CN110446184B - Multi-mode switching Internet of vehicles routing method - Google Patents

Multi-mode switching Internet of vehicles routing method Download PDF

Info

Publication number
CN110446184B
CN110446184B CN201910687760.7A CN201910687760A CN110446184B CN 110446184 B CN110446184 B CN 110446184B CN 201910687760 A CN201910687760 A CN 201910687760A CN 110446184 B CN110446184 B CN 110446184B
Authority
CN
China
Prior art keywords
rsu
node
mode
data packet
time
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201910687760.7A
Other languages
Chinese (zh)
Other versions
CN110446184A (en
Inventor
胡斌杰
林霖
温清机
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
South China University of Technology SCUT
Original Assignee
South China University of Technology SCUT
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by South China University of Technology SCUT filed Critical South China University of Technology SCUT
Priority to CN201910687760.7A priority Critical patent/CN110446184B/en
Publication of CN110446184A publication Critical patent/CN110446184A/en
Application granted granted Critical
Publication of CN110446184B publication Critical patent/CN110446184B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W28/00—Network traffic management; Network resource management
    • H04W28/02—Traffic management, e.g. flow control or congestion control
    • H04W28/08—Load balancing or load distribution
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/30—Services specially adapted for particular environments, situations or purposes
    • H04W4/40—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P]
    • H04W4/44—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P] for communication between vehicles and infrastructures, e.g. vehicle-to-cloud [V2C] or vehicle-to-home [V2H]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W40/00—Communication routing or communication path finding
    • H04W40/02—Communication route or path selection, e.g. power-based or shortest path routing
    • H04W40/04—Communication route or path selection, e.g. power-based or shortest path routing based on wireless node resources

Abstract

The invention discloses a multi-mode switching routing method for Internet of vehicles, which comprises the steps that RSU nodes carry out switching judgment according to the sending condition of data packets of the RSU nodes and the using condition of a buffer stack, and switching is carried out between a V2I2V greedy forwarding mode, a 'prediction + copy transmission' mode and a load balancing mode. The invention utilizes the RSU wired network to assist in delivering the inter-vehicle multi-hop messages, and performs load balancing according to the residual survival time of the data packets and the estimated waiting time in the RSU message waiting queue, so that the data packets originally gathered in the heavy-load RSU are distributed to each light-load RSU, the network pressure of the heavy-load RSU is reduced, and the overtime packet loss caused by long-time waiting of the data packets in the message waiting queue of the heavy-load RSU is avoided; meanwhile, three working modes are designed for the RSU, so that the routing protocol can adapt to the working environment with low vehicle node density and high vehicle node density, the end-to-end time delay of the Internet of vehicles routing is reduced, and the delivery rate is improved.

Description

Multi-mode switching Internet of vehicles routing method
Technical Field
The invention relates to the technical field of Internet of vehicles communication, in particular to a multi-mode switching Internet of vehicles routing method.
Background
The car networking technology is to realize the all-round network connection of the car, the car and the person, the car and the car, the car and the road, the car and the service platform by means of the new generation information and communication technology, to improve the intelligent level and the automatic driving ability of the car, to build the new state of the car and the traffic service, thereby improving the traffic efficiency, improving the driving feeling of the car, and providing the intelligent, comfortable, safe, energy-saving and efficient comprehensive service for the user. Road Side Units (RSUs) are an important infrastructure in the car networking, and not only are physical interfaces for connecting the car networking with an external network, but also wired and wireless hybrid heterogeneous network links are provided for long-distance, high-speed and large-capacity V2V (Vehicle to Vehicle) communication.
The routing technology is the key for realizing reliable multi-hop transmission in the vehicle networking communication, a routing protocol based on the road side units can assist remote multi-hop transmission by utilizing a wired network formed by the road side units, and meanwhile, the peripheral node information collected by the road side units can also provide a basis for routing judgment. Currently, a typical road side unit-based routing protocol in the internet of vehicles is V2I2V and the like.
The routing mode of V2I2V is divided into three steps: (1) the message source vehicle node forwards the data packet to a roadside unit inlet node closest to the message source vehicle node through multi-hop wireless communication; (2) the road side unit entrance node forwards the data packet to a road side unit exit node closest to the message destination vehicle node; (3) and the exit node of the road side unit forwards the data packet to a message destination vehicle node through multi-hop wireless communication, and the delivery is finished. However, since the vehicle nodes are unevenly distributed on the road, when high-density traffic flows are formed near some road side units, the communication load of the road side units is increased, so that the risk of overtime packet loss caused by overlong waiting time of data packets in the message waiting queue is increased; when the traffic density near some road side units is extremely low, routing topology holes are easily generated near the RSUs, the local maximum problem is caused, the delivery rate is low, and the time delay is high.
Therefore, a plurality of routing working modes are designed for the RSU node, so that the RSU node can better adapt to the characteristics of large and frequent vehicle density change in the vehicle networking environment, and the RSU node can still keep lower end-to-end time delay and higher delivery success rate under the scene of extremely high or extremely low vehicle node density.
Disclosure of Invention
The invention aims to overcome the defects of low delivery success rate and large end-to-end time delay caused by extremely high or extremely low vehicle node density in the existing vehicle networking technology, and provides a multi-mode switching vehicle networking routing method.
The purpose of the invention can be achieved by adopting the following technical scheme:
a multi-mode switching Internet of vehicles routing method comprises the following steps:
s1, the message source node generates a data packet, an RSU node closest to the message source node is set as a first-stage delivery target and is marked as an RSU entrance node, and the data packet is forwarded to the RSU entrance node through multi-hop V2V communication;
s2, the RSU entrance node judges whether the RSU entrance node is in a load balancing mode, if so, the step S3 is executed; otherwise, go to step S4;
s3, when the RSU entrance node is in the load balancing mode, executing the following steps:
s3.1, deleting a scheme containing RSU nodes in a 'prediction + copy transmission' mode from the RSU entry and exit scheme set phi by the central server;
s3.2, the central server predicts the time duration TimeThrough (i, j) of the data packet passing through the RSU network, wherein the variable represents the time duration required by the data packet to pass through the RSU network when the RSU with the number of i is used as an inlet and the RSU with the number of j is used as an outlet;
s3.3, the central server checks whether TimeThrough (i, j) meets the estimated passing time constraint, the constraint is used for judging whether the data packet has enough remaining life time to pass through the RSU wired network:
TimeThrough(i,j)≤α×TimeRest
the main reason for designing the coefficient alpha is that the data packet may reach the destination node only after passing through the RSU wired network and possibly through multi-hop V2V communication, so that time needs to be reserved for the multi-hop transmission when the decision is made.
If not, the step S3.4 is carried out, and if yes, the step S3.5 is carried out;
s3.4, the central server deletes the scheme phi (i, j) which does not meet the constraint from the RSU entry and exit scheme set phi, wherein the phi (i, j) represents the scheme which takes the RSU with the number of i as an entry and the RSU with the number of j as an exit in the scheme set phi, selects a new group of schemes from the rest RSU entry and exit scheme sets phi, and executes the step S3.2 again;
s3.5, the central server informs the RSU entrance nodes of phi (i, j) passing the constraint, and then the data packet enters the RSU network according to the specified RSU entrance nodes;
s3.6, the RSU entrance node forwards the data packet to the designated RSU exit node, and then the step S5 is carried out;
s4, forwarding the data packet to the RSU exit node closest to the destination node by the RSU entrance node according to a V2I2V greedy forwarding mode;
s5, the RSU exit judges whether the RSU exit is in a 'prediction + copy transmission' mode, if so, the step S6 is executed, and if not, the step S7 is executed;
s6, when the RSU exit node is in a 'prediction + copy transmission' mode, executing the following steps:
s6.1, predicting a PRsu node set of the RSU which is possibly reached by a target node in the next step by the central server by utilizing a Markov chain;
s6.2, deleting the RSU node in the load balancing mode from the PRsu by the central server;
s6.3, the RSU exit node trapped in the local maximum sends a data packet copy to the PRsu according to the copy transmission strategy, and then the step S7 is carried out.
And S7, after the data packet leaves the RSU network, continuing to forward according to a geography greedy strategy until the data packet reaches a destination node.
Further, in the step S3, the step S4 and the step S6, three operation modes of the RSU are: the V2I2V greedy forwarding mode, the 'prediction + replica transmission' mode and the load balancing mode are switched according to the following rules:
switching rule a: assuming that the RSU node is currently operating in a V2I2V greedy forwarding mode, if the RSU node falls into a local maximum problem, the RSU node switches to a "prediction + replica transmission" mode, and the state is maintained for a time window T; assuming that the RSU node currently works in a V2I2V greedy forwarding mode, if an overtime packet loss phenomenon occurs in a cache stack of the RSU node, switching to a load balancing mode, and maintaining a time window T in the state;
switching rule b: assuming that the RSU node currently works in a 'prediction + copy transmission' mode, if the phenomenon of overtime packet loss occurs in a cache stack of the RSU node, switching to a load balancing mode, and maintaining a time window T in the state; assuming that the RSU node is currently operating in the "predict + copy transfer" mode, if the mode retention time T has expired, the RSU node switches to the V2I2V greedy forwarding mode, and the state is maintained until other switching conditions occur;
switching rule c: assuming that the RSU node is currently operating in a load balancing mode, if the RSU node falls into a local maximum problem, it will switch to a "prediction + copy transmission" mode, and the state will maintain a time window T; assuming that the RSU node is currently operating in the load balancing mode, if the mode maintaining time T has expired, the RSU node switches to the V2I2V greedy forwarding mode, and the state is maintained until other switching conditions occur.
Further, the communication between the RSUs operating in different modes of the car networking routing method has the following rule:
limitation 1: when the RSU node in the load balancing mode executes the load balancing policy, the RSU node in the "prediction + copy transmission" mode should be deleted from the RSU entry and exit scheme set Φ in advance, that is, the RSU node in the "prediction + copy transmission" mode should not be selected as the RSU entry or exit node;
limitation 2: when the RSU node in the "prediction + duplicate transmission" mode executes the duplicate transmission scheme, the RSU node in the load balancing mode should be deleted from the prediction RSU set PRsu, that is, the RSU node in the load balancing mode should not transmit the data packet duplicate to the RSU node in the load balancing mode;
and (3) limitation: when a certain algorithm flow in the RSU is in a stage where a decision is completed but not yet executed, the change of the working mode of other RSU nodes does not affect the decision made, for example, when a certain RSU executes a load balancing algorithm to determine RSU entry and exit nodes, but a data packet is still waiting to be transmitted in a buffer stack, and at this time, the load balancing algorithm will not be executed again even if the working modes of the RSU entry and exit nodes change. The packet will be forwarded according to the given RSU ingress and egress.
Further, the step S3.2 is that the process of the central server predicting the time length of the data packet passing through the RSU network is as follows:
s3.2.1, uploading information including self message queue load capacity and message queue average throughput rate to a central server by the RSU node periodically, and summarizing and counting the information into an RSU cache stack summary table by the central server;
s3.2.2, assuming that the RSU with number i is currently selected as the RSU entry node and the RSU with number j is selected as the RSU exit node, the time wait _ entry (i) is used to indicate that the data packet is at the entry node RSUiThe waiting service time in the buffer stack is represented by TimeWait _ exit (j) for the data packet at the egress node RSUjCaching wait service time in a stack;
s3.2.3, the central server calculates the TimeWait _ entry (i) according to the information in the RSU cache stack summary table by the following formula:
Figure BDA0002146903360000061
wherein load (i) represents the current time RSUiThe central server counts two fields of the cache stack where the central server is located and the size of the data packet in the RSU cache stack summary table to obtain the total size of the data packet cached in the central server; (i) represents RSU in a period of timeiAverage throughput Rate of (RSU) by central serveriObtaining the pop information;
s3.2.4, the central server predicts the time of TimeWait _ entry (i) according to the information in the RSU cache stack summary tablejTotal packet size futureload (j) in cache of (a):
Figure BDA0002146903360000062
wherein load (j) represents the current time RSUjThe total size of the buffered packets, Throughput (j), indicates the RSU for a period of timejRepresents the RSU during TimeWait _ entry (i) timekIs expected to be sent to the RSUjThe central server counts the cache stacks in the RSU cache stack summary table,The four fields of the size of the data packet, the stack entry timestamp and the delivery target are calculated according to the following formula:
Figure BDA0002146903360000063
wherein Load (k, j, t) is indicated in RSUkIn the cache stack of (2), with RSUjThe total size of all packets that are delivery targets and the estimated waiting service time is t, the result is larger between 0 and the calculated value because futureload (j) is not negative;
s3.2.5, the center server estimates TimeWait _ exit (j):
Figure BDA0002146903360000064
s3.2.6, the central server calculates TimeThrough (i, j), because the time consumption of the transmission of the data packets of several kilobytes on the optical fiber is very short, the time consumed by the transmission of the data packets on the optical fiber is negligible compared with the time _ entry (i) and the time _ exit (j), so that the following steps are provided:
TimeThrough(i,j)＝TimeWait_entry(i)+TimeWait_exit(j)+TimeTrans
≈TimeWait_entry(i)+TimeWait_exit(j)
wherein TimeThrough (i, j), i.e. a certain data packet is in the selected RSUiAs RSU entry node, selecting RSUjIn the case of an RSU egress node, the time required to traverse the RSU wired network is expected.
Further, in the step S6.1, the central server predicts the RSU node set PRsu that the destination node may reach next by using the markov chain as follows:
s6.1.1, leaving an access record when the vehicle node accesses any one RSU, periodically uploading the vehicle access record to the central server by each RSU, and creating and updating a vehicle access history record table by the central server;
s6.1.2, establishing a state transition probability matrix P, describing the motion trail of the vehicle nodes on the road as a plurality of change processes for transferring from one RSU to another RSU, and assuming that m RSUs exist in the topology, the state transition probability matrix P is an m × m matrix:
Figure BDA0002146903360000071
wherein p isijRepresents a vehicle node from RsuiTransfer to RsujProbability of pijCalculating through historical traffic data:
Figure BDA0002146903360000072
wherein N isijIndicating a period of time from RsuiTransfer to RsujNumber of vehicles, Σj∈RsuNijIndicating a period of time from RsuiThe number of vehicles transferred to any RSU node is sigmaj∈RsuNijWhen 0, then p is notedij＝0；
S6.1.3, determining the current state of the target vehicle node, and if the current time is t and the number of the target vehicle node is u, searching the vehicle access history table, finding a record with the timestamp closest to t, and recording as the kth record, and the recorded current RSU number is v, then the current state of the target node is:
Su(k)＝[0 0 … 1 … 0 0]
wherein S isu(k) Is a 1 x m vector with the v-th bit of 1 and the remaining bits of 0, representing VehuAt step k at RsuvThe position of (a);
s6.1.4, predicting the next state of the target vehicle node:
Su(k+1)＝Su(k)P＝[pv1 pv2 … pvm]
wherein S isu(k +1) is a 1 × m vector, p in the vectorvjIs shown from RsuvTransfer to RsujThe probability of (d);
s6.1.5, determining a set of predicted RSU nodes,
the definition set R is a subset of Rsu and satisfies:
Figure BDA0002146903360000081
the set of predicted RSU nodes PRsu is a minimum set of R:
PRsu＝min_set(R)
wherein S isu(k+1)jIs SuThe j-th element of (k +1), i.e. pvjPRsu means: when the sum of the prediction probabilities is not less than the probability threshold PαThe selected RSU node set having the least number of elements.
Further, the copy transmission policy is specifically as follows:
the local maximum RSU copies L copies according to a prediction result PRsu issued by the central server, wherein L is the number of elements in the PRsu, namely L is | PRsu |; then the L copies are transmitted to each RSU in the PRsu through a wired network, the local maximum RSU still holds the original of the data packet, and the data packet is waited by a time delay tolerance mechanism, namely the RSU is allowed to cache the data packet for a long time until the RSU is separated from the local maximum problem before the survival time of the data packet expires; and after receiving the copy, the RSU node in the PRsu forwards the copy according to a geographical greedy strategy until the copy is delivered to a target vehicle node, if the target vehicle node is in the local maximum problem again, the RSU node waits by a time delay tolerance mechanism, and the data packet copy is not allowed to generate the copy again.
Further, the types of the central server include a cloud computing central server, an edge server and/or an RSU node.
Compared with the prior art, the invention has the following advantages and effects:
1. the invention combines the estimated waiting time in the RSU message waiting queue to execute load balancing according to the residual survival time of the data packet, so that the data packet originally gathered in the heavy-load RSU is distributed to each light-load RSU, the network pressure of the heavy-load RSU is reduced, and overtime packet loss caused by long-time waiting of the data packet in the message waiting queue of the heavy-load RSU is avoided, thereby reducing the end-to-end time delay and improving the delivery success rate.
2. When a certain RSU node falls into the local maximum problem, the invention predicts the motion track of a target vehicle node by using the Markov chain and executes a copy transmission strategy, so that the copy of the data packet can be efficiently transmitted to the target node, thereby reducing the end-to-end time delay and improving the delivery rate.
3. The invention designs three working modes for the RSU node, and can obtain better delivery performance under low vehicle density and high vehicle density.
Drawings
FIG. 1 is a flow chart of a multi-mode handover routing method in the present invention;
FIG. 2 is a block diagram of a vehicle node and RSU node communication system in accordance with the present invention;
FIG. 3 is a schematic switching diagram of three modes of the multi-mode routing switching method of the present invention;
FIG. 4 is a schematic diagram of an urban Internet of vehicles communication environment in accordance with the present invention;
FIG. 5 is a diagram illustrating delivery rate comparison of an embodiment of the present invention with GPSR and V2I2V routing protocols;
fig. 6 is a diagram illustrating end-to-end delay comparison between an embodiment of the present invention and GPSR and V2I2V routing protocols.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Examples
In this embodiment, as shown in fig. 4, RSUs are deployed at an urban road intersection. The RSU and the RSU are connected through optical fibers, and the RSU and the central server are connected through the optical fibers, so that high-speed and high-capacity wired communication can be provided. The vehicle-to-vehicle communication adopts V2V communication based on DSRC, and the vehicle-to-RSU communication adopts V2I communication based on DSRC.
With reference to fig. 1 and fig. 2, the internet of vehicles routing method with multi-mode switching disclosed in this embodiment includes the following steps:
and S1, the message source node generates a data packet, and sets the RSU node closest to the message source node as a first-stage delivery target which is recorded as an RSU entrance node. The packet is forwarded to the RSU ingress node via multi-hop V2V communication.
S2, the RSU entrance node judges whether the RSU entrance node is in a load balancing mode, if so, the step S3 is executed; otherwise, the process proceeds to step S4.
S3, when the RSU entrance node is in the load balancing mode, executing the following steps:
s3.1, the central server deletes the scheme containing the RSU nodes in the 'prediction + copy transmission' mode from the RSU entrance and exit scheme set phi.
And S3.2, the central server estimates the time duration TimeThrough (i, j) of the data packet passing through the RSU network, and the variable represents the time duration required by the data packet to pass through the RSU network when the RSU with the number of i is taken as an inlet and the RSU with the number of j is taken as an outlet.
S3.3, the central server checks whether TimeThrough (i, j) meets the estimated passing time constraint, the constraint is used for judging whether the data packet has enough remaining life time to pass through the RSU wired network:
TimeThrough(i,j)≤α×TimeRest
the main reason for designing the coefficient alpha is that the data packet may reach the destination node only after passing through the RSU wired network and possibly through multi-hop V2V communication, so that time needs to be reserved for the multi-hop transmission when the decision is made.
If not, the step S3.4 is carried out; if yes, go to step S3.5.
And S3.4, deleting the scheme phi (i, j) which does not meet the constraint from the RSU entry and exit scheme set phi by the central server, wherein the phi (i, j) represents the scheme which takes the RSU with the number of i as an entry and the RSU with the number of j as an exit in the scheme set phi, and reselecting a new group of schemes from the rest RSU entry and exit scheme sets phi. Step S3.2 is performed again.
S3.5, the central server informs the RSU entrance node of phi (i, j) passing the constraint, and then the data packet enters the RSU network according to the designated RSU entrance node.
S3.6, the RSU ingress node forwards the packet to the designated RSU egress node, and then proceeds to step S5.
And S4, forwarding the data packet to the RSU exit node closest to the destination node by the RSU entrance node according to a V2I2V greedy forwarding mode.
S5, the RSU exit judges whether the RSU exit is in a 'prediction + copy transmission' mode, if so, the step S6 is executed; otherwise, the process proceeds to step S7.
S6, when the RSU exit node is in a 'prediction + copy transmission' mode, executing the following steps:
s6.1, the central server predicts the PRsu node set of the RSU which is possibly reached by the target node in the next step by using the Markov chain.
S6.2, deleting the RSU nodes in the load balancing mode from the PRsu by the central server.
And S6.3, the RSU exit node trapped in the local maximum sends a data packet copy to the PRsu according to the copy transmission strategy. Then, the process proceeds to step S7.
And S7, after the data packet leaves the RSU network, continuing to forward according to a geography greedy strategy until the data packet reaches a destination node. And finishing the delivery process.
In this embodiment, as shown in fig. 3, three operation modes of the RSU are: the V2I2V greedy forwarding mode, the 'prediction + replica transmission' mode and the load balancing mode are switched according to the following rules:
switching rule a: assuming that the RSU node is currently operating in the V2I2V greedy forwarding mode, if the RSU node is stuck with a local maximum problem, it will switch to the "predict + copy transfer" mode, and the state will remain for a time window T. Assuming that the RSU node currently operates in the greedy forwarding mode of V2I2V, if an overtime packet loss occurs in the RSU node cache stack, the RSU node switches to the load balancing mode, and the state is maintained for a time window T.
Switching rule b: assuming that the RSU node currently operates in the "prediction + copy transmission" mode, if the phenomenon of timeout and packet loss occurs in the RSU node cache stack, the RSU node will switch to the load balancing mode, and the state will maintain a time window T. Assuming that the RSU node is currently operating in "predict + copy transfer" mode, if the mode hold time T has expired, it switches to V2I2V greedy forwarding mode, and the state is maintained until other switch conditions occur.
Switching rule c: assuming that the RSU node is currently operating in load balancing mode, if the RSU node is experiencing a local maximum problem, it will switch to "predict + duplicate transfer" mode, and this state will remain for a time window T. Assuming that the RSU node is currently operating in the load balancing mode, if the mode maintaining time T has expired, the RSU node switches to the V2I2V greedy forwarding mode, and the state is maintained until other switching conditions occur.
The method of the invention and a typical routing protocol V2I2V based on a road side unit in the Internet of vehicles are simulated and compared on an NS3 network simulation platform respectively, the simulation parameter setting is shown in Table 1, and the simulation result is a delivery rate comparison graph shown in figure 5 and an end-to-end time delay comparison graph shown in figure 6.
TABLE 1 simulation parameter setup table
Simulation parameters Value of
Map size 2450m*3700m
Simulation time 2000s
Number of vehicle nodes [800；1000；1200；1400；1600]An
Vehicle speed 10～100km/h
Transmission rate 6Mbps
Vehicle node transmission range 300m
Packet size [250；500；1000；2000]B
Data packet lifetime [250；500；1000；2000]ms
Packet generation rate 16/s
Number of road side units 24
Road side unit coverage 300m
Fig. 5 shows that with the change of the number of vehicle nodes, compared with the classic routing protocols GPSR and V2I2V in the internet of vehicles, the method has obvious improvement in delivery rate, and optimizes network performance.
Fig. 6 shows that with the change of the number of vehicle nodes, compared with the classic routing protocols GPSR and V2I2V in the internet of vehicles, the method has a significant reduction in end-to-end delay, and optimizes network performance.
To sum up, the above embodiment discloses a multi-mode switching routing method for internet of vehicles, including that an RSU node performs switching judgment according to the sending condition of its own data packet and the use condition of a cache stack, and performs switching in three working modes; when the RSU node is in a V2I2V greedy forwarding mode, the message source node forwards the data packet to an RSU inlet node closest to the message source node according to the greedy forwarding mode, then forwards the data packet to an RSU outlet node closest to the destination node through an RSU wired network, then the RSU outlet node forwards the data packet to vehicle nodes within the coverage range of the RSU outlet node, then forwards the data packet to the destination node through multiple times of greedy forwarding among the vehicle nodes, and finally the data packet reaches the destination node; when the RSU node is in a 'prediction + copy transmission' mode, a Markov chain is used for predicting an RSU set to which a target vehicle node is transferred next time, a data packet copy is copied and transmitted to the prediction RSU set, the prediction RSU set continues to perform greedy forwarding by taking a target vehicle as a target after receiving the copy, meanwhile, the local maximum RSU still holds a data packet original, the data packet original continues to wait until the local maximum problem is solved according to a 'delay tolerance' mechanism, and when the data packet original or any copy reaches the target vehicle node, delivery is declared to be successful; when the RSU node is in a load balancing mode, the central server utilizes the RSU cache stack summary table to estimate the time required by the data packet to pass through the RSU wired network, and combines the remaining survival time of the data packet to be transmitted to carry out load balancing judgment and search proper RSU inlet and outlet nodes for the data packet, so that the data packet is forwarded by selecting a light-load RSU. The invention utilizes the RSU wired network to assist in delivering the multi-hop messages among the vehicles, and simultaneously designs three working modes for the RSU, so that the routing protocol can better adapt to the working environments with low vehicle node density and high vehicle node density, the end-to-end time delay of the routing of the Internet of vehicles is reduced, and the delivery success rate is improved.
The above embodiments are preferred embodiments of the present invention, but the present invention is not limited to the above embodiments, and any other changes, modifications, substitutions, combinations, and simplifications which do not depart from the spirit and principle of the present invention should be construed as equivalents thereof, and all such changes, modifications, substitutions, combinations, and simplifications are intended to be included in the scope of the present invention.

Claims (7)

1. A multi-mode switching Internet of vehicles routing method is characterized by comprising the following steps:
s1, the message source node generates a data packet, an RSU node closest to the message source node is set as a first-stage delivery target and is marked as an RSU entrance node, and the data packet is forwarded to the RSU entrance node through multi-hop V2V communication;
s2, the RSU entrance node judges whether the RSU entrance node is in a load balancing mode, if so, the step S3 is executed; otherwise, go to step S4;
s3, when the RSU entrance node is in the load balancing mode, executing the following steps:
s3.1, deleting a scheme containing RSU nodes in a 'prediction + copy transmission' mode from the RSU entry and exit scheme set phi by the central server;
s3.2, the central server predicts a time duration TimeThrough (i, j) of the data packet passing through the RSU network, wherein i is used as an RSU inlet node, and j is used as an RSU outlet node;
s3.3, the central server checks whether TimeThrough (i, j) meets the estimated passing time constraint, the constraint is used for judging whether the data packet has enough remaining life time to pass through the RSU wired network:
TimeThrough(i,j)≤α×TimeRest
wherein alpha is a constant and satisfies 0 < alpha ≦ 1, and TimeRest is the remaining lifetime of the packet;
if not, the step S3.4 is carried out, and if yes, the step S3.5 is carried out;
s3.4, the central server deletes the scheme phi (i, j) which does not meet the constraint from the RSU entry and exit scheme set phi, wherein the phi (i, j) represents the scheme which takes the RSU with the number of i as an entry and the RSU with the number of j as an exit in the scheme set phi, selects a new group of schemes from the rest RSU entry and exit scheme sets phi, and executes the step S3.2 again;
s3.5, the central server informs the RSU entrance nodes of phi (i, j) passing the constraint, and then the data packet enters the RSU network according to the specified RSU entrance nodes;
s3.6, the RSU entrance node forwards the data packet to the designated RSU exit node, and then the step S5 is carried out;
s4, forwarding the data packet to the RSU exit node closest to the destination node by the RSU entrance node according to a V2I2V greedy forwarding mode;
s5, the RSU exit judges whether the RSU exit is in a 'prediction + copy transmission' mode, if so, the step S6 is executed, and if not, the step S7 is executed;
s6, when the RSU exit node is in a 'prediction + copy transmission' mode, executing the following steps:
s6.1, the central server predicts the PRsu of the RSU node set which is possibly reached by the target node in the next step by using the Markov chain, wherein the PRsu means that: when the sum of the prediction probabilities is not less than the probability threshold PαOn the premise of (1), the selected RSU node set with the least number of elements;
s6.2, deleting the RSU node in the load balancing mode from the PRsu by the central server;
s6.3, the RSU exit node trapped in the local maximum sends a data packet copy to the PRsu according to a copy transmission strategy, and then the step S7 is carried out;
and S7, after the data packet leaves the RSU network, continuing to forward according to a geography greedy strategy until the data packet reaches a destination node.
2. The method as claimed in claim 1, wherein in the steps S3, S4 and S6, the RSU operates in three modes: the V2I2V greedy forwarding mode, the 'prediction + replica transmission' mode and the load balancing mode are switched according to the following rules:
switching rule a: assuming that the RSU node currently works in a V2I2V greedy forwarding mode, if the RSU node falls into a local maximum problem, the RSU node is switched to a 'prediction + copy transmission' mode, and after the V2I2V greedy forwarding mode is switched to the 'prediction + copy transmission' mode, the state of the 'prediction + copy transmission' mode maintains a time window T; assuming that the RSU node currently works in a V2I2V greedy forwarding mode, if an overtime packet loss phenomenon occurs in a cache stack of the RSU node, switching to a load balancing mode, and maintaining a time window T in the state of the load balancing mode after the V2I2V greedy forwarding mode is switched to the load balancing mode;
switching rule b: assuming that the RSU node currently works in a 'prediction + copy transmission' mode, if the phenomenon of overtime packet loss occurs in a cache stack of the RSU node, the RSU node is switched to a load balancing mode, and after the 'prediction + copy transmission' mode is switched to the load balancing mode, the state of the load balancing mode maintains a time window T; assuming that the RSU node is currently operating in the "prediction + replica transmission" mode, if the mode retention time T has expired, the mode is switched to the V2I2V greedy forwarding mode, and after the "prediction + replica transmission" mode is switched to the V2I2V greedy forwarding mode, the state of the V2I2V greedy forwarding mode is maintained until other switching conditions occur;
switching rule c: assuming that the RSU node works in a load balancing mode currently, if the RSU node falls into a local maximum problem, the RSU node is switched to a 'prediction + copy transmission' mode, and after the load balancing mode is switched to the 'prediction + copy transmission' mode, the state of the 'prediction + copy transmission' mode maintains a time window T; assuming that the RSU node is currently operating in the load balancing mode, if the mode maintaining time T has expired, the mode is switched to the V2I2V greedy forwarding mode, and after the load balancing mode is switched to the V2I2V greedy forwarding mode, the state of the V2I2V greedy forwarding mode is maintained until other switching conditions occur.
3. The Internet of vehicles routing method of claim 1, wherein the communication between RSUs operating in different modes is limited by the following rules:
limitation 1: when the RSU node in the load balancing mode executes the load balancing policy, the RSU node in the "prediction + copy transmission" mode should be deleted from the RSU entry and exit scheme set Φ in advance, that is, the RSU node in the "prediction + copy transmission" mode should not be selected as the RSU entry or exit node;
limitation 2: when the RSU node in the "prediction + duplicate transmission" mode executes the duplicate transmission scheme, the RSU node in the load balancing mode should be deleted from the prediction RSU set PRsu, that is, the RSU node in the load balancing mode should not transmit the data packet duplicate to the RSU node in the load balancing mode;
and (3) limitation: when one algorithm process in the RSU is in a stage of completing judgment but not executing, the change of the working mode of other RSU nodes can not influence the judgment made.
4. The method for routing the internet of vehicles through multi-mode switching according to claim 1, wherein the step S3.2 of the central server estimating the time length of the data packet passing through the RSU network comprises the following steps:
s3.2.1, uploading information including self message queue load capacity and message queue average throughput rate to a central server by the RSU node periodically, and summarizing and counting the information into an RSU cache stack summary table by the central server;
s3.2.2, assuming that the RSU with number i is currently selected as the RSU entry node and the RSU with number j is selected as the RSU exit node, the time wait _ entry (i) is used to indicate that the data packet is at the entry node RSUiThe waiting service time in the buffer stack is represented by TimeWait _ exit (j) for the data packet at the egress node RSUjCaching wait service time in a stack;
s3.2.3, the central server calculates the TimeWait _ entry (i) according to the information in the RSU cache stack summary table by the following formula:
Figure FDA0002980752140000041
wherein load (i) represents the current time RSUiThe central server counts two fields of the cache stack where the central server is located and the size of the data packet in the RSU cache stack summary table to obtain the total size of the data packet cached in the central server; throughput (i) indicates the RSU over a period of timeiAverage throughput Rate of (RSU) by central serveriObtaining the pop information;
s3.2.4, the central server predicts the time of TimeWait _ entry (i) according to the information in the RSU cache stack summary tablejTotal packet size futureload (j) in cache of (a):
Figure FDA0002980752140000051
wherein load (j) represents the current time RSUjThe total size of the buffered packets, Throughput (j), indicates the RSU for a period of timejRepresents the RSU during TimeWait _ entry (i) timekIs expected to be sent to the RSUjThe central server counts four fields of a cache stack in which the RSU cache stack summary table is positioned, the size of a data packet, a stacking timestamp and a delivery target, and calculates according to the following formula:
Figure FDA0002980752140000052
wherein Load (k, j, t) is indicated in RSUkIn the cache stack of (2), with RSUjThe total size of all packets that are delivery targets and the estimated waiting service time is t, the result is larger between 0 and the calculated value because futureload (j) is not negative;
s3.2.5, the center server estimates TimeWait _ exit (j):
Figure FDA0002980752140000053
s3.2.6, the central server calculates TimeThrough (i, j), because the time consumption of the transmission of the data packets of several kilobytes on the optical fiber is very short, the time consumed by the transmission of the data packets on the optical fiber is ignored compared with TimeWait _ entry (i) and TimeWait _ exit (j), so that:
TimeThrough(i,j)＝TimeWait_entry(i)+TimeWait_exit(j)+TimeTrans
≈TimeWait_entry(i)+TimeWait_exit(j)
wherein TimeThrough (i, j), i.e. a certain data packet is in the selected RSUiAs RSU entry node, selecting RSUjIn the case of an RSU egress node, the time required to traverse the RSU wired network is expected.
5. The multi-mode switching internet-of-vehicles routing method according to claim 1, wherein the step S6.1, the process that the central server predicts the RSU node set pru that the destination node may arrive next by using the markov chain, is as follows:
s6.1.1, leaving an access record when the vehicle node accesses any one RSU, periodically uploading the vehicle access record to the central server by each RSU, and creating and updating a vehicle access history record table by the central server;
s6.1.2, establishing a state transition probability matrix P, describing the motion trail of the vehicle nodes on the road as a plurality of change processes for transferring from one RSU to another RSU, and assuming that m RSUs exist in the topology, the state transition probability matrix P is an m × m matrix:
Figure FDA0002980752140000061
wherein p isijRepresents a vehicle node from RsuiTransfer to RsujProbability of pijCalculating through historical traffic data:
Figure FDA0002980752140000062
wherein N isijIndicating a period of time from RsuiTransfer to RsujNumber of vehicles, Σj∈RsuNijIndicating a period of time from RsuiThe number of vehicles transferred to any RSU node is sigmaj∈RsuNijWhen 0, then p is notedij＝0；
S6.1.3, determining the current state of the target vehicle node, and if the current time is t and the number of the target vehicle node is u, searching the vehicle access history table, finding a record with the timestamp closest to t, and recording as the kth record, and the recorded current RSU number is v, then the current state of the target node is:
Su(k)＝[0 0…1…0 0]
wherein S isu(k) Is a 1 x m vector with the v-th bit of 1 and the remaining bits of 0, representing VehuAt step k at RsuvThe position of (a);
s6.1.4, predicting the next state of the target vehicle node:
Su(k+1)＝Su(k)P＝[pv1 pv2…pvm]
wherein S isu(k +1) is a 1 × m vector, p in the vectorvjIs shown from RsuvTransfer to RsujThe probability of (d);
s6.1.5, determining a set of predicted RSU nodes,
the definition set R is a subset of Rsu and satisfies:
Figure FDA0002980752140000071
the set of predicted RSU nodes PRsu is a minimum set of R:
PRsu＝min_set(R)
wherein S isu(k+1)jIs SuThe j-th element of (k +1), i.e. pvj。
6. The multi-mode switching internet of vehicles routing method according to claim 1, wherein the replica transmission strategy is as follows:
the local maximum RSU copies L copies according to a prediction result PRsu issued by the central server, wherein L is the number of elements in the PRsu, namely L is | PRsu |; then the L copies are transmitted to each RSU in the PRsu through a wired network, the local maximum RSU still holds the original of the data packet, and the data packet is waited by a time delay tolerance mechanism, namely the RSU is allowed to cache the data packet for a long time until the RSU is separated from the local maximum problem before the survival time of the data packet expires; and after receiving the copy, the RSU node in the PRsu forwards the copy according to a geographical greedy strategy until the copy is delivered to a target vehicle node, if the target vehicle node is in the local maximum problem again, the RSU node waits by a time delay tolerance mechanism, and the data packet copy is not allowed to generate the copy again.
7. The multi-mode switching internet-of-vehicles routing method according to claim 1, wherein the type of the central server comprises a cloud computing central server, an edge server and/or an RSU node.
CN201910687760.7A 2019-07-29 2019-07-29 Multi-mode switching Internet of vehicles routing method Active CN110446184B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910687760.7A CN110446184B (en) 2019-07-29 2019-07-29 Multi-mode switching Internet of vehicles routing method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910687760.7A CN110446184B (en) 2019-07-29 2019-07-29 Multi-mode switching Internet of vehicles routing method

Publications (2)

Publication Number Publication Date
CN110446184A CN110446184A (en) 2019-11-12
CN110446184B true CN110446184B (en) 2021-06-08

Family

ID=68432007

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910687760.7A Active CN110446184B (en) 2019-07-29 2019-07-29 Multi-mode switching Internet of vehicles routing method

Country Status (1)

Country Link
CN (1) CN110446184B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112004197B (en) * 2020-08-06 2022-03-22 重庆邮电大学 Heterogeneous Internet of vehicles switching method based on vehicle track prediction
CN112020115A (en) * 2020-08-11 2020-12-01 广州杰赛科技股份有限公司 Network access method and system based on intelligent internet
CN113507503B (en) * 2021-06-16 2022-06-14 华南理工大学 Internet of vehicles resource allocation method with load balancing function
CN113613209B (en) * 2021-07-02 2023-05-26 重庆邮电大学 Pre-switching method based on RSU grade and content popularity in ICN Internet of vehicles

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104363626A (en) * 2014-11-07 2015-02-18 中国人民武装警察部队工程大学 Bayesian theory based credible routing method for VANET (vehicular ad-hoc network)
CN106535280A (en) * 2016-11-29 2017-03-22 华南理工大学 Internet of vehicle opportunistic routing method based on geographic position
CN106899668A (en) * 2017-02-23 2017-06-27 同济大学 Information Push Service processing method in car networking
US9832705B1 (en) * 2016-09-02 2017-11-28 The University Of North Carolina At Chapel Hill Methods, systems, and computer readable media for topology management and geographic routing in mobile ad-hoc networks
CN108768852A (en) * 2018-04-02 2018-11-06 华南理工大学 Chance routing algorithm based on more copy Congestion Avoidances in a kind of car networking
CN109660465A (en) * 2019-02-14 2019-04-19 华南理工大学 A kind of car networking load balancing method for routing of roadside unit auxiliary

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104363626A (en) * 2014-11-07 2015-02-18 中国人民武装警察部队工程大学 Bayesian theory based credible routing method for VANET (vehicular ad-hoc network)
US9832705B1 (en) * 2016-09-02 2017-11-28 The University Of North Carolina At Chapel Hill Methods, systems, and computer readable media for topology management and geographic routing in mobile ad-hoc networks
CN106535280A (en) * 2016-11-29 2017-03-22 华南理工大学 Internet of vehicle opportunistic routing method based on geographic position
CN106899668A (en) * 2017-02-23 2017-06-27 同济大学 Information Push Service processing method in car networking
CN108768852A (en) * 2018-04-02 2018-11-06 华南理工大学 Chance routing algorithm based on more copy Congestion Avoidances in a kind of car networking
CN109660465A (en) * 2019-02-14 2019-04-19 华南理工大学 A kind of car networking load balancing method for routing of roadside unit auxiliary

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
《Infrastructure-Assisted Markov Prediction Routing Protocol for Urban Scenarios》;Lin Lin;《 2018 10th International Conference on Communication Software and Networks (ICCSN)》;20181011;全文 *
《车载自组织网络机会路由协议研究》;刘晓涛;《中国优秀硕士学位论文（电子期刊）》;20190131;全文 *

Also Published As

Publication number Publication date
CN110446184A (en) 2019-11-12

Similar Documents

Publication Publication Date Title
CN110446184B (en) Multi-mode switching Internet of vehicles routing method
CN109660465B (en) Road side unit assisted Internet of vehicles load balancing routing method
CN110351200B (en) Opportunistic network congestion control method based on forwarding task migration
CN101854307B (en) Processing method of network node memory congestion in delay-tolerant network
CN110708736A (en) Dynamic routing method and system based on energy efficiency relay selection
CN111130853B (en) Future route prediction method of software defined vehicle network based on time information
CN107105389B (en) Geographic information routing method based on road topological structure in vehicle-mounted network
CN110248392B (en) Opportunity forwarding method based on node efficiency in Internet of vehicles
CN108366340B (en) Urban Internet of vehicles routing method based on bus track and ant colony optimization
CN108768852B (en) Opportunistic routing method based on multi-copy congestion avoidance in Internet of vehicles
CN112020117B (en) Routing method based on transmission speed and node capacity in low-earth-orbit satellite communication network
CN107995114A (en) Delay Tolerant Network method for routing based on Density Clustering
CN103532865A (en) Congestion control method based on socially aware in delay tolerant network
WO2020186655A1 (en) Content forwarding node selection method in vehicle-mounted ad hoc network based on service type
CN116528313A (en) Unmanned aerial vehicle low-energy-consumption rapid routing method for task collaboration
Yang et al. A review of UAV ferry algorithms in delay tolerant network
Lai et al. PBMP: priority‐based multi‐path packet routing for vehicular ad hoc network system in city environment
Fekair et al. An efficient fuzzy logic-based and bio-inspired QoS-compliant routing scheme for VANET
CN103260270B (en) A kind of base station
CN107634906B (en) Cluster-based multi-attribute decision VDTN routing method
CN106254250A (en) Car networking short-distance networking method
CN113395739A (en) Improved self-repairing routing strategy based on ant colony optimization in urban Internet of vehicles
CN106304237A (en) Terminal vehicle communication method based on wireless protocols
Tang et al. Knowledge‐based replica deletion scheme using directional anti‐packets for vehicular delay‐tolerant networks
CN108848534A (en) The method for routing of node load is adaptively reduced in mobile self-grouping network

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant