CN108667668A - The access method for routing of base is connected under a kind of urban road scene based on car networking - Google Patents

The access method for routing of base is connected under a kind of urban road scene based on car networking Download PDF

Info

Publication number
CN108667668A
CN108667668A CN201810488153.3A CN201810488153A CN108667668A CN 108667668 A CN108667668 A CN 108667668A CN 201810488153 A CN201810488153 A CN 201810488153A CN 108667668 A CN108667668 A CN 108667668A
Authority
CN
China
Prior art keywords
network
node
access
routing
crossing
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.)
Granted
Application number
CN201810488153.3A
Other languages
Chinese (zh)
Other versions
CN108667668B (en
Inventor
程久军
黄震华
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.)
Tongji University
Original Assignee
Tongji University
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 Tongji University filed Critical Tongji University
Priority to CN201810488153.3A priority Critical patent/CN108667668B/en
Publication of CN108667668A publication Critical patent/CN108667668A/en
Application granted granted Critical
Publication of CN108667668B publication Critical patent/CN108667668B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14—Network analysis or design
    • H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/09—Arrangements for giving variable traffic instructions
    • G08G1/096—Arrangements for giving variable traffic instructions provided with indicators in which a mark progresses showing the time elapsed, e.g. of green phase
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/12—Discovery or management of network topologies
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
    • H04L43/0805—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters by checking availability
    • H04L43/0811—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters by checking availability by checking connectivity
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/02—Topology update or discovery
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/12—Shortest path evaluation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/16—Multipoint routing
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/74—Address processing for routing
    • H04L45/745—Address table lookup; Address filtering

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Environmental & Geological Engineering (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

The access method for routing of base is connected under a kind of urban road scene based on car networking, is related to car networking complex network field.Increase packet loss to solve the no longer effective property of data packet in data forwarding process, eventually leads to the inaccessible problem of network.The present invention is in view of the above problems, the access method for routing of urban road scene is provided, this method gives the concept at central crossbar crossing based on the connection base component to interconnect, to obtain larger range of network topological information, more preferably data transmission routing can be thus constructed.The method for routing can fast implement car networking network interconnection intercommunication for urban road complex scene and keep relative stability, to which the real-time stabilization transmission of the data informations such as traffic accident information, real time media data be effectively ensured.

Description

The access method for routing of base is connected under a kind of urban road scene based on car networking
Technical field
The present invention relates to car networking complex network fields.
Background technology
The problems such as road structure is complicated in City scenarios, existing most of Routing Protocols are all based on greedy algorithm With the stateless route agreement for carrying forwarding or periphery forwarding mechanism, that is, in repeating process, source node only focuses on purpose Outlet is just done according to certain route metric when data packet will be forwarded to next-hop in the position of node and node adjacent thereto By decision.For example GPSR, GSR and GPCR, according to next forward node of Euclidean distance selection routing, distance is closer selected general Rate is higher, and GyTAR, A-STAR, RBVT are connected to preferable section with IGRP selections and carry out data packet forwarding.However pass through these Algorithm builds routing in extensive car networking and will appear local optimum phenomenon, as shown in Figure 1, A, B, C, G, E, F are four crossway Mouthful, data packet is sent to D by S.Most of routing algorithms all can forward data along A → B → paths C, because of section SA,BWith SB,CNode it is most intensive, connected probability is maximum.But crossing C is once reached, it just will appear local optimum, because just sending out at this time Existing C and F is disjunct.There are two types of the data packets for reaching C selects：1, in SC,FData biography is carried out in a manner of carrying and forward Defeated, i.e., a certain vehicle carries data packet, and the vehicle found closer to mesh destination node is forwarded；2, by data packet retransmission to B Find other paths.Both modes can all improve data packet end-to-end delay, and finally because transmission time is long, data packet loses Timeliness and increase packet loss, it is not sensible to eventually lead to network.Local optimum phenomenon is because above-mentioned Routing Protocol only passes through office Portion network topological information does routing decision.If having overall cognitive to network topology, then A → G → paths E → F will be used Data forwarding is carried out, such data packet can successfully arrive at destination node, and shorten the time that data packet reaches destination node, carry High data packet reaches success rate.
In conclusion the present invention proposes extensive access method for routing --- the AMRP for being connected to base based on car networking (Accessibility model-based Routing Protocol), by calculating the access of section, and according to net Network topology information calculates maximum access routing.
Invention content
The present invention is based on the connection base components to interconnect (filed in 11 days October in 2017 of the inventors such as Cheng Jiujun 《The connection base component building method that car networking large scale network interconnects》(applicant：Tongji University, number of patent application 201710397807 7), give for the access method for routing of car networking large scale network under urban road complex scene.
Technical solution of the present invention is：
One kind is based on car networking connection base (filed in 11 days October in 2017 of the inventors such as Cheng Jiujun《The big rule of car networking The connection base component building method that lay wire network interconnects》(applicant：Tongji University, number of patent application 201710397807 is 7) The access method for routing of urban road scene, which is characterized in that specific method includes the following steps：
One, bridging nodes are screened
If bridging nodes will sail out of crossing region, a substitute node is begun look for.It is every that bridging nodes are calculated first A neighbours' connected member node passes through the time t used in the region of crossroadcross, then by tcrossWith green light remaining time tremPhase Compare,
trem=tg-[(tamodtC)-(tC-tg)] (1)
Wherein taFor the time used in node exterior traffic lamp, tCFor the period of traffic lights, tgFor green light stage duration.When trem＞ tcrossWhen, bridging nodes can select tcrossMaximum neighbours' connected member node is as next bridging nodes.Work as trem≤ tcrossWhen, it can select a connected member node being parked at red light.Can ideally select one from crossroad region Nearest node.
Two, the access assessment of section network
Network access assessment in section refers to weighing the sensible situation in a section, this stage will be finally in intersection Bridging nodes generate routing table, routing table includes to be connected with the intersection the access value in section and the volume in each section Number.It can determine whether bridging nodes can be connected with destination node according to routing table.Once there is connected member node quilt at crossing Bridging nodes are selected as, then start the access assessment of road network.First a section probe data packet is generated by bridging nodes (Road Section Detect Packet, RSDP) and to all connected member sections in all sections being connected with intersection Point broadcast.The connected member node for receiving RSDP starts to calculate link existence duration between neighbours' connected member node, and most The network that whole section is calculated eventually is access.
Three, center crossing is screened
The concept at introducing center crossing.Whole network is divided into multiple blocks first, and it is screened for each block Center crossing.Center crossing is defined as the most intersection of the crossing number communicated therewith in block.Center crossing has area The path of arbitrary node in this block is known at the overview of block network topology, that is, center crossing.It is sent out to destination node When sending data packet, the center crossing of block, the bridge in this center crossing is given data packet where can first finding destination node The optimal path of destination node can be found by connecing node.
The screening of center crossing is the process of a distributed dynamic:Each bridging nodes appoints the intersection where it first Crossing centered on crossing, and ρ=0 is set, that is, the crossroad number that this center crossing is connected is 0.Then each Connected member starts to exchange RSDP data packets.After having exchanged the RSDP data packets of the first round, bridging nodes can be according to REP Value updates the ρ at its affiliated center crossing.Before second exchanges RSDP, bridging nodes route it table and are put into RSDP first (use alternative area), then connected member node broadcasts into section again.After receiving RSDP, bridging nodes can be with oneself Routing table and ρ be compared, if ρ is more than itself ρ, where the bridging nodes for center crossing being set as sending this RSDP Intersection.
For irregular extensive road topology, multiple blocks can be divided into radius l, then according to above Algorithm carries out the screening of center crossing.When the size of l is arranged, it is contemplated that the convergence time and routing cost of algorithm.If setting A prodigious l value is set, is represent comprising many crossings in a block, under this l value, the expense for establishing routing is smaller. If the value of opposite l is smaller, center crossing screening time is shorter, and Route establishment then consumes the longer time.
Four, Route establishment
When establishing routing, source node firstly generates " Route Query " (RQ) data packet, contains source node ID, the geographical location of destination node ID and destination node.RQ is forwarded to nearest bridging nodes by source node first.Bridging nodes After receiving RQ, start to check whether this destination node is included in the routing table of oneself.If so, bridging nodes are built " Route Result " (RR) data packet, RR contain routing of the source node to destination node.If in bridging nodes routing table not Including destination node, then be forwarded to center crossing by RQ, the bridging nodes at the crossing of center then first check destination node whether In this block, and if it exists, the access highest that destination node is filtered out using existing Bellman-ford algorithms is route, and And it adds it to and sends back source node in RR.If being not present, source node is calculated to the access highest road of this bridging nodes Diameter is entered to be added in RQ；Then forwarding RQ continues this process until RQ data packets reach mesh to its neighbours crossroad Node.
In step 2, routing table is the format of RSDP data packets, and wherein intersection ID is to generate this RSDP data packet The ID at the crossing where bridging nodes, timestamp are the time that RSDP is generated, and access section network is access according to network Model (see embodiment 2) calculates the gained access value of section network, and section node hop count is that thus intersection is handed over to its neighbour The hop count in the path of cross road mouth.
The access model of the network is a kind of access model of large scale network being connected to base based on car networking, It is characterized in that, car networking network is access, i.e., fast implements network interconnection intercommunication according to current task, and keeps stablizing shape in real time State is for the purpose of completing current task；Access network refers to the network that i.e. connection is stablized again, and the access model of network is then to use To integrate the connectivity and stability of weighing target network.For any two node v in GAAnd vB, access network is vA And vBBetween connection base net network it is access, then network G it is access then be its be connected to base it is access, it can be deduced that network Access calculation formula：
WhereinFor the connection base of network G, CothFor network connectivty threshold value, according to the car networking application program of source vehicle Depending on demand.
Advantageous effect：
The problems such as complicated for road structure in City scenarios, existing most of Routing Protocols are all based on greedy algorithm With the stateless route agreement for carrying forwarding or periphery forwarding mechanism, that is, in repeating process, source node only focuses on purpose Outlet is just done according to certain route metric when data packet will be forwarded to next-hop in the position of node and node adjacent thereto By decision, however routing is built in extensive car networking by these algorithms and will appear local optimum phenomenon, to improve number According to packet end-to-end delay, finally because transmission time is long, the no longer effective property of data packet and increase packet loss, eventually lead to network not The problems such as sensible, the present invention is based on the connection base component to interconnect, (inventors such as Cheng Jiujun applied on October 11st, 2017 's《The connection base component building method that car networking large scale network interconnects》(applicant：Tongji University, number of patent application 7) 201710397807 give the concept at central crossbar crossing, to obtain larger range of network topological information, thus may be used To construct more preferably data transmission routing.The method for routing can fast implement car networking net for urban road complex scene Network interconnects and keeps relative stability, to which the reality of the data informations such as traffic accident information, real time media data be effectively ensured Shi Wending is transmitted.
Description of the drawings
Fig. 1 conventional routing protocols local optimum phenomenons
Fig. 2 interleaving route bridging nodes Filtering systems
Fig. 3 first round exchanges each crossing routing table after RSDP
The wheels of Fig. 4 second exchange each crossing routing table after RSDP
Fig. 5 third rounds exchange each crossing routing table after RSDP
Fig. 6 fourth rounds exchange each crossing routing table after RSDP
Fig. 7 different routing protocols be averaged end-to-end delay with traffic density variation relation
Fig. 8 different routing protocols be averaged end-to-end delay with data package transmission velocity variation relation
Fig. 9 different routing protocol data packet delivery fractions with traffic density variation relation
Figure 10 different routing protocol data packet delivery fractions with data package transmission velocity variation relation
Figure 11 different routing protocols routing cost with traffic density variation relation
Figure 12 different routing protocols routing cost with data package transmission velocity variation relation
Figure 13 is the method for the present invention flow chart
Figure 14 is 2 communication path schematic diagram of embodiment
Figure 15 is 2 one hop link model of embodiment
Specific implementation mode
The specific implementation process of the present invention is as shown in figure 13, including following 8 aspects：
1. bridging nodes are screened
2. the access assessment of section network
3. center crossing is screened
4. Route establishment
5. testing
6. averagely end-to-end delay
7. data packet delivery fraction
8. routing cost
1. bridging nodes are screened
After car networking connection base is established in each section respectively, AMRP is connected by screening bridging nodes in intersection The connection base in each section.The process of entire bridging nodes screening as can be seen from Figure 2.When initial, crossing region (Intersection zone) all vehicles (including A, B, C, D and E) are likely to be chosen as bridging nodes.It excludes first Crossing center (Intersection center) vehicle (including B, D and E) is passed over, because they will leave this road at once Mouth region domain.Then, crossing region will be entered by being filtered out from remaining node, but be not passed through the vehicle at crossing center, Namely A and C.In the two nodes, check which is connected member node.If being connected member node there are one only, This node is just confirmed as bridging nodes.If there is multiple connected member nodes, the most slow node of breakneck acceleration is as bridging nodes. In fig. 2, if A is unique connected member node, its state is set as bridging nodes.And if A and E all be connected member node, Then screen the most slow conduct bridging nodes of its medium velocity.It can be in crossing region when longer this is because being selected this node Between.
If bridging nodes will sail out of crossing region, a substitute node is begun look for.It is every that bridging nodes are calculated first A neighbours' connected member node passes through the time t used in the region of crossroadcross, then by tcrossWith green light remaining time tremPhase Compare,
trem=tg-[(tamodtC)-(tC-tg)] (1)
Wherein taFor the time used in node exterior traffic lamp, tCFor the period of traffic lights, tgFor green light stage duration.When trem＞ tcrossWhen, bridging nodes can select tcrossMaximum neighbours' connected member node is as next bridging nodes.Work as trem≤ tcrossWhen, it can select a connected member node being parked at red light.Can ideally select one from crossroad region Nearest node.
2. the access assessment of section network
Network access assessment in section refers to weighing the sensible situation in a section, this stage will be finally in intersection Bridging nodes generate routing table, routing table includes to be connected with the intersection the access value in section and the volume in each section Number.It can determine whether bridging nodes can be connected with destination node according to routing table.Once there is connected member node quilt at crossing Bridging nodes are selected as, then start the access assessment of road network.First a section probe data packet is generated by bridging nodes (Road Section Detect Packet, RSDP) and to all connected member sections in all sections being connected with intersection Point broadcast.The connected member node for receiving RSDP starts to calculate link existence duration between neighbours' connected member node, and most The network that whole section is calculated eventually is access.
Table 1 is the format of RSDP data packets, and wherein intersection ID is where generating the bridging nodes of this RSDP data packet Crossing ID, timestamp be RSDP generate time, section network it is access for according to the access model of network (see embodiment 2) calculate gained section network access value, section node hop count be thus intersection to the path of its neighbours intersection Hop count.
The format of 1 RSDP data packets of table
3. center crossing is screened
To solve the problems, such as local optimum of the conventional routing protocols in City scenarios under road structure complex environment, the present invention The concept at introducing center crossing.Whole network is divided into multiple blocks first, and its Center Road mouthful is screened for each block. In the present invention, the block that each block is one 3 × 3, in order to prevent RSDP data packets from excessively causing network storm.
Center crossing is defined as the most intersection of the crossing number communicated therewith in block.Center crossing has block The path of arbitrary node in this block is known at the overview of network topology, that is, center crossing.Therefore, using Center Road Mouth can overcome the problems, such as local optimum caused by road structure complexity in urban road.It, can when to destination node transmission data packet First to find the center crossing of destination node place block, giving the bridging nodes in this center crossing data packet can find To the optimal path of destination node.
The center crossing screening that the present invention provides is the process of a distributed dynamic, detailed process such as Fig. 3 to Fig. 6 institutes Show.Each bridging nodes appoints crossing centered on the intersection where it first, and ρ=0 is arranged, that is, this center The crossroad number that crossing is connected is 0.Then each connected member starts to exchange RSDP data packets.Exchanging the first round After RSDP data packets, bridging nodes can update the ρ at its affiliated center crossing according to REP values.RSDP is exchanged at second Before, bridging nodes route it table and are put into RSDP (use alternative area) first, then the connected member section into section again Point broadcast.After receiving RSDP, bridging nodes can be compared with the routing table and ρ of oneself, if ρ is more than itself ρ, in Heart road mouth is set as sending the intersection where the bridging nodes of this RSDP.Fig. 4 is each after being second of exchange RSDP The bridging nodes of crossing routing table state, C, the intersections G and I specify crossing centered on B, H and F, while D, F and H then specify E Centered on crossing.To A and B, the ρ values of the two intersections are identical, but because A has the larger access value of network, so A quilts It is selected as center crossing.Same process can be executed in fig. 5 and fig., but in the process of implementation, only there are new route table items Or routing table entry can be just added in REP when being updated, rather than entire routing table is all put into, can effectively it subtract in this way Small REP data package sizes.Final E will be chosen as center crossing.
For irregular extensive road topology, multiple blocks can be divided into radius l, then according to above Algorithm carries out the screening of center crossing.When the size of l is arranged, it is contemplated that the convergence time and routing cost of algorithm.If setting A prodigious l value is set, is represent comprising many crossings in a block, under this l value, the expense for establishing routing is smaller, But routing algorithm convergence rate can be very slow.If the value of opposite l is smaller, center crossing screening time is shorter, and Route establishment is then The consumption longer time.
4. Route establishment
When establishing routing, source node firstly generates " Route Query " (RQ) data packet, contains source node ID, the geographical location of destination node ID and destination node.RQ is forwarded to nearest bridging nodes by source node first.Bridging nodes After receiving RQ, start to check whether this destination node is included in the routing table of oneself.If so, bridging nodes are built " Route Result " (RR) data packet, RR contain routing of the source node to destination node.If in bridging nodes routing table not Including destination node, then be forwarded to center crossing by RQ, the bridging nodes at the crossing of center then first check destination node whether In this block, and if it exists, filter out the logical of destination node using Bellman-ford algorithms (being known in the art algorithm) It is route up to property highest, and adds it to and send back source node in RR.If being not present, source node is calculated to this bridging nodes Access highest path entered to be added in RQ；Then forwarding RQ arrive its neighbours crossroad, persistently this process until RQ data packets reach destination node.
In actual scene, for source node to destination node there may be a plurality of routing, destination node will receive multiple RQ numbers According to packet, in this case, destination node is stored in all routings in its routing table, compares the access of each routing link Value is chosen access highest routing, and is added in RR, and RR is beamed back source node.After source node receives RR packets, just Establish the access highest routing of destination node.
5. testing
(1) simulated environment
The emulation platform of the present invention still uses Veins frames to carry out traffic and network association emulation, and emulation data are to adopt With TAPASCologne large-scale datasets.The data set is to acquire Cologne, Germany urban district in OpenStreetMap databases True road topology in 400 sq-kms, and accurate vehicle mobility model is combined according to vehicular movement information, generation Include the extensive vehicle motion track of more than 70 ten thousand nodes.TAPASCologne data sets are the maximums of current retrievable scale Data set has the characteristics that ground map combining is wide, road topology is complicated, magnanimity vehicle node, the car networking suitable for City scenarios The emulation of large scale network.Since TAPASCologne data sets cover 24 hours network informations, data volume is huge, in order to Simulation efficiency is improved, present invention experiment only intercepts some periods, and counts the vehicle node number of corresponding period, grinds Large scale scene lower network performance indicator is studied carefully in true City scenarios with the variation relation of vehicle node number.
(2) experimental method
Under the large scale network scene of the cities TAPASCologne, two gateways of setting as source node and destination node, Respectively using the extensive routing mechanism AMRP of the car networking provided herein based on access model and other two kinds of routing algorithms GyTAR and GPSR structures are route, then the network index of three kinds of routing mechanisms of comparative analysis:
(1) data packet delivery fraction (Packet delivery ratio PDR)：Successfully arrive at the number-of-packet of destination node Mesh accounts for the ratio that data packet sends total number.
(2) average end-to-end delay (End-to-End delay E2ED):Data packet successfully arrives at purpose section from source node Time needed for point.
(3) routing cost (Routing overhead)：Terminate from starting to build to be routed to data packet and transmit, controls data Packet accounts for ratio shared in source data packet.
6. averagely end-to-end delay
Fig. 7 is the functional relation of average end-to-end delay and traffic density.It is clear to by figure, with being incremented by for traffic density, Network node is more and more denser, and the average end-to-end delay of tri- kinds of Routing Protocols of AMRP, GyTAR and GPSR is in large scale network field It all has a declining tendency under scape.This is because incremental node density can improve the access of road network, and then it route association It is access also stronger to discuss the routing finally constructed.By figure it is also seen that AMRP is because using the access model of the invention provided And routing is calculated by the topology of whole network, and GyTAR relies solely on Local network topology and calculates routing, will produce office Portion's greatest problem, GPSR, which is relied primarily on, carries forwarding mechanism progress data packet transmission, so the average end-to-end delay one of AMRP It is directly lower than GyTAR and GPSR, when traffic density is 0.03/meter, the average end-to-end delay of GPSR be 5 times of AMRP it More, the average end-to-end delay of GyTAR is 1.47 times of AMRP.
Fig. 8 illustrates variation tendency of the average end to section delay with data package transmission velocity, it can be seen that three kinds of routing associations The average end-to-end delay of view all increases with being incremented by for data package transmission velocity.GPSR is led to using forwarding mechanism is carried Letter, the mechanism without prevention data congestion completely, cause its average end-to-end delay in the high when ratio AMRP of data packet transmission rate and GyTAR will be high.GyTAR transmits data using greedy algorithm, and in contrast, the AMRP based on access model is in forwarding number According to when by car networking be connected to base, the number of nodes for participating in forwarding data packet is less, and can accurately calculate the access of routing, Data packet is transmitted in the more sensible routing of selection, so AMRP has the average end-to-end delay lower than GyTAR, through statistics, AMRP It is lower by 30% than the average end-to-end delay of GyTAR under different data packet transmission rates.
7. data packet delivery fraction
It is clear to by Fig. 9, the data packet delivery fraction of three kinds of algorithms all increases with being incremented by for traffic density on the whole, this is just Meet theory analysis, it is sensible between node and node because nodes are dense particularly with AMRP and GyTAR Probability is higher, and correspondingly packet loss of the data packet in transmittance process is lower.When node density is excessive, radio interference and section Point conflict increase, this explains GyTAR and GPSR algorithms node density be 0.08/meter when data packet delivery fraction start to drop Low reason.And AMRP transmits data because using the connection base having had been built up, the node for participating in forwarding data is few, even if Data packet delivery fraction is still increasing in the case that node is dense.
Figure 10 is variation relation figure of the data packet delivery fraction with data packet transmission rate.It can be obtained from the figure that going out, the data of AMRP Packet delivery fraction is highest in these three routing algorithms, is higher by GyTAR 26%, GPSR 85%.Mainly since AMRP is based on Car networking is connected to base structure routing, and routing screening concentrates on connected member node, and the node for participating in communication is less, and Communication Jamming is compared Other two kinds of routing algorithms are low.And AMRP calculates access highest routing, established routing tool according to access model There is stronger stability, link disconnection probability is low, can accordingly bring lower data packetloss rate.GyTAR and GPSR is not examined Access and excessive in the GPSR data packet transmittance process carrying forwarding for considering routing also results in its high packet loss.
8. routing cost
Figure 11 is the routing cost of three kinds of routing algorithms with the variation diagram of traffic density.It is clear to by figure, with traffic density Be incremented by, the routing cost of AMRP and GyTAR show increasing trend.This is because AMRP and GyTAR build route when to Node sends control packet, including RSDP is used for carrying out section detection, access with the network for obtaining road, and obtains in section Nodal information, also RQ and RR be used for building routing, these controls are wrapped can send in the route construction stage to each node, vehicle Number increase necessarily causes what control wrapped to increase.However AMRP is integrally smaller than the routing cost of GyTAR, this is primarily due to AMRP is route based on access model construction, fully considers the stability of network, and the link ratio GyTAR in routing has higher Stability compares GyTAR, the route maintenance expense smaller needed for AMRP.GPSR then by calculate with neighbor node it is European away from Nearest neighbor node carries out carrying forwarding with a distance from filtering out, and any control packet is not needed, so its routing cost is always 0。
Figure 12 is the routing cost of three kinds of routing algorithms with the variation diagram of data package transmission velocity.As data packet sends speed Rate is incremented by, and the routing cost of AMRP and GyTAR also show increasing trend.This is because the traffic volume of control packet is in network Node density is related, so when traffic density is constant, when data packet transmission rate is incremented by, the successful delivery rate of data packet is dropping It is low, then routing cost will rise accordingly.When data package transmission velocity is higher, the data packet success of AMRP ratios GyTAR Delivery ratio is high, so the routing cost of AMRP should be lower than GyTAR at this time, this point is also proved from figure.
Innovative point
The present invention proposes the access method for routing of car networking under a kind of urban road complex scene, fast to contribute to Speed realizes car networking network interconnection intercommunication and keeps relative stability, and the number such as traffic accident information, real time media data is effectively ensured It is believed that the real-time stabilization transmission of breath.The problems such as road structure is complicated in City scenarios, existing most of Routing Protocols are all bases In greedy algorithm and the stateless route agreement of carrying forwarding or periphery forwarding mechanism, that is, in repeating process, source node The position for only focusing on destination node and node adjacent thereto, when data packet will be forwarded to next-hop, just according to certain routing Measurement makes routing decision, however builds routing in extensive car networking by these algorithms and will appear local optimum phenomenon, To improve data packet end-to-end delay, finally because transmission time is long, the no longer effective property of data packet and increase packet loss, finally Cause network not sensible.In view of the above problems, providing the access method for routing of urban road scene, this method is based on mutual the present invention Join the connection base component of intercommunication (filed in 11 days October in 2017 of the inventors such as Cheng Jiujun《Car networking large scale network interconnects The connection base component building method of intercommunication》(applicant：Tongji University, number of patent application 201,710,397,807 7), the patent Shen The technical solution that please be provided is：Consider car networking network node redundancy properties, gives a kind of car networking network topology structure, i.e., Car networking is connected to base, and gives connection base building method using heuritic approach.), the concept at central crossbar crossing is given, To obtain larger range of network topological information, more preferably data transmission routing can be thus constructed.The method for routing can It fast implements car networking network interconnection intercommunication to be directed to urban road complex scene and keeps relative stability, handed over to be effectively ensured The real-time stabilization transmission of the data informations such as logical accident information, real time media data.
Embodiment 2
The present embodiment is based on the connection base component to interconnect (filed in 11 days October in 2017 of the inventors such as Cheng Jiujun 《The connection base component building method that car networking large scale network interconnects》(applicant：Tongji University, number of patent application 201710397807 7), give car networking large scale network access model.
One, the access analysis of car networking based on connection base.
Currently invention addresses car networking large scale network is access, because car networking connection base has only screened several connected member sections Point constitutes connection base, reduces network node redundancy, and network size is smaller compared to former network, and does not influence the normal of network Operation, based on the research of connection base, network is access can effectively reduce access computation complexity.
Here, for any given network G, wherein vehicle node integrates as V, link set E.G can regard as by network section Point device, the communication link of connecting node and the dynamic topological structure that is made of node and link form.Connect if network G exists Logical base It is connected member node and ordinary node G points, and the communication full powers between ordinary node give and dominate its company Logical member node processing, then any two points v in networki, vjBetween network it is access be vi, vjBetween connection base it is access again In addition connection base and two node adjacent links it is access, that is,：
WhereinFor vi, vjBetween connection base.
When centre forwarding connected member node is more, it is connected to the access of link between base and node and can be ignored, V at this timei, vjBetween network access be：
So the access of network G is approximately equal to its connection baseIt is access.Namely：
The access of network depends primarily on network connectivty and network stabilization, and the present invention analyzes it from these two aspects The influence access to network：
Network connectivty Co
In car networking, node wireless transmission range is limited, ability phase when being in corresponding communication radius range between node Even.Network connectivty has reacted the connected state between terminal, is the key factor for determining that network is access.
Network stabilization St
Due to the high-speed mobility of vehicle node, between node the transmission range moment change, old path disconnects and new route is built Vertical frequently alternating, network topology dynamic change is quickly.Network stabilization is to ensure that data can stablize the basis of transmission.
The connection base of network G is sought respectivelyConnectivityAnd stabilityIt can be obtained the access of G Ac (G), just like minor function relationship：
Two, the access model based on connection base
(1) joint movements model
For car networking network, the movement of vehicle has certain regularity, to study the access of mobile status lower network It needs network node motion model.Present invention assumes that the topological structure of network does not become within a shorter time Change, continuous movement discretization is described into the change of topological structure at any time using-half Markov mobility model SGM of smooth Gaussian Change.
For a given car networking network, be cut into equal short time interval Δ t first, in this way continuous time Each tempon is represented by tk=Δ t+tk-1；K=1,2 ..., n.Assuming that in network all nodes according to SGM move and The initial position of node is it is known that node viInitial position be (xi0,yi0).V can so be obtainediIn kththThe position at a moment (xik,yik) be：
Wherein VeikIndicate viIn kththThe speed at a moment, θikIndicate viThe direction of motion.In time interval Δ t, I Movement direction of nodes is regarded as it is constant.The position of node any moment in a network in this way can be solved by above formula, if Δ t Value is smaller, and the node location solved is more accurate.
It can calculate node v by formula (5)iAnd vjIn kththThe Euclidean distance at a moment is：
According to the Euclidean distance D between nodeij(tk) and node communication radius can be between decision node connection status, aijIndicate that the connection status between node, R indicate communication radius, then
According to kththThe time-varying adjacency matrix that connection status between moment arbitrary node can build network is：
A(tk) indicate kththThe whole network period is divided into n time interval by the network topology structure at moment, and network is every The adjacency matrix at a moment can be acquired according to above formula, and then understand the network connection state at each moment, be moved for establishing The access model of state lower network is significant.
(2) network connectivty
Connection base is to cover the topological structure of whole network, and ensure that each node only has one to the distance of connected member It jumps, the car networking connectivity assessment based on connection base is exactly to assess the connectivity for being connected to base.
Define 1 communication path：The alternate sequence w=v of nodes and side0e0v1e1v2e2…vkekvk+1For access Diameter.Wherein k is the hop count of communication path.If v in communication path w0=vk+1, then communication path is referred to as connected ring.
Figure 14 is communication path schematic diagram.In network a, v1And v6Between hop count be 3 the number of communication path be 4, point It is not：v1v2v3v6, v1v2v5v6, v1v4v3v6, v1v4v5v6.And in network b, v1And v5Hop count be 3 communication path number For：v1v2v4v5And v1v3v4v5.Network a contains more communication paths than network b, even if partial link disconnects in network, network A can still keep being connected to.Therefore, the connectivity ratio b of network a is strong.Therefore it may be concluded that it is alternative between nodes The number in path determines network connectivty.It is possible thereby to infer, the connectivity of the car networking based on connection base depends on connection The number size of communication path in base.For network G, connection base is The number of middle communication path is arbitrary connected member pairBetween hop count be k communication path numberSum.But in large scale networkDifficulty in computation it is larger, meter The complexity of calculation can be very high, and the number for being connected to loop can also measure the number of communication path in network, and computation complexity will It can be lower.Connection loop number be：
WhereinExpression hop count is k, and starting point isConnection loop number, NkIndicate all hop counts in connection base For the number of the connection loop of k.Sum is bigger, illustrates that the number of backup path is more, the connectivity for being connected to base is stronger.But When calculating Sum, the case where side and node repeat is contained, final Sum values may be intended to infinity.For arbitrary connected member pairThe more connection loop of hop count and the repetitive rate that other loops calculate are higher, it should reduce the connection loop more than hop count In the ratio that Sum is accounted for, therefore to NkIt is weighted, i.e.,：
Wherein λiTo be connected to the characteristic value of base adjacency matrix.
Found out by above formula, Sum ' can be calculated by the characteristic value of adjacency matrix and be obtained.But when network size is larger, Sum ' It will be a prodigious number.For the sake of convenient, by Sum ' carry out logarithm operations, connection base is obtainedConnectivity value
Formally see,It is proportional relation with Sum ',It is connection base adjacency matrixAll features The special average value of value, has only passed through an exponent arithmetic and logarithm operation.
After the motion model of node, kth can be acquired according to formula (11)thThe network connectivty at momentEqually network cycle is discrete for after multiple time interval Δ t, the network-in-dialing degree at each moment can be asked Solution.Δ t values are smaller, and obtained network connectivty is more accurate.
Specific network connectivty calculates step should be as follows：
1. structure connection baseThe initial position of acknowledging time interval of delta t, the communication radius of node, connected member, then root According to connection status of the position information confirming of connected member between it, and then build the adjacency matrix of initial time
2. according to kththThe adjacency matrix at momentIt calculatesCharacteristic value, then according to formula (11) calculate Connectivity
3. if network life cycle is also not finished, (k+1) is calculated according to SGMthThe position of each connected member in moment network It sets, and establishes adjacency matrix at this timeConnectivity is recalculated, until whole network life cycle terminates.
(3) network stabilization
Connection base be cover whole network topological structure, by be connected to base dynamic network topology stability analysis come The stability for analyzing whole network, the network stabilization Journal of Sex Research based on connection base are substantially the stability for studying connection base.
Car networking is that its topological dynamics, link are established and disconnect frequent, the link longevity with the maximum difference of traditional network Life can weigh the stability of a link.In real network, directly by sent the message between node confirm node it Between link available duration there are certain error and interference, Kalman Filter Technology can remove noise reduction truthful data.Therefore The present invention is based on node motion models and Kalman filtering to calculate the link service life, and then derive the stability of network.
For any two node vAAnd vB, enable VeAkAnd VeBkV is indicated respectivelyAAnd vBIn kththThe speed at moment.According to SGM models can be obtained node in kth+1thThe speed at moment.Further, node vAAnd vBAt (k+1)thThe speed relatively at moment Degree can be calculated as follows：
Wherein, VeRk=VeAk-VeBk, yRk=yAk-yBk, it is clear that yRkIt is 0 for mean value, variance σR 2=2 σ2Gaussian random Variable.
In order to calculate the link service life, following one hop link model, such as Figure 15 are considered.Two vehicle node vAAnd vBAccording to figure It moves in a network.Although vAAnd vBIt is moved all in accordance with SGM, it is assumed that vAIt is static, vBAccording to the relative velocity movement in formula (12). Network is regarded as a coordinate system, vehicle vAIn origin, as long as vehicle vBBetween coordinate (- R, 0) to (R, 0), vAAnd vBIt builds Vertical link.As vehicle vBInto vATransmission radius when, vABeacon message is will receive, and begins to use Kalman filtering meter Calculate vBDisplacement distance and and vBBetween relative velocity, and then calculate two nodes between the link service life.
Kalman filter is realtime recurrent algorithm, is used for carrying out optimal estimation, algorithm to the state variable of dynamical system Including two important equations：The state equation and observational equation of system.
State equation is provided first, and in the above two node links model, state variable includes vehicle vBOperating range And vBMovement speed, according to SGM models, the two variables are all related to previous moment, therefore (k+1)thThe state at moment Process equation is：
Wherein xk+1And xkRespectively refer to vehicle vBAt (k+1)thAnd kthThe position at moment；yRkAnd yxkIt is independent incoherent Gaussian random variable, mean value are 0, and variance is respectively σR 2And σx 2.The matrix form of state equation is：
Process equation can be regarded as a kind of common version X by above formulak+1=TXk+wk, Xk+1For state variable vector, T It is kththThe state-transition matrix at moment, wkBe mean value it is 0 for observation noise, covariance matrix is the Gaussian random variable of Q. Can obtain Q according to formula (14) is：
Observational equation is uniformly provided by Kalman filter, kththThe observational equation at moment is Zk=HXk+μk, wherein ZkIt is to see Direction finding amount, H are observing matrix, μkIt is observation noise, it is that mean value is 0, and covariance matrix is the gaussian variable of R.
Then it enablesFor XkPrior estimate,For XkPosterior estimator, Pk -And PkRespectively priori and posteriority covariance Matrix.Value of the initial value namely in k=0, diagonal element should be king-sized value at this time, and off-diagonal element Should be 0[65].ThereforeInitial value be：
Renewal process is updated and observed according to the time of Kalman filter algorithm, we can be to the link service life of network It is estimated.As shown in Figure 15, as vehicle vBInto vACommunication range within when, vAV can be passed throughBIt sends over Beacon information detections are to vB, vehicle v can be obtainedAWith vBAt (k+1)thThe link life estimation value at moment is：
Wherein
For network G, connection base is Network stabilization by all links of connection base weighted average Life Table Show, the weights of each of the links are the ratio for being connected to the communication path number comprising this link in base and accounting for all communication paths.Then net Network stabilityFor：
Wherein EiIndicate i-th side in connection base, niTo include link E in connection baseiCommunication path number,For even The number on side, n in logical baseiFor the number of all communication paths in connection base.It is steady to give the measurement network G a certain moment Qualitative calculation formula, according to formula (18) and Kalman filter algorithm, it can be deduced that any time network in dynamic network Stability value.
(4) the access model of network
Car networking network is access, i.e., fast implements network interconnection intercommunication according to current task, and keeps stablizing shape in real time State is for the purpose of completing current task.Access network refers to the network that i.e. connection is stablized again, and the access model of network is then to use To integrate the connectivity and stability of weighing target network.Because car networking large-scale network node is in large scale, it is directly analyzed Access difficulty is high, and founding the access model of network based on car networking connection capital construction can smaller network size.For in G arbitrary two A node vAAnd vB, access network is vAAnd vBBetween connection base net network it is access, then network G is access, be It is connected to the access of base, it can be deduced that the access calculation formula of network：
WhereinFor the connection base of network G, CothFor network connectivty threshold value, according to the car networking application program of source vehicle Depending on demand.Network-in-dialing is that access network first has to the condition met, and therefore, the access model of network should first judge net Whether network connectivity meets application demand, and when connectivity value is more than its threshold value, then network meets application program requirement, at this time network The access product for connectivity and stability.When network connectivty is less than connectivity threshold value, access is 0.

Claims (3)

1. the access method for routing of base is connected under a kind of urban road scene based on car networking, which is characterized in that specific method Include the following steps：
One, bridging nodes are screened
If bridging nodes will sail out of crossing region, a substitute node is begun look for；It is each adjacent that bridging nodes are calculated first It occupies connected member node and passes through time t used in the region of crossroadcross, then by tcrossWith green light remaining time tremIt compares Compared with,
trem=tg-[(tamodtC)-(tC-tg)] (1)
Wherein taFor the time used in node exterior traffic lamp, tCFor the period of traffic lights, tgFor green light stage duration；Work as trem＞ tcrossWhen, bridging nodes can select tcrossMaximum neighbours' connected member node is as next bridging nodes；Work as trem≤tcross When, it can select a connected member node being parked at red light；Can ideally select one it is nearest from crossroad region Node；
Two, the access assessment of section network
Network access assessment in section refers to weighing the sensible situation in a section, this stage will be finally in the bridge of intersection It connects node and generates routing table, routing table includes be connected with the intersection access value in section and the number in each section； It can determine whether bridging nodes can be connected with destination node according to routing table；Once there is connected member node to be chosen as at crossing Bridging nodes then start the access assessment of road network；First from bridging nodes generate a section probe data packet and to All connected member node broadcasts in all sections being connected with intersection；The connected member node for receiving RSDP starts to calculate and neighbour The link existence duration between connected member node is occupied, and the network that whole section is finally calculated is access；
Three, center crossing is screened
The concept at introducing center crossing；Whole network is divided into multiple blocks first, and its center is screened for each block Crossing；Center crossing is defined as the most intersection of the crossing number communicated therewith in block；Center crossing has block net The path of arbitrary node in this block is known at the overview of network topology, that is, center crossing；Number is sent to destination node When according to packet, the center crossing of block, the bridge joint section in this center crossing is given data packet where can first finding destination node Point can find the optimal path of destination node；
The screening of center crossing is the process of a distributed dynamic:Each bridging nodes appoints the intersection where it first Centered on crossing, and ρ=0 is set, that is, the crossroad number that this center crossing is connected is 0；Then each connection Member starts to exchange RSDP data packets；After having exchanged the RSDP data packets of the first round, bridging nodes can be according to REP values more The ρ at its new affiliated center crossing；Before second exchanges RSDP, bridging nodes route it table and are put into RSDP first (to be made With alternative area), then connected member node broadcasts into section again；After receiving RSDP, bridging nodes can be with oneself Routing table and ρ are compared, if ρ is more than itself ρ, where center crossing is set as sending the bridging nodes of this RSDP Intersection；
For irregular extensive road topology, multiple blocks can be divided into radius l, then according to algorithm above The crossing screening of carry out center；When the size of l is arranged, it is contemplated that the convergence time and routing cost of algorithm；If setting one A prodigious l values represent comprising many crossings in a block, and under this l value, the expense for establishing routing is smaller；On the contrary If the value of l is smaller, center crossing screening time is shorter, and Route establishment then consumes the longer time；
Four, Route establishment
When establishing routing, source node firstly generates a RQ data packet, contains source node ID, destination node ID and target section The geographical location of point；RQ is forwarded to nearest bridging nodes by source node first；Bridging nodes start to check after receiving RQ Whether this destination node is included in the routing table of oneself；If so, bridging nodes build RR data packets, RR contains source node and arrives The routing of destination node；If not including destination node in bridging nodes routing table, RQ is forwarded to center crossing, center crossing Whether the bridging nodes at place then first check destination node in this block, and if it exists, utilize existing Bellman-ford algorithms The access highest routing of destination node is filtered out, and adds it to and sends back source node in RR；If being not present, count The access highest path for calculating source node to this bridging nodes is entered to be added in RQ；Then forward RQ to its neighbours' four crossway Mouthful, continue this process until RQ data packets reach destination node.
2. the access method for routing of base, feature are connected under urban road scene as described in claim 1 based on car networking It is, in step 2, routing table is the format of RSDP data packets, and wherein intersection ID is the bridge joint for generating this RSDP data packet The ID at the crossing where node, timestamp are the time that RSDP is generated, and section network is access for according to the access model of network The gained access value of section network is calculated, section node hop count is thus jump of the intersection to the path of its neighbours intersection Number.
3. the access method for routing of base, feature are connected under urban road scene as claimed in claim 2 based on car networking It is, the access model of the network, for a kind of access model of large scale network being connected to base based on car networking, car networking Network is access, i.e., fast implements network interconnection intercommunication according to current task, and keeps stable state to complete as predecessor in real time For the purpose of business；Access network refers to the network that i.e. connection is stablized again, and the access model of network is then to weigh target for comprehensive The connectivity and stability of network；For any two node v in GAAnd vB, access network is vAAnd vBBetween be connected to base Network it is access, then the access of network G is then that it is connected to the access of base, it can be deduced that the access calculating of network is public Formula：
WhereinFor the connection base of network G, CothFor network connectivty threshold value, according to the car networking application requirement of source vehicle Depending on.
CN201810488153.3A 2018-05-21 2018-05-21 Accessibility routing method based on Internet of vehicles communication base in urban road scene Active CN108667668B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810488153.3A CN108667668B (en) 2018-05-21 2018-05-21 Accessibility routing method based on Internet of vehicles communication base in urban road scene

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810488153.3A CN108667668B (en) 2018-05-21 2018-05-21 Accessibility routing method based on Internet of vehicles communication base in urban road scene

Publications (2)

Publication Number Publication Date
CN108667668A true CN108667668A (en) 2018-10-16
CN108667668B CN108667668B (en) 2020-12-29

Family

ID=63776315

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810488153.3A Active CN108667668B (en) 2018-05-21 2018-05-21 Accessibility routing method based on Internet of vehicles communication base in urban road scene

Country Status (1)

Country Link
CN (1) CN108667668B (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109067655A (en) * 2018-10-18 2018-12-21 长安大学 A kind of car networking mixed logic dynamic forwarding strategy based on name data network
CN109996174A (en) * 2019-04-16 2019-07-09 江苏大学 A kind of real-time methods of marking in section towards the routing of vehicular ad hoc network content
CN111260947A (en) * 2020-02-20 2020-06-09 北京航空航天大学 Road network reachable range analysis method based on flooding algorithm
CN111813884A (en) * 2020-07-03 2020-10-23 李丹丹 Configuration method and configuration system for highway emergency rescue goods and materials

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106792977A (en) * 2017-01-25 2017-05-31 南通大学 Double cluster head method for routing based on traffic density in vehicular ad hoc network
WO2017133467A1 (en) * 2016-02-04 2017-08-10 中兴通讯股份有限公司 Method for implementing internet of vehicles service and localized network architecture
CN107181640A (en) * 2017-05-03 2017-09-19 同济大学 The acquisition methods of car networking connectedness index under a kind of urban road scene
CN107231408A (en) * 2017-05-03 2017-10-03 同济大学 The building method for the connectivity modeling predicted in urban road based on vehicle-state

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2017133467A1 (en) * 2016-02-04 2017-08-10 中兴通讯股份有限公司 Method for implementing internet of vehicles service and localized network architecture
CN106792977A (en) * 2017-01-25 2017-05-31 南通大学 Double cluster head method for routing based on traffic density in vehicular ad hoc network
CN107181640A (en) * 2017-05-03 2017-09-19 同济大学 The acquisition methods of car networking connectedness index under a kind of urban road scene
CN107231408A (en) * 2017-05-03 2017-10-03 同济大学 The building method for the connectivity modeling predicted in urban road based on vehicle-state

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
JIUJUN CHENG: "Connectivity Modeling and Analysis for Internet of Vehicles in Urban Road Scene", 《IEEE》 *

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109067655A (en) * 2018-10-18 2018-12-21 长安大学 A kind of car networking mixed logic dynamic forwarding strategy based on name data network
CN109996174A (en) * 2019-04-16 2019-07-09 江苏大学 A kind of real-time methods of marking in section towards the routing of vehicular ad hoc network content
CN109996174B (en) * 2019-04-16 2020-12-18 江苏大学 Road section real-time scoring method for vehicle-mounted self-organizing network content routing
CN111260947A (en) * 2020-02-20 2020-06-09 北京航空航天大学 Road network reachable range analysis method based on flooding algorithm
CN111813884A (en) * 2020-07-03 2020-10-23 李丹丹 Configuration method and configuration system for highway emergency rescue goods and materials

Also Published As

Publication number Publication date
CN108667668B (en) 2020-12-29

Similar Documents

Publication Publication Date Title
Liu et al. Distributed cooperative reinforcement learning-based traffic signal control that integrates V2X networks’ dynamic clustering
CN108667668A (en) The access method for routing of base is connected under a kind of urban road scene based on car networking
Hadded et al. A multi-objective genetic algorithm-based adaptive weighted clustering protocol in vanet
KR100673839B1 (en) Mobile Ad Hoc Network, Method for Node Advertisement in a Mobile Ad Hoc Network, and Method for Node Adevertisement by a Group of Mobile Ad Hoc Network
Vidhale et al. Performance analysis of routing protocols in realistic environment for vehicular Ad Hoc networks
Luo et al. A mobile infrastructure based VANET routing protocol in the urban environment
KR100635488B1 (en) Temporal transition network protocol in a mobile ad hoc network
CN113382382B (en) Vehicle ad hoc network routing method based on fuzzy logic optimization strategy
Dorle et al. Evaluation of multipath, unipath and hybrid routing protocols for vehicular Ad hoc networks
Inam et al. A novel protocol for vehicle cluster formation and vehicle head selection in vehicular ad-hoc networks
Elgaroui et al. New routing protocol for reliability to intelligent transportation communication
CN111683384A (en) Network optimization method for realizing dynamic weighting of communication link by artificial intelligence
CN104093182A (en) Method for acquiring a plurality of reliable communication paths based on field intensity in multi-layer wireless network
Härri et al. On meaningful parameters for routing in vanets urban environments under realistic mobility patterns
CN109743790A (en) A kind of high dynamic mobile ad-hoc network change in topology cognitive method
CN108650141B (en) Large-scale network accessibility model design method based on Internet of vehicles communication base
Coutinho et al. Performance of routing protocols for VANETs: A realistic analysis format
Christakos Sensor networks applied to the problem of building evacuation: An evaluation in simulation
Safrianti et al. Performance analysis of DSDV, AOMDV and ZRP routing protocols application simulation in pekanbaru vehicular Ad Hoc network (VANET)
Poojary et al. Ant colony optimization routing to mobile ad hoc networks in urban environments
Hong et al. Topology Change Degree: a Mobility Metric Describing Topology Changes in MANETs and Distinguishing Different Mobility Patterns.
Krishnakumar et al. QOS enabled data dissemination in hierarchical VANET using machine learning approach
Mudigonda et al. Evaluation of a methodology for scalable dynamic vehicular ad hoc networks in a well-calibrated test bed for vehicular mobility
CN106850428B (en) Opportunistic routing protocol method for perceiving link quality based on 802.11
Nurwarsito et al. Analysis of OLSR Routing Protocol Performance Based on Gauss-Markov Mobility and Random Walk in Mobile Ad-Hoc Network (MANET)

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