CN111818465B

CN111818465B - A car networking adaptive multi-hop broadcast system and method

Info

Publication number: CN111818465B
Application number: CN202010639936.4A
Authority: CN
Inventors: 毕远国; 安涛
Original assignee: Northeastern University China
Current assignee: Northeastern University China
Priority date: 2020-07-06
Filing date: 2020-07-06
Publication date: 2021-07-23
Anticipated expiration: 2040-07-06
Also published as: CN111818465A

Abstract

The invention belongs to the technical field of Internet of Vehicles information interaction, and discloses an adaptive multi-hop broadcast system and method for the Internet of Vehicles. Basic module. Filter the neighbor nodes that have been broadcasted according to the information maintained in the management information base, then select the appropriate next-hop candidate forwarding node from the remaining neighbor nodes, and finally encapsulate the selected candidate node information and all its neighbor information into the report. The message is broadcast, and the neighbor node receives the message and makes a corresponding forwarding judgment. In addition, the present invention divides IoV application messages into two categories: emergency messages and non-emergency messages, and designs two different multi-hop broadcast mechanisms for the two types of different application messages, namely fast broadcast mechanism and cooperative broadcast mechanism. The problems of high transmission delay, large network load and low reliability of the existing multi-hop broadcast strategy are improved.

Description

Internet of vehicles self-adaptive multi-hop broadcasting system and method

Technical Field

The invention belongs to the technical field of vehicle networking information interaction, and relates to an Adaptive Multi-hop Broadcast (AMHB) system and method based on a Wireless Access in Vehicular Environment (WAVE) protocol stack.

Background

With the continuous development of the automobile industry, the holding quantity of automobiles per capita is also rapidly increasing. Traffic jam in cities and highways, road safety in severe weather and traffic accidents gradually become common concerns of the whole society. By carrying out safe and effective multi-hop transmission on the Internet of vehicles information, the traffic information is quickly diffused into the possibly affected area, and the current traffic environment can be effectively improved. However, on one hand, due to the limitations of the WAVE protocol stack itself, the standard WAVE protocol stack does not forward information, and on the other hand, due to the complexity of the car networking environment, it has been a research hotspot and difficulty in this field to design a multi-hop broadcasting method that can well adapt to the current car networking environment.

The existing multi-hop broadcasting methods proposed for the internet of vehicles can be mainly divided into two categories, namely a multi-hop broadcasting method based on a sending end and a multi-hop broadcasting method based on a receiving end. The two methods are mainly different in the sender of the forwarding decision, wherein the basic idea of the method based on the sending end is to screen the relay node by the sending node and the receiving node is to transmit passively, while the method based on the receiving end means that the sending node does not select the relay node but the receiving node makes the forwarding decision. The two methods have advantages and disadvantages, and mainly show three aspects of high transmission delay, large network load and low reliability.

Disclosure of Invention

In view of the above problems, an object of the present invention is to provide a self-adaptive multi-hop broadcasting method suitable for an internet of vehicles environment, which can effectively reduce transmission delay and network load while ensuring reliability.

The AMHB method provided by the invention is structurally divided into five basic modules, namely a local information interaction module, a backoff waiting module, a relay screening module, a data forwarding module and an information maintenance module. The local information interaction module realizes information exchange among vehicles; a new back-off time calculation model is provided in the back-off module, so that information collision can be effectively reduced; a screening algorithm of the candidate nodes is provided in the relay screening module; the data forwarding module is responsible for data encapsulation and data forwarding; the information maintenance module realizes data support and data maintenance for other modules. The basic idea of the AMHB is to filter the broadcasted neighbor nodes according to the Information maintained in a Management Information Base (MIB), then select a suitable next hop candidate forwarding node from the remaining neighbor nodes, finally encapsulate the selected candidate node Information and all the neighbor Information of itself into a message and broadcast the message, and the neighbor nodes receive the message and make corresponding forwarding judgment. In addition, the invention divides the application information of the Internet of vehicles into two categories of emergency information and non-emergency information, and designs two different multi-hop broadcast mechanisms aiming at the two different types of application information, namely a quick broadcast mechanism and a cooperative broadcast mechanism.

The local information interaction module: the module is responsible for finishing data interaction work between each node and neighbor nodes in the Internet of vehicles. And the data interacted between the neighbor nodes is used as the theoretical basis of other modules and is dynamically maintained in the information maintenance module.

A back-off waiting module: the invention designs a segmented backoff waiting mechanism aiming at two conditions which can cause information conflict, wherein one condition is a plurality of Next-hop Candidate nodes (NCNs) screened by the same Node, the NCNs which are mutually neighbors can forward data at the same time, and the other condition is a plurality of NCNs screened by different nodes, the NCNs which are mutually neighbors can forward data at the same time. Aiming at the first situation, the invention provides a Back-off Wait on Priority Queue (BW-PQ) algorithm based on a Priority Queue, wherein the BW-PQ algorithm is provided on the basis of research on the existing Back-off Wait algorithm, and the BW-PQ algorithm considers that the basic idea of most existing Back-off Wait algorithms is to randomly Wait for a time, on one hand, when the number of candidate nodes is large, the random waiting method still can cause a plurality of candidate nodes to simultaneously send messages, thereby causing unnecessary information collision. On the other hand, the random waiting method may also cause a node with a lower priority to forward preferentially, thereby causing problems of reduced broadcast efficiency and increased transmission delay. The BW-PQ algorithm has the advantages that a corresponding priority queue can be generated according to the selection condition of the candidate nodes, a specific back-off time is assigned to each candidate node in the queue, and the waiting time of the node with higher priority is shorter, so that each candidate node can share the channel resource in a specific time period, the node with higher priority can be preferentially forwarded, and the data transmission efficiency is improved. The BW-PQ algorithm includes two steps of generating a priority queue and calculating a back-off time.

The method comprises the following steps: a priority queue is generated. The transmitting node inserts the screened NCN information into a Priority Queue (PQ), and then encapsulates the PQ into an MH-WSA message. Each neighbor node can obtain complete PQ information after receiving the MH-WSA message, then sequentially dequeues the NCN information in the PQ, compares the NCN information with the information of the neighbor node to determine whether the neighbor node is the NCN or not, and calculates the backoff waiting time of the neighbor node if the neighbor node is the NCN.

Step two: the back-off time is calculated. The priority queue is composed of NCN information, and only the neighbor node selected as the NCN enters a backoff waiting stage to calculate the own backoff time. The back-off time is calculated as shown in equation (1).

T＝T_max×(s-1) (1)

Where s denotes the dequeue order, T_maxRepresenting a standard one-hop maximum delay time.

For the second case, the present invention uses a default binary exponential backoff algorithm in IEEE 802.11p to perform backoff wait.

A relay node screening module: the selection of the relay node is the most core part in the invention, and an excellent relay node selection algorithm can effectively reduce the network load, avoid the broadcast storm problem and improve the data transmission efficiency. The module provides a Relay Node screening (RNS-RR) Algorithm based on Request Response, and the Algorithm is divided into a Request phase and a Response phase and is respectively responsible for NCN Selection and Relay Node (RN) Selection.

The method comprises the following steps: and selecting the NCN. Each node in the internet of vehicles maintains a respective neighbor list, and the neighbor list stores information such as position, speed, direction, signal strength and the like of neighbor nodes. Therefore, when selecting NCN, the sending node primarily calculates the priority of each neighbor node first, and assumes that the priority of the neighbor node i is denoted as p_i，i∈[1，n]N is the number of neighbor nodes, then p_iCan be obtained from equation (2).

Wherein D is_iThe distance between the neighbor node i and the sending node can be obtained by formula (3), wherein R is a one-hop communication range, V_iThe relative speed of the neighbor node i and the sending node can be obtained by the formula (4),V_maxthe maximum relative velocity, RSSI, that can be theoretically achieved_iThe signal intensity of the neighbor node i, and alpha, beta and gamma are distances D respectively_iRelative velocity V_iWeighted RSSI of sum signal strength_iAnd α + β + γ ═ 1.

D_i＝R′·arccos(cos(x)·cos(x_i)·cos(y_i-y)+sin(x)sin(x_i)) (3)

Wherein R' is the radius of the earth, x and y are longitude and latitude information of the sending node respectively, and x_i，y_iRespectively longitude and latitude information of the neighbor node i.

Where v is the velocity of the transmitting node, v_iAnd theta is the speed of the neighbor node i, and theta is the included angle between the driving direction of the sending node and the neighbor node i.

After the priorities of all the neighbor nodes are obtained, the priority threshold value threshold is calculated according to the formula (5).

Priority satisfies condition p in neighbor node_iNodes that are > threshold will be selected as NCN. Only the neighbor node selected as NCN is likely to become RN.

Step two: and selecting the RN. The neighbor node selected as the NCN enters a backoff waiting stage after receiving the message, the NCN counts repeated MH-WSA messages received in the backoff time, and stores neighbor node information of other NCNs carried in the message into a repeated MH-WSA data table of the NCN. After the back-off time is over, the NCN compares the own neighbor list with the repeated MH-WSA data table, and judges whether the neighbor node only belongs to the own neighbor list exists or not. If the area exists, the NCN considers that the area which is not broadcasted still exists in the communication range of the NCN, and at the moment, the NCN forwards the message to become the RN. If not, the NCN considers that all neighbor nodes in the current communication range of the NCN have received the broadcast message, and then the NCN gives up forwarding data.

A data forwarding module: the module is responsible for encapsulating and forwarding two messages, namely a Multi-hop WSA (MH-WSA) message and a Multi-hop WSM (MH-WSM) message. In order to ensure the compatibility with the standard WAVE protocol stack, the module encapsulates the fields related to the AMHB algorithm into the extension fields of the corresponding message.

An information maintenance module: the module is responsible for data maintenance work in the AMHB system and provides data support for other modules, and mainly comprises maintenance of a neighbor list, maintenance of a repeated MH-WSA data table and maintenance of a repeated MH-WSM data table.

Two message types: the invention divides the application messages into two categories of urgent messages and non-urgent messages according to the priority of the application messages. The priority of the application message is specified by an application layer, the value range of the priority is 1-8, and the smaller the value is, the higher the priority of the message is. Messages with priority < 4 are referred to as urgent messages, and such messages generally relate to personal safety, have high requirements on time efficiency and have small data volume; messages with priority >4 are called non-urgent messages, and are mainly used for improving the driving experience of users, and have relatively low requirements on timeliness and large data volume.

Two forwarding modes: corresponding to the two message types are two different forwarding modes, namely a fast forwarding mode and a cooperative forwarding mode. The fast forwarding mode is a forwarding mode designed for emergency messages, the mode only relates to the forwarding of MH-WSA messages, and broadcast data are carried by the MH-WSA messages; the cooperative forwarding mode is a forwarding mode designed for non-emergency messages, and relates to the forwarding of MH-WSA messages and MH-WSM messages, and broadcast data are carried by the MH-WSM messages.

Aiming at the problems of high transmission delay, large network load and low reliability of the existing multi-hop broadcasting strategy, the method of the invention designs a flexible and efficient relay node screening algorithm, provides a response backoff mechanism, improves the problems and has important reference value for the research of the multi-hop broadcasting technology in the internet of vehicles.

Drawings

Fig. 1 is a basic flow diagram of AMHB.

Fig. 2 is an overall architecture design diagram of the AMHB.

Fig. 3 is a schematic diagram of a Hello message sending flow.

Fig. 4 is a schematic diagram of a Hello packet receiving process.

Fig. 5 is a schematic diagram of MH-WSA message encapsulation flow.

Fig. 6 is a schematic diagram of MH-WSA message forwarding flow.

Fig. 7 is a schematic diagram of MH-WSM message encapsulation flow.

Fig. 8 is a schematic diagram of MH-WSM message forwarding flow.

Detailed Description

The following detailed description of embodiments of the invention refers to the accompanying drawings.

In the method of the embodiment, the software environment is Ubuntu 16.04, and the simulation environment is NS-3.

As shown in FIG. 1, the basic flow of the AMHB method designed by the invention, the normal operation of the AMHB algorithm is established on the basis of information interaction of each node in the Internet of vehicles. By maintaining the information of the neighbor nodes in the MIB, each node in the Internet of vehicles can clearly know the current network condition. Firstly, a certain number of neighbor nodes are selected as NCNs by a sending end node, and only the nodes selected as the NCNs have the permission to transmit MH-WSA messages and MH-WSM messages. Then the NCN node determines whether to execute the authority according to the forwarding condition of the current message, and finally only the NCN which finishes the action of forwarding the message is called as RN. The local information interaction runs through the whole operation period of the equipment and is not dependent on whether a multi-hop broadcast service request exists in the current Internet of vehicles or not. The non-broadcast area refers to the neighbor nodes which have not received the multi-hop broadcast message in the current NCN communication range.

An application method of an adaptive multi-hop broadcasting system of the Internet of vehicles comprises the following steps:

first, according to maintenance in the management information baseSelecting candidate nodes by the information in the step (1); each node in the Internet of vehicles maintains a respective neighbor list, and the neighbor list stores the position, speed, direction and signal strength information of neighbor nodes; when a sending node selects a candidate node, the priority of each neighbor node is mainly calculated at first, and the priority of a neighbor node i is set as p_i，i∈[1，n]N is the number of neighbor nodes, then p_iThe result is obtained by formula (2);

wherein D is_iThe distance between the neighbor node i and the sending node is obtained by formula (3), R is a one-hop communication range, V_iThe relative speed of the neighbor node i and the sending node is obtained by the formula (4), V_maxThe maximum relative velocity, RSSI, that can be theoretically achieved_iThe signal intensity of the neighbor node i, and alpha, beta and gamma are distances D respectively_iRelative velocity V_iWeighted RSSI of sum signal strength_iAnd α + β + γ ═ 1;

D_i＝R′·arccos(cos(x)·cos(x_i)·cos(y_i-y)+sin(x)sin(x_i)) (3)

wherein R' is the radius of the earth, x and y are longitude and latitude information of the sending node respectively, and x_i，y_iRespectively longitude and latitude information of a neighbor node i;

where v is the velocity of the transmitting node_iThe speed of the neighbor node i is shown, and theta is an included angle between the driving direction of the sending node and the driving direction of the neighbor node i;

after the priorities of all the neighbor nodes are obtained, calculating a priority threshold according to a formula (5);

priority satisfies condition p in neighbor node_iNodes which are more than or equal to threshold are selected as candidate nodes; only the neighbor node selected as the candidate node is possible to become the relay node;

secondly, generating a priority queue; the sending node inserts the candidate node information screened in the first step into a priority queue, and then packages the priority queue into an MH-WSA message; each neighbor node obtains complete priority queue information after receiving the MH-WSA message, then sequentially dequeues candidate node information in the priority queue, compares the candidate node information with the self information, determines whether the neighbor node is a candidate node, calculates the backoff waiting time of the neighbor node if the neighbor node is the candidate node, and abandons forwarding if the neighbor node is not the candidate node;

thirdly, selecting a relay node; the candidate node counts repeated MH-WSA messages received in the back-off time, and stores neighbor node information of other candidate nodes carried in the messages into a repeated MH-WSA data table of the candidate node; after the back-off time is over, the candidate node compares the own neighbor list with the repeated MH-WSA data table, and judges whether the neighbor node only belongs to the own neighbor list exists in the own neighbor list or not; if the message exists, the candidate node considers that the area which is not broadcasted still exists in the communication range of the candidate node, and at the moment, the candidate node forwards the message to become a relay node; if the broadcast message does not exist, the candidate node considers that all neighbor nodes in the current own communication range have received the broadcast message, and at the moment, the candidate node gives up forwarding data.

Fig. 2 is an overall architecture design diagram of the AMHB algorithm, and in consideration of compatibility of the AMHB algorithm with a standard WAVE protocol stack, an information interaction module is designed in an application layer and used as a new car networking application to complete data interaction between neighboring nodes. And the information maintenance module is used as a part of the MIB and is responsible for completing maintenance work on the neighbor list, the repeated MH-WSA data table and the repeated MH-WSM data table. The BW-PQ backoff algorithm and the screening algorithm of the relay node are realized in the WSMP protocol flow. The data forwarding module and the BEB back-off algorithm are implemented in the LLC layer and the MAC layer, respectively.

Fig. 3 and fig. 4 are respectively a flow of sending and receiving a Hello message in a local information interaction process. When the sending node filters the NCN, the sending node needs to know the current position, speed, direction, signal strength and other information of all the neighbor nodes, and the purpose of local information interaction is to complete the information collection of the neighbor nodes, and store the collected neighbor node information into the neighbor list after performing corresponding processing. The Hello message adopts a periodic broadcast mode, and the broadcast period can be adjusted by applying parameters. In addition, since the information interaction is a spontaneous action, in order to avoid unnecessary interaction delay, the sending of the Hello message does not specify a specific channel, that is, the Hello message can be sent no matter in the SCH time slot or the CCH time slot, so that the condition of data failure caused by waiting for channel switching can be reduced.

Fig. 5 and 6 are respectively a flow of MH-WSA packet encapsulation and forwarding, where in forwarding control information carried by an MH-WSA packet, a value of an FCF field is specified by an application layer and is mainly used to identify a packet type and control a forwarding hop count, a Change count field in a source MAC and a packet is used to determine whether a current MH-WSA packet is a duplicate packet, a Hello _ Number and a Candidate _ Number are respectively used to indicate the Number of neighbor nodes of a sending node of the current MH-WSA packet and the Number of selected NCNs, the neighbor MAC sequence refers to a MAC sequence of a neighbor node of the sending node of the current MH-WSA packet, and the previous Candidate _ Number in the sequence is an NCN. These pieces of information are encapsulated in the header extension section of a standard WSA message. The MH-WSA message is responsible for informing all neighbor nodes of NCN information screened by the sending node in a fast forwarding mechanism or a cooperative forwarding mechanism, and the MH-WSA message is also responsible for carrying service data in the fast forwarding mechanism, so that the MH-WSA message is of great importance in the AMHB algorithm for forwarding.

Fig. 7 and 8 are respectively an encapsulation and forwarding flow of an MH-WSM message, where forwarding control information is encapsulated in a header extension field of a standard WSM message, and fields FCF and FEI are specified by an application layer. Besides carrying service data, the MH-WSM message carries an independent forwarding control field inside, and is used for forwarding judgment of response of candidate forwarding nodes. In addition, the MH-WSM message forwarding must comply with the interaction model of the standard WAVE protocol stack, and before forwarding the MH-WSM message, the corresponding MH-WSA message must be forwarded.

The neighbor list is a data table for storing information such as the position, speed, direction, and signal strength of the neighbor node. The data stored in the neighbor list will be used as screening basis and forwarding judgment basis for the NCN. Due to the high-speed mobility of the nodes, the network environments of the nodes at each moment are different, and the neighbor nodes in the communication range are different. Therefore, maintenance of the neighbor list is important. And the overdue data in the neighbor list is deleted in time through a dynamic maintenance means, so that the condition that the neighbor list is continuously enlarged and the efficiency of the AMHB algorithm is influenced is avoided. The maintenance of the neighbor list runs through the whole operation cycle of the node, and the maintenance process mainly comprises the following three aspects:

1) maintenance is performed when data is added or updated: when receiving the Hello message, the node firstly inquires whether a node identified by the Hello message exists in a neighbor list, and deletes outdated data in the inquiry process.

2) Maintenance is performed when calculating the neighbor priority: when a multi-hop broadcast service request is generated, the node traverses the neighbor list and calculates the priority, and the expired data is deleted in the traversing process.

3) Periodic maintenance: at intervals to traverse the neighbor list and delete stale data. Because the neighbor list may be huge, a certain time is required for the system to traverse the list, and the maintenance period is not set to be too short in order to avoid system deadlock. In addition, in order to avoid the excessively large neighbor list, the maintenance period is not set to be excessively large.

The repeated MH-WSA data table is used for storing neighbor MAC sequences carried in repeated MH-WSA messages received by the NCN node within the backoff time in a multi-hop broadcast service. In order to improve the stability of broadcast transmission, the AMHB algorithm proposed herein ensures that each node in an area can receive a broadcast packet by selecting multiple NCNs. This results in that an NCN may receive MH-WSA messages from other NCNs while waiting for forwarding, nodes corresponding to neighbor MAC sequences in the messages have received multi-hop broadcast data, and the NCN determines whether there is an area that has not been broadcast in the current communication range by recording these data and comparing with its own neighbor list at the end of the waiting time. The repeated MH-WSA data table stores the node information carried by the repeated MH-WSA message received by the forwarding node within the back-off time and sent by the neighbor node. The data is only effective in the multi-hop broadcasting process, and when the node finishes forwarding judgment, the data in the table is emptied.

The repeated MH-WSM data table is mainly used for identifying MH-WSM messages received in a multi-hop broadcasting process and avoiding the repeated reception of the same data. And the repeated MH-WSM data table only identifies the receiving condition of the message and does not store the data. The maintenance of the repeated MH-WSM data table relates to each node in the Internet of vehicles, a timer is adopted for maintenance, the timer is refreshed every time the node receives an MH-WSM message with the same MAC and psid, and if the timer is overtime, all bits of corresponding bitmap in the repeated MH-WSM data table are set to be 0.

Claims

1. An adaptive multi-hop broadcast system for the Internet of Vehicles, characterized in that it comprises a local information interaction module, a backoff waiting module, a relay screening module, a data forwarding module and an information maintenance module;

Local information interaction module: used to complete the data interaction between each node and its neighbor nodes in the Internet of Vehicles. The data exchanged between neighbor nodes is used as the theoretical basis for other modules, and is dynamically maintained in the information maintenance module;

Back-off waiting module: It is used to generate a corresponding priority queue according to the selection of candidate nodes, and specify a specific back-off time for each candidate node in the queue. A candidate node can exclusively share channel resources within a specific period of time, and ensure that nodes with high priority can be forwarded first;

Relay node screening module: It is used to screen candidate nodes and relay nodes respectively according to the two stages of request and response; specifically:

Selection of candidate nodes: Each node in the Internet of Vehicles maintains its own neighbor list, and the neighbor list stores information such as the location, speed, direction, and signal strength of neighbor nodes; therefore, when the sending node selects a candidate node, First, the priority of each neighbor node is mainly calculated, and the priority of neighbor node i is denoted as p _i , i∈[1,n], n is the number of neighbor nodes, then p _i can be obtained by formula (2);

Among them, D _i is the distance between the neighbor node i and the sending node, obtained by formula (3), R is the one-hop communication range, and V _i is the relative speed between the neighbor node i and the sending node, obtained by formula (4) , V _max is the theoretically achievable maximum relative velocity, RSSI _i is the signal strength of the neighbor node i, α, β, γ are the weight RSSI _i of the distance Di, the relative velocity _Vi and the signal strength _, respectively, and α+β +γ=1;

D _i =R'·arccos(cos(x)·cos(x _i )·cos(y _i -y)+sin(x)sin(x _i )) (3)

Among them, R' is the radius of the earth, x, y are the longitude and latitude information of the sending node, respectively, x _i , y _i are the longitude and latitude information of the neighbor node i respectively;

Among them, v is the speed of the sending node, v _i is the speed of the neighbor node i, and θ is the angle between the sending node and the driving direction of the neighbor node i;

After obtaining the priorities of all neighbor nodes, calculate the priority threshold threshold according to formula (5);

The node whose priority satisfies the condition p _i ≥threshold among the neighbor nodes will be selected as the candidate node; only the neighbor node selected as the candidate node can become the relay node;

Selection of relay nodes: Neighbor nodes selected as candidate nodes will enter the back-off waiting stage after receiving the message. The neighbor node information of other candidate nodes carried is stored in its own repeated MH-WSA data table; after the back-off time expires, the candidate node compares its neighbor list with the repeated MH-WSA data table to determine whether it is in its neighbor list. There is a neighbor node that only belongs to itself; if it exists, the candidate node considers that there is still an area that has not been broadcast within its communication range. At this time, the candidate node forwards the message and becomes a relay node; if it does not exist, the candidate node considers that the current All neighbor nodes within its communication range have received the broadcast message. At this time, the candidate node gives up forwarding data;

Data forwarding module: It is used to encapsulate and forward the multi-hop WSA message and multi-hop WSM message; in order to ensure the compatibility with the standard WAVE protocol stack, this module will be related to the adaptive multi-hop broadcast algorithm. The field is encapsulated into the extension field of the corresponding message;

Information maintenance module: This module is responsible for data maintenance in the adaptive multi-hop broadcast system, and provides data support for other modules, including the maintenance of neighbor lists, the maintenance of repeated MH-WSA data tables, and the maintenance of repeated MH-WSM data tables.

2. the application method of a kind of vehicle networking adaptive multi-hop broadcast system according to claim 1, is characterized in that, comprises the steps as follows:

The first step is to select candidate nodes according to the information maintained in the management information base; each node in the Internet of Vehicles maintains its own neighbor list, and the neighbor list stores the location, speed, direction and signal of the neighbor nodes. Strength information; when the sending node selects a candidate node, it firstly calculates the priority of each neighbor node, set the priority of neighbor node i as p _i , i∈[1,n], n is the number of neighbor nodes, then p _i is obtained by formula (2);

D _i =R'·arccos(cos(x)·cos(x _i )·cos(y _i -y)+sin(x)sin(x _i )) (3)

The second step is to generate a priority queue; the sending node inserts the candidate node information screened in the first step into the priority queue, and then encapsulates the priority queue into the MH-WSA message; each neighbor node receives the MH-WSA message After obtaining the complete priority queue information, and then dequeue the candidate node information in the priority queue in turn, and compare it with its own information to determine whether it is a candidate node, if so, calculate its own backoff waiting time, if not, give up forwarding;

The third step is the selection of relay nodes; the candidate node counts the repeated MH-WSA messages received within the back-off time, and stores the neighbor node information of other candidate nodes carried in the message into its own repeated MH-WSA In the data table; after the back-off time expires, the candidate node compares its neighbor list with the duplicate MH-WSA data table to determine whether there is a neighbor node that only belongs to itself in its neighbor list; if there is, the candidate node considers its own neighbor node. There is still an area that has not been broadcasted within the communication range. At this time, the candidate node forwards the message and becomes a relay node; if it does not exist, the candidate node considers that all neighbor nodes within its current communication range have received the broadcast message. , the candidate node gives up forwarding data.

3. method according to claim 2 is characterized in that, the back-off time calculation mode in the second step is as follows:

T=T _max ×(s-1) (1)

Among them, s represents the dequeuing order, and T _max represents the standard maximum delay time of one hop.

4. The method according to claim 2, characterized in that, after the relay node is selected in the third step, the application messages are divided into two categories: emergency messages and non-emergency messages according to different priority levels of the application messages; The priority is specified by the application layer. The value of the priority ranges from 1 to 8. The smaller the value, the higher the priority of the message; the message with priority ≤ 4 is called emergency message, which usually involves personal safety. The requirements for timeliness are high and the amount of data is small; messages with priority > 4 are called non-urgent messages. These messages are mainly used to improve the user's driving experience. The requirements for timeliness are relatively low, and usually The amount of data is large; corresponding to the two message types are two different forwarding modes, namely fast forwarding mode and cooperative forwarding mode; fast forwarding mode is a forwarding method for emergency messages, and this mode only involves MH-WSA. For message forwarding, broadcast data is carried by MH-WSA messages; cooperative forwarding mode is a forwarding method for non-emergency messages. This mode involves the forwarding of MH-WSA and MH-WSM messages, and broadcast data is carried by MH-WSM. message carrying.