A self-organizing network management method based on determinism

Abstract

The present invention relates to a self-organizing network protocol based on a deterministic management method, including a gateway node and a control center connected through a network, and a self-organizing network with a tree topology structure formed between the gateway node and each child node and the parent node, each node is defined with a backup path discovery mechanism for uplink routing and downlink routing, and a network level management address is defined on each parent node connected to a child node; wherein the network level management address includes the network level, the number of forwarding nodes in the current layer, and the forwarding order of the current node. The self-organizing network protocol based on the deterministic management method provided in the above technical solution can solve the problem that the existing wireless networking technology does not optimize large-scale control scenarios and lacks a special management method for delay control.

Description

Technical Field

[0001] The present invention relates to the technical field of self-organizing network protocols, and in particular to a self-organizing network management method based on determinism. Background Art

[0002] In many scenarios in the field of the Internet of Things, objects need to be connected to each other wirelessly to obtain information. For scenarios with relatively large data throughput, such as video applications, WiFi networks, 5G / 4G and other high-speed wireless technologies will be used for connection. In some scenarios, such as environmental temperature and humidity collection, the amount of data required is very small, so low-speed wireless technology is needed for networking. At present, the more mainstream wireless low-speed networking technologies include LoRaWAN, NB-IoT, etc. The common feature of these technologies is the use of star networking solutions, that is, the terminal and the gateway or base station communicate directly. The processing power and rate of the terminal are generally low, but the processing power of the gateway and the base station is relatively high, so as to support a large number of terminal connections. However, since the distance of wireless communication is usually limited by the transmission power and environmental factors, it is impossible to ensure that the terminal and the gateway or base station always maintain a normal connection in specific applications. For example, if the LoRaWAN terminal is deployed in an underground pipeline, it is difficult to communicate with the gateway deployed on the ground. The usual solution is to increase the density of gateway deployment and install more gateways close to the terminal in a communication-restricted environment. This approach is firstly high in cost and secondly, it is impossible to ensure the installation of the gateway when the power supply is limited. Therefore, wireless ad hoc networking technology can be used. When the terminal node and the gateway cannot communicate directly, a terminal can be added between the two to automatically form a forwarding network, so that the network coverage can be expanded without relying on the base station. At the same time, the terminal can be driven by energy such as batteries, so there is no need to consider the use of wired power. Similar wireless networking technologies are also used in the market, such as ZigBee, but they are not optimized for certain scenarios, such as large-scale street light control. When street lights have the need for unified switches, ZigBee technology can be implemented, but there is no special management method for delay control and other aspects. Therefore, it is urgent to design a new technical solution to comprehensively solve the problems existing in the existing technology. Summary of the invention

[0003] The purpose of the present invention is to provide a self-organizing network management method based on determinism, which can solve the problems that the existing wireless networking technology is not optimized for large-scale control scenarios and lacks a special management method for delay control.

[0004] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0005] A self-organizing network management method based on determinism includes a gateway node and a control center connected through a network, and a self-organizing network with a tree topology structure formed between the gateway node and each child node and the parent node. Each node is defined with a backup path discovery mechanism for uplink routing and downlink routing, and a network level management address is defined on each parent node connected to a child node; wherein the network level management address includes the network level, the number of forwarding nodes in the current layer, and the forwarding order of the current node.

[0006] The process of the self-organizing network management method includes:

[0007] Network level management allocation is run on the gateway node. After a node successfully joins the network, it sends a request to the parent node. If the parent node has been assigned a network level management address, no processing is done. If the network level management address of the parent node is currently set to not forward broadcasts, the gateway node is requested to reconfigure the network level management address of the parent node.

[0008] The network downlink broadcast process is run on the branch node; after receiving the broadcast, the branch node checks the local network level management address of the branch node. If the network level management address of the branch node is not forwarded, the process ends; if the network level management address of the branch node is forwarded, the broadcast forwarding is implemented according to the network level management address setting of the branch node;

[0009] The alternative path discovery mechanism for the uplink routing runs on the branch node. After receiving the broadcast message, the branch node checks the local network level management address of the branch node and the network level management address of the sending node. If the network level of the sending node is lower than that of the branch node, the sending node is used as the forwarding node of the uplink path. If the network level of the sending node is higher than that of the branch node, no processing is performed.

[0010] The alternative path discovery mechanism for downstream routing runs on all nodes; the gateway node sends a request to the parent node that can connect to the destination terminal node. If the gateway node receives a reply within the specified time, it updates the routing table and establishes an alternative downstream path; if the gateway node does not receive a reply within the specified time, the process ends.

[0011] During the network downlink broadcast process, if the network level and forwarding order of the network level management address of the branch node are both 1, the branch node updates the local network level management address to the broadcast information and performs broadcast forwarding.

[0012] During the network downlink broadcast process, if the network level and forwarding order of the network level management address of the branch node are not 1, the delay calculation method is as follows:

[0013] Delay = (S local -Sreceived -1) * T1 (when S local > S received , L local =L received )

[0014] =(N received -S received +S local -1)*T1 (when L local = (L received +1))

[0015] Where: S local Indicates the forwarding order of the local node in the current layer in the local network hierarchical management address, S received Indicates the order in which the forwarding nodes of the received broadcast message are sent in the current layer, T1 indicates the time required to forward a broadcast message; N received Indicates how many forwarding nodes there are in the layer where the forwarding node of the received broadcast message is located, L received Indicates the network level at which the forwarding node of the received broadcast message is located, L local Indicates the network level to which the local node belongs in the local network level management address.

[0016] The deterministic self-organizing network management method of the present invention removes unnecessary network control signaling when using low-speed wireless devices for networking, and reserves the wireless bandwidth for users as much as possible; at the same time, the network can support the establishment of redundant paths (including uplink paths (from any terminal to the gateway) and downlink paths (from the gateway to any terminal)), ensuring the robustness and reliability of the network. When the user instruction requires the use of broadcast to traverse the entire network, the traversal time is deterministic and controllable.

[0017] The self-organizing network management method based on determinism provided in the above technical solution can achieve the following by establishing a broadcast forwarding management mechanism, an uplink alternative path discovery mechanism, and a downlink alternative route discovery mechanism:

[0018] (1) An orderly, conflict-free downlink broadcast mechanism ensures that control signaling or user commands can reach all nodes in the network quickly and reliably.

[0019] (2) Establishing alternative uplink and downlink paths. Alternative paths guarantee network reliability and flexibility. If the current path is unavailable, communication between nodes in the network can bypass the problem node and maintain network communication.

[0020] (3) All communications are controlled by the gateway. Nodes in the network are in a listening state most of the time and respond only when needed, making effective use of the bandwidth resources in the network.

[0021] (4) The delay of intra-network broadcast and unicast uplink and downlink communications is controllable, which is suitable for time-sensitive applications with control requirements.

[0022] (5) The delay of intra-network communication depends on the specifications of the physical communication technology used. The management within the network has little effect on the communication delay. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a system topology diagram based on tree structure;

[0024] Figure 2 This is a schematic diagram of the broadcast forwarding management mechanism within the network;

[0025] Figure 3 It is the upstream backup path discovery mechanism;

[0026] Figure 4 It is a downlink backup path discovery mechanism;

[0027] Figure 5 A flowchart for allocating network-level management addresses for gateway nodes;

[0028] Figure 6 This is a flow chart for downlink broadcast forwarding;

[0029] Figure 7 Discover a flow chart for the upstream alternative path;

[0030] Figure 8 Diagram of the process of initiating node detection and alternative downlink path discovery for the gateway;

[0031] Fig. 9 Perform detection request forwarding process for intermediate nodes;

[0032] Fig.10 Forwarding detection reply process for network nodes. DETAILED DESCRIPTION

[0033] In order to make the purpose and advantages of the present invention more clearly understood, the present invention is specifically described below in conjunction with embodiments. It should be understood that the following text is only used to describe one or several specific embodiments of the present invention, and does not strictly limit the scope of protection of the specific claims of the present invention.

[0034] The self-organizing network management method based on determinism proposed in the present invention refers to Figure 1 On the one hand, the gateway node is connected to the remote control center through a public network or a private network, and on the other hand, it forms a self-organizing network based on a tree topology with each child node and parent node. At the same time, each node is defined with a backup path discovery mechanism for uplink and downlink routing, and a network level management address (NLMADDR) is defined to each parent node that has a child node connected (such as Figure 2 ,and Figure 5 As shown); where NLMADDR includes the network level, the number of forwarding nodes in the current layer, and the forwarding order of the current node.

[0035] In the process of forming a tree network, when a node successfully joins the network, it needs to send a request to the parent node, indicating that the current node sets it as the parent node. If the parent node has been assigned NLMADDR, no processing is done. If the NLMADDR of the parent node is currently set to not forward broadcasts, it is necessary to send a request to the gateway node to request the gateway to configure NLMADDR.

[0036] In another case, the gateway sends the detection request to the terminal node in the downlink communication through the tree structure standard, that is, the detection message is forwarded by the parent node of the detected node. According to this situation, if the child node is found to be unresponsive for many times, and all the recorded child nodes are in the same situation, the parent node will also send an update NLMADDR request to the gateway, requesting that it should not be forwarded.

[0037] When the gateway receives the NLMADDR request from the node, it first checks whether the request is for application as a forwarding node. If there is no forwarding request, the NLMADDR corresponding to the node is set to not forward, the database is updated, and the updated NLMADDR is sent to the requesting node.

[0038] If the application is to be a forwarding node, the gateway node first extracts the network level L of the requesting node in the request, and then checks the database. If the database records that the number of nodes with forwarding function at this level has reached the upper limit, the forwardable NLMADDR is not allocated, but its NLMADDR is set to non-forwarding, the database is updated, and the result is sent to the requesting node. If the current level can still allocate forwarding nodes, a new NLMADDR is allocated to the applying node, the database is updated, and then sent to the applying node.

[0039] For example, the NLMADDR of node 2 is 0×010202, where 0×01 indicates that the depth of the current node in the network is 1, that is, it is in the first layer; the first 02 indicates that there are two nodes with forwarding function in the same layer, and the second 02 indicates that the current node is the second forwarding node in the current layer; similarly, the NLMADDR of node 1 is 0×010201, indicating that node 1 is in the first layer and is the first node in the first layer to forward broadcasts. The NLMADDR of node 5 is 0×020201, indicating that node 5 is in the second layer, there are two nodes with forwarding function in the current layer, and node 5 is the first node to forward in the second layer. Although node 4 is also a backbone node, it does not need to forward because it has no child nodes. Its NLMADDR is 0×020200, indicating that the node is in the second layer of the network (0×02), there are two forwarding nodes in this layer (0×02), but the current node does not need to forward (0×00). The NLMADDR address of node 7 indicates that the current node is in the second layer and is the second forwarding node. Each node reports to the gateway node based on the actual situation of whether it has child nodes connected to it, and the gateway node is responsible for allocating NLMADDR information.

[0040] Each node that forwards a broadcast message needs to include its own NLMADDR information in the message, so that the next layer of nodes with forwarding function that receive the forwarded broadcast message can instantly calculate when the message should be forwarded.

[0041] Assume that sending a broadcast message takes time T1. Figure 2 For example, when the gateway sends a broadcast:

[0042] (1) Both node 1 and node 2 receive the broadcast. Node 1 calculates NLMADDR and concludes that the message should be forwarded immediately. Node 2 calculates NLMADDR and concludes that it should delay T1*1 before forwarding the message.

[0043] (2) After T1, node 4 receives the broadcast message sent by node 1, but node 4 does not need to forward it;

[0044] (3) After T1 time has passed, node 2 starts forwarding;

[0045] (4) After 2*T1 time has passed, nodes 5 and 7 receive the broadcast message forwarded by node 2. By checking the NAMADDR of node 2 that node 2 updated into the broadcast message, node 5 calculates that the broadcast message has been forwarded at the first layer of the network, and it will immediately forward it for the first time at the second layer; node 7 calculates that the broadcast message has been forwarded at the first layer of the network, and it needs to wait for T1 time before forwarding it at the second location;

[0046] (5) After 3*T1 time has passed, node 5 completes broadcast forwarding and node 7 starts broadcast forwarding;

[0047] (6) After 4*T1 time has passed, the broadcast is completed and the gateway information reaches every node in the entire network.

[0048] The time delay is completed by the timer on the wireless communication device. Under the premise of controlling the number of forwarding nodes and the scale of the network, a certain amount of redundancy is set according to the error range of the timer. The error of the timer will not affect the broadcast forwarding.

[0049] like Figure 6 As shown in the figure, the network downlink broadcast process is executed on nodes other than the gateway. When the node receives the broadcast, it first checks the local NLMADDR. If it is set to not forward, the process ends. If the node is a broadcast forwarding node, check the level L setting and forwarding order setting S of the local NLMADDR. If L is 1 and S is 1, it means that the current node is the first node that needs to be forwarded in the first layer, that is, the broadcast is obtained directly from the gateway. Then the node updates the local NLMADDR to the broadcast message and immediately forwards the broadcast.

[0050] If the current node is not the first forwarding node in the first layer, the delay is calculated according to the NLMADDR of the sending node included in the broadcast message. The calculation process is:

[0051] Delay = (S local -S received -1) * T1 (when S local > S received , L local =L received )

[0052] =(N received -S received +S local -1)*T1 (when L local = (L received +1))

[0053] Where S local Indicates the forwarding order of the local node in the current layer indicated in the local NLMADDR, S received Indicates the order of forwarding nodes of the received broadcast message in the current layer, and T1 indicates the time required to forward a broadcast message. If the current node receives a message from a forwarding node in the same layer, and the forwarding order of the forwarding node is less than the forwarding order of the local node, the delay is calculated.

[0054] N received Indicates how many forwarding nodes there are in the layer where the forwarding node of the received broadcast message is located, L receivedIndicates the network level at which the forwarding node of the received broadcast message is located, L local Indicates that the local network level management address indicates the network level to which the local node belongs. If the network level at which the local node is located is one level higher than the level at which the forwarding node of the received broadcast message is located, the delay is calculated.

[0055] If the delay can be calculated, the local NLMADDR is updated to the broadcast message and the timer is started. When the timer expires, the broadcast is forwarded. In other cases, no calculation is performed and the process stops.

[0056] This is because if the current node and the forwarder of the intercepted broadcast message are located at the same network layer, the delay calculation is based on the difference between the order of the current node and the intercepted forwarding node.

[0057] If the network layer of the forwarder of the broadcast message detected by the current node is larger or smaller than its own (the difference is greater than 1), the current node does not need to consider delay calculation, because the message forwarding may have been completed at this layer, or has not yet reached the previous layer of this layer, and the delay cannot be calculated.

[0058] Define an upstream backup path discovery mechanism for each node such as Figure 3 , Figure 7 As shown in the figure, the gateway node periodically initiates the whole network broadcast process. When each node receives a broadcast message or a forwarded broadcast message, the current node checks the NLMADDR of the message sending node and the local NLMADDR. If the comparison finds that the sending node is at a lower network level than the current node, the sending node is used as the forwarding node of the uplink path and updated to the local routing table. If the node is not the parent node of the current node, it is equivalent to adding an alternative path.

[0059] Specifically: when the gateway node is idle and has no tasks, it periodically starts the downlink broadcast mechanism (the time interval is controlled by the gateway and depends on the total number of nodes and the busyness of user tasks). Figure 3 In the figure, the parent node of node 5 is node 2. During the operation of the broadcast mechanism, after the broadcast of node 1 is completed, if node 5 can receive the broadcast forwarding, it will find that it is in the first layer of the network by analyzing the NLMADDR of the broadcast forwarder. Because the NLMADDR of node 5 indicates that node 5 is in the second layer of the network and the broadcast forwarder is not its parent node, node 5 can update node 1 as an alternative path for the upstream route to the local routing table. In subsequent upstream communications, if node 2 cannot communicate and before node 5 starts the mechanism to relocate the parent node, node 5 can try to use node 1 as the upstream path. After each downstream broadcast mechanism is completed, all nodes in the network can discover and update the upstream alternative path.

[0060] Usually, the terminal node is required to actively report the heartbeat packet in the network and report its existence to the gateway. However, in a low-speed network, the node actively reporting the heartbeat packet will occupy the network bandwidth and affect the downlink communication of the gateway. For example, the maximum rate of the LoRa terminal in the Sub-1Ghz frequency band is only 37.5kbps. If a large number of nodes are deployed in the network, and each node independently decides to send the heartbeat packet, it is easy to cause communication conflicts. The protocol of the present invention stipulates that the detection of the terminal node is actively initiated by the gateway, that is, the gateway sends an inquiry packet to the terminal node, and the terminal node replies to the inquiry. In this way, all nodes are in a receiving state at ordinary times, and only when the gateway inquires will the transmission operation be generated, and because it is a unicast communication, multiple nodes will not send at the same time; if the gateway temporarily receives a user command and needs to send a downlink communication, it can stop terminal detection at any time until the user task is completed.

[0061] Define a downlink backup path discovery mechanism for each node such as Figure 4 , Figure 8 As shown, when the gateway is preparing to perform the node detection process, if it is desired to perform the alternative downlink path discovery process at the same time, DDS is set to 1, and then according to the requirements of the tree topology, the parent node that can be connected to the destination terminal node is selected as the forwarding node and the request is sent; after the detection request is sent, the gateway is in a waiting state. If a reply is received within the specified time, the routing table is updated. Because DDS is set to 1, if there is a path other than the downlink path specified according to the tree topology, the reply should be reported to the gateway from the available alternative path. Based on this, the gateway can establish or update the alternative downlink path to the specified destination node. If no reply is received after the timeout, the process ends.

[0062] For example Figure 4 The gateway node initiates a device detection message for node 5 and sets the downlink routing management switch parameter (DDS) to 1. The message first passes through node 2, the parent node of node 5, and then reaches node 5. After node 5 receives the detection message, it will prepare to send a reply. If the DDS in the gateway detection message is set to 0, node 5 will return the reply to the gateway through parent node 2 according to the provisions of the tree network; if DDS is set to 1, node 5 will check the routing table. If there is an alternate uplink path, the reply will be sent using the alternate path, which is node 1 in the figure above. After the gateway receives the reply from node 1, it can establish an alternate path from the gateway to node 5, that is, forwarding through node 1. At the same time, node 1 also establishes a downlink path to node 5.

[0063] The intermediate node performs the detection request forwarding process as follows Fig. 9As shown in the figure, the detection request initiated by the gateway is forwarded along the downstream route in the network. When a node receives the request, if the node is not the destination node, it selects a child node that can be connected to the destination node according to the tree topology structure for forwarding. If the node is the destination node, check the DDS setting. If DDS is set to 0, select the parent node to send a detection reply. If DDS is set to 1, detect the alternative upstream path. If there is an alternative upstream path, select the alternative path to send a reply, otherwise still select the parent node to send a reply.

[0064] The network node forwarding detection reply process is as follows Fig.10 As shown in the figure, when the network node receives the network detection reply, if the current node is the destination node, the process ends. Otherwise, check the DDS setting in the reply. If DDS is set to 0, select the parent node to forward the reply. If DDS is set to 1, check whether there is an alternative uplink path locally. If so, select the alternative path to forward the reply, otherwise, select the parent node to forward the reply.

[0065] The implementation modes of the present invention are described in detail above in conjunction with the embodiments, but the present invention is not limited to the above-mentioned implementation modes. For ordinary technicians in this technical field, after knowing the contents recorded in the present invention, they can make several equivalent changes and substitutions thereto without departing from the principle of the present invention, and these equivalent changes and substitutions should also be regarded as belonging to the protection scope of the present invention.

Claims

1. A self-organizing network management method based on determinism, characterized in that: A self-organizing network including a gateway node and a control center connected through a network, and a tree-like topology structure is formed between the gateway node and each child node and the parent node. Each node is defined with a backup path discovery mechanism for uplink routing and downlink routing, and a network level management address is defined on each parent node connected to a child node; The network level management address includes the network level, the number of forwarding nodes in the current level, and the forwarding order of the current node; The process of the self-organizing network management method based on determinism includes network level management allocation, network downlink broadcast process, an alternative path discovery mechanism for uplink routing, and a backup path discovery mechanism for downlink routing; Among them, the network downlink broadcast process is run on the branch node; after receiving the broadcast, the branch node checks the local network level management address of the branch node. If the network level management address of the branch node is not forwarded, the process ends; if the network level management address of the branch node is forwarded, the broadcast forwarding is implemented according to the network level management address setting of the branch node; During the network downlink broadcast process, if the network level and forwarding order of the network level management address of the branch node are not 1, the delay calculation method is as follows: When S local > S received , L local =L received When Delay = (S local -S received -1) * T1; When L local = (L received +1), then Delay=(N received -S received +S local -1)*T1; Where: S local Indicates the forwarding order of the local node in the current layer in the local network hierarchical management address, S received Indicates the order in which the forwarding nodes of the received broadcast message are sent in the current layer, T1 indicates the time required to forward a broadcast message; N received Indicates how many forwarding nodes there are in the layer where the forwarding node of the received broadcast message is located, L received Indicates the network level at which the forwarding node of the received broadcast message is located, L local Indicates that the local network level management address indicates the network level to which the local node belongs; If the delay can be calculated, the local network level management address is updated to the broadcast information, and the timer is started. When the timer expires, the broadcast is forwarded; otherwise, no calculation is performed and the process stops; During the network downlink broadcast process, if the network level and forwarding order of the network level management address of the branch node are both 1, the branch node updates the local network level management address to the broadcast information and performs broadcast forwarding; Among them, the alternative path discovery mechanism for the uplink route is run on the branch node; after receiving the broadcast message, the branch node checks the local network level management address of the branch node and the network level management address of the sending node. If the network level of the sending node is lower than that of the branch node, the sending node is used as the forwarding node of the uplink path; if the network level of the sending node is higher than that of the branch node, no processing is performed; The alternative path discovery mechanism for downlink routing runs on all nodes; the gateway node sends a request to the parent node that can connect to the destination terminal node. If the gateway node receives a reply within the specified time, the routing table is updated and an alternative downlink path is established; if the gateway node does not receive a reply within the specified time, the process ends.

2. The self-organizing network management method based on determinism according to claim 1, characterized in that: Network level management allocation is run on the gateway node; after a node successfully joins the network, it sends a request to the parent node. If the parent node has been assigned a network level management address, no processing is performed; if the network level management address of the parent node is currently set to not forward broadcasts, the gateway node is requested to reconfigure the network level management address of the parent node.

Images