nodes to return a sequence of historical deviation accumulation rates. ; When a communication anomaly is detected, a broadcast request is initiated to send a collection request to the surrounding neighbor nodes to obtain the communication quality fluctuation data of the neighbor nodes, that is, the R value sequence of the neighbor nodes. The number of neighbor nodes n and the historical deviation accumulation rate sequence of the previous T communications of n neighbor nodes b are obtained. . T gains experience value 10.

[0028] Among them, neighbor nodes refer to other intelligent terminal nodes that are within the broadcast coverage range and can directly receive and respond to the current node's broadcast signal under the same LoRa parameter configuration. After the broadcast response, the current node can collect the communication time deviation accumulation rate sequence of multiple neighbor nodes (that is, the R value sequence of each neighbor), which is used in the subsequent steps to determine whether the current communication anomaly has regional characteristics.

[0029] S3: Construct the channel switching probability of the water meter node based on the consistency and trend strength of the change trend of the historical deviation accumulation rate series corresponding to the water meter node and multiple neighboring nodes.

[0030] It should be noted that after obtaining the historical deviation accumulation rate sequence of neighboring nodes, in order to prevent the entire local area from switching at the same time, causing a sudden drop in the utilization rate of a single channel and affecting the stability of use, the current node needs to determine whether there is a trend consistency in the deviation accumulation rate sequence of its own node and neighboring nodes. If the communication change trend of multiple neighboring nodes is consistent with that of the current node, it can be determined that the anomaly of the current node is "regional". Prioritize switching areas with consistent communication changes, so that when switching, areas with consistent changes in the deviation accumulation rate sequence can be switched first to alleviate the communication pressure of the channel and avoid causing the value of the deviation accumulation rate to drop. The sizes are the same but the changing trends are different. Even after entering the subsequent channel switching, the regional channel pressure cannot be alleviated, resulting in ineffective switching. At the same time, low-intensity changes will also lead to the consistency of the changing trend. When considering consistency, the intensity of the changing trend also needs to be considered. The mean value of the change can be obtained from the slope of the straight line fitted by the deviation accumulation rate sequence, which is used as an index of the change intensity. Furthermore, it is necessary to comprehensively analyze the consistency and intensity of the changing trends of the deviation accumulation rate sequences of the neighbor nodes and the self-node, so as to obtain the switching probability of this node.

[0031] In an embodiment of the present application, the process of constructing the channel switching probability by the water meter self-node is as follows: 1. Obtain the historical deviation accumulation rate sequence of the first T communications of the self-node a . T takes the empirical value of 10.

[0032] 2. Calculate the similarity of the communication quality changes between the self-node and multiple neighbor nodes.

[0033] After obtaining the historical deviation accumulation rate sequences of the first T communications of the self and multiple neighbor nodes, consider their correlation, that is, the consistency of the changing trend. A high correlation indicates that the changes of the neighbor nodes are similar to those of the self, and communication anomalies have occurred. This may be due to the fact that the self-node and the neighbor nodes use the same channel for transmission during transmission, and it is possible that the channel is congested. Multiple nodes keep retransmitting because they cannot receive the Ack confirmation information of successful transmission, resulting in channel congestion. A low correlation indicates that it may be only a local problem caused by abnormal interference received by the current node, and switching at the self level can ensure transmission efficiency and stability.

[0034] The implementation steps are as follows: 1) Perform Pearson correlation calculation on the historical deviation accumulation rate sequences of the self-node a and n neighbor nodes b to obtain n Pearson correlation coefficient values , and then normalize them using the normalization method to obtain ; To determine whether there is consistency in the communication changes between two nodes, if the correlation is high (close to 1), it indicates that the communication quality changes of the two are similar, and they may both be affected by channel congestion; if the correlation is low (close to 0), it indicates that there is no same factor affecting them currently, and their changes may be affected by their own environments.

[0035] 2) Calculate the average correlation coefficient of all n neighbor nodes ; Calculate the average correlation coefficient of n neighboring nodes of node a, and consider the correlation between all neighboring nodes b and node a. If the average correlation is high, it means that most of the neighboring nodes have poor communication quality, which may be related to the fact that they all use the same channel for transmission, causing channel congestion.

[0036] 3. Calculate the trend strength of the communication quality changes of multiple neighboring nodes b.

[0037] Detailed logic: Since correlation analysis only considers the degree of correlation between sequences, low-intensity changes will also lead to a high degree of correlation between the node itself and multiple neighboring nodes. Therefore, we cannot only consider the degree of correlation, but also the intensity of communication quality changes. We use the least squares method to linearly fit the cumulative rate series of deviations of multiple neighboring nodes to obtain the trend slope to quantify the intensity of changes. Since only the deterioration is considered when calculating the R value, the slope reflects the degree of deterioration of communication quality. If the average slope of all neighboring nodes is high, it means that the degree of deterioration of the communication quality of all neighboring nodes is high, and it is necessary to switch channels for some or all nodes to ensure the real-time and stability of transmission.

[0038] The implementation steps are as follows: 1) Use the least squares method to perform linear fitting on the historical deviation accumulation rate sequence R of each neighbor node b to calculate the trend slope k.

[0039] The slope of the linear fitting of the historical deviation accumulation rate series is the degree of change of the R value, which reflects the degree of deterioration of the communication quality of each neighbor node b.

[0040] 2) Calculate the average trend strength of all n neighbor nodes .

[0041] Calculate the average correlation coefficient of the n neighboring nodes of node a, and consider the average of the communication quality deterioration trend intensity of all neighboring nodes b. If the average trend intensity is high, it means that most of the neighboring nodes have a high intensity of communication quality deterioration, which may be due to high congestion in the channel, resulting in a high degree of communication quality deterioration of multiple nodes.

[0042] 3) The average trend strength is normalized by the inverse tangent method Normalized to the range of (0,1) to obtain the normalized average trend strength .

[0043] Using the Inverse Tangent Function Average Trend Strength Convert to angle. Normalize the angle to the range of (0,1) according to the range of the value. Not only does it retain the relative size of the slope, but it also makes the result more intuitive and easier to compare through normalization.

[0044] 4. Construct the handover probability S of the current node.

[0045] When considering the handover probability of a single node a, the consistency of the change trend of communication quality and the situation of change intensity should be comprehensively considered. Construct a handover probability index to evaluate the handover probability S of the current node. The current node a constructs its handover probability S: S = Correlation coefficient with the neighbor R value sequence * Mean value of the neighbor R value trend slope.

[0046] S4: If the channel handover probability is greater than the preset handover threshold, perform channel handover on the water meter node.

[0047] In an embodiment of the present application, it is judged whether the current node needs to switch channels according to the preset handover threshold. When the channel handover probability of the node is greater than the preset handover threshold, the communication mode of the node is switched from LoRa to NB-IoT to relieve the load pressure of the LoRa network and ensure the reliability of key node communication. The threshold is set to 0.5 using an empirical value.

[0048] In summary, by implementing the solution of the present application, it is possible to utilize the current network state and the collaborative perception results of neighbor nodes, dynamically evaluate the regional congestion situation, and intelligently select some affected nodes to switch to the alternative NB-IoT communication channel, effectively ensuring the real-time performance, continuity of key meter reading data, and the transmission reliability of the overall communication network.

[0049] The above are all preferred embodiments of the present application, and the protection scope of the present application is not limited thereby. Therefore, all equivalent changes made according to the structure, shape, and principle of the present application should be covered within the protection scope of the present application.

Claims

1. A multi-channel switching method for intelligent water meter communication, characterized in that, It includes the following steps: Obtain the round-trip time observation data of the water meter node during the communication process; If the deviation accumulation rate corresponding to the round-trip time observation data is greater than the preset communication anomaly threshold, request and obtain the historical deviation accumulation rate sequences returned by multiple neighbor nodes; Construct the channel switching probability of the water meter node according to the change trend consistency and trend strength of the historical deviation accumulation rate sequences corresponding to the water meter node and the multiple neighbor nodes; If the channel switching probability is greater than the preset switching threshold, perform channel switching on the water meter node; Among them, constructing the channel switching probability of the water meter node includes: performing Pearson correlation calculation on the historical deviation accumulation rate sequences of the water meter node and the multiple neighbor nodes to obtain n Pearson correlation coefficients, calculating the average correlation coefficient of the n Pearson correlation coefficients, performing linear fitting on the historical deviation accumulation rate sequence of each neighbor node and calculating the trend slope, and calculating the average trend strength of the multiple neighbor nodes according to the trend slope. The channel switching probability of the water meter node is equal to the product of the average correlation coefficient and the average trend strength.

2. The method according to claim 1, wherein Calculating the deviation accumulation rate corresponding to the round-trip time observation data includes: Obtain the communication round-trip time when the water meter node sends data for the i-th time in the round-trip time observation data; Select the communication round-trip time values of the water meter node's historical consecutive M successful communications, and calculate the mean of the communication round-trip time values as the reference communication round-trip time; Calculate the corresponding deviation accumulation rate by continuously monitoring the relative deviation between the communication round-trip time of the water meter node and the reference communication round-trip time.

3. The method according to claim 2, wherein The calculating the corresponding deviation accumulation rate by continuously monitoring the relative deviation between the communication round-trip time of the water meter node and the reference communication round-trip time includes: Obtain the first N successful communication round-trip time values before the i-th data transmission of the water meter node as a sliding observation window, and calculate the deviation accumulation rate between the communication round-trip time corresponding to the sliding observation window and the reference communication round-trip time, where the window length is N.

4. The method according to claim 1, wherein Obtaining the round-trip time observation data of the water meter node during the communication process includes: Each water meter node records its own transmission time and the time when the gateway returns the Ack message indicating receipt during communication; Calculate the observed round-trip time data of the water meter node a during the i-th data transmission .

5. The method according to claim 1, characterized in that Requesting and obtaining the historical deviation accumulation rate sequences returned by multiple neighbor nodes includes: Set the communication quality exception judgment threshold according to the deviation accumulation rate of the current water meter node during the communication process ; If the deviation accumulation rate when the current water meter node sends data satisfies then it is determined that the communication of the current water meter node is in an abnormal state; Based on the abnormal state, initiate a broadcast request, send a collection request to the surrounding neighbor nodes, and obtain the historical deviation accumulation rate sequences of the neighbor nodes.

6. The method according to claim 1, wherein The slope of the linear fitting of the historical deviation accumulation rate sequence is the degree of change of the corresponding value of the historical deviation accumulation rate sequence, and the degree of change of the corresponding value of the historical deviation accumulation rate sequence reflects the degree of deterioration of the communication quality of each neighbor node.

7. The method according to claim 1, characterized in that, After calculating the average trend strength of the multiple neighbor nodes, it further includes: Normalize the average trend strength to the range of (0, 1) by the arctangent normalization method to obtain the normalized average trend strength.

8. The method according to claim 1, characterized in that, If the channel switching probability is greater than the preset switching threshold, performing channel switching on the water meter node includes: When the channel switching probability is greater than a preset switching threshold, switch the communication mode of the water meter node from LoRa to NB-IoT.

Images