CN114423040A

CN114423040A - Delay tolerant network infectious routing method based on congestion control

Info

Publication number: CN114423040A
Application number: CN202210020724.7A
Authority: CN
Inventors: 郭永安; 李陈浩; 孙洪波
Original assignee: Nanjing University of Posts and Telecommunications
Current assignee: Nanjing University of Posts and Telecommunications
Priority date: 2022-01-10
Filing date: 2022-01-10
Publication date: 2022-04-29

Abstract

The invention discloses a delay tolerant network infection routing method based on congestion control, which comprises the steps of calculating the request bandwidth of a node; and calculating the weighted average value of the residual bandwidth in the path to obtain the available residual bandwidth in the link. Computing a requested bandwidth for a node, comprising: a source node sends a bandwidth request to a one-hop neighbor node; the source node calculates the request bandwidth; the method for sending the bandwidth request to the one-hop neighbor node by the source node comprises the following steps: the source node sends a Hello data packet at intervals; if the one-hop neighbor node receives the Hello data packet, the one-hop neighbor node creates an entry in a routing table of the one-hop neighbor node; and if the one-hop neighbor node does not send the control data packet to the source node within the set time interval, the source node broadcasts the Hello data packet to other one-hop neighbor nodes. The invention can greatly avoid network congestion, reduce end-to-end delay, increase throughput, improve the transmission rate of data packets and enable the transmission of information to be quicker.

Description

Delay tolerant network infectious routing method based on congestion control

Technical Field

The invention relates to a delay tolerant network infection routing method based on congestion control, and belongs to the technical field of wireless communication networks.

Background

The delay tolerant network DTN is a dynamic network with limited resources, is mainly applied to a network environment with high delay and lack of continuous connection, and is an architecture facing an unreliable network. In such networks, there is no end-to-end path available between the source node and the destination node, and network outages occur frequently. Meanwhile, the nodes in the DTN are dynamic, and depending on the mobility of the nodes, messages may be delivered to any other node within communication range of the node for the opportunity to deliver messages between the source node and the target node. The mobile node then transmits the message to the destination in a "carry-store-forward" mode, effectively solving the communication problem in the restricted network. The DTN has wide application in the scenes of vehicle-mounted sensing networks, military tactical communication networks, deep space exploration, wild animal monitoring and the like.

Compared with the conventional network, various resources such as cache, bandwidth, node processing capability and the like of the DTN network are very limited, and in order to improve the message delivery rate, the DTN usually duplicates multiple message copies of the message and sends the message copies through multiple paths respectively, so that the reliability of message delivery is ensured, and thus, the load pressure of nodes in the DTN network becomes very large. Traditional routing protocols are unable to deliver messages in challenging environments, delay tolerant networks help overcome difficulties in providing network access in challenging environments, and require different routing protocols than traditional routing protocols.

Epidemic is one of the more classical routing methods in DTN, also called infectious routing, and the routing algorithm adopts a message flooding concept. In the routing algorithm, each DTN node holds a summary vector list recording messages carried by the node, the message summary vector lists of the nodes are exchanged when the nodes meet each other, then the messages carried by the node but not carried by the node are copied to the node, the messages are rapidly diffused in the DTN, and finally all the messages are transmitted to all the nodes in the network. The Epidemic routing algorithm does not require any network state related information and may, in some extreme cases, be the only way to successfully deliver a message to a destination node. However, the Epidemic routing method generates a large number of message copies in the DTN network, which is very easy to exhaust network resources and then cause network congestion, resulting in reduced network performance.

Disclosure of Invention

The technical problem to be solved by the invention is to overcome the defects of the prior art and provide a delay tolerant network infection routing method based on congestion control.

In order to achieve the above object, the present invention provides a delay tolerant network infection routing method based on congestion control, including:

calculating the request bandwidth of the node;

and calculating the weighted average value of the residual bandwidth in the path to obtain the available residual bandwidth in the link.

Preferably, the computing node's requested bandwidth includes:

a source node sends a bandwidth request to a one-hop neighbor node;

the source node calculates the request bandwidth;

the method for sending the bandwidth request to the one-hop neighbor node by the source node comprises the following steps:

the source node sends a Hello data packet at intervals;

if the one-hop neighbor node receives the Hello data packet, the one-hop neighbor node creates an entry in a routing table of the one-hop neighbor node;

if the one-hop neighbor node does not send the control data packet to the source node within the set time interval, the source node broadcasts the Hello data packet to other one-hop neighbor nodes;

the Hello message comprises Hello INTERVAL and ALLOWED Hello LOSS, wherein the Hello INTERVAL represents the maximum time INTERVAL between two times of sending the Hello message, and the ALLOWED Hello LOSS represents the INTERVAL time of waiting for receiving the Hello message by a one-hop neighbor node under the condition that the source node is not disconnected with the one-hop neighbor node;

the control packet includes determination information that the Hello packet reaches the one-hop neighbor node.

Preferably, the source node calculates the requested bandwidth, including:

the one-hop neighbor node extracts header information from the Hello data packet, wherein the header information comprises whether the one-hop neighbor node is a destination node or not;

if the one-hop neighbor node is the destination node of the Hello data packet, the destination node updates the session cache;

the destination node sends a HELLOACK data packet to the source node, and the source node records the time T of receiving the HELLOACK data packet_r(ii) a The Hello packet includes a time T at which the Hello message is sent_s；

The source node calculates the requested bandwidth on the link between the source node and the one-hop neighbor node:

T＝T_r-T_s，

P_size＝2(P_RTS+P_CTS)P_R+P_H-Ack，

in the formula, P_sizeT is the total number of bytes of the HELLO data packet, T is the total round trip time of the source node for sending the Hello data packet and receiving the HELLOACK data packet, and T is the total round trip time of the source node for sending the Hello data packet and receiving the HELLOACK data packet_rIs the time T when the source node receives the HELLOACK packet_sIs the time when the source node sends a HELLO packet;

P_His the total number of bytes, P, of the HELLO data packet_H-AckIs the total byte number, P, of the HELLOACK data packet_RTSSend requests, P, generated separately for HELLO and HELLOACK packets_CTSTransmission permission generated separately for HELLO packets and HELLOACK packets.

Preferably, calculating a weighted average of the remaining bandwidth in the path to obtain the remaining bandwidth available in the link includes:

the source node selects a route, and broadcasts the BRREQ data packet to an intermediate node or a destination node;

the BRREQ packet includes a route to the destination node, a minimum required bandwidth sent from the source node, and a session id with a source address; if the intermediate node receives the BRREQ data packet, executing the steps 1-4:

step 1, if the intermediate node receives the BRREQ data packet, the intermediate node checks whether the BRREQ data packet has a route to a destination node;

step 2, if the route to the destination node exists, comparing the minimum required bandwidth with the residual bandwidth of the intermediate node;

step 3, if the minimum required bandwidth is smaller than the residual bandwidth of the intermediate node, the intermediate node establishes a reverse route and rebroadcasts a BRREQ data packet, and the BRREQ data packet records the minimum bandwidth value required by the next link;

and 4, executing the steps 1-3 until the BRREQ data packet reaches a destination node.

Preferably, if there are more than two paths with the same distance to the destination node, and multiple source nodes are using one of the paths to transmit the link, the source node transmits the BRREQ packet on the link with higher bandwidth.

Preferably, if the destination node receives the BRREQ data packet, the average value of the end-to-end residual bandwidth from the source node to the destination node is calculated

In the formula, t_iThe time period of the BRREQ data packet which is broadcast from the source node and is counted from 0 and received by the destination node is the time period;

is a weighted average of the remaining bandwidth, BW, of the current time of the path_Res(t_i) Is a period of time t_iThe actual value of the remaining bandwidth of (a),

weighted averaging of remaining bandwidth for a point in time on a pathThe mean, α, is the weight of the current remaining bandwidth.

Preferentially, if the destination node receives the BRREQ data packet, the total bandwidth BW consumed on the transmission path from the source node to the destination node is calculated_con：

BW_con＝HC_max×BW_1cons

In the formula (BW)_1consBandwidth value, HC, consumed by a one-hop neighbor node that is a source node_maxIs the maximum number of hops on the path;

the target node compares the total bandwidth consumed by the source node with the average value of the end-to-end residual bandwidth, and if the total bandwidth consumed by the source node is larger than the average value of the end-to-end residual bandwidth, the source node is informed to reduce the data rate to

BW_avgIs the average value of the end-to-end residual bandwidth from the source node to the destination node.

Preferentially, the destination node sends a BRREP message to the source node;

when the BRREP message passes through each intermediate node, all the intermediate nodes verify the current residual bandwidth by using the bandwidth information contained in the BRREP message;

if the current residual bandwidth of the intermediate node is less than the bandwidth information contained in the BRREP message, replacing the bandwidth information contained in the BRREP message with the current residual bandwidth of the intermediate node;

returning according to a path from a source node to a destination node, forwarding a BRREP message to the source node by each intermediate node through a reverse route, checking whether each intermediate node is the source node by the BRREP message, comparing a requested bandwidth value with an average value of end-to-end residual bandwidths fed back by the source node if the intermediate node is the source node, and adjusting the rate of the source node to be equal to the rate of the source node if the requested bandwidth value is smaller than the average value of the end-to-end residual bandwidths fed back by the source node

Each intermediate node checks the reverse path entry in the routing table if the intermediate node is in the routeIf the table has a reverse path entry, the intermediate node forwards the BRREP message according to the specified path.

Preferentially, in the process of transmitting the Hello data packet from the source node to the destination node, calculating the transmission possibility of the data packet, and sequentially discarding the data packets cached by the nodes according to the sequence from low to high of the transmission possibility, the method comprises the following steps:

calculating data packet P cached by source node or intermediate node_iThe possibility of direct delivery to the destination node is W_id，i∈[1,n]：W_id＝1-e^-λT,

Where T is the corresponding packet P_iλ is an exponential parameter, e is a constant;

calculating P_iProbability of second transmission reaching destination node W_ij：

In the formula, TC_iThe total number of copies of the data packet i is N, and the total number of nodes in the network is N;

data packet p_iPossibility of finally reaching the destination node W_iComprises the following steps:

W_i＝βW_id+γW_ij，

in which beta is W_idThe weight of the ratio, gamma is W_ijA proportional weight, β + γ ═ 1;

comparing each packet p in the buffer in the event that the buffer is not sufficiently cached to support storing new packets_iW of (2)_iDiscard W_iLowest data packet p_i。

Preferably, W is calculated based on an analytic hierarchy process_idThe ratio weights beta and W_ijThe ratio weight γ, the decision matrix is obtained:

calculating the maximum eigenvalue of the judgment matrix to be 2, and the corresponding eigenvector to be [0.1240,0.9923 ];

normalizing the judgment matrix to obtain W_idAnd W_ijWeight (β, γ) ═ 0.1111, 0.8889.

The invention achieves the following beneficial effects:

the bandwidth feedback infectious routing strategy EBFRS provided by the invention can greatly avoid network congestion, reduce end-to-end delay, increase throughput, improve the transmission rate of data packets and enable information transmission to be quicker by using the bandwidth feedback and a new packet loss method.

Drawings

FIG. 1 is a flow chart of the present invention.

Detailed Description

The following examples are only for illustrating the technical solutions of the present invention more clearly, and the protection scope of the present invention is not limited thereby.

The invention provides an epidemic routing strategy EBFRS with bandwidth feedback aiming at the problem of network congestion of an epidemic routing. In total, three stages are provided, as shown in fig. 1. In the first phase, the requested bandwidth for each node is calculated based on two steps: step one, each node in the network sends a bandwidth request to each neighbor node, and step two, the requested bandwidth is calculated locally, so that the transmission time and the cost of message transmission are saved;

in a second phase, calculating a weighted average of the residual bandwidths in the paths so as to inform the source node of the latest residual bandwidth available in the network for each link;

in the third stage, the packet with the lowest discarding possibility is selected by calculating the transmission possibility of the packet.

The first stage of the technical scheme adopted by the invention specifically comprises the following two steps:

(1) source node bandwidth request

Before starting the bandwidth calculation process, each source node finds its one-hop neighbor nodes by sending Hello packets at intervals, including:

when a one-hop neighbor node receives the Hello packet, the one-hop neighbor node creates an entry in its routing table. The control message is stored in a Hello data packet, and in order to maintain the connection with the one-hop neighbor node, if the one-hop neighbor node does not send any control data information back within a set time interval, the source node broadcasts the Hello data packet to other neighbor nodes.

The control messages include messages of the network itself, whether the network is clear, whether packets are reachable, and whether routing is available.

If the neighboring node does not receive any Hello packets within a specified time interval, it means that the neighboring node is not within the transmission range of the source node and the source node's connection to the neighboring node has been lost.

The Hello packet uses Hello INTERVAL and ALLOWED Hello LOSS to determine connectivity between the source node and its one-hop neighbor nodes. HELLO INTERVAL represents the maximum time INTERVAL between two transmissions of the HELLO packet. ALLOWED HELLO LOSS represents the interval between a one-hop neighbor node waiting to receive a HELLO packet without the source node disrupting its connection to the one-hop neighbor node. The recommended value of HELLO INTERVAL is one second and the recommended value of ALLOWED HELLO LOSS is two seconds, which means that if a source node cannot receive control data information from a one-hop neighbor node within two seconds of sending the last HELLO packet, the source node loses connection with the one-hop neighbor node.

The Hello packet is also used to compute a requested bandwidth between the source node and its one-hop neighbor node. In order to calculate the requested bandwidth on the one-hop neighbor node of each source node, the time T of sending the Hello message is recorded in the Hello data packet_s. Each source node appends a T to the Hello packet_sAnd transmits it to the one-hop neighbor node.

When the Hello packet reaches a directly connected one-hop neighbor node, the one-hop neighbor node extracts header information from the Hello packet. The header information includes information for determining whether the one-hop neighbor node is a destination node, and if the one-hop neighbor node is a destination node of the Hello packet, the destination node updates the session cache. In order to notify the source node that the Hello packet has successfully arrived at the destination node, the destination node sends an acknowledgement packet back to the source node, with a time value for the transmission also being appended to the acknowledgement packet. After that, the HELLOACK packet is transmitted from the destination node.

(2) Source node calculates requested bandwidth

Let the requested bandwidth be equal to the maximum throughput between the directly connected source node and the one-hop neighbor node. When the source node receives the HELLOACK data packet, recording the receiving time T of the HELLOACK data packet_r. Requested bandwidth BW on links between source node computation and one-hop neighbor nodes_Req，

T＝T_r-T_s，

P_sizeincluding the size of other MAC messages transmitted by the source node to the one-hop neighbor node, as shown in equation (1):

P_size＝2(P_RTS+P_CTS)+P_H+P_H-Ack (1)

in the formula (1), P_HAnd P_H-AckIndicating the size of the HELLO and HELLOACK packets, P_RTSRTS (request to Send), P, generated separately for HELLO and HELLOACK packets_CTSCTS (Clear to send, transmission permission) represented as a HELLO packet and a HELLOACK packet generated separately.

In the second stage of the technical scheme adopted by the invention, a routing Request BRREQ (Bandwidth based Route Request) packet is assumed to contain the minimum required Bandwidth sent from the source node. After the BRREQ packet arrives at the destination node, it is retransmitted from the destination node back to the source node to create a reverse route.

For an intermediate node connecting a source node and a destination node, the intermediate node transmits a data packet if the requested bandwidth of the intermediate node is less than the remaining bandwidth on the link. In this way, appropriate routes may be created to transmit BRREQ packets according to available bandwidth, thereby avoiding congestion. The method specifically comprises the following four steps:

(1) transmitting BRREQ packets from a source node

In EBFRS, routes are selected according to the source node requirements. The source node gives the minimum requested bandwidth that must be guaranteed to be reached, and this packet with the minimum requested bandwidth is called a bandwidth-oriented route request (BRREQ). The new BRREQ packet with bandwidth extension includes the session id (sid) with the source address. Each session id is unique and is used for identifying each flow, and the timer is incremented by 1 each time a new BRREQ data packet is generated. When an intermediate node receives a BRREQ packet, it creates a reverse route and rebroadcasts the BRREQ packet. The same process continues until the BRREQ packet arrives at the destination node.

(2) Receiving and looking up bandwidth requests of intermediate nodes

Each node receiving a BRREQ packet first extracts the header from the received BRREQ packet: when the intermediate node receives the BRREQ data packet, the intermediate node checks whether the BRREQ data packet has a route to the destination node. If the intermediate node has a route to the destination node, the value of the requested bandwidth is compared with the remaining bandwidth of the intermediate node. If the requested bandwidth is less than the intermediate node's remaining bandwidth, the intermediate node forwards the BRREQ packet, which may be accomplished by sending an immediate cache update message to the last node that includes a bandwidth value for the next link that is greater than the previous link.

If there are two paths to the destination node that are the same distance away and multiple source nodes are using one of the outgoing links, a BRREQ packet loss may occur on that link. In this case, the source node transmits the BRREQ packet on a link with higher bandwidth to avoid congestion.

(3) Receiving and searching bandwidth request of target node

After receiving the BRREQ packet, the node extracts the header from the packet. In this case, if the receiving node is the destination node, the following two calculations are performed. First, a weighted average of the end-to-end residual bandwidth

Calculated as shown in equation (2):

when t is_iBW when 0_Res(t_i) When t is_i>Use at 0 time

Computing

As shown in the formula (3), wherein,

a weighted average representing the remaining bandwidth on the path, time period t_iActual remaining bandwidth value of BW_Res(t_i) And (4) showing. The weighted average of the final residual bandwidth values is expressed as

α is the weight of the current residual bandwidth, which is set to 0.8 in this scenario, 1- α is used to weight the weighted average of the residual bandwidth values. Equation (3) shows that the current remaining bandwidth is allocated with a higher priority to take over the impact of the current bandwidth value.

Secondly, the destination node also needs to estimate the Bandwidth (BW) to be consumed by the source node_con) The value of (c). To countIt is most important to consider intra-stream interference, also known as mutual interference, in terms of the bandwidth consumed by the source node. The parameter in the EBFRS that calculates the in-stream interference is the Hop Count (HC). The HC is determined by measuring the distance of each node along the path from the source node to the target node. By maximum number of hops on path (HC)_max) To calculate the consumed bandwidth, BW at the target node_con＝HC_max×BW_reqWherein BW_reqIs the bandwidth requested by the source node, the destination node compares the total bandwidth consumed with the end-to-end remaining bandwidth. If the total bandwidth consumed is greater than the remaining bandwidth, the source node is notified to reduce the data rate to

(4) Bandwidth feedback from a destination node to a source node

After completing all bandwidth calculation, the destination node sends a BRREP (Bandwidth based Route reply) message to the source node. When passing through each intermediate node, all intermediate nodes verify the current remaining bandwidth with the bandwidth contained in the header. If the remaining bandwidth is less than the bandwidth given in the header, the node replaces the bandwidth value in the header with the remaining bandwidth. Each intermediate node forwards the BRREP message back to the source node via reverse routing. The process of forwarding the BRREP message on the reverse path is the same as the path of the message from the source node to the destination node.

Each intermediate node checks the reverse path entry in the routing table and if the intermediate node has a reverse path entry in the routing table, the intermediate node forwards the packet along the specified path. The BRREP message containing the feedback will check whether each intermediate node is the source node, and if so, extract all headers from the packet. The source node compares the requested reduced data rate with a feedback value sent by the destination node according to the consumed bandwidth and adjusts the data rate of the source node.

Since Epidemic is mainly concerned with discarding packets that have been discarded recently when the buffer is full, there are three ways to discard packets for the first time: discard head, discard tail and random discard. These effects, which are expressed in DTN, are not very good and the third stage of the invention proposes a new packet loss strategy depending on the possibility of delivery to the destination node.

Assuming that a node's buffer is full of n packets, from p₁To p_nThe latest packet will be stored at the end of the buffer. A data packet, or a copy thereof, is defined as a "useful packet" when it is transmitted to its destination node.

At this point the packet should be stored in a buffer defining packet p_iThe probability of the last arrival at the destination node in the buffer is W_iComprising two parts, the first part being p_iPossibility of direct delivery to destination node W_idThe second part is p_iLikelihood of forwarding to next hop and reaching destination W_ijThe "useful package" needs to satisfy at least one of the above two parts. In DTN, it is not possible to predict whether a particular packet can satisfy one of these two conditions, and it is necessary to give them a certain weight to represent the possibility of prediction, indicating that the packet is a "useful packet", and then decide which packet should be discarded.

The mutual contact time of the nodes is a time period from the time when the nodes enter the communication range of each other to the time when the nodes move out of the communication range of each other for the last time, and the mutual contact time of the node i and the node j is distributed exponentially. W_idIs calculated as shown in formula (4), by obtaining the contact time of node i and node j, the corresponding λ is obtained, and T is the corresponding data packet p_xResidual survival time of

W_id＝1-e^-λT (4)

λ can be obtained by obtaining contact time of the node i and the node j, specifically, mutual contact time of the node i and the node j is in exponential distribution, and λ is a parameter in exponential distribution; w_idThe larger the size, the more desirable it is to keep the corresponding packet in the buffer.

For calculating W_ijHow many copies of this packet of information that need to be obtained by the node are transmitted into the network. The node records the number of copies of packet x it creates, with the NC_xRepresent while at the same timeThe total number of copies TC of a packet is also recorded_xWhen two nodes meet, each node will use the NC of the other node_xUpdate its TC_x. If the encountered node does not have packet x, then after packet x is sent to the node, its TC_xTo be inherited, NC_xIs set to zero. In this way, a node can obtain information that a partial copy of each packet in its buffer has been sent into the network. Suppose that there is a particular packet x, TC in a buffer at a node_xHigher means that the likelihood of a copy of the packet being sent to the destination is lower.

W_ijThe calculation method of (2) is shown in formula (5), wherein N represents the total number of nodes, and the formula (4) and the formula (5) are combined,

in the manner shown in formula (6), wherein β and γ are W_idAnd W_ijThe weight of the ratio, β + γ, is 1. When the buffer is full, the node

W_i＝βW_id+γW_ij (6)

In the event that the buffer is not sufficient to support the storage of new data packets, the W for each packet in its buffer will be compared_iAnd discarding W_iThe lowest packet. Calculating two parameters of the ratio according to the analytic hierarchy process to obtain a judgment matrix as shown in formula (7), calculating the maximum eigenvalue to be 2, and calculating the corresponding eigenvectors to be [0.1240,0.9923]]，

And normalizing the obtained product to obtain W_idAnd W_ijThe formula (8) is obtained from the test coefficient CR with the weight (β, γ) of (0.1111,0.8889), and as can be seen from the formula (8), CR is 0<0.1, the decision matrix is considered to pass the consistency test, its normalized characteristics

The vectors may be used as weights.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

The above description is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, several modifications and variations can be made without departing from the technical principle of the present invention, and these modifications and variations should also be regarded as the protection scope of the present invention.

Claims

1. A delay tolerant network epidemic routing method based on congestion control is characterized by comprising the following steps:

calculating the request bandwidth of the node;

2. The method of claim 1, wherein the delay tolerant network infection routing method based on congestion control,

computing a requested bandwidth for a node, comprising:

a source node sends a bandwidth request to a one-hop neighbor node;

the source node calculates the request bandwidth;

the source node sends a Hello data packet at intervals;

3. The method of claim 2, wherein the source node calculates the requested bandwidth, and comprises:

T＝T_r-T_s，

P_size＝2(P_RTS+P_CTS)+P_H+P_H-Ack，

4. The method of claim 1, wherein the delay tolerant network infection routing method based on congestion control,

calculating a weighted average of the residual bandwidth in the path to obtain the residual bandwidth available in the link, including:

5. The method of claim 4, wherein the delay tolerant network infection routing method based on congestion control,

if more than two paths which reach the destination node and have the same distance exist, and a plurality of source nodes use one path to transmit a link, the source nodes transmit BRREQ data packets on the link with higher bandwidth.

6. The method of claim 4, wherein the delay tolerant network infection routing method based on congestion control,

if the destination node receives the BRREQ data packet, calculating the average value of the end-to-end residual bandwidth from the source node to the destination node

alpha is the weight of the current remaining bandwidth, which is a weighted average of the remaining bandwidth at a point in time on the path.

7. The method of claim 6, wherein if the BRREQ packet is received by the destination node, the method calculates the total bandwidth BW consumed on the transmission path from the source node to the destination node_con：

BW_con＝HC_max×BW_1cons，

8. The method according to claim 7, wherein the destination node sends a BRREP message to the source node;

Each intermediate node checks the reverse path item in the routing table, and if the intermediate node has a reverse path entry in the routing table, the intermediate node forwards the BRREP message according to the specified path.

9. The method according to claim 1, wherein in the process of transmitting Hello packets from a source node to a destination node, the transmission probability of the packets is calculated, and the packets cached by the nodes are discarded in sequence from low to high according to the transmission probability, the method comprises:

calculating data packet P cached by source node or intermediate node_iThe possibility of direct delivery to the destination node is W_id，i∈[1,n]：

W_id＝1-e^-λT,

W_i＝βW_id+γW_ij，

10. The method of claim 9, wherein the W is calculated based on an analytic hierarchy process_idThe ratio weights beta and W_ijThe ratio weight γ, the decision matrix is obtained: