CN110650510A

CN110650510A - Low-power-consumption wide-area multi-hop networking method and system based on software definition

Info

Publication number: CN110650510A
Application number: CN201810667098.4A
Authority: CN
Inventors: 罗海林; 孙严智; 崔晨; 吴怡; 马文凤; 范新东; 刘旋; 温泉
Original assignee: Yunnan Power Grid Co Ltd
Current assignee: Yunnan Power Grid Co Ltd
Priority date: 2018-06-26
Filing date: 2018-06-26
Publication date: 2020-01-03

Abstract

The invention introduces Software Defined Network (SDN) technology into a low-power-consumption wide area multi-hop network, and completes dynamic networking and flexible transmission of the network through design and realization of links such as topology discovery and maintenance, flow table generation and issuing and the like. The network realizes software definition, flexibility and controllability in the network transmission process by control and data transmission decoupling, and effectively reduces the routing overhead of common sensing nodes by adopting a centralized control algorithm, thereby improving the routing success rate and the network stability. The networking and transmission method provided by the invention is essentially different from the traditional routing-based networking method. The software defined network thought based on the separation of centralized control and numerical control can well accord with the advantages of the traditional software defined network and the requirements of the wireless multi-hop network, and brings convenience to the aspects of network management, network centralized configuration and the like of the wireless multi-hop network.

Description

Low-power-consumption wide-area multi-hop networking method and system based on software definition

Technical Field

The invention relates to the field of low-power-consumption wide area network networking, in particular to a low-power-consumption wide area multi-hop networking method and system based on software definition.

Background

The Low-Power Wide-Area Network (LPWAN) technology is used as a new hotspot technology in the field of Internet of things, short boards of existing communication technologies such as Wi-Fi and Zigbee are made up in links such as Power consumption, transmission distance and Network cost, the system receiving sensitivity reaches-148 dBm, long-distance communication exceeding 15km is supported, the equipment receiving current is only 10mA, the dormancy current is less than 200nA, the service life of an equipment battery can reach more than 5 years, and powerful technical support and a brand-new solution are provided for Low-cost interconnection of large areas.

The general low-power-consumption wide area communication network is mainly formed by a star-shaped structure, the terminal and the gateway are directly communicated in a one-hop direct connection mode, the system is simple in networking and is suitable for a common communication scene with perfect infrastructure. However, for emergency service, post-disaster rescue and other scenes, the network lacks available infrastructure, or the infrastructure has a large-area fault, the network nodes are required to be independent of the infrastructure, the network nodes can realize rapid deployment and use through a wireless multi-hop technology, and a communication network with rapid deployment, flexible networking and dynamic extension is provided for emergency and emergency work.

With the development of the internet of things, the functions of the terminal nodes of the wireless multi-hop network are continuously upgraded and innovated, the supported service range is greatly expanded, and the network application is diversified. In the face of the terminal revolution and the explosive growth of application, the traditional networking and transmission method based on the routing protocol cannot realize the flexible control and management of the network and cannot perform the complex configuration and interactive operation of the terminal and the protocol.

In summary, the software defined network technology is introduced into the low-power-consumption wide-area multi-hop network aiming at the scenes of emergency service, post-disaster rescue and the like, so that the advantages of the traditional software defined network and the requirements of the wireless multi-hop network can be well met, and convenience in network management, network centralized configuration and the like is brought to the wireless multi-hop network.

Disclosure of Invention

In view of this, the present invention provides a low-power-consumption wide-area multi-hop networking method and system based on software definition, which complete dynamic networking and flexible transmission of a network and facilitate management and control of the network through design and implementation of links such as topology discovery and maintenance, flow table generation and issue. Meanwhile, the routing overhead of the common sensing nodes is effectively reduced by adopting a centralized control algorithm, so that the routing success rate and the network stability are improved.

In this embodiment, the acquisition of topology information of the entire network is completed by designing topology discovery, flow table generation, flow table forwarding, and network maintenance, the collected information is used for calculation and processing, a flow table containing control information of each node is made and distributed to each sensing node, and dynamically changing network information is monitored and maintained in real time.

The topology discovery is to collect topology information of a network, and is responsible for a network operating system layer, so that a common node can interact with a controller. For this purpose, the control node periodically broadcasts a topology request message into the entire multi-hop network. Each hop topology request message contains some information, such as the number of hops from the current node to the control node, the current node power level, etc. After a node receives the topology request message, adding 1 to the hop count information in the topology request message, changing the electric quantity of the previous hop node into the electric quantity of the current node, and then forwarding the topology request message. In addition, after receiving the topology request message, the node also measures a current Received Signal Strength Indicator (RSSI). In this way, each node can obtain some information about the link state to ensure that the node can find a reliable path to the controller. The node can determine a reliable backhaul path according to the number of hops, power and signal strength. In order to complete the selection, each common node records a neighbor table, and the table records surrounding nodes and signal strength values thereof. And transmitting the neighbor table back to the control node through the reliable path in the form of a topology reply message.

The flow table generation needs to comprehensively consider information such as link quality information (signal strength, topology information and the like), node information (node residual capacity, storage space and the like), service types and the like, and flexible flow table rules are formulated according to different requirements on node functions under different scenes.

The flow table forwarding comprises two situations, namely, aiming at the conditions of small network scale, single service type, fixed network topology and stable wireless channel, an active forwarding strategy is adopted, namely, a central control node directly issues a flow table to each node in a network, and a sensing node only needs to execute corresponding processing according to rules in the flow table; secondly, aiming at the scenes that the network scale is large, the service types are various, the network topology is changeable, and the wireless channel is unstable, a strategy of requesting forwarding is adopted, namely, a sensing node sends a request to a central control node when forwarding data, and after a control node judges the current network information, a flow table suitable for the node is made to control the node to complete the forwarding of the data information. The two strategies are mutually matched to ensure that the system can operate efficiently and orderly under different scenes.

The network maintenance, that is, topology maintenance, refers to the periodic detection of topology change by a controller, which includes the detection of the original topology change and the detection of whether a new node is added, the monitoring of the flow of each node in the network, the monitoring of the residual energy of a sensing node, and the update and maintenance of a flow table. If the state of the network is monitored to be changed, topology adjustment is needed, and the process of generating the flow table is executed again.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.

Fig. 1 is a network flow chart of a low-power-consumption wide-area multi-hop networking system based on software definition according to an embodiment of the present invention;

fig. 2 is a diagram of a topology discovery and maintenance phase SDL of a low-power-consumption wide-area multi-hop networking system based on software definition according to an embodiment of the present invention;

fig. 3 is a flowchart of a flow table issuing process according to an embodiment of the present invention;

fig. 4 is a flow table generation and delivery stage SDL diagram of a low-power-consumption wide-area multi-hop networking system based on software definition according to an embodiment of the present invention;

Detailed Description

The embodiment of the invention provides a low-power-consumption wide-area multi-hop networking method and system based on software definition, and the flow design of the system refers to fig. 1. The multi-hop networking process is divided into two stages of topology discovery and maintenance, flow table generation and issuing according to functions, and each stage is designed, wherein the specific design is as follows:

the topology discovery and maintenance phase is mainly used for collecting multi-dimensional information such as position information and electric quantity information of nodes and link quality between adjacent nodes, so that the control nodes can generate an optimal flow table by using the global information, the topology discovery and maintenance phase comprises two processes of topology discovery and topology maintenance, when a controller does not know the topology of a network, namely, a process of acquiring the topology when the network is deployed for the first time is called a topology discovery process, and a process of updating the topology later is called a topology maintenance process.

In the topology discovery and topology maintenance phases, the control node and the general node will perform different operations, and the specific process refers to fig. 2.

The controller side topology discovery and topology maintenance SDL graph includes 3 states, which are power on, topology discovery, and topology maintenance, respectively, and triggers other operations when a clock is overtime or a topology reply message (nm _ topo _ rp) and a topology repair message (nm _ topo _ repair) of a terminal node are received, and the processing procedure is relatively simple: after the controller is started and finishes a series of initialization work, the controller broadcasts a topology discovery message (nm _ topo _ rq) in the whole network and then enters a topology discovery state, and if a topology reply message sent by a terminal node is received, the received message is unpacked to obtain a neighbor table for perfecting the topology table (namely an adjacent matrix); setting a certain time interval, updating the topology mark once the timer is overtime, entering a topology maintenance state to wait for receiving the topology maintenance message, extracting information in the message and updating a topology table if the topology maintenance message is received, increasing the topology update mark, and entering the topology maintenance state again to wait.

The terminal topology discovery and topology maintenance SDL graph also contains 3 states of power-on, topology discovery and topology maintenance. The terminal node needs to broadcast the hello request message (hello _ rq) periodically in a single hop, the neighbor node returns a hello reply message (hello _ rp) when receiving the hello request message, the terminal node can obtain which neighbors exist and the link quality between the neighbor node by counting the number of the hello _ rq message and the hello _ rp message, and the neighbor information is written into a neighbor list for storage. Meanwhile, the terminal node receives a topology request message (nm _ ropo _ rq) sent by the controller, and the terminal node needs to mark the topology request message and record the forwarding times of the topology request message. And when the forwarding times are not more than the upper forwarding limit (set to 50 here), reversely sending the neighbor list (namely the topology reply message nm _ topo _ rp) maintained by the node to the control node, and continuously broadcasting the topology request message. Once the terminal node sends the neighbor table to the controller, the terminal node enters a topology maintenance state, the processing process is similar to the topology discovery state, except that the terminal node needs to periodically check whether the original neighbor table and the new neighbor table are changed, and once the original neighbor table and the new neighbor table are changed, the neighbor table is updated and sent to the controller in the form of a topology maintenance message. When the terminal node receives the abnormal message, only discarding is carried out.

Nodes in the network are dynamically changed, and a link between any two nodes is still in a connected state at the last moment but can be disconnected at the next moment. When the change of the network topology is detected, the adjacency matrix needs to be modified, and the flow table generation and issuing stage is re-entered.

The flow table generating and issuing stage refers to fig. 3, and this part needs to use the neighbor matrix M stored in the flow table generating process. Firstly, setting hop count hops to be 1, namely, firstly, issuing a flow table to a hop node around a controller, wherein the hop count farthest from the controller node is Maxhop, the controller node needs to obtain a sequence number of the hop count hops node, and takes out a corresponding routing entry (rt _ entry), then encapsulates the sequence number into a flow table (rt _ packet), and takes out an IP address of the node to be issued, sends out the flow table through a socket port, increments the retransmission times resend, and then waits for receiving an ACK message reply; and setting a certain timeout time and the maximum allowable retransmission number (MAXNUM), and if the ACK message reply is not received due to timeout and the retransmission number is less than MAXNUM, repeatedly executing the process.

Otherwise, if the retransmission number (resend) is equal to MAXNUM, which indicates that the link between two nodes in the network is disconnected, i.e. the topology structure of the network is changed, the current flow table issuing process needs to be skipped, the corresponding adjacency matrix is modified, and the flow table generating process is entered again; if the controller node receives the ACK message reply quickly, the data update _ n of the routing update node needs to be increased progressively, and then the flow table is issued to the neighbor node with hop count of hops + +. And completing the flow table issuing process once till the ACK message reply of the neighbor node with the hop count of Maxhop is received.

The use specification SDL map of flow table generation and flow table issue refers to fig. 4.

For the controller, each row of the switching matrix R is taken out in sequence to obtain a routing entry of each terminal node, then a function is called to convert the routing entry into a corresponding destination IP address and a next hop IP address, the controller node needs to modify a routing table of the controller node firstly, and then the controller encapsulates routing information into the routing table according to a flow table issuing strategy to issue all the terminal nodes.

For the terminal, the terminal node does not have a distributed routing mechanism any more because the task of route generation is completed by the controller. The terminal node always monitors the port, waits for receiving the flow table, when the flow table is received, firstly decapsulates according to a packet format to obtain a corresponding destination IP address and a next hop IP address, then calls a system command to modify a routing table of the terminal node, and in order to inform a controller that the flow table is successfully received and the routing table is successfully modified, the terminal node needs to encapsulate an ACK (acknowledgement) reply message and send the ACK (acknowledgement) reply message to the controller according to the routing table through a socket.

Claims

1. A low-power-consumption wide-area multi-hop networking method and a system based on software definition complete dynamic networking and flexible transmission of a network through design and realization of links such as topology discovery and maintenance, flow table generation and issuing and the like, and are convenient for management and control of the network. Meanwhile, the routing overhead of the common sensing nodes is effectively reduced by adopting a centralized control algorithm, so that the routing success rate and the network stability are improved.

2. The method and system for low power consumption wide area multi-hop networking based on software definition according to claim 1, wherein the topology discovery link collects topology information of the network, and is responsible for a network operating system layer, so that common nodes can interact with the controller.

3. The method and system for low power consumption wide area multi-hop networking based on software definition according to claim 2, wherein in the topology discovery process, the control node periodically broadcasts topology request messages to the whole multi-hop network. Each hop topology request message contains some information, such as the number of hops from the current node to the control node, the current node power level, etc.

4. The software-definition-based low-power-consumption wide-area multi-hop networking method and system according to claim 2, wherein after receiving the topology request message, the terminal node adds 1 to the hop count information in the topology request message, changes the previous-hop node electric quantity to the current-node electric quantity, and forwards the topology request message. In addition, after receiving the topology request message, the node also measures a current Received Signal Strength Indicator (RSSI). Each common node records a neighbor table, and the nodes around and the signal strength values thereof are recorded in the table. The node can determine a reliable return path according to the hop count, the electric quantity and the signal strength, and return the neighbor table to the control node through the path in the form of a topology reply message.

5. The software-definition-based low-power-consumption wide-area multi-hop networking method and system according to claim 1, wherein the flow table generation step synthetically considers link quality information (signal strength, topology information, etc.), node information (node remaining capacity, storage space, etc.), service types, etc. to formulate flexible flow table rules.

6. The low-power-consumption wide-area multi-hop networking method and system based on software definition according to claim 1, wherein the flow table issuing link adopts two forwarding modes: firstly, aiming at a network with small scale, single service, fixed topology and stable channel, an active forwarding strategy is adopted, namely, a central control node directly issues a flow table to each node in the network, and a sensing node only needs to execute corresponding processing according to rules in the flow table; secondly, aiming at a network with large scale, various services, variable topology and unstable channel, a strategy of requesting forwarding is adopted, namely, a sensing node sends a request to a central control node when forwarding data, and after the control node judges the current network information, a flow table suitable for the node is made to control the node to complete the forwarding of the data information.

7. The software-definition-based low-power-consumption wide-area multi-hop networking method and system according to claim 1, wherein in the topology maintenance step, the controller periodically detects topology changes, including detection of original topology changes and detection of whether new nodes are added, and monitoring of flow of each node in the network, monitoring of residual energy of the sensing nodes, updating and maintaining of flow tables, and the like. When the state of the network is monitored to be changed, the topology needs to be adjusted, and the process of generating the flow table is executed again.