Routing method of high-speed mobile node based on ST-GPSR protocol

Abstract

The invention belongs to the technical field of ultra-short wave communication, and particularly relates to a routing method of a high-speed mobile node based on an ST-GPSR protocol, which comprises the following steps: constructing an ultra-short wave communication network consisting of a high-speed mobile node and a base station; gathering all high-speed mobile nodes of the ultra-short wave communication network; periodically broadcasting a HELLO data packet by each high-speed mobile node in the high-speed mobile node set, and constructing a neighbor list according to the HELLO data packet; screening each node in the neighbor list by adopting an ST-GPSR protocol to obtain a candidate node set; constructing a credibility function, screening out an optimal node from the candidate node set through the credibility function, and carrying out routing; according to the method, factors of node link survival time and forwarding deflection angle are comprehensively considered, and in the process of selecting the next hop, the data transmission efficiency is ensured, and the link connection stability is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of ultra-short wave communication, and particularly relates to a routing method for high-speed mobile nodes based on the ST-GPSR protocol. Background Art

[0002] In recent years, with the increasingly prominent position of the ocean in China's development strategy and the continuous development of modern maritime communication technologies, ultra-short wave ad-hoc network, as a new type of wireless communication technology, has gradually become an important part of maritime communication. Ultra-short wave ad-hoc network is a wireless communication network for data transmission based on the ultra-short wave frequency band, which can support the nodes in the network to automatically establish an overall communication network through a self-organization mechanism and integrated radio anti-jamming technology without pre-establishing infrastructure, so it is widely used in fields such as emergency communication, maritime operations, and maritime cooperative operations.

[0003] Traditional ultra-short wave communication itself belongs to line-of-sight transmission. When the operation platform is far from the coast, it can only rely on other mobile platforms for networking. And maritime communication is prone to be affected by uncertain factors such as weather, and the movement trajectories of nodes are random. Therefore, under the networking conditions where the environment is harsh and the operation platform is in a real-time moving state, the stability of the communication link is greatly affected. At the same time, the communication bandwidth resources of the maritime ultra-short wave wireless ad-hoc network are relatively tight, and it is impossible to allocate independent channels for each user when the nodes are densely distributed, resulting in the inability to convey information in the communication network in a timely manner. In order to achieve high-efficiency and high-reliability data communication services in the maritime communication environment, it is necessary to optimize the routing decision of the routing protocol.

[0004] The GPSR routing protocol based on geographical location has the advantages of simple routing forwarding mechanism, strong topology adaptability, and low routing overhead, so it is widely used in mobile ad hoc networks. However, in the scenario where nodes move frequently and rapidly, the forwarding mechanism of the GPSR routing protocol is prone to problems such as packet forwarding falling into a local optimal solution and routing holes. In response to the application limitations of the GPSR routing protocol in the mobile ad hoc network scenario, scholars at home and abroad have conducted explorations in different aspects. In the prior art, SILVAA, REZA N, OLIVEIRA A, Improvement and performance evaluation of GPSR-based routing techniques for vehicular ad hoc networks[J]. IEEE Access, 2019(7):21722-21733 excluded the nodes that had used the perimeter forwarding mode and the nodes that were currently performing data forwarding work during routing decision-making to avoid falling into a routing hole again. In BENGAG A, BENGAG A, BOUKHARI M E, Enhancing GPSR routing protocol based on velocity and density for real-time urban scenario[C] / / 2020 International Conference on Intelligent Systems and Computer Vision(ISCV), Fez, Morocco, 2020:1-5, the node movement speed and node degree were used as the next-hop selection criteria to solve the problems of link instability and routing holes during data transmission. The literature YANG X, LI M, QIAN Z, et al. Improvement of GPSR protocol in vehicular ad hoc network[J]. IEEE Access, 2018, 6:39515-39524 proposed the MM-GPSR routing protocol (Maxduration-Minangle GPSR), which selects the neighbor node with the largest cumulative communication duration as the next-hop node. However, the above improvement schemes still have limitations such as being unable to adapt to the random direction movement of nodes, large network overhead, and generating routing loops in the maritime VHF ad hoc network scenario. Summary of the Invention

[0005] To solve the problems existing in the above prior art, the present invention proposes a routing method for high-speed mobile nodes based on the ST-GPSR protocol. The method includes: constructing a VHF communication network composed of high-speed mobile nodes and base stations; aggregating all high-speed mobile nodes in the VHF communication network; each high-speed mobile node in the high-speed mobile node aggregation periodically broadcasts HELLO packets, and constructs a neighbor list according to the HELLO packets; screening each node in the neighbor list by using the ST-GPSR protocol to obtain a candidate node set; constructing a trust function, screening out the optimal node from the candidate node set through the trust function, and performing routing.

[0006] Advantages of the present invention:

[0007] The ST-GPSR routing protocol of the present invention selects neighbor nodes that meet the link establishment conditions within the communication range through a node prediction model, calculates the comprehensive trust degree of nodes by combining the link survival time and the forwarding deflection angle, so as to select the optimal next hop. At the same time, the ST-GPSR protocol defines the node movement tendency degree according to the node movement state, performs perimeter forwarding by selecting neighbor nodes with the best movement tendency degree, and in addition, avoids duplicate forwarding of the same packet by establishing a DTN table. The ST-GPSR protocol of the maritime VHF ad hoc network of the present invention improves the packet delivery rate by about 4%, the network throughput by about 20%, and the average end-to-end delay by about 13% compared with the traditional MM-GPSR and GPSR protocols. Description of the drawings

[0008] Figure 1 It is a schematic diagram of GPSR greedy forwarding;

[0009] Figure 2 It is a schematic diagram of GPSR perimeter forwarding;

[0010] Figure 3 It is a schematic diagram of the defects of GPSR greedy forwarding;

[0011] Figure 4 It is a schematic diagram of the defects of GPSR perimeter forwarding;

[0012] Figure 5 It is a model diagram of a maritime VHF mobile ad hoc network;

[0013] Figure 6 It is a diagram of the decomposition of the movement speed vector between nodes;

[0014] Figure 7 It is a diagram of the movement trend between nodes when θ ∈ (0, π / 2);

[0015] Figure 8 It is a diagram of the movement trend between nodes when θ ∈ (π / 2, π);

[0016] Figure 9 is |v nx |<|v s |Time node - to - node relative velocity decomposition diagram;

[0017] Figure 10 is |v nx |>|v s |Time node - to - node relative velocity decomposition diagram;

[0018] Figure 11 Schematic diagram of forwarding deflection angle;

[0019] Figure 12 Schematic diagram of moving tendency degree;

[0020] Figure 13 DTN table working flow chart;

[0021] Figure 14 Group delivery rate effect comparison diagram;

[0022] Figure 15 Network throughput comparison diagram;

[0023] Figure 16 Average end - to - end delay comparison diagram;

[0024] Figure 17 Overall flow chart of the present invention. Detailed implementation manners

[0025] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0026] A routing method for high - speed mobile nodes based on the ST - GPSR protocol, as Figures 1 - 4 and Figure 17 shown. The method includes: constructing a VHF communication network, which consists of high - speed mobile nodes and base stations; aggregating all high - speed mobile nodes in the VHF communication network; each high - speed mobile node in the high - speed mobile node aggregation periodically broadcasts HELLO data packets, and constructs a neighbor list according to the HELLO data packets; screening each node in the neighbor list by using the ST - GPSR protocol to obtain a candidate node set; constructing a trust function, and screening out the optimal node from the candidate node set through the trust function and performing routing.

[0027] The GPSR routing protocol, whose full name is Greedy Perimeter Stateless Routing, is a routing algorithm based on geographical location information. The GPSR routing protocol only needs to save the attributes and status of adjacent one-hop neighbor nodes, without establishing, updating, and maintaining the routing table of the global network. Therefore, compared with table-driven routing protocols and on-demand routing protocols, the network overhead is smaller. As the number of network nodes increases, the network scalability of the GPSR routing protocol is stronger than that of distance-vector routing or link-state routing. Even if the nodes in the network move quickly and randomly, the GPSR protocol can quickly find an alternative route for packet forwarding based on the information of one-hop neighbor nodes, so it is widely used in wireless ad hoc networks.

[0028] Using the geographical location information of network nodes, the next-hop path is selected through a greedy algorithm for packet forwarding. In the mobile ad hoc network framework, each network node carries its own GPS positioning system and is uniformly addressed, which enables the node to make path decisions according to the GPSR routing protocol and forward the data to the next hop with the smallest Euclidean distance between the destination node within its communication range. When the data forwarding falls into a routing hole and the greedy algorithm cannot be used, the GPSR routing protocol will use the perimeter forwarding algorithm to transmit the data to the nodes around the hole according to the right-hand rule, helping the data to get out of the routing hole, thereby improving the data transmission efficiency and network reliability.

[0029] When performing data packet forwarding, the GPSR protocol mainly uses the greedy forwarding mode, that is, selects the neighbor node with the smallest distance from the destination node as the next-hop forwarding node, as Figure 1 shown. When the node does not meet the greedy forwarding condition, that is, there is no neighbor node other than itself with the smallest distance from the destination node within the communication range, it is considered that the data forwarding has fallen into a "routing hole" at this time. The GPSR routing protocol switches to perimeter forwarding and determines the next-hop forwarding node through the right-hand rule. As Figure 2 shown, in order to avoid routing holes, the perimeter forwarding state will forward data in the order of S→X→W→K→V→Y→S.

[0030] In the greedy forwarding mode, the next-hop node selected by the GPSR protocol is usually located at the edge of the communication range of the forwarding node. In the maritime VHF ad hoc network scenario, the network topology changes with high dynamics, making the next-hop node may move outside the communication range of the forwarding node before receiving the data packet, resulting in the breakage of the network communication link, an increase in the probability of packet retransmission and loss, and thus reducing the overall network efficiency. As Figure 3As shown, node A selects node F as the next hop at this moment. However, at the next moment, node F moves outside the communication range of A. If the original selection is still maintained, it is very likely that the established communication link will have poor stability, resulting in problems such as communication link breakage and data loss.

[0031] When data transmission falls into a routing hole, the GPSR protocol switches to the perimeter forwarding mode. The next-hop forwarding node is selected through the right-hand rule to help the data packet transmission jump out of the routing hole. However, the traditional perimeter forwarding mode not only requires a relatively large number of forwarding times to find a node that meets the greedy forwarding condition, but also the found forwarding path may have a "detour" phenomenon, which leads to an increase in end-to-end delay. In the case where the distribution of network nodes is relatively sparse, data packet transmission will frequently encounter routing holes, so the transmission mode will be continuously switched, resulting in more redundant hops. As shown in the figure, when sending data from node S to node D, according to the right-hand rule, the data forwarding order is S→A→B→C→D. At this time, if S→C→D is selected as the forwarding path, the number of hops can be effectively reduced.

[0032] Aiming at the problems of poor stability of the communication link established by the GPSR protocol in the maritime VHF ad hoc network and easy to fall into routing holes, this paper proposes an improved ST-GPSR routing protocol. For the greedy forwarding mode, the protocol estimates the moving distance and coordinate changes of nodes through a node position prediction model, which is used as a criterion for whether to become a candidate node. At the same time, based on the link survival time and forwarding deflection angle, the trust degree of candidate nodes is calculated regularly. Based on the optimal comprehensive trust degree criterion, the node with the strongest stability is selected as the next hop.

[0033] As Figure 5 shown, the maritime VHF communication network system consists of a base station and several ship nodes. The set of ship nodes is defined as V C ={v1, v2, v3,......, v9}, and each node is in a uniformly variable motion in real time, and the motion direction is random. All ships are equipped with a GPS satellite positioning system and a radio communication system, which can record their own geographical location, motion speed and direction in real time; all nodes can send and receive data, and the transmission power is the same. Remote scheduling and monitoring are realized through sea-to-sea and shore-to-sea information interaction.

[0034] In this embodiment, the node position prediction model includes: Let the position coordinates of the source node S be (x s , y s ), and the velocity vector be v s . The position coordinates of the neighbor node N are (x n , y n ), and the velocity vector be v n , and the included angle between the velocity directions of the nodes is θ, as Figure 6As shown in the figure, the Euclidean distance between the source node S and the neighbor node N is shown in Equation (1).

[0035]

[0036] The vector decomposition of the velocity of node N can be obtained from Equations (2) and (3).

[0037] v nx = |v n cosθ|i (2)

[0038] v ny = |v n sinθ|j (3)

[0039] Since the nodes in the VHF mobile ad-hoc network have the characteristics of randomly changing velocity and moving direction, it is necessary to study in combination with the included angle of the velocity directions between nodes. Let the included angle of the moving directions be θ, and the following will be discussed according to the value of θ.

[0040] (1) When θ ∈ (0, π / 2), the moving trend of the nodes is as Figure 7 shown. At this time, the x-axis component of the velocity of node S is in the same direction as the velocity component of node N. Therefore, it can be determined that nodes S and N are in the same-direction motion state. At this time, the relative velocity between the nodes can be obtained from Equation (4):

[0041]

[0042] where v r is the relative moving velocity of neighbor node N with respect to source node S. Then, the predicted moving distance D r of the neighbor node within the time interval t can be obtained from Equation (5):

[0043] D r = |v r | × t (5)

[0044] In the case of θ ∈ (0, π / 2), the magnitude of the lateral coordinate component of the velocity between nodes has little influence on the relative motion trend of the nodes. Therefore, the included angle φ value can be known from Equation (6).

[0045]

[0046] (2) When θ ∈ (π / 2, π), the moving trend between nodes is as Figure 8 shown. At this time, the x-axis component of the velocity of node S is in the opposite direction to the velocity component of node N. Therefore, it can be determined that nodes S and N are in the opposite-direction motion state. Thus Figure 3 the relative velocity between the nodes in can be obtained from Equation (7).

[0047]

[0048] At this time, the magnitude of the horizontal coordinate component of the velocity between nodes has a greater impact on the relative motion trend of the nodes, and it is necessary to conduct a classification discussion again.

[0049] When |v nx |<|v s |, the relative motion of the nodes at this time is as Figure 9 shown. Although the motion directions of node N and node S tend to be opposite, according to the principle of relative motion, the relative motion direction of node N tends to be the same as that of node S. Therefore, the specific calculation of the included angle φ can be obtained from formula (8).

[0050]

[0051] When |v nx |>|v s |, the relative motion of the nodes at this time is as Figure 10 shown. At this time, the motion directions of node N and node S tend to be opposite. The predicted moving distance D r of the neighbor node S within the time interval t can be obtained from formula (5), and the specific calculation of the included angle φ can be obtained from formula (9).

[0052]

[0053] According to the cosine theorem, the estimated distance between the predicted neighbor node and the source node is as shown in formula (10).

[0054]

[0055] In summary, before forwarding the data packet, the source node carrying the data packet will traverse the neighbor table and select the nodes in the neighbor table that are likely to establish a stable communication link according to formula (10). The node N i that satisfies formula (11) can enter the candidate node set SN.

[0056]

[0057] To reduce the impact brought by the model prediction error, it is necessary to introduce ε as the error adjustment parameter, and the determination formula is as shown in formula (12), where v max is the upper limit of the node movement speed.

[0058]

[0059] In this embodiment, the link survival time is the total time that the communication link between the data sending end and the data receiving end is maintained during the network communication process, which is an important indicator for evaluating the stability of the communication link. The source node establishes a communication link with the neighbor nodes within the communication range. When a neighbor node moves outside the effective communication range, the link communication effect begins to decline until it fails. Therefore, the longer the link survival time between nodes, the better the survival performance of the network link built by the node. In a complex terrain environment, the link connection time is related to the node position, speed magnitude, and direction. Therefore, in combination with the above position prediction model, the link connection time can be defined.

[0060] Let the geographical coordinates of the source node S be (x s , y s ), and the position of the candidate node N be (x n , y n ). Then, according to formula (1), the distance D between the candidate node and the target node can be obtained SN . In the GPSR routing protocol, in order to determine and update the working status and geographical location of its own neighbor nodes, the source node needs to periodically generate and send HELLO packets. After the neighbor nodes successfully receive them, the link survival time is calculated and the neighbor table is updated. As Figure 7 shown, assume that at time t, the relative velocity vector of the candidate node N and the source node S is v r , and the nodes move at a constant speed until the edge of the communication range at a certain time interval. According to formula (13), the link connection time t link can be obtained.

[0061]

[0062] In this embodiment, the forwarding deflection angle is the included angle between the source node, the forwarding node, and the destination node. The smaller the forwarding deflection angle, the closer the selected forwarding path is to a straight line. Selecting a path with an appropriate forwarding deflection angle for data transmission can ensure that the data packet is transmitted to the target node along a straight line as much as possible, thereby reducing the routing hops and improving the data transmission efficiency. As Figure 11 shown, when the geographical location coordinates of the source node S, the candidate node N, and the target node G are known, the forwarding deflection angle θ N can be obtained according to the cosine theorem, and the specific calculation is shown in formula (14).

[0063]

[0064] where d ij is the distance between node i and node j. The forwarding deflection angle θ NThe smaller the value, the more the forwarding path tends to be a straight line, and the more suitable the node is as the next hop. To sum up, a mathematical model of a weighted distribution trust function is constructed by combining the link connection time and the forwarding deflection angle, as shown in formula (15).

[0065] Q N =γ * t link +(1 - γ) * cosθ N (15)

[0066] When a data packet is sent to a certain node in the network, the trust values of candidate nodes within the communication range of this node are calculated through the above formula, and the node with the largest trust value is selected as the next hop. By continuously repeating this process, the data is finally transmitted to the destination node, thereby improving the stability of the overall self-organizing network for data transmission.

[0067] In this embodiment, the perimeter forwarding algorithm includes: during the process of breaking out of the routing hole, the traditional perimeter forwarding algorithm is too simple, with high randomness in the selection of the next-hop node, and prone to problems such as detouring and even routing loops. In the maritime VHF self-organizing network, the random movement of nodes will also lead to problems such as an increased probability of packet forwarding encountering a routing hole and an increased number of forwarding hops. Based on this, the ST-GPSR protocol uses the node movement tendency index to evaluate the degree of approximation of the movement states of neighbor nodes and the target node, and selects the neighbor node with the optimal tendency value for perimeter forwarding to reduce the probability of data packets falling into the routing hole during the movement of future node positions.

[0068] The node movement tendency is an index that measures the degree of approximation between the movement direction of a candidate node and its geographical direction relative to the target node at a certain moment. The larger the value of the node movement tendency, the more the movement direction and distance of the node at the next moment are closer to the target node, and the higher the probability of converting to greedy forwarding during the next data transmission.

[0069] The node movement tendency is calculated based on the movement direction of neighbor nodes and the relative movement distance to the destination node at the next moment. As Figure 12 shown, S is the current node carrying the data packet, G is the target node, A, B, and C are the neighbor nodes within one hop of node S. Assume that the position coordinates of the source node S are (x s , y s ), and the speed coordinates are (v sx , v sy ); the position coordinates of neighbor node A are (x a , y a ), and the speed coordinates are (v ax , v ay ); the position coordinates of the target node G are (x g , y g) Then, the moving speed coordinates of node N relative to node S are shown by formula (16).

[0070]

[0071] Then, it can be known that the relative moving speed v of the node r and the position vector d AG The included angle δ A between them can be obtained from formula (17).

[0072]

[0073] From the above formula, it can be seen that when δ A The smaller the value, the more the moving direction of node A tends to the geographical location of the target node G. During the subsequent data transmission process, it increases the probability that neighbor node A finds a next-hop forwarding node closer to G, thereby minimizing the average number of hops of the overall transmission route as much as possible and optimizing the data transmission efficiency.

[0074] To meet the optimization conditions of the ST-GPSR routing protocol, it is necessary to modify the HELLO packet and neighbor table format of the routing protocol. At the same time, in order to minimize the network overhead and the probability of data getting into a routing loop, it is necessary to design a DTN table to record the forwarding history of the data.

[0075] The HELLO packet is the main means in the GPSR routing protocol to obtain neighbor information. During the operation of the ultra-short wave ad hoc network, nodes discover neighbor nodes by periodically generating and broadcasting HELLO packets, and obtain their geographical locations and related information. To ensure the accuracy of the node neighbor list, the HELLO packet will only be broadcast within one-hop range adjacent to the source node, and other nodes cannot forward it.

[0076] Let the sending interval of the HELLO packet be t, the jitter delay interval be [0.5t, 1.5t], and the difference between the waiting time for receiving the response and the sending time be Δt. Then, the source node will broadcast the HELLO packet within the communication range every t time period and wait for the neighbor node to respond. If the source node detects that Δt < 3t, it is determined that the node is a neighbor node, and the corresponding neighbor table data is updated according to the response return information. Otherwise, it is determined that the node does not exist within the communication range, and the node is excluded from the neighbor table.

[0077] The HELLO data information of the traditional GPSR routing protocol only contains the geographical location and speed magnitude of the nodes. To implement an optimized solution for the greedy forwarding and perimeter forwarding modes, the ST-GPSR protocol modifies the format of the HELLO packet controlled by the protocol and the neighbor information table. The modified HELLO packet format includes the node position, the modulus value of the velocity component, and the moving tendency degree, where the minimum bandwidth is represented by the available waiting queue length. The specific format design description is shown in Table 2.

[0078] Table 2 HELLO Packet Format

[0079]

[0080] In the improved GPSR routing protocol solution, after a networking node receives the HELLO data information broadcast by a neighbor node, in addition to recording the ID, location information, and the timestamp of receiving the HELLO message of the neighbor node, it also needs to record the link connection time and the moving tendency degree of the neighbor node. The format of the modified neighbor list is shown in Table 3. Among them, field 2 is responsible for recording the ID number of the node in the neighbor table, and fields 3 and 4 record the geographical location coordinates of the corresponding node; field 5 records the timestamp of receiving the HELLO message. The LLT field represents the link connection time, and the MT field represents the moving tendency degree.

[0081] Table 3 Neighbor Table Format

[0082]

[0083] To reduce the phenomenon of increased network overhead and routing decision falling into a loop caused by the GPSR routing protocol during the transmission process. This paper uses the idea of data forwarding record to count the forwarding path of the packet and the target node ID, so as to minimize the probability of the packet being repeatedly sent to the same node. The specific format of the DTN table is shown in Table 4.

[0084] Table 4 DTN Table Format

[0085]

[0086] Among them, the data packet identifier uses the identifier field in the IPv4 header. The packet identifier is an independent and unique number in each data transmission process, so as to distinguish different data packets.

[0087] The establishment and maintenance process of the transmission history table is as Figure 13As shown in the figure. In the ad-hoc network system using this routing protocol, when the source node S sends a packet to the neighbor node K, it first obtains the relevant information of the data packet header, and then queries the transmission history table maintained by itself to check whether there is an entry that matches the packet header information. If there is, it means that the data packet has been forwarded by the neighbor node K, and the node directly abandons sending the packet; otherwise, the node will forward the packet to the neighbor node K, and at the same time update the packet forwarding information to its own data sending table, recording the ID of the previous hop node, the data packet identifier, and the ID information of the destination node. When the source node S receives the HELLO message from the neighbor node K or the information entry about node K in the neighbor table expires and becomes invalid, the relevant entry of the neighbor node K in the DTN table is deleted accordingly. By statistically grouping the sending of historical information, duplicate data packets can be identified and filtered out in a timely manner, thereby reducing network overhead. At the same time, discarding duplicate data packets can effectively prevent the occurrence of network loop routing and improve the efficiency of routing forwarding.

[0088] This paper uses the network simulation simulator NS-3 to build a VHF communication environment to simulate the communication process of the ST-GPSR routing protocol, and compares the simulation results with those of the improved MM-GPSR and traditional GPSR routing protocols in the same environment. By increasing the number of nodes in steps of 20 and increasing the node speed in steps of 5, the performance of the routing protocol before and after improvement is compared and analyzed in turn.

[0089] In this simulation experiment, three metrics, namely packet delivery ratio, network throughput, and end-to-end delay, are selected to evaluate the performance of the routing protocol.

[0090] (1) Packet delivery ratio: The packet delivery ratio (PDR) refers to the proportion of valid data packets successfully received by the destination node in the total number of data packets sent during network transmission. In a dynamic wireless network, due to the mobility of nodes and frequent topological changes, the transmission success rate of data packets often fluctuates. Therefore, PDR is of great significance in dynamic environments such as wireless networks, especially mobile ad-hoc networks. Its expression formula (18) is as follows, where num recv is the total number of data packets received by all network nodes, and num send is the total number of data packets successfully sent by all network nodes.

[0091]

[0092] (2) Average end-to-end delay: This parameter refers to the time it takes for a data packet to travel from the source node to the destination node. End-to-end delay is an important indicator for evaluating whether the network meets the quality of service requirements. End-to-end delay usually consists of propagation delay, transmission delay, queuing delay, and processing delay. To simplify the model, this paper only considers the transmission delay, and its calculation process is shown in Equation (19), where n is the number of successfully transmitted and received data packet groups, d i is the moment when the i-th data packet is successfully forwarded to the next-hop node, and s i is the moment when the source node successfully sends the i-th data packet.

[0093]

[0094] (3) Network throughput: This parameter represents the number of data bits successfully transmitted by the overall network system per unit time, usually measured in bits per second. In a VHF mobile ad-hoc network, network throughput is usually affected by factors such as changes in network topology, signal interference, and limited bandwidth resources. Therefore, optimizing the network throughput indicator is of great significance for optimizing the transmission efficiency, communication quality, and reliability of the network. Its calculation process is shown in Equation (20), where num recv is the total number of data packets received by all network nodes, P size is the size of a single data packet, and T is the experimental simulation time.

[0095]

[0096] After the simulation is completed, the comparison results of the network performance between ST-GPSR and the routing protocol control group are as follows.

[0097] Figure 13 are the data packet delivery rate performance curves of the GPSR, GPSR-MM, and ST-GPSR protocols under different network simulation scenarios. In the simulation environment, the packet delivery rates of the ST-GPSR, MM-GPSR, and GPSR protocols increase with the expansion of the network scale and decrease with the increase in node movement speed. When the number of nodes is small and the movement speed is the fastest, the quality of the network link built is the worst. Therefore, the phenomenon of packet loss and retransmission during data packet transmission will be more serious. Under the same network scale and speed change, the packet delivery rate of ST-GPSR is on average 3.8% higher than that of MM-GPSR and 5.6% higher than that of GPSR on average. This is because the ST-GPSR protocol can select the next-hop node within the set security area, thus avoiding to the greatest extent the phenomenon of the next-hop node moving out of the communication area during the forwarding process, resulting in the breakage of the communication link and ensuring the stability of communication.

[0098] Figure 14It is the network throughput performance curves of three protocols, GPSR, MM-GPSR, and ST-GPSR, under different network simulation scenarios. As the network scale expands, the network throughputs of the ST-GPSR, MM-GPSR, and GPSR routing protocols are continuously increasing. However, phenomena such as multipath propagation effects and natural electromagnetic interference occurring during the rapid movement of the offshore platform will also reduce the network throughput. Under the same network environment, the network throughput of the ST-GPSR protocol is on average 19.1% higher than that of MM-GPSR and 24.4% higher than that of GPSR. Different from the traditional GPSR protocol that uses distance as the next-hop selection criterion, the ST-GPSR protocol comprehensively considers the node link survival time and forwarding angle, thereby selecting a more stable next-hop node. Therefore, the selected data transmission link is more stable than the traditional GPSR protocol, and thus the optimization effect on the network throughput is better.

[0099] Figure 15 It is the average end-to-end delay performance curves of three protocols, GPSR, MM-GPSR, and ST-GPSR, under different network simulation scenarios. When the number of nodes is small, data packets are prone to getting stuck in routing holes during the forwarding process and cannot reach the source node in time, which not only increases the average number of hops for data transmission but also continuously increases the end-to-end delay. In a high-speed mobile network environment, the TCP protocol may perform frequent retransmissions and congestion, resulting in higher delays.

[0100] As Figure 16 shown, it can be seen from the simulation results that the end-to-end delay of the ST-GPSR protocol is on average 12.8% lower than that of MM-GPSR and 20.5% lower than that of GPSR. This is because when the ST-GPSR enters the perimeter forwarding state, it uses the node movement tendency to select the next-hop node, making the data packet forwarding direction converge with the geographical location of the target node, thereby increasing the probability of data packets jumping out of the routing hole and effectively reducing the frequency of switching between forwarding states, thus effectively reducing the end-to-end delay of data transmission.

[0101] The above-mentioned embodiments further elaborate on the purpose, technical solutions, and advantages of the present invention. It should be understood that the above-mentioned embodiments are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made to the present invention within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A routing method for high-speed mobile nodes based on the ST-GPSR protocol, characterized in that, Including: Construct a VHF communication network, which consists of high-speed mobile nodes and base stations; aggregate all high-speed mobile nodes of the VHF communication network; Each high-speed mobile node in the set of high-speed mobile nodes periodically broadcasts HELLO packets and constructs a neighbor list based on the HELLO packets; Use the ST-GPSR protocol to screen each node in the neighbor list to obtain a set of candidate nodes; construct a trust function, and screen out the optimal node from the set of candidate nodes through the trust function and perform routing.

2. The routing method of a high-speed mobile node based on the ST-GPSR protocol according to claim 1, wherein The HELLO packet consists of a message type, a timestamp, the current node ID, node coordinates, and the speed component of the node.

3. The routing method of a high-speed mobile node based on the ST-GPSR protocol according to claim 1, characterized in that Constructing a neighbor list includes: the HELLO packet sending interval is t, the jitter delay interval is [0.5t, 1.5t], the difference between the waiting reception response time and the sending time is Δt, the source node broadcasts HELLO packets within the communication range every t time period and waits for neighbor nodes to respond. If the source node detects that Δt < 3t, it determines that the node is a neighbor node and updates the corresponding neighbor table data according to the response return information, otherwise it deletes the node from the neighbor list.

4. A routing method for a high-speed mobile node based on the ST-GPSR protocol according to claim 1, characterized in that Using the ST-GPSR protocol to screen each node in the neighbor list includes: determining the forwarding mode of the ST-GPSR protocol. If the ST-GPSR protocol is the greedy forwarding mode, use the GPSR protocol based on the greedy forwarding mode to screen neighbor nodes; if the ST-GPSR protocol is the perimeter forwarding mode, use the GPSR protocol based on the perimeter forwarding mode to screen neighbor nodes.

5. The routing method of a high-speed mobile node based on the ST-GPSR protocol according to claim 4, characterized in that, Using the GPSR protocol based on the greedy forwarding mode to screen neighbor nodes includes: obtaining the location information of each neighbor node; inputting the location information into the node location prediction model to predict the moving distance and coordinate change of the node; judging whether the node is a candidate node according to the predicted moving distance and coordinate change.

6. The routing method of a high-speed mobile node based on the ST-GPSR protocol according to claim 5, characterized in that, The node location prediction model processes the input data including: calculating the Euclidean distance between it and adjacent nodes according to the node location coordinates; performing vector decomposition on the speed of the node; obtaining the moving direction angle of the moving node according to the vector decomposition result; estimating the distance between the neighbor node and the source node according to the Euclidean distance and the moving direction angle; setting a threshold, comparing the distance between the neighbor node and the source node with the threshold. If it is less than the set threshold, the node is used as a candidate node, otherwise it is not used as a candidate node.

7. The routing method of a high-speed mobile node based on the ST-GPSR protocol according to claim 6, characterized in that, The threshold is εR, where R is the node communication range and ε is the error adjustment parameter, and its expression is: where v max is the upper limit of the node movement speed, and t is the time interval.

8. A routing method for a high-speed mobile node based on the ST-GPSR protocol according to claim 4, characterized in that, The screening of neighbor nodes using the GPSR protocol based on the perimeter forwarding mode includes: obtaining the position coordinates of the source node, the position coordinates of the neighbor nodes, the velocity coordinates of the neighbor nodes, the position coordinates of the target node, and the velocity coordinates of the target node; calculating the moving direction of the neighbor nodes and the predicted distance between the neighbor nodes and the target node at the next moment according to the position coordinates of the source node, the position coordinates of the neighbor nodes, the velocity coordinates of the neighbor nodes, and the position coordinates of the target node; calculating the moving angle of the neighbor nodes according to the moving direction of the neighbor nodes; calculating the moving tendency degree of the neighbor nodes according to the moving angle and the predicted distance; and screening out the neighbor nodes according to the moving tendency degree to obtain the candidate node set.

9. The routing method of a high-speed mobile node based on the ST-GPSR protocol according to claim 1, wherein, The trust function includes the link connection time and the forwarding deflection angle, and its expression is: Q N = γ * t link + (1 - γ) * cosθ N Among them, Q N is a trust function, γ is a weight parameter, where 0 < γ < 1, t link is the link connection time, θ N is the forwarding deflection angle, R is the node communication range, D SN is the Euclidean distance between the candidate node N and the source node S, v r is the relative velocity vector between the candidate node N and the source node S, and φ is the angle between nodes.

Images