Optimal routing selection method based on routing cost

Abstract

The invention belongs to the technical field of Internet of Things, particularly relates to an optimal routing selection method based on routing cost, and aims to solve the routing selection problem in a multi-routing cluster tree wireless MESH network. According to the method, the local routing table containing the destination node address, the next hop address, the routing cost and the updating time is maintained at each routing node, so that the function of dynamically updating the routing information is realized. The routing cost is a result obtained by adding the cost factors (such as electric quantity, link quality, available load and the like) of each node on the message transmission path according to a preset weight. The mechanism ensures that various influence factors can be comprehensively considered during route selection, so that the optimal path is selected. No matter the uplink message forwarding or the downlink message forwarding, the method can guide the routing selection through the routing cost updated in real time, thereby effectively improving the success rate of message transmission and reducing the transmission time delay.

Description

Technical Field

[0001] The invention belongs to the technical field of Internet of Things, and in particular relates to an optimal route selection method based on route cost. Background Art

[0002] The cluster-tree topology wireless MESH network is an important networking method for wireless MESH networks. Compared with the fully connected MESH network, the cluster-tree MESH network uses the tree topology to solidify the uplink and downlink routes, so the cost of route maintenance is lower, and the nodes have lower requirements for power consumption, computing power, memory, and bandwidth. It is widely used in data collection and monitoring of the Internet of Things.

[0003] The cluster tree wireless MESH network is composed of several routing nodes and leaf nodes through a multi-level cluster topology. A single cluster consists of a head node and several member nodes. The member nodes can be routing nodes or leaf nodes. The head node is generally a routing node. In a single cluster, the packet sent by the member node to the head node is an uplink packet, and the packet sent by the head node to the member node is a downlink packet. Multiple single clusters are cascaded to form a tree network. The root node of the tree, i.e., the gateway node, is a special routing node (Border Router). It forwards the uplink packets within the network to the external network, and forwards the downlink packets from the external network to the nodes within the network. In the data collection and monitoring Internet of Things, the uplink packets are generally the collected data, and the downlink packets are generally the query, configuration and control instructions.

[0004] If each member node is associated with only one head node in a cluster, the topology is relatively simple, but reliability cannot be guaranteed due to the single route. Especially in scenarios where nodes move or the wireless environment changes greatly, the topological relationship between nodes needs to be dynamically rebuilt, and the routes also change accordingly. The network topology and route reconstruction takes a long time and a high cost, which affects the real-time and reliability of data transmission. Therefore, in actual deployment, in order to enhance network reliability, the cluster tree network can allow a routing node to be associated with the head nodes of multiple clusters or be associated with the member nodes of multiple clusters. This cluster tree network is called a multi-route cluster tree wireless MESH network.

[0005] Figure 1 This is a typical example of a multi-route cluster tree wireless MESH network. The network contains 8 communication nodes, forming four clusters. Among them, routing node e is associated with two head nodes: a and b. Therefore, there are two uplink routes from routing node e to gateway node g, and there are two downlink routes from gateway node g to routing node e. Multi-route cluster tree MESH network brings about the routing selection problem, that is, the routing node needs to select an optimal route when forwarding uplink and downlink packets.

[0006] For example, the patent "Downlink Routing System and Method for Wireless Self-Organizing MESH Network", Publication (Announcement) No.: CN118400785B, proposes a specific method for constructing, updating and using routing tables in routing nodes to solve the inefficiency problem caused by downstream flooding. The patent proposes to update the downstream route by comparing the QoS, hop count and timestamp of the upstream packet, but does not take into account the resource status of each node on the path passed by the packet (such as load, power, link quality, etc.), so it is impossible to obtain the optimal routing judgment based on the real-time status of the entire route. At the same time, the patent only points out the relevant methods of the downstream routing table, and does not mention the relevant methods of the upstream routing table. Summary of the invention

[0007] The purpose of the present invention is to provide an optimal route selection method based on routing cost, by weighting the cost factors of all routing nodes on the message forwarding path and adding them together, the routing cost from the source node to the current routing node is obtained and stored in the routing table as the basis for route selection when forwarding reverse direction message packets, so as to solve the optimal route selection problem in a multi-route cluster tree wireless MESH network.

[0008] To achieve the above-mentioned purpose, the present invention adopts the following technical scheme: an optimal route selection method based on routing cost, there are multiple routes between each routing node, each routing node maintains a local routing table for route selection when forwarding uplink and downlink messages, the key fields of each route in the routing table include the destination node address, the next hop node address, the routing cost and the update time, the routing cost is obtained by accumulating the cost factors of all routing nodes on the message forwarding path after assigning weights to each of them, and updating them into the routing table as the basis for route selection when the routing node forwards the message in the reverse direction, that is, if there are multiple routes to the destination node specified by the message in the routing table, the routing costs are compared to select the optimal route.

[0009] Preferably, the routing cost is a number between 0 and MaxCost, and is calculated as follows: routing cost = (MaxCost - minimum value of the amount of electricity passing through the node) * W1 + (MaxCost - minimum value of the link quality passing through the node) * W2 + (MaxCost - minimum value of the available load passing through the node) W3 + number of hops passing through the node (MaxCost / maximum allowed number of hops) * W4.

[0010] Preferably, W1, W2, W3, and W4 are weights of respective cost factors, reflecting the influence of respective cost factors on routing cost, and satisfying W1+W2+W3+W4=1.

[0011] Preferably, the MaxCost is the upper limit value of the cost, reflecting the quantization accuracy of the routing cost, and the larger the value, the higher the accuracy.

[0012] Preferably, the four cost factors are respectively the minimum value of the amount of electricity passing through the node, the minimum value of the link quality passing through the node, the minimum value of the available load passing through the node, and the number of hops passing through the node; The minimum value of the power of the nodes passed is the minimum value of the power of all the forwarding nodes that the message passes through; The minimum value of the link quality of the node passed is the minimum value of the link quality of all forwarding nodes that the message passes through; The minimum value of the available load of the node through which the message passes refers to the minimum value of the available load of all forwarding nodes through which the message passes; The number of node hops refers to the number of all forwarding nodes that the message passes through.

[0013] Preferably, the four cost factors are required to be carried in the headers of all uplink and downlink message packets. When the routing node receives each uplink and downlink message, it calculates a new routing cost and updates it to the routing table.

[0014] Preferably, the routing node needs to recalculate the four cost factors before forwarding the message and update them to the header of the message packet. The calculation method is: The minimum power value of the new node passed = Min (the minimum power value of the old node passed, the power value of this node); The minimum available load of the new passing node = Min (the minimum available load of the old passing node, the available load of this node); The minimum link quality of the new passing node = Min (the minimum link quality of the old passing node, the link quality of this node); The number of new node hops = the number of old node hops + 1.

[0015] Preferably, the routing table is queried according to the source node address and the previous hop node address information carried in the message packet header; If there is no route with the corresponding destination node or next hop node as the index in the routing table, a new route is added; otherwise, the route cost and update time of the route are updated. The downstream route is updated according to the upstream message, and the upstream route is updated according to the downstream message.

[0016] Preferably, a periodic scanning and routing cost decay processing mechanism is provided in the routing table, which is used to periodically scan the entire routing table for the routing node, using the time decay function Process the route cost decay, that is, recalculate the new route cost = original route cost * ;in, = , a is the decay intensity factor, t0 is the update time field of the route, and t is the current system time.

[0017] Preferably, the routing node queries the routing table according to the destination node address information carried in the message, and handles three situations respectively: If the route does not exist, if it is an uplink message forwarding, any head node is selected as the next hop node; if it is a downlink message forwarding, the broadcast channel is used for broadcast forwarding; If there is a single route, obtain the next hop node address in the route for forwarding; If there are multiple routes, obtain the next hop node address in the route with the smallest routing cost for forwarding.

[0018] Technical effects and advantages of the present invention: Compared with the prior art, the optimal route selection method based on route cost proposed by the present invention has the following advantages: The present invention obtains the routing cost from the source node to the current routing node by weighting the cost factors of all routing nodes on the message forwarding path and then adding them up, and saves the cost in the routing table as the basis for routing selection when forwarding message packets in the reverse direction; the method is applicable to both uplink and downlink routing, and the routing cost reflects the real-time status of the entire routing. As the basis for optimal routing selection, it can significantly improve the success rate of message routing and reduce message transmission delay. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 This is an example diagram of a traditional multi-route cluster tree wireless MESH network; Figure 2 This is an example diagram of the routing table in this embodiment; Figure 3 This is an example diagram of a message packet header in this embodiment; Figure 4 is a flow chart of the optimal route selection method based on route cost in this embodiment; Figure 5 Schematic diagram of the hardware functions of the routing node in this embodiment; Figure 6 FIG. 4 is a schematic diagram of software functions of a routing node in this embodiment. DETAILED DESCRIPTION

[0020] The following will be combined with the accompanying drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. The specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0021] The present invention provides an optimal route selection method based on route cost. There are multiple routes between each route node. Each route node maintains a local route table for route selection when forwarding uplink and downlink messages. The key fields of each route in the route table include the destination node address, the next hop node address, the route cost and the update time. The route cost is obtained by accumulating the cost factors of all route nodes on the message forwarding path after weighting them respectively, and is updated in the route table as the basis for route selection when the route node forwards the message in the reverse direction. That is, if there are multiple routes to the destination node specified by the message in the route table, the route costs are compared to select the optimal route.

[0022] In order to solve the problem of optimal route selection, the present invention provides a route selection method based on route cost, that is, by weighting the cost factors of all route nodes on the message forwarding path and adding them together, the route cost from the source node to the current route node is obtained and stored in the routing table as the basis for route selection when forwarding reverse direction message packets. This method is applicable to both uplink and downlink routes. The route cost reflects the real-time status of the entire route. As the basis for optimal route selection, it can significantly improve the success rate of message routing and reduce message transmission delay.

[0023] The following will be described in detail with reference to the accompanying drawings: Before describing the routing method provided by the present invention, several key data structures need to be described: Routing table, such as Figure 2 shown. Figure 2 A routing table example is provided. Each routing table contains several routes, each of which includes at least a sequence number, a destination node address, a next-hop node address, a routing cost, and an update time. When the destination node address is the same, the next-hop node address cannot be the same. That is, the destination node address and the next-hop node address of each route are uniquely determined. The update time field is used to indicate the system time of the route update.

[0024] Message packet header, such as Figure 3 The header fields of the uplink and downlink message packets carry information used to calculate the routing cost, namely, the cost factor. Figure 3Several typical cost factors are given in the message packet header example: the minimum value of the power of the node passed, the minimum value of the link quality of the node passed, the minimum value of the available load of the node passed, and the number of hops of the node passed. Among them, the "minimum value of the power of the node passed" refers to the minimum value of the power of all forwarding nodes passed by the message. The power value can be graded and quantized into integers between [0, MaxCost], and the smaller the value, the less power. Generally, if the routing node is a non-battery-powered device, the power value is MaxCost. If the routing node is a battery-powered device, the power value can be a function of the power supply voltage, and then normalized to an integer between 0-MaxCost. When the message packet is forwarded, the "minimum value of the power of the node passed" field needs to be updated.

[0025] Among them, "the minimum value of the link quality of the node passed" refers to the minimum value of the link quality of all forwarding nodes passed by the message. The link quality value can be hierarchically quantified into integers between [0, MaxCost]. The smaller the value, the worse the link quality. The link refers to the link between this node and the associated head node. The link quality can be obtained by multiple measurements and hierarchical quantification of indicators of different protocol layers of the wireless link, such as RSSI and SNR indicators of the physical layer, PER indicators of the link layer, etc., or the above indicators can be comprehensively considered to obtain the final link quality. When a message is forwarded, the "minimum value of the power of the node passed" field needs to be updated.

[0026] Among them, "the minimum available load value of the node through which the message passes" refers to the smallest value among the available loads of all forwarding nodes through which the message passes. The available load value can be graded and quantified into integers between [0, MaxCost]. The smaller the value, the less available load. The available load can be obtained by measuring the resource idle rate of the routing node and quantifying it in grades, such as the idle rate of the cache queue, the busy-idle ratio of the air interface resources, etc., or the above resources can be comprehensively considered to obtain the final node load. When a message is forwarded, the "minimum available load value of the node through which the message passes" field needs to be updated.

[0027] Finally, "Number of Node Hops Passed" refers to the number of all forwarding nodes that the message passes through. When a message is forwarded, the "Number of Node Hops Passed" field needs to be updated.

[0028] The parameter MaxCost used above is the upper limit of the cost, which reflects the quantization accuracy of the routing cost. The larger the value, the higher the accuracy. MaxCost is a global parameter of the network, which satisfies , typical value 255.

[0029] The routing selection method provided by the present invention includes three processes, the process contents and processing relationships are as follows: Figure 4 When the routing node receives the message, it calls three processes in sequence: updating the routing table, routing selection, and updating the message header: 1) Update routing table process: Before updating the routing table, the routing cost needs to be calculated. The calculation of routing cost requires the input of each cost factor in the message header information, routing cost = (MaxCost-minimum value of the power of the node passed)*W1+(MaxCost-minimum value of the link quality of the node passed)*W2+(MaxCost-minimum value of the available load of the node passed)*W3+number of hops passed*(MaxCost / maximum allowed hops)*W4. Among them, w1-w4 are weight factors, which are determined by the system configuration and satisfy w1+w2+w3+w4 =1. Since the value range of each cost factor is 0-MaxCost, it is guaranteed that the final routing cost value is also in the range of 0-MaxCost.

[0030] W1-W4 are the weights of each cost factor, reflecting the impact of each cost factor on the routing cost. They are global parameters of the network and satisfy W1+W2+W3+W4 = 1. Principles for setting weight values: If the energy consumption balance of the entire network is given priority, W1 can be set greater than other weights; if the transmission reliability of the entire network is given priority, W2 can be set greater than other weights; if the load balance of the entire network is given priority, W3 can be set greater than other weights; if the transmission delay of the entire network is given priority, W4 can be set greater than other weights; if you want to balance all indicators of the entire network, all weights can be set equal.

[0031] The routing table is queried according to the source node address and the previous hop node address information carried in the message packet. If there is no corresponding route with the index (destination node, next hop node) in the routing table, a new route is added; otherwise, the route cost and update time of the route are updated. It should be noted that the downstream route is updated according to the upstream message, and the upstream route is updated according to the downstream message.

[0032] exist Figure 4 In the example, the routing node receives a message packet from the previous hop node B, calculates the new routing cost as 13 based on the packet header information, queries the routing table based on the address pair (A, B), and updates the corresponding routing cost.

[0033] 2) Routing selection process: The routing node queries the routing table based on the destination node address information carried in the message, considering three situations: There is no routing. If it is an uplink message, any head node is selected as the next hop node; if it is a downlink message, it is broadcast forwarded using the broadcast channel; If there is a single route, obtain the next hop node address in the route for forwarding; If there are multiple routes, obtain the next hop node address in the route with the smallest routing cost for forwarding.

[0034] It should be noted that if the number of message hops reaches an upper limit, such as MaxCost, the message will be discarded.

[0035] exist Figure 4 In the example, the routing node queries the routing table according to the address (E), compares the routing costs, and obtains the next hop address C.

[0036] 3) Update message header process: To update the packet header information, you need to enter the test values ​​of each cost factor of this routing node. These measured values ​​are smoothed and quantized to ensure that the values ​​are within the range of 0-MaxCost. The update algorithm is as follows: Update the next hop address C with the result of the routing selection process; The new minimum power value of the passing node = min (the old minimum power value of the passing node, the power value of this node); The new minimum available load of the passing node = min (the old minimum available load of the passing node, the available load of the current routing node); The new minimum link quality of the passing nodes = min (the old minimum link quality of the passing nodes, the link quality of the current routing node); The new number of passed nodes = the old number of passed nodes + 1.

[0037] In actual application, there are several auxiliary processes for routing table update: 1. The root node periodically rolls over the update frequency of the route cost field in the routing table: If there is no business data in the network for a long time, the network needs to stimulate the update of the routing table of each node through spontaneous messages to maintain the validity of the route. The root node of the cluster tree network periodically sends broadcast or multicast messages, such as status query, to stimulate each destination node to return an uplink message, and the routing cost in the routing table is updated.

[0038] 2. Periodic scanning and decay processing of the routing cost field in the routing table: If the routing cost is not updated for a long time, the credibility will drop rapidly. Provide a decay processing mechanism. Each time it is scanned, the cost value will increase gradually until it fails and the memory space is released. The specific implementation is that the node periodically scans the routing table and uses the time decay function The route cost of each route is decayed, that is, the new route cost is recalculated = the original route cost * .in, = , a is the decay intensity factor, t0 is the update time of the route in the routing table. The value range of a is (0,1).

[0039] The routing node is the core device for building a cluster tree network and implementing the routing method. Its typical function implementation is given below.

[0040] 1. Hardware features: Figure 5 This is a typical hardware function diagram of a routing node, which includes three subsystems. The power supply subsystem supports battery power supply and external power supply; the antenna subsystem is responsible for wireless signal modulation and demodulation; the computing subsystem includes a computing unit and a volatile / non-volatile storage unit, and is the carrier of the routing method of the present invention. The computing subsystem communicates with other nodes through the wireless subsystem to jointly build a cluster tree network and implement the routing system mentioned in the present invention.

[0041] 2. Software features: Figure 6 This is a typical software function diagram of a routing node. The software runs on the computing subsystem, and its functional modules are divided into three parts: message processing, routing maintenance, and cost factor measurement. The message processing module is responsible for sending and receiving uplink and downlink messages, querying routing tables, refreshing packet header information, and maintaining wireless links; the routing maintenance module load calculates routing costs, updates routing based on message packets, and updates routing based on decay mechanisms; the cost factor measurement module provides quantitative measurement data for routing, including power measurement, link quality measurement, and available load measurement. The three functional modules cooperate with each other to implement the routing selection method based on routing cost given in the present invention.

[0042] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An optimal route selection method based on route cost, wherein there are multiple routes between each route node, each route node maintains a local route table for route selection when forwarding uplink and downlink messages, and the key fields of each route in the route table include the destination node address, the next hop node address, the route cost and the update time, characterized in that: The routing cost is obtained by weighting the cost factors of all routing nodes on the message forwarding path and summing them up, and updating them in the routing table as the basis for route selection when the routing node forwards messages in the reverse direction. That is, if there are multiple routes to the destination node specified by the message in the routing table, the routing costs are compared to select the optimal route.

2. The optimal route selection method based on route cost according to claim 1, characterized in that: The routing cost is a number between 0 and MaxCost, and is calculated as follows: routing cost = (MaxCost - minimum value of the amount of electricity passing through the node) * W1 + (MaxCost - minimum value of the link quality passing through the node) * W2 + (MaxCost - minimum value of the available load passing through the node) W3 + number of hops passing through the node (MaxCost / maximum allowed number of hops) * W4.

3. The optimal route selection method based on route cost according to claim 2, characterized in that: The W1, W2, W3, and W4 are weights of the cost factors, respectively, reflecting the impact of the cost factors on the routing cost, and satisfying W1+W2+W3+W4=1.

4. The optimal route selection method based on route cost according to claim 3, characterized in that: The MaxCost is the upper limit of the cost, reflecting the quantization accuracy of the routing cost. The larger the value, the higher the accuracy.

5. The optimal route selection method based on route cost according to claim 4, characterized in that: The four cost factors are respectively the minimum value of the amount of electricity passing through the node, the minimum value of the link quality passing through the node, the minimum value of the available load passing through the node, and the number of hops passing through the node; The minimum value of the power of the nodes passed is the minimum value of the power of all the forwarding nodes that the message passes through; The minimum value of the link quality of the node passed is the minimum value of the link quality of all forwarding nodes that the message passes through; The minimum value of the available load of the node through which the message passes refers to the minimum value of the available load of all forwarding nodes through which the message passes; The number of node hops refers to the number of all forwarding nodes that the message passes through.

6. The optimal route selection method based on route cost according to claim 5, characterized in that: The four cost factors are required to be carried in the headers of all uplink and downlink message packets. When the routing node receives each uplink and downlink message, it calculates a new routing cost and updates it to the routing table.

7. The optimal route selection method based on route cost according to claim 6, characterized in that: Before forwarding the message, the routing node needs to recalculate the four cost factors and update them to the header of the message packet. The calculation method is: The minimum power value of the new node passed = Min (the minimum power value of the old node passed, the power value of this node); The minimum available load of the new passing node = Min (the minimum available load of the old passing node, the available load of this node); The minimum link quality of the new passing node = Min (the minimum link quality of the old passing node, the link quality of this node); The number of new node hops = the number of old node hops + 1.

8. The optimal route selection method based on route cost according to claim 7, characterized in that: Query the routing table based on the source node address and previous hop node address information carried in the message packet header; If there is no route with the corresponding destination node or next hop node as the index in the routing table, a new route is added; otherwise, the route cost and update time of the route are updated. The downstream route is updated according to the upstream message, and the upstream route is updated according to the downstream message.

9. The optimal route selection method based on route cost according to claim 8, characterized in that: The routing table is equipped with a periodic scanning and routing cost decay processing mechanism, which is used to periodically scan the entire routing table for the routing node, using the time decay function Process the route cost decay, that is, recalculate the new route cost = original route cost * ;in, = , a is the decay intensity factor, t0 is the update time field of the route, and t is the current system time.

10. The optimal route selection method based on route cost according to claim 9, characterized in that: The routing node queries the routing table according to the destination node address information carried in the message, and handles three situations separately: If the route does not exist, if it is an uplink message forwarding, any head node is selected as the next hop node; if it is a downlink message forwarding, the broadcast channel is used for broadcast forwarding; If there is a single route, obtain the next hop node address in the route for forwarding; If there are multiple routes, obtain the next hop node address in the route with the smallest routing cost for forwarding.

Images