Real-time communication network construction method and system based on loRa ad hoc network technology

Abstract

The application discloses a real-time communication network construction method and system based on LoRa ad hoc network technology, wherein a core layer gateway extracts service flow features, carries out spatial clustering, and determines a cluster head node; the cluster head node broadcasts a clustering result; service flow classification is carried out, and an access mode of different service flow classifications is determined; expected transmission times and node residual energy are calculated through bidirectional link probe packets, and a multi-hop path is dynamically optimized; the cluster head node receives link quality data reported by a terminal, and dynamically adjusts a spread spectrum factor; the cluster head node and the core layer gateway carry out key negotiation, and generate a session key; and the core layer gateway converts a LoRa data packet into a standard message; the cluster head node compresses and converges node data of terminals in a cluster, and transmits the node data to the core layer gateway through a multi-hop mode. The scheme can solve the problems of intense channel competition, rigid routing and insufficient clock synchronization accuracy in the prior art, and meets the business demand of high real-time performance and high reliability of a power system.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of communication technology, in particular to a real-time communication network construction method and system based on LoRa ad hoc network technology. BACKGROUND

[0002] With the large-scale access of distributed photovoltaic, energy storage, electric vehicle charging piles and other resources in the power distribution network, the traditional centralized communication architecture cannot realize millisecond-level end-to-end communication and massive terminal collaborative control in scenarios such as rapid fault isolation and power dynamic balance. The LoRa-based ad hoc network technology becomes the preferred solution due to its long-distance and low-power characteristics, but the existing LoRaWAN protocol has problems such as intense channel competition, rigid routing, and insufficient clock synchronization accuracy, which makes it difficult to meet the high real-time and high reliability business needs of the power system. SUMMARY

[0003] In view of the above defects, the present application provides a real-time communication network construction method and system based on LoRa ad hoc network technology, which solves the problems of intense channel competition, rigid routing and insufficient clock synchronization accuracy of the prior art, and meets the high real-time and high reliability business needs of the power system.

[0004] The embodiment of the present application provides a real-time communication network construction method based on LoRa ad hoc network technology, which comprises the following steps:

[0005] The core layer gateway receives the registration request sent by the terminal, extracts the business flow characteristics, performs spatial clustering according to the extracted characteristics, forms a local cluster, and determines the cluster head node;

[0006] The cluster head node broadcasts the clustering result to all nodes in the local cluster; classifies the business flow, determines the access mode of different business flow classification; calculates the expected transmission times and node residual energy through bidirectional link probe packets, and dynamically optimizes the multi-hop path according to the expected transmission times and the node residual energy;

[0007] The cluster head node receives the link quality data reported by the terminal, dynamically adjusts the spreading factor according to the link quality data, and updates the communication parameters of the cluster head node according to the adjusted spreading factor;

[0008] The cluster head node and the core layer gateway perform key negotiation to generate a session key; and convert the LoRa data packet into a standard message through the core layer gateway;

[0009] The cluster head node compresses and aggregates the node data of the terminals in the cluster, and transmits it to the core layer gateway through the multi-hop mode.

[0010] Preferably, the registration request comprises device ID, GPS coordinates and business type.

[0011] The service flow feature includes a service flow density, the service flow density ; is a number of data packets reported by a node within a time window, and T is a length of a sampling time window.

[0012] Preferably, the spatial clustering according to the extracted features, forming local clusters, and determining cluster head nodes, comprises:

[0013] The core layer gateway randomly selects a center point position by a maximum distance principle, and performs clustering based on a K-means clustering algorithm to distribute different nodes into different local clusters.

[0014] The comprehensive capability score of each node in each local cluster is calculated, and the node with the highest comprehensive capability score is selected as a cluster head node.

[0015] Preferably, the clustering loss function of the K-means clustering algorithm is ;

[0016] The comprehensive capability score of the i-th node ;

[0017] wherein, is a loss value, and k is a number of local clusters; is a feature vector of the j-th node; denotes the i-th local cluster; is a center point position of the i-th local cluster; is a residual energy ratio of the i-th node, is a link quality index of the i-th node, is a computing capability index of the i-th node, and α, β and γ are weighting coefficients.

[0018] Preferably, the service flow classification is performed to determine an access mode of different service flow classifications, comprising:

[0019] The service flow is classified into real-time service flow and non-real-time service flow according to performance requirements of the service flow;

[0020] The fixed time slot access mode is adopted for the real-time service flow, and the contention access mode is adopted for the non-real-time service flow.

[0021] Preferably, the multi-hop path is dynamically optimized according to the expected transmission times and the residual energy of the nodes, comprising:

[0022] The comprehensive path cost value of different paths is calculated according to the expected transmission times and the residual energy of the nodes;

[0023] The path with the minimum comprehensive path cost value is selected as an optimal path.

[0024] wherein the integrated path cost value from the ith node to the jth node ; and are weight coefficients, and are the node residual energy ratios of the jth node and the ith node, respectively, is the expected number of transmissions for successful transmission on the path from the ith node to the jth node.

[0025] Preferably, the link quality data comprises a current signal-to-noise ratio and a current packet loss rate.

[0026] The dynamically adjusting the spreading factor according to the link quality data comprises:

[0027] when the current signal-to-noise ratio is greater than a preset first signal-to-noise ratio threshold, and the current packet loss rate is less than a preset first packet loss rate threshold, reducing the current spreading factor by a preset first amplitude;

[0028] when the current signal-to-noise ratio is less than a preset second signal-to-noise ratio threshold, and the current packet loss rate is greater than a preset second packet loss rate threshold, increasing the current spreading factor by a preset second amplitude.

[0029] wherein the first signal-to-noise ratio threshold is greater than the second signal-to-noise ratio threshold, and the first packet loss rate threshold is less than the second packet loss rate threshold, and the initial spreading factor is a preset value.

[0030] Preferably, the cluster head node and the core layer gateway perform key agreement to generate a session key; and the LoRa data packet is converted into a standard message by the core layer gateway, comprising:

[0031] The cluster head node generates a node private key and a node public key, and sends the node public key to the core layer gateway.

[0032] The core layer gateway generates a gateway private key and a gateway public key, and generates a shared key according to the node public key and the gateway public key; generates an AES key through a key derivation algorithm according to the shared key; encrypts each data block according to the AES key and a pre-generated random initial counter; and takes the encrypted ciphertext and the random initial counter as the standard message.

[0033] Preferably, the cluster head node compresses and aggregates node data of terminals within the cluster, and transmits the node data to the core layer gateway through a multi-hop mode, comprising:

[0034] The cluster head node reports complete initial data to the core layer gateway, and subsequently calculates the change amount of current data and the initial data in each period; when the change amount is less than a preset threshold, the current data is uploaded to the core layer gateway; when the change amount is not less than the threshold, the current data is uploaded to the core layer gateway.

[0035] The core layer gateway detects the performance index of the cluster head node in real time, and issues a network optimization instruction to the corresponding cluster when a performance anomaly is detected.

[0036] Another embodiment of the application provides a real-time communication network construction system based on LoRa ad hoc network technology, which comprises a core layer gateway and a cluster head node.

[0037] The core layer gateway receives a registration request sent by a terminal, extracts traffic characteristics, performs spatial clustering according to the extracted characteristics, forms a local cluster, and determines a cluster head node.

[0038] The cluster head node broadcasts the clustering result to all nodes in the local cluster, classifies the business flow, determines the access mode of different business flow classifications, calculates the expected transmission times and node residual energy through bidirectional link probe packets, and dynamically optimizes the multi-hop path according to the expected transmission times and the node residual energy.

[0039] The cluster head node receives link quality data reported by the terminal, dynamically adjusts the spreading factor according to the link quality data, and updates the communication parameters of the cluster head node according to the adjusted spreading factor.

[0040] The cluster head node and the core layer gateway perform key negotiation to generate a session key, and convert a LoRa data packet into a standard message through the core layer gateway.

[0041] The cluster head node compresses and aggregates the node data of the terminals in the cluster, and transmits it to the core layer gateway through a multi-hop mode.

[0042] The application provides a real-time communication network construction method and system based on LoRa ad hoc network technology. BRIEF DESCRIPTION OF DRAWINGS

[0043] Figure 1 is a flowchart of a real-time communication network construction method based on LoRa ad hoc network technology provided by an embodiment of the application. DETAILED DESCRIPTION

[0044] The technical solutions in the embodiments of the application will be clearly and completely described in combination with the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by a person of ordinary skill in the art without creative labor fall within the protection scope of the application.

[0045] Reference Figure 1 is a flowchart of a real-time communication network construction method based on LoRa ad hoc network technology provided by an embodiment of the application. The method comprises steps S1-S5.

[0046] In step S1, a core layer gateway receives a registration request sent by a terminal, extracts service flow features, performs spatial clustering according to the extracted features, forms a local cluster, and determines a cluster head node.

[0047] In step S2, the cluster head node broadcasts clustering results to all nodes in the local cluster, classifies service flows, determines access modes of different service flow classifications, calculates expected transmission times and node residual energy through bidirectional link probe packets, dynamically optimizes a multi-hop path according to the expected transmission times and the node residual energy.

[0048] Step S3, the cluster head node receives the link quality data reported by the terminal, dynamically adjusts the spreading factor according to the link quality data, and updates the communication parameters of the cluster head node according to the adjusted spreading factor;

[0049] Step S4, the cluster head node and the core layer gateway perform key negotiation to generate a session key, and convert the LoRa data packet into a standard message through the core layer gateway;

[0050] Step S5, the cluster head node compresses and aggregates the node data of the terminals in the cluster, and transmits them to the core layer gateway through multi-hop mode.

[0051] In the implementation of the embodiment, first, terminal registration and feature collection are performed, the terminal node sends a registration request (containing device ID, GPS coordinates, and service type) to the core layer gateway, then the gateway analyzes the service traffic density (such as the number of times of reporting per second of a photovoltaic inverter) through an edge computing module, and finally the geographical position and service weight of the node are stored as clustering feature parameters.

[0052] Dynamic clustering and cluster head election are performed: the core layer gateway clusters the nodes based on the K-means algorithm according to the characteristics of geography and service density, then elects a cluster head according to a cluster head scoring model (residual energy, link quality, and calculation capacity weighting), and finally the cluster head node broadcasts the clustering result to the members of the cluster.

[0053] A mixed access mechanism is scheduled: the cluster head node classifies the service flow, and determines the access mode of different service flow classifications;

[0054] By way of example, the service is divided into real-time type (such as protection signal) and non-real-time type (such as electric meter data), then a fixed TDMA time slot (20msx4 per frame) is allocated for the real-time type, and finally the non-real-time type uses the CSMA / CA contention access with a dynamic backoff window.

[0055] Dynamic routing optimization is performed: the cluster head node calculates the expected transmission times and the residual energy of the node through bidirectional link probe packets, then selects the optimal path based on a path cost formula, and triggers a traffic bypass mechanism when congestion occurs.

[0056] Adaptive spreading factor adjustment is performed: the terminal periodically reports link quality data including SNR (signal-to-noise ratio) and PLR (packet loss rate) to the cluster head node, the cluster head node dynamically switches the spreading factor according to a preset threshold, and updates the LoRa communication parameters of the nodes in the whole network.

[0057] The encryption transmission and protocol mapping are performed, the cluster head node and the core layer gateway perform key negotiation, a session key is generated, and the LoRa data packet is converted into an IEC 61850 MMS message through the core layer gateway.

[0058] Data aggregation and feedback are performed: the cluster head node aggregates the cluster data (Delta encoding compression), then transmits the data to the core layer gateway through multi-hop routing, and finally the core layer gateway feeds back the network state (such as congestion alarm) to the terminal node to trigger parameter adjustment.

[0059] The present application proposes a distributed resource real-time communication network construction method based on LoRa ad hoc network to solve the problems of poor real-time performance, poor scalability, weak protocol compatibility and the like of the existing power distribution network communication system, and effectively improves the overall performance of the communication network in the power distribution network scenario through dynamic clustering, hybrid access scheduling, dynamic routing optimization, adaptive spread spectrum adjustment, full-link encryption and standard protocol mapping.

[0060] In another embodiment provided by the present application, when the system is initialized or the terminal node is newly accessed to the LoRa ad hoc network of the power distribution network, the node needs to complete the registration process first when the step S2 is specifically implemented. The registration process not only ensures the terminal identity management, but also extracts the geographical location information and service characteristics (service flow density) of the node through edge computing, thereby laying a data foundation for subsequent dynamic clustering.

[0061] The terminal node sends a registration request to the core layer gateway through the LoRa communication link, and the registration request includes the following information:

[0062] Device ID (DeviceID): used for uniquely identifying the terminal node.

[0063] GPS coordinates (GPS): representing the geographical location of the node.

[0064] Service type (Type): such as photovoltaic monitoring, load monitoring, protection device, etc.

[0065] The gateway side analyzes the request content and calls the edge computing module to extract the service flow characteristics:

[0066] The flow density (FlowDensity, FD) index is defined, and the greater the FD index, the higher the node data activity, and more resources are needed to ensure, and the calculation formula is as follows: ;

[0067] Wherein, is the number of data packets reported by node i in the time window; is the sampling time window length.

[0068] The node characteristics are recorded into a feature library for dynamic clustering.

[0069] In another embodiment provided by the application, the step S2 is specifically implemented as follows:

[0070] According to the node characteristic parameters, spatial clustering is performed to form a local cluster structure, and a cluster head node is elected according to energy, link quality and computing capacity to optimize the network structure.

[0071] The clustering is performed based on a K-means clustering algorithm, and nodes are distributed to the nearest cluster center according to a clustering loss function;

[0072] The comprehensive capability score of each node in each local cluster is calculated according to the node energy, link quality and computing capacity, and the node with the highest comprehensive capability score is selected as the cluster head node.

[0073] In another embodiment provided by the application, the clustering loss function of the K-means clustering algorithm is ;

[0074] The comprehensive capability score of the i th node ;

[0075] Wherein, is a loss value, k is the number of local clusters; is the feature vector of the j th node; represents the i th local cluster; is the center point position of the i th local cluster; is the residual energy ratio of the i th node, is the link quality index of the i th node, is the computing capacity index of the i th node, and α, β and γ are weighting coefficients.

[0076] The clustering is performed based on the K-means clustering algorithm, the target is to minimize the difference between nodes in the cluster, the initial center point is randomly selected by maximizing the distance principle, and the algorithm avoids random selection of the center point in initialization, adopts the probability far away principle: a random point is initially selected as the first cluster center, and then a point far away from the existing cluster center is selected as a new cluster center each time to increase the stability of clustering, and to realize the minimization of the deviation of the initial cluster center selection and improve the clustering effect.

[0077] In another embodiment provided by the application, the classification process of the service flow in the step S2 is specifically as follows:

[0078] Different access mechanisms are adopted according to different service types (real-time and non-real-time) to ensure the dual targets of low delay and high resource utilization.​

[0079] The cluster head service flow is classified into a real-time type (such as a fault tripping protection, a service flow requiring extremely low delay) and a non-real-time type (such as power data, load recording, and a service flow acceptable to a certain delay).

[0080] Two access mechanisms are adopted for different types of services, in which TDMA ensures collision-free transmission of key data, is suitable for a high reliability scenario, and CSMA / CA utilizes channel idle sensing to improve overall resource utilization, is suitable for a lighter load service. The specific service access mechanism is as follows.

[0081] For the real-time type service, TDMA access (fixed time slot mode) is adopted.

[0082] The overall time is framed, each frame is 80 ms, and is subdivided into 4 20 ms sub-slots; the cluster head pre-allocates a time slot number corresponding to each node; the node only sends in the specified time slot to avoid collision.

[0083] For the non-real-time type service, CSMA / CA access (contention-based access mode) is adopted.

[0084] The node listens to whether the channel is idle; if the channel is idle, a backoff time B~U(0,CW) is randomly generated, the channel is occupied after countdown; after sending is completed, the channel is released; and the window CW is dynamically increased when collision occurs.

[0085] The two access modes are scheduled in parallel in one cluster and do not interfere with each other, so that shunt management is achieved.

[0086] In another embodiment provided in the application, when dynamic path optimization is performed, specifically:

[0087] According to a link transmission success rate and a node energy remaining condition, a multi-hop path is dynamically optimized, energy consumption is reduced, and the success rate is improved.

[0088] Link probe packets are periodically exchanged, and an expected transmission number ETX is calculated, the ETX represents an expected transmission number required for successful transmission of data on a link, the smaller the ETX, the better the link, and the more inclined to select the link, and the expected transmission number for successful transmission on a path from an i-th node to a j-th node is .

[0089] Wherein, is a single-hop transmission success probability from the i-th node to the j-th node.

[0090] The node remaining energy proportion is recorded.

[0091] According to a comprehensive path cost calculation formula, an optimal path is selected, and the calculation mode is as follows: the path with the minimum comprehensive path cost value is selected as the optimal path.

[0092] The comprehensive path cost value from the i-th node to the j-th node is: .

[0093] wherein, , is a weight coefficient, and are the node residual energy proportions of the j-th node and the i-th node, respectively.

[0094] In yet another embodiment provided by the present application, when the adaptive spreading factor adjustment is performed, specifically:

[0095] The link quality data includes the current signal-to-noise ratio and the current packet loss rate;

[0096] According to the real-time link quality (SNR, PLR), the LoRa terminal spreading factor (SF) is dynamically adjusted to adapt to the change of the communication environment.

[0097] The node periodically reports the current signal-to-noise ratio SNR and the current packet loss rate PLR, and the calculation method is as follows:

[0098] ;

[0099] ;

[0100] wherein, is the received useful signal power, is the received background noise power; is the total number of data packets sent by the node, is the number of data packets successfully received by the gateway.

[0101] The LoRa terminal spreading factor SF is dynamically adjusted based on the following judgment mechanism:

[0102] When the current signal-to-noise ratio is greater than a preset first signal-to-noise ratio threshold, and the current packet loss rate is less than a preset first packet loss rate threshold, the current spreading factor is reduced by a preset first amplitude;

[0103] When the current signal-to-noise ratio is less than a preset second signal-to-noise ratio threshold, and the current packet loss rate is greater than a preset second packet loss rate threshold, the current spreading factor is increased by a preset second amplitude;

[0104] Exemplarily, ;

[0105] wherein, is the spreading factor currently used by the node, is the adjusted spreading factor, is the first signal-to-noise ratio threshold, ​is a second signal-to-noise ratio threshold, is a first packet loss rate threshold, is a second packet loss rate threshold. is negotiated by the node according to the link quality at the first access.

[0106] It should be noted that in the embodiment, 1 is taken as the first amplitude and the second amplitude, and in other embodiments, other values can be set.

[0107] By periodically updating the node spreading parameter, the communication efficiency is improved.

[0108] In still another embodiment provided by the application, when performing encrypted transmission and protocol mapping, specifically:

[0109] To ensure the security of data transmission between nodes in the power distribution network LoRa ad hoc network, the ECDH protocol is introduced for key negotiation, and the AES-128-CTR encryption mode is used for end-to-end protection of data. At the same time, the gateway encapsulates the encrypted data in a standard protocol to support compatible access of the main station system.

[0110] ECDH (Elliptic Curve Diffie-Hellman) is a secure key agreement protocol based on elliptic curve cryptography (ECC), and in specific applications:

[0111] On the node side, a pair of keys is generated for each node, including a node private key and a node public key , and the node public key is sent to the gateway.

[0112] On the gateway side, the gateway also generates its own gateway private key and gateway public key ; after receiving the node public key, the shared key is calculated, and the calculation formula is as follows:

[0113] ;

[0114] AES (Advanced Encryption Standard) is a widely used symmetric encryption algorithm, in which AES-128-CTR (Counter Mode) belongs to a stream encryption mode, which is suitable for fast encryption of continuous transmission data, and the specific encryption process is as follows:

[0115] Session key generation, using the obtained by ECDH negotiation in the previous step, to generate the AES key through a key derivation algorithm.

[0116] Initialize the counter, the system generates a random initial counter Nonce for each session.

[0117] For each data block, encryption is performed using the following formula: ;

[0118] Wherein, is the encrypted i-th block of ciphertext, is the i-th block of plaintext data, is the counter value input to the AES encryption, is the bitwise XOR operation.

[0119] Send ciphertext And initial Nonce, the receiver synchronously increments the counter to decrypt and recover the original text.

[0120] After the gateway receives the encrypted data, the same And the counter are used for decryption; the decrypted original data is encapsulated into IEC 61850 MMS protocol standard messages according to the preset mapping rule; the standard messages are uploaded to the power grid dispatching control system through a special interface, realizing seamless integration.

[0121] In another embodiment provided by the application, when data aggregation and feedback are performed, specifically:

[0122] In order to improve communication efficiency and reduce energy consumption, the cluster head node compresses and aggregates the data of the terminal nodes in the cluster, and transmits them to the core layer gateway through a multi-hop mode. At the same time, the core layer gateway feeds back network control instructions according to the overall network operation state, so as to realize adaptive optimization and dynamic adjustment of the communication network.

[0123] Data aggregation and Delta encoding compression are performed, the amount of redundant data is reduced, the burden of wireless transmission is reduced, the overall network energy efficiency is improved, and the battery life of the terminal node is prolonged; large-scale nodes can be simultaneously accessed, and the system scalability is improved.

[0124] The principle and process of Delta encoding compression are as follows:

[0125] The complete data is reported for the first time , and only the change amount is sent in each subsequent period , if the change amount is below the set threshold, only the difference is uploaded, if the change amount is abnormal or serious packet loss is detected, the complete data is uploaded to re-synchronize.

[0126] Network state feedback and adaptive control are performed, which are used to maintain the stability, reliability and high efficiency of the communication network in a dynamic environment; avoid network paralysis caused by node energy depletion or link congestion.

[0127] The core layer gateway real-time monitoring node energy, channel state, delay, packet loss rate and other indicators; when detecting abnormalities, network optimization instructions are issued to the corresponding cluster;

[0128] The optimization instructions include adjusting the packet frequency, dynamic re-routing, re-election of cluster head, and changing the spreading factor.

[0129] The scheme obtains a hierarchical communication network based on LoRa wireless technology through the above processing procedure, organizes nodes in units of clusters, performs multi-hop forwarding through cluster head nodes, and finally aggregates all terminal data to the core gateway.

[0130] In the application of power distribution network, the network can be understood as: terminal nodes, head nodes, cluster head relay hops, core layer gateways, and main network, a total of 5 layers, each layer has a special role division, the terminal is responsible for data collection, the cluster head is responsible for collection and relay, and the gateway is responsible for accessing the main network.

[0131] Node clustering: The nodes in each part of the area will be automatically divided into a cluster according to the location and business density.

[0132] Cluster head management: A cluster head node is selected in each cluster, which is responsible for communication with other nodes in the cluster, data management, and information forwarding.

[0133] Data compression upload: Nodes do not directly upload all data as is, but use Delta encoding (only changes are transmitted), and the cluster head packages again.

[0134] Multi-hop relay: Cluster heads can relay and forward each other to form a hop-by-hop transmission link, so that even if the nodes are far away, they can also upload through the relay.

[0135] Real-time monitoring and adjustment: The gateway will continuously monitor the network situation and automatically command each node to adjust the packet frequency, switch the route, and even reselect the cluster head.

[0136] To verify the effect of the proposed method, simulation tests are performed based on the Network Simulator 3 platform, and the experimental settings are as shown in Tables 1-6:

[0137] Table 1 Experimental Setting Table

[0138]

[0139] After the simulation test, the experimental comparison results are as follows:

[0140] Table 2 End-to-end delay comparison table

[0141]

[0142] Table 3 Network throughput comparison table

[0143]

[0144] Table 4 node energy consumption comparison table

[0145]

[0146] Table 5 packet loss rate comparison table

[0147]

[0148] Table 6 data compression efficiency table

[0149]

[0150] Based on the above simulation test, the scheme has the following significant effects and advantages compared with the traditional communication network method:

[0151] The traditional LoRaWAN or pure CSMA / CA access mechanism has a channel competition when the number of nodes exceeds 1000, and the end-to-end service delay is usually higher than 200ms, which is difficult to meet the control requirement of 50ms level of power distribution system. The application adopts TDMA fixed time slot scheduling for real-time service, and adopts CSMA / CA competition access for non-real-time service, and the high priority data transmission delay is controlled within 85ms (5000 node scale), which meets the real-time control requirement of IEC 61850-7-420 standard.

[0152] The traditional static clustering scheme has serious imbalance of cluster head load when the number of nodes exceeds 3000, and the network throughput decreases by more than 60%. The application introduces a dynamic clustering and cluster head election mechanism based on node geographical position and service density, and combines ETX and energy weighted dynamic routing optimization strategy to support dynamic access of more than 5000 nodes, and the overall network throughput is improved by more than 40%, and the cluster head switching and routing repair mechanism guarantees the communication stability in large-scale dynamic environment.

[0153] The traditional LoRa network usually adopts simple encryption or no encryption, which is easy to be threatened by man-in-the-middle attack or replay attack. The application introduces ECDH key negotiation mechanism between nodes and gateway, and combines AES-128-CTR high-efficiency stream encryption to realize end-to-end full-link encryption protection, effectively prevent data leakage, forgery and tampering, and meet the demand of power grid for high security of data communication.

[0154] Traditional AODV or static routing protocol recovers slowly when the link quality changes dramatically, and is easy to form a loop or link collapse. The present application realizes real-time monitoring and adaptive spreading factor (SF) adjustment through SNR and PLR, combines dynamic optimal routing reconstruction and cluster head reselection mechanism, so that the network can quickly adapt to the complex environment of link quality fluctuation, node energy decline, local congestion, etc. and maintain stable and reliable communication.

[0155] Traditional transmission usually adopts complete data upload, resulting in wireless link congestion and increased energy consumption. The present application introduces Delta encoding compression, uploads only the data change amount, combines multi-hop cluster head data aggregation, greatly reduces the link load (compression rate can reach 30%-70%), and at the same time, through the adaptive adjustment strategy of packet sending frequency, effectively prolongs the battery life of terminal node by more than 20%.

[0156] The embodiment of the present application also provides a real-time communication network construction system based on LoRa ad hoc network technology, which comprises a core layer gateway and a cluster head node.

[0157] The core layer gateway receives the registration request sent by the terminal and extracts the traffic feature; according to the extracted feature, spatial clustering is carried out to form a local cluster, and a cluster head node is determined;

[0158] The cluster head node broadcasts the clustering result to all nodes in the local cluster; classifies the service flow, determines the access mode of different service flow classifications; calculates the expected transmission times and node residual energy through bidirectional link probe packets, and dynamically optimizes the multi-hop path according to the expected transmission times and the node residual energy;

[0159] The cluster head node receives the link quality data reported by the terminal, dynamically adjusts the spreading factor according to the link quality data, and updates the communication parameters of the cluster head node according to the adjusted spreading factor;

[0160] The cluster head node and the core layer gateway perform key negotiation to generate a session key; and convert the LoRa data packet into a standard message through the core layer gateway;

[0161] The cluster head node compresses and aggregates the node data of the terminals in the cluster, and transmits it to the core layer gateway through the multi-hop mode.

[0162] It should be noted that the real-time communication network construction system based on LoRa ad hoc network technology provided by the embodiment of the present application can perform the real-time communication network construction method based on LoRa ad hoc network technology described in any of the above embodiments, and the specific functions of the real-time communication network construction system based on LoRa ad hoc network technology will not be repeated here.

[0163] The above is the preferred embodiment of the present application, it should be pointed out that, for those skilled in the art, without departing from the principles of the present application, can also make a number of improvements and refinements, these improvements and refinements are also considered to be within the scope of the present application.

Claims

1. A method for constructing a real-time communication network based on LoRa self-organizing network technology, characterized in that: The method comprises: The core layer gateway receives the registration request sent by the terminal and extracts the service traffic characteristics; it performs spatial clustering based on the extracted characteristics to form local clusters and determine the cluster head node; The cluster head node broadcasts the clustering result to all nodes in the local cluster; classifies the service flow and determines the access mode of different service flow classifications; calculates the expected number of transmissions and the remaining energy of the node through a bidirectional link detection packet, and dynamically optimizes the multi-hop path based on the expected number of transmissions and the remaining energy of the node; The cluster head node receives the link quality data reported by the terminal, dynamically adjusts the spreading factor according to the link quality data, and updates the communication parameters of the cluster head node according to the adjusted spreading factor; The cluster head node and the core layer gateway perform key negotiation to generate a session key; and convert the LoRa data packet into a standard message through the core layer gateway; The cluster head node compresses and aggregates the node data of the terminals in the cluster, and transmits the data to the core layer gateway in a multi-hop manner.

2. The method for constructing a real-time communication network based on LoRa self-organizing network technology as claimed in claim 1, wherein: The registration request includes device ID, GPS coordinates and service type; The business flow characteristics include business flow density, business flow density ; is the number of data packets reported by the node in the time window, and T is the length of the sampling time window.

3. The method for constructing a real-time communication network based on LoRa self-organizing network technology as claimed in claim 1, wherein: The spatial clustering is performed based on the extracted features to form local clusters and determine cluster head nodes, including: The core layer gateway randomly selects the center point position by maximizing the spacing principle, and performs clustering based on the K-means clustering algorithm to assign different nodes to different local clusters; Calculate the comprehensive ability scores of different nodes in each local cluster, and select the node with the highest comprehensive ability score as the cluster head node.

4. The method for constructing a real-time communication network based on LoRa self-organizing network technology as claimed in claim 3, wherein: The clustering loss function of the K-means clustering algorithm is ; The comprehensive capability score of the i-th node ; in, is the loss value, k is the number of local clusters; is the feature vector of the jth node; represents the i-th local cluster; is the center point position of the i-th local cluster; is the remaining energy ratio of the i-th node, is the link quality indicator of the i-th node, is the computing capability index of the i-th node, α, β and γ are weighting coefficients.

5. The method for constructing a real-time communication network based on LoRa self-organizing network technology as claimed in claim 2, wherein: Classify service flows and determine access methods for different service flow categories, including: Based on the performance requirements of business flows, business flows are divided into real-time business flows and non-real-time business flows; A fixed time slot access method is used for real-time service flows, and a competitive access method is used for non-real-time service flows.

6. The method for constructing a real-time communication network based on LoRa self-organizing network technology as claimed in claim 1, wherein: The dynamically optimizing the multi-hop path according to the expected number of transmissions and the remaining energy of the node includes: Calculating comprehensive path cost values ​​of different paths according to the expected number of transmissions and the remaining energy of the node; Select the path with the smallest comprehensive path cost as the optimal path; Among them, the comprehensive path cost value from the i-th node to the j-th node is ; and is the weight coefficient, and are the node residual energy ratios of the j-th node and the i-th node, is the expected number of successful transmissions on the path from the i-th node to the j-th node.

7. The method for constructing a real-time communication network based on LoRa self-organizing network technology as claimed in claim 1, wherein: The link quality data includes a current signal-to-noise ratio and a current packet loss rate; The dynamically adjusting the spreading factor according to the link quality data includes: When the current signal-to-noise ratio is greater than a preset first signal-to-noise ratio threshold, and the current packet loss rate is less than a preset first packet loss rate threshold, reducing the current spreading factor by a preset first amplitude; When the current signal-to-noise ratio is less than a preset second signal-to-noise ratio threshold, and the current packet loss rate is greater than a preset second packet loss rate threshold, increasing the current spreading factor by a preset second amplitude; The first signal-to-noise ratio threshold is greater than the second signal-to-noise ratio threshold, the first packet loss rate threshold is less than the second packet loss rate threshold, and the initial spreading factor is a preset value.

8. The method for constructing a real-time communication network based on LoRa self-organizing network technology as claimed in claim 1, wherein: The cluster head node and the core layer gateway perform key negotiation to generate a session key; The core layer gateway converts the LoRa data packet into a standard message, including: The cluster head node generates a node private key and a node public key, and sends the node public key to the core layer gateway; The core layer gateway generates a gateway private key and a gateway public key, and generates a shared key based on the node public key and the gateway public key; generates an AES key based on the shared key through a key derivation algorithm; encrypts each data block based on the AES key and a pre-generated random initial counter; and uses the encrypted ciphertext and the random initial counter as the standard message.

9. The method for constructing a real-time communication network based on LoRa self-organizing network technology according to claim 1, wherein: The cluster head node compresses and aggregates the node data of the terminals in the cluster and transmits the data to the core layer gateway via a multi-hop method, including: The cluster head node reports the complete initial data to the core layer gateway, and calculates the difference between the current data and the initial data in each subsequent cycle; when the difference is less than a preset threshold, the current data is uploaded to the core layer gateway; when the difference is not less than the threshold, the current data is uploaded to the core layer gateway; The core layer gateway detects the performance indicators of the cluster head nodes in real time, and sends network optimization instructions to the corresponding cluster when a performance anomaly is detected.

10. A real-time communication network construction system based on LoRa self-organizing network technology, characterized in that: The system includes: a core layer gateway and a cluster head node; The core layer gateway receives the registration request sent by the terminal, extracts the service traffic characteristics; performs spatial clustering based on the extracted characteristics to form a local cluster, and determines the cluster head node; The cluster head node broadcasts the clustering result to all nodes in the local cluster; classifies the service flow and determines the access mode of different service flow classifications; calculates the expected number of transmissions and the remaining energy of the node through a bidirectional link detection packet, and dynamically optimizes the multi-hop path based on the expected number of transmissions and the remaining energy of the node; The cluster head node receives the link quality data reported by the terminal, dynamically adjusts the spreading factor according to the link quality data, and updates the communication parameters of the cluster head node according to the adjusted spreading factor; The cluster head node and the core layer gateway perform key negotiation to generate a session key; and convert the LoRa data packet into a standard message through the core layer gateway; The cluster head node compresses and aggregates the node data of the terminals in the cluster, and transmits the data to the core layer gateway in a multi-hop manner.

Images