AODV routing protocol method based on joint optimization of three-dimensional weight and hop count penalty mechanism

Abstract

The invention relates to an AODV routing protocol method, in particular to an AODV routing protocol method based on joint optimization of a three-dimensional weight hop count penalty mechanism. The invention aims to solve the problem that a single path selection standard neglects node residual energy, link quality and topology stability. The method comprises the following steps: based on position weight calculation of inter-node distance dynamic attenuation, suppressing a long-distance low-quality link; an energy weight is constructed by multiplying the ratio of the residual energy and the initial energy, and network load balancing is optimized; a movement weight in a speed difference square root reciprocal form is introduced, and a link life cycle is predicted; and a hop number punishment mechanism is newly added, so that nonlinear adjustment of multi-hop path quality is realized. And intelligent optimization of comprehensive score of the path is realized by combining a request message RREQ of an improved AODV routing protocol through normalization constraint and a hop punishment mechanism of a three-dimensional coefficient. The invention belongs to the technical field of wireless ad hoc networks.

Description

TECHNICAL FIELD

[0001] The application relates to an AODV routing protocol method and belongs to the technical field of wireless self-organizing networks. BACKGROUND

[0002] Mobile Ad-Hoc Network (MANET) as a kind of distributed communication technology without fixed infrastructure and supporting dynamic networking has important application value in civil fields such as field rescue, fire monitoring, traffic assistance, and military fields such as reconnaissance, cruise and monitoring. In recent years, as a derivative form of MANET, Flying Ad-Hoc Network (FANET) shows a significant growth trend in the research and application of unmanned aerial vehicle cluster cooperation. The network architecture generally adopts low-cost and miniaturized node devices, which are limited by multiple constraints such as power supply, sensing accuracy, communication bandwidth and computing resources, and it is difficult to effectively support multi-task cooperation requirements in complex environments. In addition, due to the dynamic nature of the network topology and the limited nature of the network resources in MANET, the design and optimization of its routing protocol have always been a problem to be solved.

[0003] The commonly used routing protocols in MANET are mainly divided into two categories: active routing protocol and passive routing protocol, the latter is also commonly known as responsive routing protocol. The active routing protocol adopts the traditional routing table update mechanism, which needs nodes to broadcast update information frequently to maintain routing, which will cause a large amount of communication overhead, and this shortcoming is more prominent in the scene of node-intensive or frequent network topology changes. In contrast, the responsive routing protocol only finds the route when there is a communication demand, so it has a great advantage in communication overhead. Because the nodes in the network are mostly small in size and limited in computing power, the responsive routing protocol is preferred in route selection.

[0004] Ad-Hoc On-Demand Distance Vector Routing (AODV) is a kind of responsive routing. AODV realizes low-overhead communication through a routing discovery mechanism based on hop count, but its core defect is to take the minimum hop count as the only optimization target, and the single path selection standard ignores key problems such as node residual energy, link quality and topology stability, resulting in prominent network performance bottlenecks. Therefore, constructing an intelligent routing protocol with multi-objective optimization to overcome the key problems ignored by AODV and improve the network routing quality is an urgent research direction at present. SUMMARY

[0005] The application discloses a three-dimensional weight hop penalty mechanism combined optimization AODV routing protocol method.

[0006] The application discloses a three-dimensional weight hop penalty mechanism combined optimization AODV routing protocol method. Step 1, in a route request stage, a source node broadcasts a route request message RREQ carrying dynamic parameters, wherein the dynamic parameters include node position coordinates, a real-time speed vector, a residual energy value and a hop count value; Step 2, after an intermediate node receives the route request message RREQ, link quality is evaluated through a three-dimensional weight calculation module of the distance, energy and movement of the node; Step 3, a forwarding node is screened through a weight threshold comparison mechanism, nodes meeting a condition participate in route forwarding, and step 2 is continued; Step 4, after a destination node receives multiple route requests, an optimal path is selected through a path comprehensive score obtained based on three-dimensional weights and path penalties; Step 5, a reverse route unicast RREP is sent to the source node, and a route is established.

[0007] Further, the dynamic parameters are carried in an extended RREQ packet structure, that is, a new field of node position, a speed vector component and an energy state identifier is added in the RREQ packet.

[0008] Further, in step 2, the three-dimensional weight calculation formula of the node and the node is as follows: (1), in formula (1), denotes the three-dimensional weight between the node and the node , denotes a distance weight, denotes an energy weight, denotes a movement weight, denotes a normalization coefficient of the distance weight, denotes a normalization coefficient of the energy weight, and denotes a normalization coefficient of the movement weight.

[0009] Further, the calculation formula of the distance weight is as follows: (2), in formula (2), denotes the node , Euclidean distance between them, represents the maximum communication distance of the node.

[0010] Further, the energy weight The calculation formula is: (3), In formula (3), represents the residual energy of the node represents the residual energy of the node represents the residual energy of the node represents the initial energy of the node represents the initial energy of the node represents the initial energy of the node represents the initial energy of the node represents the initial energy of the node

[0011] Further, the calculation formula of the moving weight (4), In formula (4), represents the velocity vector of the node represents the velocity vector of the node represents the velocity vector of the node , , , , represents the velocity component of the node in the x direction, represents the velocity component of the node in the y direction, represents the velocity component of the node in the z direction, represents the velocity component of the node in the x direction, represents the velocity component of the node in the y direction, represents the velocity component of the node in the z direction, represents the maximum moving speed of the node.

[0012] Further, the three-dimensional weight coefficient , and satisfies the normalization constraint condition: (5).

[0013] Further, the specific method of screening the forwarding node by using the weight threshold in step 3 is to ensure that only the intermediate node satisfying participates in the routing establishment.

[0014] ​The value of the weight threshold is usually between 0.4 and 0.7. There are usually two methods for setting the weight threshold: one is a fixed threshold method; and one is a method of dynamically adjusting the threshold according to the network density.

[0015] Further, the path synthesis score calculation formula in step 4 is: (6), In formula (6), denotes the path synthesis score, denotes the sum of three-dimensional weights in the route, denotes the total hop number of the path with a dimension of 1, denotes a hop penalty coefficient; The value range of is usually between 0 and 1. The optimal route is the route with the highest path synthesis score.

[0016] The beneficial effects of the present application are: the present application fuses distance, energy and movement three-dimensional weight and hop penalty mechanism, constructs a link quality evaluation model, effectively overcomes the defects of the traditional AODV protocol relying on single hop number. The distance square reciprocal inhibits long distance low quality link, the energy product normalization optimizes load balancing, the speed difference square root reciprocal predicts the link lifetime, and the normalized weight coefficient dynamically adapts to the scene demand. And combined with the hop penalty mechanism to adjust the multi-hop path quality, improve the network stability and energy efficiency, prolong the network life cycle, suitable for unmanned aerial vehicle cluster and other ad hoc network scenes. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 is a flow chart of the present application; Figure 2 is a process schematic diagram of the routing method in the embodiment of the present application; Figure 3 is an extended RREQ message schematic diagram in a flight ad hoc network. DETAILED DESCRIPTION

[0018] Detailed implementation one: an AODV routing protocol method based on three-dimensional weight and hop penalty mechanism joint optimization, the core improvement of the present application is to construct a three-dimensional weight and hop penalty mechanism joint decision model, through the extension of the routing request data packet structure, the link quality evaluation mechanism and the path synthesis score method, realize multi-dimensional routing optimization. As shown in Figure 1 The method comprises the following steps: As shown in Figure 2As shown in the network of the node 1 is the source node, and the node 19 is the target node, the positions, velocities and residual energy percentages of the 20 nodes are initialized. The circle represents the node, and the two groups of numbers in the circle represent the residual energy and the velocity component of the node respectively. The gray dashed line represents that the two nodes can communicate with each other, the gray arrow represents the RREQ message, and the four groups of numbers on the gray arrow represent the distance between the nodes , the residual energy of the broadcasting node , the velocity component of the broadcasting node , , and the current hop count, and the black arrow represents the RREP message. The communication range of the node is 100 m, and the maximum moving speed of the node is 20 m / s. The following will specifically explain the generation process of a path.

[0019] Step 1, in the route request phase, the source node broadcasts the route request message RREQ carrying dynamic parameters, the dynamic parameters including the node position coordinates, the real-time velocity vector, the residual energy value and the hop count cumulative value. The node 1 initiates the communication request to the node 19, and starts the route finding process. The node 1 broadcasts the extended RREQ message to the surrounding nodes.

[0020] Therefore, the method needs to construct an extended route request (RREQ) data packet. As shown in Figure 3 , three dynamic parameter fields are added to the RREQ data packet format of the standard AODV protocol: 1. The node position coordinates in the flying ad hoc network are stored in the format of 3x16-bit floating point numbers, each direction can cover 65504 meters, which can cover the flying ad hoc network operation.

[0021] 2. The velocity vector component records the real-time speed value of the node in the x, y and z directions. The velocity component is represented by an 8-bit floating point number, which can cover a speed range of m / s, which is suitable for the flying ad hoc network scene and can adapt to most mobile ad hoc network application scenarios.

[0022] 3. The energy state identifier stores the residual energy percentage of the node (i.e. the ratio of the residual energy to the initial energy, 0-100%) in the form of an 8-bit floating point number, which is used to evaluate the energy consumption state of the node.

[0023] Step 2, after receiving the route request message RREQ, the intermediate node performs link quality evaluation through the three-dimensional weight calculation module of the node distance, energy and movement. The nodes 2, 3, 4 and 5 respectively evaluate the link quality from the node 1 to the node after receiving the RREQ message from the node 1.

[0024] Specifically, the node​​ and node The three-dimensional weight calculation formula between two nodes is:

[0025] wherein, is the node and node The three-dimensional weight between two nodes, , and The three weight components are distance weight, energy weight and movement weight, , , The normalization coefficients of distance weight, energy weight and movement weight are respectively.

[0026] In the three-dimensional weight calculation formula, the distance weight The calculation formula is as follows:

[0027] wherein, is the Euclidean distance between the node and node , is the maximum communication distance of the node.

[0028] Based on the free space path loss model, the path loss of two nodes in air communication is proportional to the square of the distance between the nodes, and the greater the distance, the higher the path loss. In order to avoid , infinite problem, the constant term 1 in the denominator is introduced. Considering the order of magnitude of the communication distance of the node, the is used instead of , so that the order of magnitude of the distance weight is comparable to that of the energy weight and the movement weight. The value range of is . Through the reciprocal square function, the communication weight of the long-distance node is suppressed, the adjacent node is preferentially selected, the unstable link is avoided from the source, the signal attenuation and network coverage range are balanced, the packet loss rate is reduced, and the stability of the network can be improved.

[0029] In practical application, the node position needs to be obtained by GPS or self-positioning algorithm, and the accuracy is usually in meter level. At this time, the calculation formula of Euclidean distance is:

[0030] wherein, , , , , , and 、 、 are the position components of node and node in x, y, z directions, and the six quantities are stored in the extended RREQ.

[0031] In the three-dimensional weight calculation formula, the calculation formula of energy weight is as follows:

[0032] wherein, 、 respectively represent the residual energy of node and node , 、 are the initial energy of node and node .

[0033] The value range of is , the normalization processing of the product of the residual energy of the two nodes, and the square root characteristic makes the change of energy weight more gentle, preferentially selects the path with balanced energy, prevents the path from breaking due to the early failure of low-energy nodes, and prolongs the life cycle of the network.

[0034] In practical applications, the residual energy percentage of nodes is stored in the extended RREQ, and the calculation formula of energy weight can be simplified as:

[0035] wherein, , are the residual energy percentages of node and node .

[0036] In the three-dimensional weight calculation formula, the calculation formula of movement weight is as follows:

[0037] wherein, 、 respectively represent the velocity vector of node and node , , , , 、 、 and 、 、 are nodes and nodes Velocity components in x, y, z directions, is the maximum moving speed of the node.

[0038] When the velocity difference tends to 0, the infinite problem, the constant term 1 in the denominator is introduced. Considering the order of magnitude of the node velocity, the moving weight is replaced by instead of , so that the order of magnitude of the moving weight is comparable to the position weight and the energy weight. The value range of is . Based on the square sum of the velocity vector difference, the moving difference between nodes is quantified, the weight of the high-speed moving node is dynamically reduced, the highest priority of the cooperative motion link is preserved, the link with small mobility difference is preferentially selected, and the link stability is improved. At the same time, when the signal source and the receiver move relative to each other, there will be a difference between the received signal frequency and the transmitted frequency, that is, there will be a Doppler shift. The greater the velocity difference between the signal source and the receiver, the more significant the frequency shift, and high Doppler shift will lead to an increase in signal demodulation error rate, which proves that the velocity difference between nodes will affect the link stability.

[0039] In practical applications, the velocity component of the node needs to be obtained in real time through an inertial measurement unit or a velocity sensor.

[0040] The three-dimensional weight coefficient , and satisfies the normalization constraint condition:

[0041] wherein, , , can ensure that key factors such as energy and mobility can participate in decision-making.

[0042] In practical applications, the priority of the three-dimensional weight is adjusted by the normalization coefficient to adapt to different network scenarios. For example, in a self-organizing network with high-speed node motion, can be increased to 0.4-0.5 to avoid link breakage caused by high-speed node motion; in a general flight ad hoc network, can be increased to 0.3-0.4 to prolong the survival time of low-power nodes, while moderately allowing long-distance communication to cover sparse nodes. Figure 2 The three coefficients in the equation are 0.3, 0.5 and 0.2, respectively.

[0043] Step 3, the forwarding node is screened through the weight threshold comparison mechanism, the node meeting the condition participates in routing forwarding, and step 2 is continued.

[0044] The specific method of screening the forwarding node by the weight threshold is to ensure that only the node meeting the condition Intermediate nodes participate in route establishment. The weight threshold is typically between 0.4 and 0.7.

[0045] Because of the normalization constraint , The range of values ​​is within The weight threshold is set to 0.4, which allows for average performance in some dimensions of the network; the weight threshold is set to 0.7, which allows for excellent performance in at least one dimension of the link between two nodes. Figure 2 The network is in good condition and the activity area of ​​nodes is small, so the weight threshold is set to 0.7.

[0046] There are generally two methods for setting weight thresholds. One is a fixed threshold method, and the other is a method that dynamically adjusts the threshold based on network density. The former provides a lightweight solution for resource-constrained networks and is suitable for networks with limited computing power. The latter typically increases the weight threshold to reduce redundant forwarding at high network density and decreases the weight threshold to expand coverage at low network density. The dynamic threshold, adjusted through network congestion feedback, allows the routing protocol to maintain a balance between stability and efficiency during sparse / dense scenario switching, avoiding insufficient coverage or resource waste caused by fixed thresholds, and improving network lifetime and adaptability.

[0047] Node 5 was calculated , Node 5 can participate in route establishment. Node 5 continues to broadcast RREQ messages to surrounding nodes.

[0048] Step 4: After receiving multiple routing requests, the destination node selects the optimal path based on the comprehensive path score obtained from the three-dimensional weights and path penalties.

[0049] The formula for calculating the comprehensive path score based on three-dimensional weights and path penalties is as follows:

[0050] in, The overall score represents the path. This represents the sum of the three-dimensional weights in the routing. This represents the total number of hops in a path with a dimension of 1. This represents the hop count penalty coefficient. The optimal route is the route with the highest overall path score. Node 19 receives the routing request and calculates the path. The overall path score is taken. , It is the highest-scoring path among all routing requests received by node 19 from node 1, and is selected. As a communication path.

[0051] The path hop count has been stored in the traditional RREQ message, which can be directly obtained in practical application.

[0052] The numerator term is the comprehensive weight value of the three dimensions of distance, energy and mobility, reflecting the overall stability and reliability of the path; the denominator term applies a nonlinear penalty to the total hop count , which suppresses the transmission delay and cumulative error caused by excessive hop count.

[0053] The path loss is related to the path loss exponent as . Assuming a path with hops, the distance of the kth path is , and the total end-to-end signal attenuation is . If the distance per hop is approximately equal , then each distance is also approximately equal to the average distance , so .

[0054] Let to compensate for the physical layer attenuation characteristics. Generally, the default network is in free space, and the path loss exponent is 2; in an obstacle-dense environment, is 3-4. Therefore, from theoretical derivation, the value range is usually . When , the model degenerates to linear hop count penalty; when , the penalty increases super-linearly with the increase of hop count, and short paths are preferred to reduce end-to-end delay.

[0055] In practical application, the hop count penalty mechanism can significantly reduce the score of long paths, prefer short hop count paths, suppress long paths, effectively reduce communication delay, indirectly reduce the dependence of the network on unstable links with high hop count, and improve the overall routing robustness. The specific setting of needs to be combined with network density, business demand and device capability, through theoretical range constraint and scenario-based dynamic adjustment, to achieve the global optimum of routing efficiency and stability. It is recommended to use as the benchmark value in the free space when first deployed, and then fine-tune according to the actual performance monitoring results.

[0056] Step 5, establish routing according to the reverse route unicast RREP to the source node. Node 19 broadcasts RREP according to the path of , and node 1 receives the RREP message from node 19, and the routing is established.

[0057] The application solves the limitation of the single hop index of the traditional AODV protocol by the method of the joint decision of the three-dimensional weight and the hop penalty mechanism. The real-time parameters carried by the extended RREQ packet provide data basis for weight calculation; the distance weight suppresses long-distance low-quality links, the energy weight optimizes load balancing, and the mobile weight improves the link survival period; the dynamic threshold screening reduces the redundant routing discovery overhead; the hop penalty mechanism improves the link stability; and the path comprehensive score realizes global optimization. The above improvements significantly improve the network stability in a high dynamic environment, and are suitable for scenes such as unmanned aerial vehicle cluster cooperation and emergency communication network.

[0058] Table 1 symbol explanation table

[0059] The above is only a preferred embodiment of the present application, and does not limit the present application in any form. Although the present application has been disclosed as above with a preferred embodiment, it is not intended to limit the present application. Any person skilled in the art can make some changes or modifications to the above disclosed technical content without departing from the scope of the technical solution of the present application, and any simple modification, equivalent replacement and improvement of the above embodiments within the scope of the technical solution of the present application, the technical essence of the present application, and the spirit and principles of the present application, are all within the protection scope of the present application.

Claims

1. A method for AODV routing protocol jointly optimized based on a three-dimensional weighted hop count penalty mechanism, characterized in that, The specific steps include: Step 1: During the routing request phase, the source node broadcasts a routing request message RREQ carrying dynamic parameters, including node location coordinates, real-time velocity vector, remaining energy value, and cumulative hop count. Step 2: After receiving the routing request message RREQ, the intermediate node performs link quality assessment through the three-dimensional weight calculation module of the node's distance, energy, and mobility. Step 3: Filter forwarding nodes through a weight threshold comparison mechanism. Nodes that meet the conditions participate in routing and forwarding, and continue to step 2. Step 4: After receiving multiple route requests, the destination node selects the optimal path based on the comprehensive path score obtained from the three-dimensional weights and path penalties. Step 5: unicast RREP to the source node according to the reverse route to establish a route.

2. The AODV routing protocol method based on joint optimization of a three-dimensional weighted hop count penalty mechanism according to claim 1, characterized in that, The dynamic parameter carrying method adopts an extended RREQ data packet structure, that is, adding fields to the RREQ data packet: node position, velocity vector component and energy state identifier.

3. The AODV routing protocol method based on joint optimization of a three-dimensional weighted hop count penalty mechanism according to claim 1, characterized in that, Nodes in step 2 and nodes The formula for calculating the three-dimensional weight of a two-node component is as follows: （1）， In formula (1), Represents a node and nodes The three-dimensional weights between the two nodes Indicates distance weight, Indicates energy weight, Indicates the moving weight. This represents the normalized coefficient of the distance weight. The normalization coefficient representing the energy weight. This represents the normalization coefficient of the moving weights.

4. The AODV routing protocol method based on joint optimization of a three-dimensional weighted hop count penalty mechanism according to claim 3, characterized in that, Distance weight The calculation formula is: （2）， In formula (2), Represents a node With nodes The Euclidean distance between them Indicates the maximum communication distance of the node.

5. The AODV routing protocol method based on joint optimization of a three-dimensional weighted hop count penalty mechanism according to claim 3, characterized in that, Energy weight The calculation formula is: （3）， In formula (3), Represents a node The remaining energy, Represents a node The remaining energy, Represents a node initial energy, Represents a node The initial energy.

6. The AODV routing protocol method based on joint optimization of a three-dimensional weighted hop count penalty mechanism according to claim 3, characterized in that, Shift weights The calculation formula is: （4）， In formula (4), Represents a node The velocity vector, Represents a node The velocity vector, , , , Represents a node The velocity component in the x-direction, Represents a node The velocity component in the y-direction Represents a node The velocity component in the z-direction, Represents a node The velocity component in the x-direction, Represents a node The velocity vector in the y-direction. Represents a node The velocity vector in the z-direction, This indicates the maximum movement speed of the node.

7. The AODV routing protocol method based on joint optimization of a three-dimensional weighted hop count penalty mechanism according to claim 3, characterized in that, Three-dimensional weighting coefficients And it satisfies the normalization constraint: （5）。 8. The AODV routing protocol method based on joint optimization of a three-dimensional weighted hop count penalty mechanism according to claim 1, characterized in that, The specific method for filtering forwarding nodes using weighted thresholds in step 3 is to ensure that only nodes that meet the weighted thresholds are forwarded. Intermediate nodes participate in route establishment.

9. The AODV routing protocol method based on joint optimization of a three-dimensional weighted hop count penalty mechanism according to claim 1, characterized in that, The formula for calculating the overall path score in step 4 is as follows: （6）， In formula (6), The overall score represents the path. This represents the sum of the three-dimensional weights in the routing. This represents the total number of hops in a path with a dimension of 1. This represents the penalty coefficient for the number of jumps; The value range is usually in ; The optimal route is the route with the highest overall path score.

Images