WO2021036055A1 - 一种应用在集群无人机通信中的路由方法 - Google Patents

一种应用在集群无人机通信中的路由方法 Download PDF

Info

Publication number
WO2021036055A1
WO2021036055A1 PCT/CN2019/121119 CN2019121119W WO2021036055A1 WO 2021036055 A1 WO2021036055 A1 WO 2021036055A1 CN 2019121119 W CN2019121119 W CN 2019121119W WO 2021036055 A1 WO2021036055 A1 WO 2021036055A1
Authority
WO
WIPO (PCT)
Prior art keywords
node
cluster
cluster head
auxiliary
head node
Prior art date
Application number
PCT/CN2019/121119
Other languages
English (en)
French (fr)
Inventor
姜培
Original Assignee
中科芯(苏州)微电子科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中科芯(苏州)微电子科技有限公司 filed Critical 中科芯(苏州)微电子科技有限公司
Publication of WO2021036055A1 publication Critical patent/WO2021036055A1/zh

Links

Images

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04WWIRELESS COMMUNICATION NETWORKS
    • H04W40/00Communication routing or communication path finding
    • H04W40/02Communication route or path selection, e.g. power-based or shortest path routing
    • H04W40/04Communication route or path selection, e.g. power-based or shortest path routing based on wireless node resources
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04WWIRELESS COMMUNICATION NETWORKS
    • H04W40/00Communication routing or communication path finding
    • H04W40/24Connectivity information management, e.g. connectivity discovery or connectivity update
    • H04W40/248Connectivity information update
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04WWIRELESS COMMUNICATION NETWORKS
    • H04W40/00Communication routing or communication path finding
    • H04W40/24Connectivity information management, e.g. connectivity discovery or connectivity update
    • H04W40/30Connectivity information management, e.g. connectivity discovery or connectivity update for proactive routing
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04WWIRELESS COMMUNICATION NETWORKS
    • H04W40/00Communication routing or communication path finding
    • H04W40/24Connectivity information management, e.g. connectivity discovery or connectivity update
    • H04W40/32Connectivity information management, e.g. connectivity discovery or connectivity update for defining a routing cluster membership
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04WWIRELESS COMMUNICATION NETWORKS
    • H04W84/00Network topologies
    • H04W84/18Self-organising networks, e.g. ad-hoc networks or sensor networks

Definitions

  • the present invention is suitable for the field of cluster drones, and the specific content relates to a routing method applied in cluster drone communications.
  • the mobile ad hoc network is developed from the basis of the wireless packet network. It is an unlimited network with the advantages of "no center, self-organization, multi-hop routing, and dynamic topology". All nodes in the network have equal status, and there is no need to set up any central control node. In other words, when most of the nodes are damaged, the remaining nodes can still self-organize to complete the link transmission.
  • the routing protocol is to quickly and efficiently establish a communication path for the nodes that need to communicate.
  • the routing protocol is one of the main problems in the study of mobile ad hoc networks.
  • Mobile autonomous networks are mainly divided into flat routing protocols and hierarchical routing protocols.
  • the idea of flat routing is relatively simple. It is mainly that all nodes have equal status and no special nodes, but its network scale will be limited and its scalability is not good.
  • the main structure of hierarchical routing is the clustering structure. The main idea is to divide each node into different clusters, and the nodes inside and outside the cluster specify different functions. The cluster head nodes need to be selected in different clusters, and nodes belonging to multiple clusters at the same time are gateway nodes. Generally, active routing protocols are used in areas, and reactive routing protocols are used between areas.
  • the transmission path is that the nodes in the cluster send to the cluster head, the cluster head sends to the gateway, and the gateway sends to the cluster heads of other clusters again, until it is sent to the target node.
  • cluster heads are pre-designated or elected by nodes in the cluster, and multiple cluster heads can form a cluster again upwards to form a higher-level network structure.
  • the purpose of the present invention is to propose a routing method applied in cluster UAV communication, and a hierarchical routing method that can perform efficient and reliable communication in large-scale mobile nodes.
  • the present invention provides the following technical solutions.
  • a routing method applied in cluster UAV communication includes the following steps:
  • each cluster is weighted by using energy, node connectivity, relative mobility, and relative distance, and cluster heads are elected according to the weighted results, and the cluster head node with the largest weight is selected as the cluster head node;
  • the clustering algorithm based on weights for delimiting clusters specifically includes the following steps:
  • Each node broadcasts a HELLO message to obtain neighbor information.
  • the energy, node degree, relative mobility and relative distance of the node are used for weighting.
  • the calculation of the weight is as follows:
  • w p + w c + w m + w d 1
  • w p is the weight of node energy
  • P is the node energy
  • w c is the weight of node connectivity
  • C is the number of neighboring nodes
  • w m is the relative node
  • M is the value of relative mobility
  • the value of M is calculated according to the GPS of the drone. If it is similar to the movement direction of the surrounding n nodes, the higher the value, otherwise the lower the value, the value of M
  • D is set to half of the number of nearest nodes
  • w d is the sum of the distances of n nodes
  • the method for selecting the auxiliary cluster head node is specifically as follows:
  • the cluster head After the cluster head is selected successfully, the cluster head sends a HELLO message, and other nodes wait for the message:
  • the value of C is set to 10, that is, when the number of connections around the cluster head exceeds 10, the selection of auxiliary cluster head nodes is performed;
  • the auxiliary cluster head node sends a HELLO message to the cluster to ensure that the nodes in the cluster are connected to the auxiliary cluster head node;
  • Reactive routing lookups are performed between the two cluster heads, and if data packets from the same node are received, they are discarded;
  • the nodes in the cluster adopt flat-type proactive routing as the local route, and the cluster head node adopts the DV mechanism, that is, each node in the cluster not only saves the distance to all other cluster heads and vector routing information, but also retains the local routing table information;
  • each node in the cluster When searching for routes, if the destination node is in the local area, it will be sent directly by the local route; if there is a neighboring node in the cluster, it will be sent to the neighboring node; otherwise, the cluster head information sent to the destination node will be sent by the AODV route to make the cluster
  • the first information is relayed; if the cluster of the destination node contains two cluster heads, they are sent separately; when the relay node receives the same information, the current information is discarded; if the destination node receives the information from the cluster head node, The current route is retained, and the route of other nodes in the cluster is discarded; the steps of updating the cluster head node are also included:
  • the new node When the new node receives the HELLO broadcast from the cluster head node, it actively joins the cluster;
  • the node and the auxiliary cluster head node are compared with the weight, and the one with the larger weight is regarded as the new auxiliary cluster head node, and the auxiliary cluster head node becomes the node in the cluster; if the auxiliary cluster head node is updated, the new auxiliary cluster head The node broadcasts within the cluster to determine the node that assists the cluster head node to communicate;
  • the auxiliary cluster head node When the weight of the auxiliary cluster head node is greater than the weight of the cluster head node, the auxiliary cluster head node becomes the cluster head node, finds the next new auxiliary cluster head node, and broadcasts within the cluster to complete the update of the cluster head node;
  • the cluster head node moves to other clusters, the cluster head node with the larger weight is updated to the new cluster head node, and the cluster head node with the smaller weight is added to the cluster as a node in the cluster; the cluster's auxiliary The cluster head node compares with the node to generate a new auxiliary cluster head node; if the auxiliary cluster head node is updated, it will broadcast in the cluster to determine the node that the auxiliary cluster head node communicates with.
  • the present invention also provides a routing method applied in cluster UAV communication, including the following steps:
  • each cluster is weighted by using energy, node connectivity, relative mobility, and relative distance, and cluster heads are elected according to the weighted results, and the cluster head node with the largest weight is selected as the cluster head node;
  • Reactive routing lookups are performed between the two cluster heads, and if data packets from the same node are received, they are discarded.
  • the clustering algorithm based on weights for delimiting clusters specifically includes the following steps:
  • Each node broadcasts a HELLO message to obtain neighbor information.
  • the energy, node degree, relative mobility and relative distance of the node are used for weighting.
  • the calculation of the weight is as follows:
  • w p + w c + w m + w d 1
  • w p is the weight of node energy
  • P is the node energy
  • w c is the weight of node connectivity
  • C is the number of neighboring nodes
  • w m is the relative node
  • M is the value of relative mobility
  • the value of M is calculated according to the GPS of the drone. If it is similar to the movement direction of the surrounding n nodes, the higher the value, otherwise the lower the value, the value of M
  • D is set to half of the number of nearest nodes
  • w d is the sum of the distances of n nodes
  • weight values use machine learning training to give different values, and the node with the higher weight is selected as the cluster head to form a cluster.
  • the method for selecting the auxiliary cluster head node is specifically as follows:
  • the cluster head After the cluster head is selected successfully, the cluster head sends a HELLO message, and other nodes wait for the message:
  • the auxiliary cluster head node sends a HELLO message to the cluster to ensure that the nodes in the cluster are connected to the auxiliary cluster head node.
  • the value of C is set to 10, that is, when the number of connections around the cluster head exceeds 10, the selection of auxiliary cluster head nodes is performed.
  • the nodes in the cluster adopt flat proactive routing as the local routing, and the cluster head node adopts the DV mechanism, that is, each node in the cluster saves the distance to all other cluster head nodes and vector routing information, as well as the local routing table information ;
  • the route is searched, if the destination node is local, the local route is used to send; if there is a neighboring node in the cluster, it is sent to the neighboring node; otherwise, the AODV route is used to send the cluster head information to the destination node to make
  • the cluster head information is relayed; if the cluster of the destination node contains two cluster heads, they are sent separately; when the relay node receives the same information, the current information is discarded; if the destination node receives the information from the cluster head node , The current route is retained, and the route of other nodes in the cluster is discarded.
  • the new node When the new node receives the HELLO broadcast from the cluster head node, it actively joins the cluster;
  • the node and the auxiliary cluster head node are compared with the weight, and the one with the larger weight is regarded as the new auxiliary cluster head node, and the auxiliary cluster head node becomes the node in the cluster; if the auxiliary cluster head node is updated, the new auxiliary cluster
  • the head node performs intra-cluster broadcast to determine the node that assists the cluster head node to communicate;
  • the auxiliary cluster head node When the weight of the auxiliary cluster head node is greater than the weight of the cluster head node, the auxiliary cluster head node becomes the cluster head node, finds the next new auxiliary cluster head node, and broadcasts within the cluster to complete the update of the cluster head node;
  • the cluster head node moves to other clusters, the cluster head node with a large weight is updated to be a new cluster head node, and the cluster head node with a small weight is added to the cluster as a node in the cluster; the auxiliary cluster head of the cluster The node compares with the node to generate a new auxiliary cluster head node; if the auxiliary cluster head node is updated, it will broadcast in the cluster to determine the node that the auxiliary cluster head node can communicate with.
  • the present invention has the following beneficial effects:
  • the routing of the present invention integrates mature plane routing in multi-level routing, adopts routing ideas based on multi-level routing protocols, integrates mature plane routing in multi-level routing, and makes routing easy to understand and apply.
  • the node weight is calculated by weighting the energy, node connectivity, relative mobility and relative distance of the node, and the largest weight is selected as the cluster head.
  • the present invention creatively proposes that the auxiliary cluster head node is activated only when the adjacent nodes of the cluster head node are greater than a certain number. On the one hand, the robustness of the communication system is improved. On the other hand, the auxiliary cluster head also reduces the update of the cluster head and avoids the overhead.
  • the routing method uses auxiliary cluster head nodes, which can improve the robustness of the communication system and reduce the overhead of updating cluster heads; use weights to elect cluster head nodes to obtain better communication; this method can be used in UAV ad hoc networks Enough to provide better communication performance. It provides better communication performance for the unmanned aerial vehicle ad hoc network with rapidly changing topological structure.
  • the use of node energy with different weights, node connectivity, relative mobility of nodes and relative distance of nodes further enables the routing method to cope with different complex topologies and improves the robustness of the system.
  • the present invention has good communication performance, and is suitable for large-scale topologically changing scenes such as drone clusters.
  • Figure 1 is a schematic diagram of a routing method used in cluster UAV communication.
  • Figure 2 is a schematic diagram of three different routing modes of the present invention.
  • the routing method of the present invention is applied to cluster UAV communication, including the use of composite routing, each cluster is weighted by using energy, node connectivity, relative mobility and relative distance, and under the result of weighting processing, Election is used to select the cluster head, and the node with the largest weight is selected as the cluster head node. When the number of nodes in the cluster is greater than a certain number, a cluster head is used again through the election just now. Reactive routing lookups are performed between the two cluster heads, and if data packets from the same node are received, they are discarded.
  • the present invention mainly integrates plane routing in multi-level routing and uses weights to elect cluster heads, and introduces auxiliary cluster head nodes. It provides better communication performance for the unmanned aerial vehicle ad hoc network with rapidly changing topological structure.
  • the use of node energy with different weights, node connectivity, relative mobility of nodes and relative distance of nodes further enables the routing method to cope with different complex topologies and improves the robustness of the system.
  • the delineation of clusters is based on the clustering algorithm of weights.
  • Each node broadcasts a HELLO message to obtain neighbor information.
  • the calculation of the weight is:
  • w p +w c +w m +w d 1.
  • w p is the weight of the node energy
  • P is the node energy.
  • w c is the weight of node connectivity
  • C is the number of neighboring nodes
  • w m is the weight of the relative mobility of the node
  • M is the value of relative mobility
  • the value of M can be calculated according to the GPS of the drone. If it is similar to the movement direction of n nodes around, the value is higher, otherwise the value is lower, and M is the sum of n values.
  • D can be set to half the number of nearest nodes.
  • w d is the sum of the distances of n nodes.
  • Four of the weight values can be given different values using machine learning training. The node with the highest weight is selected as the cluster head. Form a cluster.
  • the cluster head After the cluster head is selected successfully, the cluster head sends a HELLO message, and other nodes wait for the message. If others receive the HELLO message, they will join the cluster to complete the construction of the cluster. Calculate the value of the cluster head C. If the value of C of the cluster head is greater than 1, the node with the second highest weight before is selected as the auxiliary cluster head node. The value of C is set to 10, that is, when the number of connections around the cluster head exceeds 10, the selection of auxiliary cluster head nodes is performed. The auxiliary cluster head node sends a HELLO message to the cluster to ensure that the nodes in the cluster can be connected to the auxiliary cluster head node.
  • the nodes in the cluster adopt flat-type proactive routing as the local routing, and the cluster head node adopts the DV mechanism, that is, each node in the cluster saves the distance to all other cluster heads and vector routing information, as well as the local routing table. information.
  • searching for a route if the destination node is local, it will be sent directly by the local route. If there is a neighboring node in the cluster, it will be sent to the neighboring node. Otherwise, use AODV to route the cluster head information sent to the destination node to facilitate the relay of the cluster head information. If the cluster group of the destination node contains two cluster heads, they are sent separately. When the relay node receives the same information, it discards the current information. If the destination node receives the information from the cluster head node, the current route is retained, and the route of another cluster point is discarded.
  • a new node When a new node receives the broadcast HELLO from the cluster head node, it actively joins the cluster. The node and the auxiliary cluster head node are compared with the weight, and the one with the larger weight is regarded as the new auxiliary cluster head node, and the auxiliary cluster head node becomes the node in the cluster group. If the auxiliary cluster head node is updated, the new auxiliary cluster head node will broadcast within the cluster to determine the nodes that the auxiliary cluster head node can communicate with.
  • the auxiliary cluster head node becomes the cluster head node, find the next new auxiliary cluster head node, and broadcast in the cluster to complete the update of the cluster head node .
  • the cluster head node with the larger weight is updated to the new cluster head node, and the cluster head node with the smaller weight is added to the cluster as a node in the cluster.
  • the auxiliary cluster head node of the cluster is compared with the node, and a new auxiliary cluster head node is generated. If the auxiliary cluster head node is updated, it is broadcast within the cluster to determine the nodes that the auxiliary cluster head node can communicate with.
  • This routing method elects cluster heads based on weights to form clusters.
  • the dual cluster head design enhances the reliability of communication and reduces the update frequency of cluster head nodes. Its characteristics are suitable for UAV cluster scenarios with rapidly changing topologies.
  • Figure 1 shows the cluster head update when the weight of the cluster head node is smaller than the auxiliary cluster head node.
  • the square is the node in the new cluster group, and the star represents the previous auxiliary cluster head, and it is the current cluster head.
  • the diamond represents the previous cluster head. Circles indicate nodes leaving the cluster.
  • Figure 2 shows three different routing methods.
  • the triangles represent nodes belonging to two clusters. The first one is the local route in the cluster. The second is to find the neighboring node as the target cluster node through the local route, and reach it directly without passing through the cluster head node. The third is to use neighboring nodes to relay. The neighboring nodes send to the cluster head and the auxiliary cluster head at the same time. The auxiliary cluster head cannot be reached and the line is interrupted. The cluster head continues to forward to the next relay node, and then completes the link.
  • the invention adopts the routing idea based on the multi-level routing protocol, and integrates the mature plane routing in the multi-level routing, so that the routing is easy to understand and apply.
  • the node weight is calculated by weighting the energy, node connectivity, relative mobility and relative distance of the node, and the largest weight is selected as the cluster head.
  • the present invention creatively proposes that the auxiliary cluster head node is activated only when the adjacent nodes of the cluster head node are greater than a certain number. On the one hand, the robustness of the communication system is improved. On the other hand, the auxiliary cluster head also reduces the update of the cluster head and avoids the overhead.
  • the invention has good communication performance and is suitable for scenes with large topological changes such as drone clusters and the like.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本发明属于无线通信领域,具体为一种应用在集群无人机通信中的路由方法。本发明方法采用基于多级路由协议的路由思想,在多级路由中融合了成熟的平面路由,使得路由易于理解应用。通过对节点的能量,节点连接度,相对运动性以及相对距离加权来计算节点权值,选取权值最大的作为簇首节点。提出辅助簇首节点,是在簇首节点的临节点大于一定数量的时候才会启用。这一方面提高的通信系统的鲁棒性,另一方面辅助簇首节点也减少了对簇群簇首的更新,避免了开销。本发明具有良好的通信性能,适用在无人机集群等大型拓扑变化迅速的场景中。

Description

一种应用在集群无人机通信中的路由方法 技术领域
本发明适用于集群无人机领域,具体内容涉及一种应用在集群无人机通信中的路由方法。
背景技术
移动自组网为无线分组网基础发展而来,具有“无中心,自组织,多跳路由,动态拓扑”的优点的无限网络。网络中所有的节点地位平等,无需设置任何的中心控制节点。也就是说,在大部分节点收到破坏时,剩余节点依旧可以自组织完成链路的传输。而路由协议就是为需要通信的节点来快速高效的建立一条通信路径,路由协议正是研究移动自组网的主要问题之一。
移动自主网主要分为平面路由协议和分级路由协议。平面路由的思想比较简单,其主要是所有节点地位平等,没有特殊节点,但是其网络规模会受到限制,并且拓展性不好。分级路由的主要结构即为分簇结构,其主要思想就是把各个节点划分为不同的簇群,并且群内群外的节点指定不同的功能。不同的簇群内需要选出簇首节点,同时属于多个簇群的节点为网关节点。一般区域内使用主动路由协议,区域间使用反应式路由协议。通常传递路径即为簇群内节点发送给簇首,簇首发给网关,网关再次发送给其他簇群的簇首,直到发送给目标节点。通常簇首预先指定或者由群内节点选举出,而多个簇首可以向上再次形成簇群,组成更高一级的网络结构。
现代社会中无人机的重要性愈发凸显。在集群无人机中,分级路由更适合使用在集群无人机自组网中。由于无人机多方向快速移动从而导致拓扑结构变化快,以及能量的局限都会导致链路的中断。因此现有的自组网路由协议性能较差,丢包率较高,并不适用在集群无人机通信中。
发明内容
为了解决现有技术的问题,本发明的目的在于提出一种应用在集群无人机通信中的路由方法,可以在大规模移动节点中进行高效可靠通信的分级路由方法。
为了实现上述目的,本发明提供了如下的技术方案。
一种应用在集群无人机通信中的路由方法,包括以下步骤:
使用复合路由,对每个簇群通过使用能量、节点连接度、相对运动性以及相对距离进行加权,依据加权结果选举出簇首,选取权值最大的作为簇首节点;
所述簇群的划定基于权值的分簇算法,具体包括以下步骤:
每个节点广播HELLO消息,用来获取邻居信息,使用节点的能量、节点度、相对运动性以及相对距离进行加权,其中权值的计算为:
W=w p*P+w c*C+w m*M+w d*D
其中w p+w c+w m+w d=1,w p为节点能量的权重,P为节点能量,w c为节点连接度的权重,C是临近节点的个数,w m为节点相对运动性的权重,M为相对运动性的值,M的值根据无人机的GPS来计算得出,若与周围n个节点运动方向相似,则值越高,否则值越低,M值为n个值的和;D设定为最近节点数目的一半;w d为n节点的距离的和;
其中四个权重值使用机器学习训练来给定不同的值,选定权值高的节点的为簇首,构成一个簇群;
当簇群内节点大于设定数目时,再选举出一个辅助簇首节点;
所述辅助簇首节点的选择方法具体为:
在簇首选定成功后,簇首发送HELLO消息,其他节点等待消息:
若其他接收到HELLO消息,加入该簇,完成簇的构建,计算簇首C的值,若簇首的C值大于l,选择之前权值次高的节点作为辅助簇首节点;
其中,所述的C值设定为10,即当簇首周围连接数目超过10个,进行辅助簇首节点的选择;
辅助簇首节点往簇内发送HELLO消息,确保簇内节点与辅助簇首节点相连;
两个簇首之间分别进行反应式路由查找,若接收到来自相同节点的数据包,则丢弃;
所述反应式路由查找具体步骤如下:
簇群内节点采用平面式先应式路由作为本地路由,簇首节点采用DV机制,即每个簇群内节点既保存到其他所有簇首的距离以及矢量路由信息,也保留本地路由表信息;当路由寻找时,若目的节点在本地,则直接利用本地路由发送;若有临近节点为该簇群,则发送给临近节点;否则,利用AODV路由发送到目的节点的簇首信息,以使簇首信息进行中继;若目的节点的簇群包含有两个簇首,则分别发送;当中继节点接收到相同的信息时,则抛弃当前信息;若目的节点接收的来自簇首节点的信息,则保留当前路由,且抛弃另外簇内节点的路由;还包括簇首节点的更新步骤:
当新节点接收到有簇头节点的广播HELLO的时候,主动加入该簇群;
将该节点与辅助簇首节点做权值比价,将权值大的作为新的辅助簇首节点,辅助簇首节点成为该簇群内节点;若更新辅助簇首节点,则新的辅助簇首节点作簇内广播,确定辅助簇首节点通信的节点;
当辅助簇首节点的权值大于簇首节点的权值时,辅助簇首节点成为簇首节点,查找下一个新的辅助簇首节点,并作簇内广播以完成簇首节点的更新;
当簇首节点移动到其他簇时,将权值较大的簇首节点更新为新的簇首节点,簇首节点权值较小的作为该簇群内的节点加入该簇;该簇的辅助簇首节点与该节点比较,产生新的辅助簇首节点;若更新辅助簇首节点则作簇内广播,以确定辅助簇首节点通信的节点。
本发明还提供一种应用在集群无人机通信中的路由方法,包括以下步骤:
使用复合路由,对每个簇群通过使用能量、节点连接度、相对运动性以及相对距离进行加权,依据加权结果选举出簇首,选取权值最大的作为簇首节点;
当簇群内节点大于设定数目时,再选举出一个辅助簇首节点;
两个簇首之间分别进行反应式路由查找,若接收到来自相同节点的数据包, 则丢弃。
作为本发明的进一步改进,簇群的划定基于权值的分簇算法,具体包括以下步骤:
每个节点广播HELLO消息,用来获取邻居信息,使用节点的能量、节点度、相对运动性以及相对距离进行加权,其中权值的计算为:
W=w p*P+w c*C+w m*M+w d*D
其中w p+w c+w m+w d=1,w p为节点能量的权重,P为节点能量,w c为节点连接度的权重,C是临近节点的个数,w m为节点相对运动性的权重,M为相对运动性的值,M的值根据无人机的GPS来计算得出,若与周围n个节点运动方向相似,则值越高,否则值越低,M值为n个值的和;D设定为最近节点数目的一半;w d为n节点的距离的和;
其中四个权重值使用机器学习训练来给定不同的值,选定权值高的节点的为簇首,构成一个簇群。
作为本发明的进一步改进,辅助簇首节点的选择方法具体为:
在簇首选定成功后,簇首发送HELLO消息,其他节点等待消息:
若其他接收到HELLO消息,加入该簇,完成簇的构建,计算簇首C的值,若簇首的C值大于l,选择之前权值次高的节点作为辅助簇首节点;
辅助簇首节点往簇内发送HELLO消息,确保簇内节点与辅助簇首节点相连。
作为本发明的进一步改进,所述的C值设定为10,即当簇首周围连接数目超过10个,进行辅助簇首节点的选择。
作为本发明的进一步改进,反应式路由查找具体步骤如下:
簇群内节点采用平面式先应式路由作为本地路由,簇首节点采用DV机制,即每个簇群内节点既保存到其他所有簇首节点的距离以及矢量路由信息,也保留本地路由表信息;当路由寻找时,若目的节点在本地,则直接利用本地路由发送;若有临近节点为该簇群,则发送给临近节点;否则,利用AODV路由发送到目的节点的簇首信息,以使簇头信息进行中继;若目的节点的簇群包含有 两个簇首,则分别发送;当中继节点接收到相同的信息时,则抛弃当前信息;若目的节点接收的来自簇首节点的信息,则保留当前路由,且抛弃另外簇内节点的路由。
作为本发明的进一步改进,还包括簇头节点的更新步骤:
当新节点接收到有簇头节点的广播HELLO的时候,主动加入该簇群;
将该节点与辅助簇首节点做权值比价,将权值大的作为新的辅助簇首节点,辅助簇首节点成为该簇群内节点;若更新辅助簇簇首节点,则新的辅助簇首节点作簇内广播,确定辅助簇首节点通信的节点;
当辅助簇首节点的权值大于簇首节点的权值时,辅助簇首节点成为簇首节点,查找下一个新的辅助簇首节点,并作簇内广播以完成簇首节点的更新;
当簇首节点移动到其他簇时,将权值大的簇首节点更新成为新的簇首节点,簇首节点权值小的作为该簇群内的节点加入该簇;该簇的辅助簇首节点与该节点比较,产生新的辅助簇首节点;若更新辅助簇首节点则作簇内广播,确定辅助簇首节点可以通信的节点。
与现有技术相比,本发明具有以下有益效果:
本发明的路由在多级路由中融合了成熟的平面路由,采用基于多级路由协议的路由思想,在多级路由中融合了成熟的平面路由,使得路由易于理解应用。通过对节点的能量,节点连接度,相对运动性以及相对距离加权来计算节点权值,选取权值最大的作为簇首。本发明创造性的提出辅助簇头节点,是在簇头节点的临节点大于一定数量的时候才会启用。这一方面提高的通信系统的鲁棒性,另一方面辅助簇头也减少了对簇群簇头的更新,避免了开销。该路由方法使用辅助簇头节点,可以提高通信系统的鲁棒性,减少更新簇头开销;使用权重来选举簇头节点,可以得到较好的通信;该方法能够在无人机自组网中够提供较好的通信性能。为拓扑结构变化迅速的无人机自组网提供了较好的通信性能。使用不同权重的节点能量,节点连接度,节点相对运动性以及节点的相对距离进一步使得路由方法可以应对不同复杂的拓扑,提高了系统的鲁棒性。本 发明具有良好的通信性能,适用在无人机集群等大型拓扑变化迅速的场景中。
附图说明
在此描述的附图仅用于解释目的,而不意图以任何方式来限制本发明公开的范围。另外,图中的各部件的形状和比例尺寸等仅为示意性的,用于帮助对本发明的理解,并不是具体限定本发明各部件的形状和比例尺寸。本领域的技术人员在本发明的教导下,可以根据具体情况选择各种可能的形状和比例尺寸来实施本发明。在附图中:
图1为一种应用在集群无人机通信中的路由方法示意图。
图2为本发明三种不同的路由方式示意图。
具体实施方式
为了使本技术领域的人员更好地理解本发明中的技术方案,下面将结合本发明实施例中的附图,对本发明实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅仅是本发明一部分实施例,而不是全部的实施例。基于本发明中的实施例,本领域普通技术人员在没有做出创造性劳动的前提下所获得的所有其他实施例,都应当属于本发明保护的范围。
本发明一种应用在集群无人机通信中的路由方法,包括使用复合路由,每个簇群通过使用能量,节点连接度,相对运动性以及相对距离进行加权,在加权处理的结果下,通过选举来选簇首,选取权值最大的作为簇首节点。当簇群内节点大于一定数目时,通过刚才的选举再次使用一个簇首。两个簇首之间分别进行反应式路由查找,若接收到来自相同节点的数据包,则丢弃。
本发明主要在多级路由中融合了平面路由结合使用权重来选举簇首,并引入辅助簇首节点。为拓扑结构变化迅速的无人机自组网提供了较好的通信性能。使用不同权重的节点能量,节点连接度,节点相对运动性以及节点的相对距离进一步使得路由方法可以应对不同复杂的拓扑,提高了系统的鲁棒性。
本发明的具体实施细节包括以下步骤:
1)簇群的划定基于权值的分簇算法。每个节点广播HELLO消息,用来获取邻居信息。使用节点的能量,节点度,相对运动性,以及相对距离进行加权。其中权值的计算为:
W=w p*P+w c*C+w m*M+w d*D
其中w p+w c+w m+w d=1。w p为节点能量的权重,P为节点能量。w c为节点连接度的权重,C是临近节点的个数,w m为节点相对运动性的权重,M为相对运动性的值,M的值可以根据无人机的GPS来计算得出,若与周围n个节点运动方向相似,则值越高,否则值越低,M值为n个值的和。D可以设定为最近节点数目的一半。w d为n节点的距离的和。其中四个权重值可以使用机器学习训练来给定不同的值。选定权值高的节点的为簇首。构成簇。
2)在簇首选定成功后,簇首发送HELLO消息,其他节点等待消息。若其他接收到HELLO消息,加入该簇,完成簇的构建。计算簇首C的值,若簇首的C值大于l,选择之前权值次高的节点作为辅助簇首节点。其中C值设定为10,即当簇首周围连接数目超过10个,进行辅助簇首节点的选择。辅助簇首节点往簇内发送HELLO消息,确保簇内节点可以与辅助簇首节点相连。
3)簇群内节点采用平面式先应式路由作为本地路由,簇首节点采用DV机制,即每个簇群内节点既保存到其他所有簇首的距离以及矢量路由信息,也保留本地路由表信息。当路由寻找时,若目的节点在本地,则直接利用本地路由发送。若有临近节点为该簇群,则发送给临近节点。否则,利用AODV路由发送到目的节点的簇首信息,以使利于簇首信息进行中继。若目的节点的簇群包含有两个簇首,则分别发送。当中继节点接收到相同的信息时,则抛弃当前信息。若目的节点接收的来自簇首节点的信息,则保留当前路由,且抛弃另外簇点的路由。
4)还还包括簇头节点的更新步骤:
4.1)当新节点接收到有簇头节点的广播HELLO的时候,主动加入该簇群。 将该节点与辅助簇首节点做权值比价,将权值大的作为新的辅助簇首节点,辅助簇首节点成为该簇群内节点。若更新辅助簇首节点,则新的辅助簇首节点作簇内广播,确定辅助簇首节点可以通信的节点。
4.2)当辅助簇首节点的权值大于簇首节点的权值时,辅助簇首节点成为簇首节点,查找下一个新的辅助簇首节点,并作簇内广播以完成簇首节点的更新。
4.3)当簇头节点移动到其他簇时,将权值较大的簇首节点更新为新的簇首节点,簇首节点权值较小的作为该簇群内的节点加入该簇。该簇的辅助簇首节点与该节点比较,产生新的辅助簇首节点。若更新辅助簇首节点则作簇内广播,以确定辅助簇首节点可以通信的节点。
本路由方法基于权值选举簇头形成簇群,同时双簇头设计增强了通信的可靠性并减少了簇头节点的更新频次,其特点适用于拓扑快速变化的无人机群场景。
实施例1
图1为当簇头节点权值小于辅助簇头节点的时候,更新簇头。其中正方形为新簇群内节点,星形表示之前的辅助簇头,且为现在的簇头。菱形表示之前的簇头。圆形表示离开簇群的节点。
图2为三种不同的路由方式。其中三角形代表了属于两个簇群的节点。第一个为簇内的本地路由。第二个为通过本地路由找到临近节点为目标簇群的节点,不经过簇头节点直接到到。第三种为利用临近节点中继,临近节点同时发送给簇头和辅助簇头,其中辅助簇头因无法到达,此条线路中断。簇头继续转发给下一中继节点,然后完成链路。
本发明采用基于多级路由协议的路由思想,在多级路由中融合了成熟的平面路由,使得路由易于理解应用。通过对节点的能量,节点连接度,相对运动性以及相对距离加权来计算节点权值,选取权值最大的作为簇首。本发明创造性的提出辅助簇头节点,是在簇头节点的临节点大于一定数量的时候才会启用。 这一方面提高的通信系统的鲁棒性,另一方面辅助簇头也减少了对簇群簇头的更新,避免了开销。本发明具有良好的通信性能,适用在无人机集群等大型拓扑变化迅速的场景中。
以上所述仅为本发明的较佳实施例,并不用以限制本发明,凡在本发明的精神和原则之内,所作的任何修改、等同替换、改进等,均应包含在本发明的保护范围之内。
需要说明的是,在本发明的描述中,术语“第一”、“第二”等仅用于描述目的和区别类似的对象,两者之间并不存在先后顺序,也不能理解为指示或暗示相对重要性。此外,在本发明的描述中,除非另有说明,“多个”的含义是两个或两个以上。
应该理解,以上描述是为了进行图示说明而不是为了进行限制。通过阅读上述描述,在所提供的示例之外的许多实施例和许多应用对本领域技术人员来说都将是显而易见的。因此,本教导的范围不应该参照上述描述来确定,而是应该参照前述权利要求以及这些权利要求所拥有的等价物的全部范围来确定。出于全面之目的,所有文章和参考包括专利申请和公告的公开都通过参考结合在本文中。在前述权利要求中省略这里公开的主题的任何方面并不是为了放弃该主体内容,也不应该认为申请人没有将该主题考虑为所公开的发明主题的一部分。

Claims (7)

  1. 一种应用在集群无人机通信中的路由方法,其特征在于,包括以下步骤:
    使用复合路由,对每个簇群通过使用能量、节点连接度、相对运动性以及相对距离进行加权,依据加权结果选举出簇首,选取权值最大的作为簇首节点;
    所述簇群的划定基于权值的分簇算法,具体包括以下步骤:
    每个节点广播HELLO消息,用来获取邻居信息,使用节点的能量、节点度、相对运动性以及相对距离进行加权,其中权值的计算为:
    W=w p*P+w c*C+w m*M+w d*D
    其中w p+w c+w m+w d=1,w p为节点能量的权重,P为节点能量,w c为节点连接度的权重,C是临近节点的个数,w m为节点相对运动性的权重,M为相对运动性的值,M的值根据无人机的GPS来计算得出,若与周围n个节点运动方向相似,则值越高,否则值越低,M值为n个值的和;D设定为最近节点数目的一半;w d为n节点的距离的和;
    其中四个权重值使用机器学习训练来给定不同的值,选定权值高的节点的为簇首,构成一个簇群;
    当簇群内节点大于设定数目时,再选举出一个辅助簇首节点;
    所述辅助簇首节点的选择方法具体为:
    在簇首选定成功后,簇首发送HELLO消息,其他节点等待消息:
    若其他接收到HELLO消息,加入该簇,完成簇的构建,计算簇首C的值,若簇首的C值大于l,选择之前权值次高的节点作为辅助簇首节点;
    其中,所述的C值设定为10,即当簇首周围连接数目超过10个,进行辅助簇首节点的选择;
    辅助簇首节点往簇内发送HELLO消息,确保簇内节点与辅助簇首节点相连;
    两个簇首之间分别进行反应式路由查找,若接收到来自相同节点的数据包,则丢弃;
    所述反应式路由查找具体步骤如下:
    簇群内节点采用平面式先应式路由作为本地路由,簇首节点采用DV机制,即每个簇群内节点既保存到其他所有簇首的距离以及矢量路由信息,也保留本 地路由表信息;当路由寻找时,若目的节点在本地,则直接利用本地路由发送;若有临近节点为该簇群,则发送给临近节点;否则,利用AODV路由发送到目的节点的簇首信息,以使簇首信息进行中继;若目的节点的簇群包含有两个簇首,则分别发送;当中继节点接收到相同的信息时,则抛弃当前信息;若目的节点接收的来自簇首节点的信息,则保留当前路由,且抛弃另外簇内节点的路由;还包括簇首节点的更新步骤:
    当新节点接收到有簇头节点的广播HELLO的时候,主动加入该簇群;
    将该节点与辅助簇首节点做权值比价,将权值大的作为新的辅助簇首节点,辅助簇首节点成为该簇群内节点;若更新辅助簇首节点,则新的辅助簇首节点作簇内广播,确定辅助簇首节点通信的节点;
    当辅助簇首节点的权值大于簇首节点的权值时,辅助簇首节点成为簇首节点,查找下一个新的辅助簇首节点,并作簇内广播以完成簇首节点的更新;
    当簇首节点移动到其他簇时,将权值较大的簇首节点更新为新的簇首节点,簇首节点权值较小的作为该簇群内的节点加入该簇;该簇的辅助簇首节点与该节点比较,产生新的辅助簇首节点;若更新辅助簇首节点则作簇内广播,以确定辅助簇首节点通信的节点。
  2. 一种应用在集群无人机通信中的路由方法,其特征在于,包括以下步骤:
    使用复合路由,对每个簇群通过使用能量、节点连接度、相对运动性以及相对距离进行加权,依据加权结果选举出簇首,选取权值最大的作为簇首节点;
    当簇群内节点大于设定数目时,再选举出一个辅助簇首节点;
    两个簇首之间分别进行反应式路由查找,若接收到来自相同节点的数据包,则丢弃。
  3. 根据权利要求2所述的应用在集群无人机通信中的路由方法,其特征在于,簇群的划定基于权值的分簇算法,具体包括以下步骤:
    每个节点广播HELLO消息,用来获取邻居信息,使用节点的能量、节点度、 相对运动性以及相对距离进行加权,其中权值的计算为:
    W=w p*P+w c*C+w m*M+w d*D
    其中w p+w c+w m+w d=1,w p为节点能量的权重,P为节点能量,w c为节点连接度的权重,C是临近节点的个数,w m为节点相对运动性的权重,M为相对运动性的值,M的值根据无人机的GPS来计算得出,若与周围n个节点运动方向相似,则值越高,否则值越低,M值为n个值的和;D设定为最近节点数目的一半;w d为n节点的距离的和;
    其中四个权重值使用机器学习训练来给定不同的值,选定权值高的节点的为簇首,构成一个簇群。
  4. 根据权利要求3所述的应用在集群无人机通信中的路由方法,其特征在于,辅助簇首节点的选择方法具体为:
    在簇首选定成功后,簇首发送HELLO消息,其他节点等待消息:
    若其他接收到HELLO消息,加入该簇,完成簇的构建,计算簇首C的值,若簇首的C值大于l,选择之前权值次高的节点作为辅助簇首节点;
    辅助簇首节点往簇内发送HELLO消息,确保簇内节点与辅助簇首节点相连。
  5. 根据权利要求4所述的应用在集群无人机通信中的路由方法,其特征在于,所述的C值设定为10,即当簇首周围连接数目超过10个,进行辅助簇首节点的选择。
  6. 根据权利要求2所述的应用在集群无人机通信中的路由方法,其特征在于,反应式路由查找具体步骤如下:
    簇群内节点采用平面式先应式路由作为本地路由,簇首节点采用DV机制,即每个簇群内节点既保存到其他所有簇首的距离以及矢量路由信息,也保留本地路由表信息;当路由寻找时,若目的节点在本地,则直接利用本地路由发送;若有临近节点为该簇群,则发送给临近节点;否则,利用AODV路由发送到目 的节点的簇首信息,以使簇头信息进行中继;若目的节点的簇群包含有两个簇首,则分别发送;当中继节点接收到相同的信息时,则抛弃当前信息;若目的节点接收的来自簇首节点的信息,则保留当前路由,且抛弃另外簇内节点的路由。
  7. 根据权利要求2所述的应用在集群无人机通信中的路由方法,其特征在于,还包括簇首节点的更新步骤:
    当新节点接收到有簇头节点的广播HELLO的时候,主动加入该簇群;
    将该节点与辅助簇首节点做权值比价,将权值大的作为新的辅助簇首节点,辅助簇首节点成为该簇群内节点;若更新辅助簇首节点,则新的辅助簇首节点作簇内广播,确定辅助簇首节点通信的节点;
    当辅助簇首节点的权值大于簇首节点的权值时,辅助簇首节点成为簇首节点,查找下一个新的辅助簇首节点,并作簇内广播以完成簇首节点的更新;
    当簇首节点移动到其他簇时,将权值较大的簇首节点更新成为新的簇首节点,簇首节点权值较小的作为该簇群内的节点加入该簇;该簇的辅助簇首节点与该节点比较,产生新的辅助簇首节点;若更新辅助簇首节点则作簇内广播,以确定辅助簇首节点可以通信的节点。
PCT/CN2019/121119 2019-08-27 2019-11-27 一种应用在集群无人机通信中的路由方法 WO2021036055A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910798394.2 2019-08-27
CN201910798394.2A CN110519820B (zh) 2019-08-27 2019-08-27 一种应用在集群无人机通信中的路由方法

Publications (1)

Publication Number Publication Date
WO2021036055A1 true WO2021036055A1 (zh) 2021-03-04

Family

ID=68628273

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/121119 WO2021036055A1 (zh) 2019-08-27 2019-11-27 一种应用在集群无人机通信中的路由方法

Country Status (2)

Country Link
CN (1) CN110519820B (zh)
WO (1) WO2021036055A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114885379A (zh) * 2022-04-29 2022-08-09 西北核技术研究所 一种大规模无人机集群自适应分簇组网方法

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111556550B (zh) * 2020-03-17 2022-03-15 西安电子科技大学 无人机网络通信的路由方法

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140335839A1 (en) * 2012-06-13 2014-11-13 All Purpose Networks LLC Wireless network based sensor data collection, processing, storage, and distribution
CN104754683A (zh) * 2015-04-02 2015-07-01 西北工业大学 基于多跳路由与移动元素的无线传感器网络数据收集方法
US20180376357A1 (en) * 2017-06-27 2018-12-27 Veniam, Inc. Self-organized fleets of autonomous vehicles to optimize future mobility and city services
CN109905164A (zh) * 2019-03-11 2019-06-18 国网甘肃省电力公司信息通信公司 集群范围内无人机的通信与数据回传方法

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103209454A (zh) * 2013-01-06 2013-07-17 南昌大学 一种基于节点竞争值及软阈值的双簇头选举方法
CN103269507B (zh) * 2013-03-05 2016-02-10 江苏科技大学 双簇头无线传感器网络路由方法
CN104301864B (zh) * 2014-07-23 2017-09-29 浙江工业大学 电动汽车充电桩集群的无线通信组网方法
CA3041916A1 (en) * 2016-10-31 2018-05-03 Atomic Energy Of Canada Limited/Energie Atomique Du Canada Limitee System and method for indirectly monitoring one or more environmental conditions
CN106900025B (zh) * 2017-03-15 2020-11-06 重庆邮电大学 一种基于双簇首的无线传感器网络分簇路由方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140335839A1 (en) * 2012-06-13 2014-11-13 All Purpose Networks LLC Wireless network based sensor data collection, processing, storage, and distribution
CN104754683A (zh) * 2015-04-02 2015-07-01 西北工业大学 基于多跳路由与移动元素的无线传感器网络数据收集方法
US20180376357A1 (en) * 2017-06-27 2018-12-27 Veniam, Inc. Self-organized fleets of autonomous vehicles to optimize future mobility and city services
CN109905164A (zh) * 2019-03-11 2019-06-18 国网甘肃省电力公司信息通信公司 集群范围内无人机的通信与数据回传方法

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114885379A (zh) * 2022-04-29 2022-08-09 西北核技术研究所 一种大规模无人机集群自适应分簇组网方法

Also Published As

Publication number Publication date
CN110519820A (zh) 2019-11-29
CN110519820B (zh) 2021-06-15

Similar Documents

Publication Publication Date Title
CN106054875B (zh) 一种分布式多机器人动态网络连通性控制方法
CN108337166B (zh) 一种面向航空集群网络的低时延高可靠路由方法
CN110995333B (zh) 一种分簇QoS路由设计方法
CN110167097B (zh) 基于加权度量转发和路径规划的移动机器人中继路由方法
WO2021036055A1 (zh) 一种应用在集群无人机通信中的路由方法
Cheng et al. A design framework and taxonomy for hybrid routing protocols in mobile ad hoc networks
CN108366093A (zh) 一种基于“虚拟中心”和“移动代理”的分布式无线自组网
He et al. A fuzzy logic reinforcement learning-based routing algorithm for flying ad hoc networks
CN110139335A (zh) 一种基于节点位置信息和可用容量的移动Ad Hoc网络路由方法
Toorchi et al. Skeleton-based swarm routing (SSR): Intelligent smooth routing for dynamic UAV networks
CN108234316A (zh) 一种基于立方体空间区域划分的无人机网络路由选取方法
Bhardwaj et al. An efficient routing protocol for FANET based on hybrid optimization algorithm
CN114641049A (zh) 一种基于模糊逻辑的无人机自组网分层路由方法
CN114449608A (zh) 一种基于Q-Learning的无人机自组网自适应路由方法
CN112423356B (zh) 基于能量均衡的无人装备集群aodv路由方法
CN114867081A (zh) 基于中继无人机节点的移动自组网多源传输路由方法
Zhao et al. A Grid-Based Virtual Routing Algorithm for UAV Systems in Three-Dimensional Space
Wang et al. Elastic routing mechanism for flying ad hoc network
Chenghai et al. Research on dynamic routing algorithm of the combat collaboration communication network
Zheng et al. Simulation Research of UAANET Based on BATMAN-ADV Routing Protocol
Santhi et al. A self-organized location aware energy efficient protocol for wireless sensor networks
Banerjee et al. Study on Proactive and Reactive Routing Approaches for Flying Ad-hoc Networks
Xie et al. A Hierarchical Routing Protocol Based on AODV for Unmanned Aerial Vehicle Swarm Network
Zhou et al. A Geographic Location Prediction-based Routing Algorithm for Flying Ad Hoc Networks
CN106792745B (zh) 一种面向无线自组织网的节点分层路由调度方法及系统

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19943187

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19943187

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 11/08/2022)

122 Ep: pct application non-entry in european phase

Ref document number: 19943187

Country of ref document: EP

Kind code of ref document: A1