Passive routing method, hybrid routing method and hybrid routing system

Abstract

The invention belongs to the technical field of wireless communication ad hoc networks, and discloses a passive routing method, a hybrid routing method and a hybrid routing system.The passive routing method comprises the steps that geographic position information and equipment capacity information of a node and other nodes in a network are obtained; calculating link states among nodes in the network; generating a network topology structure diagram; calculating an optimal path to form an optimal path set; when a routing missing event occurs or a data message needs to be forwarded across clusters, triggering a source node to send a routing discovery request message; after receiving the route discovery request message, the destination node sends a route discovery response message; and the source node generates a routing table item after receiving the routing discovery response message and adds the routing table item to the passive routing table. According to the method, routing holes are avoided, routing loops are eliminated, routing efficiency and routing reliability are improved, occupation of wireless channel resources by routing discovery service broadcast information is reduced, and the method has good applicability in an unmanned aerial vehicle group scene.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless communication ad hoc networks, and in particular relates to a passive routing method, a hybrid routing method and a hybrid routing system. The passive routing method is a self-organizing network routing method that integrates geographic location information and device capability information. The hybrid routing method integrates the passive routing method and the active routing method. The hybrid routing system implements the hybrid routing method. Background Art

[0002] In the prior art, active routing strategies have the advantages of good first packet data immediacy and accurate routing information, and have been well applied in small-scale networking scenarios. However, with the expansion of the scale of unmanned clusters, the number of wireless ad hoc network nodes has also increased, and the service overhead of active routing strategies has increased significantly. Excessive service overhead will reduce network transmission performance and even cause network congestion. In this case, a logical clustering method can be used to limit the scope of routing service information to a cluster domain to limit service overhead. Nodes in the same cluster domain establish routes through active routing strategies, and then open up routes between nodes in different cluster domains through the cluster head area, thereby completing the establishment of the entire network route. The clustered active routing strategy retains the advantages of active routing while also limiting the scale of service overhead.

[0003] However, clustering active routing strategy also brings new problems. Due to the constraints of logical clustering, even if two nodes in different cluster domains have the conditions and physical capabilities for single-hop transmission, message transmission still needs to be forwarded through the cluster head domain for multi-hop forwarding, resulting in a decrease in network bandwidth utilization efficiency. Summary of the invention

[0004] In order to solve the above-mentioned problems existing in large-scale self-organizing network routing, the present invention proposes a passive routing method, which integrates geographic location information and device capability information, including the following steps:

[0005] Step 1: Obtain the geographic location information of the node and other nodes in the network. The nodes in the network can interact with other nodes in geographic location information based on the topological structure established by active routing, and form a geographic location information set {G1, G2....G i ...}, where G i is the geographical location information of the ith node in the network, the geographical location information is used to calculate the distance between nodes, the geographical location information is the latitude and longitude information and elevation information or the projection coordinate position and elevation information;

[0006] Step 2: Obtain device capability information of the node and other nodes in the network. The nodes in the network can interact with other nodes based on the topology structure established by active routing to generate a device capability information set {A1, A2....A i ...}, where A i is the device capability information of the i-th node in the network. The device capability information is used as a parameter of the channel propagation model for estimating the link status between nodes. The device capability information includes the device's transmit power, antenna gain, operating frequency, and demodulation sensitivity;

[0007] Step 3: Calculate the link status between nodes in the network based on the geographic location information and device capability information to form a link status set {L 1,2 ,L 1,3 ....L i,j …}, where L i,j is the link state between node i and node j. The link state is divided into no connection state N, unidirectional connection state H and bidirectional connection state S. The link state has a weight M;

[0008] Step 4, generate a network topology diagram: take the current node as the source node, select all links in the bidirectional connection state as edges, and generate a network topology diagram, where the link weight M is the edge weight of the topology diagram;

[0009] Step 5: Based on the network topology diagram generated in step 4, the current node is used as the source node, and other nodes in the network are used as the destination nodes. The optimal path is calculated using the Dijkstra algorithm or other path selection algorithms to form a set of preferred paths.

[0010] Step 6: When a route loss event occurs or a data message needs to be forwarded across clusters, the source node is triggered to send a route discovery request message. The route discovery request message carries the preferred path information generated in step 5. The nodes along the way use the preferred path as the forward propagation path and propagate the route discovery request information to the destination node.

[0011] Step 7: After receiving the route discovery request message, the destination node sends a route discovery response message, which is propagated back to the source node along the preferred link. During the reverse propagation process, the nodes along the way add the link quality information of themselves and their neighbor nodes to the route discovery response message, wherein the link quality information includes signal strength and signal-to-noise ratio.

[0012] Step 8: After receiving the route discovery response message, the source node generates a routing table entry and adds the routing table entry to the passive routing table, wherein the routing table entry includes destination node information, next hop node information, hop count information, and link quality information of each hop.

[0013] Furthermore, in step 3, the process of calculating the link status of node i and node j according to the geographical location information and the device capability information is as follows:

[0014] Step 3.1: From the geographical location information set {G1, G2....G i ...} to obtain G i and G j , and calculate the distance d between node i and node j i,j ;

[0015] Step 3.2: Use the channel propagation model algorithm to calculate the maximum communication distance C when node i sends and node j receives i,j , on the contrary, calculate the maximum communication distance C when node j sends and node i receives j,i ;

[0016] Step 3.3: Determine the link status:

[0017] When i，j ≤min(C i，j , C j，i ), the link state is a bidirectional connection state;

[0018] When i,j >max(C i,j ,C j,i ), the link state is disconnected;

[0019] When i,j >min(C i,j ,C j,i ) and d i,j ≤max(C i,j ,C j,i ), the link state is a unidirectional connection state;

[0020] Step 3.4: Calculate the link weight based on the device capability information of node i and node j and the distance d i,j , and combined with the channel propagation model algorithm to calculate the receiving power margin of node i and node j respectively, and select the minimum value as the link weight.

[0021] The present invention also proposes a hybrid routing system, including a cluster active routing maintenance module, a passive routing maintenance module and a routing fusion module, wherein:

[0022] The cluster active routing maintenance module is used to establish and maintain an active routing table, and uses an active routing method to complete the establishment of intra-cluster and inter-cluster routing. The generated routing table entries contain destination node information, next hop node information, hop count information, and link quality information of each hop, and the link quality information includes signal strength and signal-to-noise ratio;

[0023] The passive routing maintenance module is used for establishing and maintaining a passive routing table, using the perceived geographical location information and device capability information to pre-establish link state information and network topology, and selecting one or more preferred paths based on the pre-established network topology, and then using the preferred paths to guide the propagation of passive routing discovery protocol messages to complete the establishment of a passive routing table, and the generated routing table entries include destination node information, next hop node information, hop count information, and link quality information of each hop, and the link quality information includes signal strength and signal-to-noise ratio;

[0024] The routing fusion module is used for fusion of active routing table and passive routing table. When both active routing table and passive routing table have paths to the destination node, the path with the least hops or the largest link weight is selected as the routing result and a forwarding routing table is generated.

[0025] The present invention also proposes a hybrid routing method based on the above hybrid routing system, comprising the following steps:

[0026] Step 1: When active routing or passive routing generates a routing update, a routing update message is sent to the routing fusion module;

[0027] Step 2: After receiving the routing update message, the routing fusion module merges the active routing table entries and the passive routing table entries to obtain a set of all reachable destination nodes;

[0028] Step 3: traverse all reachable destination nodes, check whether there is a routing table entry in the active routing table and the passive routing table for the destination node, and decide how to add the routing table entry to the forwarding routing table according to steps 4 and 5;

[0029] Step 4: If there is only an entry to the destination node in the active routing table or the passive routing table, the routing entry is directly added to the forwarding routing table;

[0030] Step 5: If there are routing entries to the destination node in both the active routing table and the passive routing table, the weights Q of the two routing entries are calculated respectively, the path with the larger weight is selected as the optimal path, and the corresponding entry is added to the forwarding routing table.

[0031] Further, in step 5, the weight Q of the routing table entry is calculated as:

[0032]

[0033] Where N is the number of hops in the route, SNR i is the signal-to-noise ratio of the i-th hop, i=1,2,…,N.

[0034] While generating the preferred path, the present invention avoids routing holes and eliminates routing loops, thereby improving routing efficiency and routing reliability. At the same time, because the routing discovery protocol message is propagated along the preferred path in a unicast manner, there is no need to use a broadcast propagation method, which effectively reduces the occupation of wireless channel resources by routing discovery service broadcast information. The present invention has good applicability in scenarios where channel propagation models are easy to predict, especially in drone swarm scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 Optimize the schematic diagram for hybrid routing;

[0036] Figure 2 It is a structural diagram of the hybrid routing system of the present invention;

[0037] Figure 3 It is a basic flow chart of the passive routing method of the present invention;

[0038] Figure 4 is a flow chart of the hybrid routing method of the present invention;

[0039] Figure 5 is a network structure diagram of an embodiment of the present invention;

[0040] Figure 6 A clustered active routing topology diagram of an embodiment of the present invention;

[0041] Figure 7 A pre-built network topology diagram. DETAILED DESCRIPTION

[0042] The technical solution of the present invention is further described below through the accompanying drawings and embodiments.

[0043] The basic process of the passive routing method of the present invention integrating geographic location information and device capability information is as follows: Figure 3 As shown, each step is explained as follows.

[0044] Step 1: Obtain the geographic location information of the node and other nodes in the network. The nodes in the network can rely on the topological structure established by active routing to complete the interaction of geographic location information with other nodes, and can form a geographic location information set {G1, G2....G i ...}, where G i is the geographical location information of the ith node in the network. The geographical location information is used to calculate the distance between nodes. It can be longitude and latitude information and elevation information, or it can be projection coordinate position and elevation information;

[0045] Step 2: Obtain device capability information of this node and other nodes in the network. Nodes in the network can interact with other nodes through the topology structure established by active routing to complete device capability information exchange. Generate device capability information set {A1, A2....A i ...}, where A i is the device capability information of the i-th node in the network. The device capability information is used as a parameter of the channel propagation model to estimate the link status between nodes, and may include but is not limited to the device's transmit power, antenna gain, operating frequency, demodulation sensitivity and other information;

[0046] Step 3: Calculate the link status between nodes in the network based on the geographic location information and device capability information to form a link status set {L 1，2 , L 1，3 ....L i，j ...}, where L i，j Indicates the link status between node i and node j. Link status can be divided into unconnected state, unidirectional connection state and bidirectional connection state, represented by letters N, H and S respectively. Link status can have a weight M. The higher the weight, the better the link quality.

[0047] For example: L i，j The value of (S,5) indicates that the link state between node i and node j is a bidirectional connection state with a weight of 5.

[0048] Based on the geographic location information and device capability information, the process of calculating the link status of node i and node j is as follows:

[0049] 1) Get G in the geographic location information collection i and G j , and calculate the distance d between node i and node j i，j ;

[0050] 2) Select a suitable channel propagation model algorithm (for example, the free space propagation model algorithm can be selected in the high-altitude line-of-sight UAV swarm wireless communication scenario) to calculate the maximum communication distance C when node i sends and node j receives i，j , on the contrary, calculate the maximum communication distance C when node j sends and node i receives j，i ;

[0051] 3) Determine the connection status of the link:

[0052] When i，j ≤min(C i，j , C j，i ), the link state is a bidirectional connection state;

[0053] When i，j >max(Ci，j , C j，i ), the link state is disconnected;

[0054] When i,j >min(C i,j ,C j,i ) and d i,j ≤max(C i,j ,C j,i ), it is a one-way connection state;

[0055] 4) Calculate the link weight.

[0056] In a specific embodiment, the device capability information of node i and node j and the distance d may be used to determine the distance between nodes i and j. i，j , and combined with the channel propagation model selection algorithm to calculate the receiving power margin of node i and node j respectively, and select the minimum value as the link weight.

[0057] Step 4: Generate a network topology diagram.

[0058] Taking this node as the source node, all links in the bidirectional connection state are selected as edges to generate a network topology graph, where the weight M of the link is the weight of the edge of the topology graph.

[0059] Step 5: Based on the network topology diagram, with this node as the source node and other nodes in the network as the destination nodes, the Dijkstra algorithm or other path selection algorithms are used to calculate the optimal path to each destination node to form an optimal path set.

[0060] Preferably, multiple preferred paths with less correlation can be selected for each destination node. Multiple preferred paths can improve the success rate of route discovery and establishment.

[0061] Step 6: When a route missing event occurs or a data message needs to be forwarded across clusters, the source node is triggered to send a route discovery request message.

[0062] The preferred path information generated in step 5 is carried in the route discovery request message, and the nodes along the way propagate the route discovery request information to the destination node using the preferred path as the forward propagation path.

[0063] Step 7: After receiving the route discovery request message, the destination node sends a route discovery response message.

[0064] The route discovery response message is back-propagated to the source node along the preferred path. During the back-propagation process, the nodes along the way add the link quality information of themselves and their neighbor nodes to the route discovery response message. The link quality information may include signal strength, signal-to-noise ratio (SNR), and other information.

[0065] Step 8: After receiving the route discovery response message, the source node generates a routing table entry and adds it to the passive routing table. The routing table entry includes the destination node information, the next hop node information, the number of hops information, and the link quality information of each hop.

[0066] In the present invention, nodes in the network use perceived geographic location information and device capability information to complete the pre-construction of link status and network topology. Based on the pre-constructed network topology, one or more preferred paths are selected. The routing discovery protocol message is propagated using the preferred path as the forward propagation path. The present invention can effectively avoid routing holes, routing loops, and broadcasting of routing discovery protocols, improve the reliability, stability, and timeliness of routing protocols, and reduce the occupancy of network bandwidth.

[0067] Based on the passive routing method of the present invention, combined with the active routing method, the present invention also proposes a hybrid routing method that can take into account the advantages of the two routing strategies. The basic process is as follows: Figure 4 shown.

[0068] For the active routing method, a known active routing strategy (such as OLSR) can be selected to complete the construction of the active routing table. The generated routing table contains the destination node information, the next hop node information, the number of hops information, and the link quality information of each hop. Among them, the link quality information may include information such as signal strength and signal-to-noise ratio SNR.

[0069] Step 1: When active routing or passive routing generates a routing update, a routing update message is sent to the routing fusion module;

[0070] Step 2: After receiving the routing update message, the routing fusion module merges the active routing table entries and the passive routing table entries to obtain a set of all reachable destination nodes;

[0071] Step 3: traverse all reachable destination nodes, check whether there are routing table entries for the destination node in the active routing table and the passive routing table, and decide how to add the routing table entries to the forwarding routing table according to steps 4 and 5.

[0072] Step 4: If there is only an entry to the destination node in the active routing table or the passive routing table, the routing entry is directly added to the forwarding routing table.

[0073] Step 5: If there are routing entries to the destination node in both the active routing table and the passive routing table, the weights Q of the two routing entries are calculated respectively, the path with the larger weight is selected as the optimal path, and the corresponding entry is added to the forwarding routing table.

[0074] In a specific embodiment of the present invention, the formula for calculating the weight Q of the routing table entry by knowing the hop count information and the signal-to-noise ratio (SNR) information of each hop is:

[0075]

[0076] Where N is the number of hops in the route, SNR i is the signal-to-noise ratio of the i-th hop, i=1,2,…,N.

[0077] The hybrid routing method of the present invention is based on cluster active routing. When a cross-cluster domain message arrives, it can be immediately transmitted according to the multi-hop path generated by the active routing strategy, and at the same time trigger the execution of the passive routing strategy, and try to find a better cross-cluster transmission path through the passive routing strategy. This strategy takes into account the advantages of active routing first packet transmission without waiting for routing discovery, and at the same time can optimize the multi-hop transmission path between clusters through the passive routing strategy. The hybrid routing optimization diagram is as follows Figure 1 shown.

[0078] The present invention also proposes a hybrid routing system, the composition architecture of which is as follows: Figure 2 As shown, it includes a cluster active routing maintenance module, a passive routing maintenance module and a routing fusion module.

[0079] The cluster active routing maintenance module is used to establish and maintain the active routing table. The existing known active routing strategy (such as OLSR protocol) can be used to complete the establishment of intra-cluster and inter-cluster routing. The generated routing table entry contains the destination node information, the next hop node information, the number of hops information, and the link quality information of each hop. Among them, the link quality information may include information such as signal strength and signal-to-noise ratio SNR.

[0080] The passive routing maintenance module based on geographic location information and device capability information is used for establishing and maintaining the passive routing table. The module uses the perceived geographic location information and device capability information to pre-establish link state information and network topology, and selects one or more preferred paths based on the pre-built network topology, and then uses the preferred paths to guide the propagation of the passive routing discovery protocol message to complete the establishment of the passive routing table. The generated routing table entry contains the destination node information, the next hop node information, the number of hops information, and the link quality information of each hop. Among them, the link quality information may include information such as signal strength and signal-to-noise ratio SNR. The passive routing maintenance module is used to implement the passive routing method of the present invention described above.

[0081] The routing fusion module is used to merge the active routing table and the passive routing table. When both the active routing table and the passive routing table have paths to the destination node, the path with fewer hops or larger link weight is selected as the routing result and a forwarding routing table is generated. The forwarding routing table is the result of routing optimization and is used to control the forwarding path of the data message.

[0082] Example

[0083] like Figure 5 There are 8 communication device nodes in the exemplary network structure shown. The dotted line connection between the communication device nodes indicates that the communication device nodes have the physical capability to complete wireless communication.

[0084] After implementing the cluster active routing strategy, the generated topology is as follows: Figure 6 As shown, nodes 2, 3, 4 and 5 belong to cluster domain 1; nodes 6, 7 and 8 belong to cluster domain 2; and nodes 1, 2 and 6 belong to the cluster head domain.

[0085] In this embodiment, node 5 is taken as an example, and its active routing table entry example is shown in the following table, where the i-th hop link information is represented by (ID, SNR), where ID is the ID of the i-th hop node, and SNR represents the SNR value of the i-th hop. NAN means that the information does not exist.

[0086] Destination Node Next hop node Hop Count One-hop information Second hop information Three-hop information Four-hop information 1 4 3 (4,10) (2,3) (1,6) NAN 2 4 2 (4,10) (2,3) NAN NAN 3 3 1 (3,4) NAN NAN NAN 4 4 1 (4,10) NAN NAN NAN 6 4 3 (4,10) (2,3) (6,5) NAN 7 4 4 (4,10) (2,3) (6,5) (7,2) 8 4 4 (4,10) (2,3) (6,5) (8,4)

[0087] Node 5 completes the interaction of geographic location information with other nodes through the topological structure established by active routing, and obtains the geographic location coordinates of other nodes as shown in the following table:

[0088] node Projection coordinate position information (x, y, z) 1 (2510,6700,0) 2 (6210,4510,0) 3 (9180,3740,0) 4 (8880,5960,0) 5 (10280,6300,0) 6 (6060,9450,0) 7 (9310,8050,0) 8 (8600,10290,0)

[0089] Node 5 completes the exchange of device capability information with other nodes through the topology structure established by active routing, and obtains the device capability information of other nodes as shown in the following table:

[0090] node Transmit power(dBm) Center frequency(MHz) Demodulation sensitivity(dBm) 1 15 1400 -98 2 15 1400 -98 3 10 1400 -98 4 10 1400 -98 5 10 1400 -98 6 15 1400 -98 7 10 1400 -98 8 10 1400 -98

[0091] Node 5 calculates the link status between nodes in the network based on the acquired geographic location information and device capability information. The process of link status calculation is as follows:

[0092] 1) Calculate the distance d between node i and node j i，j , the calculation formula is:

[0093]

[0094] 2) Calculate the maximum communication distance C when node i sends and node j receives i，j , and then calculate the maximum communication distance C when node j sends and node i receives j，i . With C i，j Taking calculation as an example, the calculation formula is:

[0095]

[0096] Among them, P i is the transmission power of node i, S jis the receiving sensitivity of node j.

[0097] 3) Determine the connection status of the link.

[0098] When i，j ≤max(C i，j , C j，i ), the link state is a bidirectional connection state;

[0099] When i，j >max(C i，j , C j，i ), the link state is disconnected;

[0100] When i,j >min(C i,j ,C j,i ) and d i,j ≤max(C i,j ,C j,i ), it is a one-way connection state;

[0101] 4) Assuming that the link weight of each bidirectional link is 1, the link status information is as shown in the following table:

[0102]

[0103]

[0104] Taking node 5 as the source node and selecting all links in the bidirectional connection state as edges, we can generate Figure 7 The network topology diagram shown.

[0105] Taking node 5 as the source node, the one-hop optimal path to node 8 is calculated using the Dijkstra algorithm as {node 5, node 7, node 8}.

[0106] When a cross-cluster domain service with node 5 as the source node and node 8 as the destination node is generated, the execution of the passive routing protocol will be triggered. Node 5 embeds the preferred path into the route discovery request information and sends the route discovery request information to node 7. After receiving the route discovery request information, node 7 parses the preferred path in the information header to obtain the next hop node information and sends the route discovery request information to node 8. After receiving the route discovery request information, node 8 finds that it is the destination node, and then terminates the propagation of the route discovery request information.

[0107] After receiving the route discovery request message, node 8 adds the link quality information with node 7 to the route discovery response message and sends the route discovery response message to node 7. After receiving the route discovery response message, node 7 adds the link quality information with node 5 and sends the route discovery response message to node 5.

[0108] After receiving the route discovery response message, node 5 generates a route table entry to node 8 and adds it to the passive routing table. At the same time, node 7, as a relay node in the routing path, is also added to the passive routing table. The generated passive routing table results are shown in the following table.

[0109] Destination Node Next hop node Hop Count One-hop information Second hop information Three-hop information Four-hop information 8 7 2 (7,10) (8,8) NAN NAN 7 7 1 (7,10) NAN NAN NAN

[0110] After completing the update of the passive routing table entry, node 5 sends a routing update message to the routing fusion module.

[0111] After receiving the routing update message, the routing fusion module of node 5 merges the active routing table entries and the passive routing table entries to obtain a set of reachable destination nodes {node 1, node 2, node 3, node 4, node 6, node 7, node 8};

[0112] All reachable destination nodes are traversed, among which the routing table entries reaching nodes 1, 2, 3, 4, and 6 only exist in the active routing table and are directly added to the forwarding routing table.

[0113] The routing table entry to node 8 exists in both the active routing table and the passive routing table. The weights of the active routing table item and the passive routing table item are calculated to be 1.24 and 4.47 respectively. The passive routing table item has a higher weight, so the passive routing table item is added to the forwarding routing table.

[0114] The final merged forwarding routing table result is shown in the following table.

[0115] Destination Node Next hop node Hop Count One-hop information Second hop information Three-hop information Four-hop information 1 4 3 (4,10) (2,3) (1,6) NAN 2 4 2 (4,10) (2,3) NAN NAN 3 3 1 (3,4) NAN NAN NAN 4 4 1 (4,10) NAN NAN NAN 6 4 3 (4,10) (2,3) (6,5) NAN 7 7 1 (7,10) NAN NAN NAN 8 7 2 (7,10) (8,8) NAN NAN

[0116] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A passive routing method, characterized in that: The method integrates geographic location information and device capability information, and includes the following steps: Step 1: Obtain the geographic location information of the node and other nodes in the network. The nodes in the network can interact with other nodes in geographic location information based on the topological structure established by active routing, and form a geographic location information set {G1, G2....G i ...}, where G i is the geographical location information of the ith node in the network, the geographical location information is used to calculate the distance between nodes, and the geographical location information is latitude and longitude information and elevation information or projection coordinate position and elevation information; Step 2: Obtain device capability information of the node and other nodes in the network. The nodes in the network can interact with other nodes based on the topology structure established by active routing to generate a device capability information set {A1, a2....A i ...}, where A i is the device capability information of the i-th node in the network. The device capability information is used as a parameter of the channel propagation model for estimating the link status between nodes. The device capability information includes the device's transmit power, antenna gain, operating frequency, and demodulation sensitivity; Step 3: Calculate the link status between nodes in the network based on the geographic location information and device capability information to form a link status set {L 1，2 , L 1，3 ....L i，j ...}, where L i，j is the link state between node i and node j. The link state is divided into no connection state N, unidirectional connection state H and bidirectional connection state S. The link state has a weight M; Step 4, generate a network topology diagram: take the current node as the source node, select all links in the bidirectional connection state as edges, and generate a network topology diagram, where the link weight M is the edge weight of the topology diagram; Step 5: Based on the network topology diagram generated in step 4, the current node is used as the source node, and other nodes in the network are used as the destination nodes. The optimal path is calculated using the Dijkstra algorithm or other path selection algorithms to form a set of preferred paths. Step 6: When a route loss event occurs or a data message needs to be forwarded across clusters, the source node is triggered to send a route discovery request message. The route discovery request message carries the preferred path information generated in step 5. The nodes along the way use the preferred path as the forward propagation path and propagate the route discovery request information to the destination node. Step 7: After receiving the route discovery request message, the destination node sends a route discovery response message, which is propagated back to the source node along the preferred link. During the reverse propagation process, the nodes along the way add the link quality information of themselves and their neighbor nodes to the route discovery response message, wherein the link quality information includes signal strength and signal-to-noise ratio. Step 8: After receiving the route discovery response message, the source node generates a routing table entry and adds the routing table entry to the passive routing table, wherein the routing table entry includes destination node information, next hop node information, hop count information, and link quality information of each hop.

2. A passive routing method according to claim 1, characterized in that: In step 3, the process of calculating node i and node,link status according to geographic location information and device capability information is as follows: Step 3.1: From the geographical location information set {G1, G2....G i ...} to obtain G i and G j , and calculate the distance d between node i and node j i，j ; Step 3.2: Use the channel propagation model algorithm to calculate the maximum communication distance C when node i sends and node i receives i，j , on the contrary, calculate the maximum communication distance C when node j sends and node i receives j，i ; Step 3.3: Determine the link status: When i，j ≤min(C i，j , C j，i ), the link state is a bidirectional connection state; When i，j >max(C i，j , C j，i ), the link state is disconnected; When i，j >min(C i，j , C j，i ) and d i，j ≤max(C i，j , C j，i ), the link state is a unidirectional connection state; Step 3.4, calculate the link weight, based on the device capability information of node i and node , and the distance d i，j , and combined with the channel propagation model algorithm to calculate the receiving power margin of node i and node j respectively, and select the minimum value as the link weight.

3. A hybrid routing system, characterized in that: It includes cluster active routing maintenance module, passive routing maintenance module and routing fusion module, among which: The cluster active routing maintenance module is used to establish and maintain an active routing table, and uses an active routing method to complete the establishment of intra-cluster and inter-cluster routing. The generated routing table entries contain destination node information, next hop node information, hop count information, and link quality information of each hop, and the link quality information includes signal strength and signal-to-noise ratio; The passive routing maintenance module is used for establishing and maintaining a passive routing table, using the perceived geographical location information and device capability information to pre-establish link state information and network topology, and selecting one or more preferred paths based on the pre-established network topology, and then using the preferred paths to guide the propagation of passive routing discovery protocol messages to complete the establishment of a passive routing table, and the generated routing table entries include destination node information, next hop node information, hop count information, and link quality information of each hop, and the link quality information includes signal strength and signal-to-noise ratio; The routing fusion module is used for fusion of active routing table and passive routing table. When both active routing table and passive routing table have paths to the destination node, the path with the least hops or the largest link weight is selected as the routing result and a forwarding routing table is generated.

4. A hybrid routing method based on the hybrid routing system according to claim 3, characterized in that: The steps include: Step 1: When active routing or passive routing generates a routing update, a routing update message is sent to the routing fusion module; Step 2: After receiving the routing update message, the routing fusion module merges the active routing table entries and the passive routing table entries to obtain a set of all reachable destination nodes; Step 3: traverse all reachable destination nodes, check whether there is a routing table entry in the active routing table and the passive routing table for the destination node, and decide how to add the routing table entry to the forwarding routing table according to steps 4 and 5; Step 4: If there is only an entry to the destination node in the active routing table or the passive routing table, the routing entry is directly added to the forwarding routing table; Step 5: If there are routing entries to the destination node in both the active routing table and the passive routing table, the weights Q of the two routing entries are calculated respectively, the path with the larger weight is selected as the optimal path, and the corresponding entry is added to the forwarding routing table.

5. The hybrid routing method according to claim 4, characterized in that: In step 5, the weight Q of the routing table entry is calculated as: Where N is the number of hops in the route, SNR i is the signal-to-noise ratio of the i-th hop, i=1,2,…,N.

Images