CN110784852B - V2V routing method based on online link duration prediction - Google Patents
V2V routing method based on online link duration prediction Download PDFInfo
- Publication number
- CN110784852B CN110784852B CN201910978117.XA CN201910978117A CN110784852B CN 110784852 B CN110784852 B CN 110784852B CN 201910978117 A CN201910978117 A CN 201910978117A CN 110784852 B CN110784852 B CN 110784852B
- Authority
- CN
- China
- Prior art keywords
- node
- link duration
- duration prediction
- neighbor
- data
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000004891 communication Methods 0.000 claims abstract description 36
- 230000005540 biological transmission Effects 0.000 claims abstract description 27
- 230000033001 locomotion Effects 0.000 claims description 29
- 238000012549 training Methods 0.000 claims description 25
- 238000013528 artificial neural network Methods 0.000 claims description 15
- 238000006073 displacement reaction Methods 0.000 claims description 12
- 230000006855 networking Effects 0.000 claims description 11
- 230000001133 acceleration Effects 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 claims description 4
- 230000010365 information processing Effects 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 9
- 230000008569 process Effects 0.000 description 7
- 239000011159 matrix material Substances 0.000 description 6
- 230000004913 activation Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 3
- 238000002372 labelling Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 239000008186 active pharmaceutical agent Substances 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 231100000279 safety data Toxicity 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- 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]
- H04W4/46—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P] for vehicle-to-vehicle communication [V2V]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Traffic Control Systems (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention belongs to the technical field of information processing, and particularly relates to a method, a system and a device for selecting a V2V route based on online link duration prediction, aiming at solving the problem of low success rate of data transmission caused by unstable route selection in Internet of vehicles communication. The method of the system comprises the steps that when a first node requests a data packet to a second node, whether a neighbor table of the second node contains the first node is judged; acquiring characteristic data based on corresponding data difference values in the data packets of the second node and each adjacent node; obtaining a link duration prediction value through a link duration prediction model based on the characteristic data; selecting an adjacent node corresponding to the maximum reachability coefficient as a next hop node based on the link duration predicted value; and circularly judging whether the neighbor table of the next hop node contains the first node or not until the data packet is sent to the first node. The invention improves the stability of routing in the communication of the Internet of vehicles and improves the success rate of data transmission.
Description
Technical Field
The invention belongs to the technical field of information processing, and particularly relates to a V2V routing method, system and device based on online link duration prediction.
Background
The vehicle networking technology is a technology for sharing road traffic information and environmental information based on communication between vehicles or between the vehicles and road side infrastructure, and further enhancing perception of the vehicles on the traffic information and coordination and cooperation between traffic participation elements. The Vehicle networking can be mainly divided into two forms of V2V (Vehicle-to-Vehicle) for communication between vehicles and V2I (Vehicle-to-Infrastructure) for communication between vehicles and Infrastructure, wherein the V2V is mainly used for sharing state information of neighbor vehicles based on Vehicle-mounted sensors and communication units, so that the applications of collision avoidance, accident early warning and the like are realized, and the safety and the efficiency of transportation are improved. V2V does not need to deploy a lot of expensive roadside infrastructure, and vehicle nodes act as sensor units and communication nodes by themselves, and can cover a wider range, thus having more advantages and being widely researched by academia and industry.
Compared with the traditional mobile ad hoc network MANET, the nodes in the vehicle networking network have higher mobility, the network topology changes faster, and the communication links among the nodes have higher instability. In order to ensure the accessibility of safety data and reduce the loss rate of data packets, and further ensure the safety of transportation, researchers carry out a great deal of research and test on routing algorithms in the car networking environment. Currently, routing protocols related to car networking can be mainly classified into topology-based routing, location-based routing, clustering routing, multicast routing, broadcast routing, and the like. The routing protocol which is generally most widely applied and has the best effect is a routing protocol based on location, such as gpsr (greedy Perimeter Stateless routing), which broadcasts a discovery message including location information with a peripheral vehicle node, maintains a neighbor node table, and then selects a next-hop node based on greedy strategies such as nearest or farthest node priority when data needs to be transmitted, thereby transmitting the data to a destination node. However, these protocols do not provide an effective assessment of the stability of the communication link and selecting a node based solely on the current location message is still likely to result in a transmission failure.
In order to improve a routing protocol of the Internet of vehicles and improve the success rate of data transmission, the historical tracks of neighbor nodes and road environment data can be analyzed, the link duration between the neighbor nodes and candidate neighbor nodes is predicted, and therefore the most stable node is selected as the next hop. However, the related methods are mainly based on mathematical models, and they mainly construct a multivariate function based on parameters such as relative position and relative speed through theoretical analysis, and predict the corresponding link duration by inputting the current parameter level. Such methods cannot take personalized features of the vehicle into account, and the assumed road scene is also generally simple and single, such as a straight highway scene, but lacks good robustness in a real and complex scene, and can only predict a short-term link state based on current state data. The artificial neural network is a powerful machine learning method, can fit a complex function system based on various personalized features, and achieves better effects in the fields of classification, prediction and the like. The deep neural network can accurately predict the vehicle displacement state in a longer time in the future based on historical track data, so that a more accurate basis is provided for link state analysis, and the neural network with the deeper layers can obtain higher prediction accuracy through good training. However, the training of the deep neural network not only requires a large amount of data, but also is time-consuming, and it usually takes tens of minutes or even days to train a reliable model under the general problem, and the demand for computing resources is high. In the scene of link prediction of the internet of vehicles, although the method based on deep learning can obtain higher precision, the feasibility is lower, and the practical value is lacked. The breadth learning is used as a new machine learning algorithm, is characterized by transverse expansion of shallow neuron nodes, can quickly learn, has higher generalization capability, and is suitable for intelligent scenes with higher real-time requirements.
Disclosure of Invention
In order to solve the above problems in the prior art, that is, to solve the problem of low success rate of data transmission due to unstable routing in communication of internet of vehicles, a first aspect of the present invention provides a V2V routing method based on online link duration prediction, which is applied to communication between vehicle nodes in the internet of vehicles, and the method includes:
step S100, when a first node requests a data packet from a second node, if a neighbor table of the second node contains the first node, sending the corresponding data packet to the first node; otherwise, executing step S200; the data packet comprises a position, a speed, an acceleration, a motion direction and an MAC address;
step S200, respectively obtaining characteristic data of the second node and each neighbor node based on the difference value of corresponding data in the data packet of the second node and the data packet of each node in the neighbor node group; the neighbor node group comprises one or more neighbor nodes which are adjacent nodes of the second node; the characteristic data comprises distance difference, speed difference, motion direction difference, space similarity and relative displacement;
step S300, based on the characteristic data, obtaining link duration prediction values of the second node and each neighbor node through a link duration prediction model;
step S400, based on the link duration predicted value, obtaining the reachability coefficient between the second node and each neighbor node; selecting a neighbor node corresponding to the maximum reachability coefficient as a next hop node of the second node;
step S500, if the neighbor table of the next hop node contains the first node, the data packet is sent to the first node, otherwise, the steps S200 to S400 are executed in a circulating manner until the data packet is sent to the first node;
the link duration prediction model is constructed based on a width neural network and used for obtaining link duration prediction values of the two nodes according to the characteristic data.
In some preferred embodiments, the link duration prediction model includes a forward link duration prediction model and a reverse link duration prediction model.
In some preferred embodiments, in step S300, "obtaining the link duration prediction values of the second node and each neighboring node through a link duration prediction model based on the feature data" includes:
if the motion direction difference is smaller than 90 degrees, obtaining link duration prediction values of the second node and each neighbor node through a same-direction link duration prediction model according to the characteristic data; otherwise, link duration prediction values of the second node and each neighbor node are obtained through a reverse link duration prediction model.
In some preferred embodiments, in step S400, "obtaining the reachability coefficient between the second node and each neighboring node based on the predicted link duration value" includes:
acquiring the size of the second node data packet and the transmission rate of the second node data packet to each neighbor node to obtain the transmission time of the second node data packet to each neighbor node;
dividing the link duration predicted value with the transmission time to obtain a link value factor of the second node and each neighbor node;
and obtaining the reachability coefficients of the second node and each neighbor node according to the link value factor and the distance difference between the second node and each neighbor node.
In some preferred embodiments, the method of obtaining the reachability coefficients of the second node and the neighboring nodes according to the link cost factor and the distance difference between the second node and the neighboring nodes includes:
wherein AC is a reachability coefficient, alpha is a weight coefficient, VF is a link cost factor, max (d) is the maximum distance difference between the second node and each neighbor node, diAnd i is the subscript value of the neighbor node.
In some preferred embodiments, in step S100, "if the neighbor table of the second node includes the first node, the corresponding data packet is sent to the first node", and the method includes: and if the neighbor table of the second node contains the MAC address of the first node, sending a corresponding data packet to the first node.
In some preferred embodiments, the link duration prediction model is trained by:
respectively acquiring link duration and characteristic data of every two adjacent nodes as training data;
dividing the training data into homodromous training data and reverse training data according to the motion direction difference;
and respectively inputting the two groups of training data into the link duration prediction model to obtain a syntropy link duration prediction model and a reverse link duration prediction model.
In a second aspect of the invention, a V2V routing system based on online link duration prediction is provided, which is applied to communication between vehicle nodes in a vehicle networking, and comprises a request judging module, a characteristic data acquiring module, a link duration prediction module, a reachability coefficient acquiring module and a cycle output module;
the request judging module is configured to send a corresponding data packet to a first node if a neighbor table of a second node contains the first node when the first node requests the data packet from the second node; otherwise, executing the characteristic data acquisition module; the data packet comprises a position, a speed, an acceleration, a motion direction and an MAC address;
the characteristic data acquisition module is configured to acquire characteristic data of the second node and each neighbor node based on a difference value between the data packet of the second node and corresponding data in the data packet of each node in the neighbor node group; the neighbor node group comprises one or more neighbor nodes which are adjacent nodes of the second node; the characteristic data comprises distance difference, speed difference, motion direction difference, space similarity and relative displacement;
the link duration prediction module is configured to obtain link duration prediction values of the second node and each neighbor node through a link duration prediction model based on the characteristic data;
the reachability coefficient acquisition module is configured to acquire reachability coefficients of the second node and each neighbor node based on the link duration predicted value; selecting a neighbor node corresponding to the maximum reachability coefficient as a next hop node of the second node;
the loop output module is configured to send the data packet to the first node if the neighbor table of the next hop node includes the first node, and otherwise, to loop the characteristic data obtaining module and the reachability coefficient obtaining module until the data packet is sent to the first node;
the link duration prediction model is constructed based on a width neural network and used for obtaining link duration prediction values of the two nodes according to the characteristic data.
In a third aspect of the present invention, a storage device is provided, in which a plurality of programs are stored, the programs being loaded and executed by a processor to implement the above-mentioned online link duration prediction-based V2V routing method.
In a fourth aspect of the present invention, a processing apparatus is provided, which includes a processor, a storage device; a processor adapted to execute various programs; a storage device adapted to store a plurality of programs; the program is adapted to be loaded and executed by a processor to implement the above-described online link duration prediction based V2V routing method.
The invention has the beneficial effects that:
the invention improves the stability of routing in the communication of the Internet of vehicles and improves the success rate of data transmission. According to the method, a link duration prediction model is constructed based on the width neural network, data characteristics are learned quickly, and the real-time performance and efficiency of link duration prediction are improved. Meanwhile, a same-direction link duration prediction model and a reverse-direction link duration prediction model are respectively trained according to the included angle of the movement direction of the node, so that the problem of interference of different modal data on the models when the movement directions of the nodes are the same or different is solved, and the accuracy of link duration prediction is improved. The stability and the accessibility of all the neighbor nodes are estimated based on the link duration prediction, and the optimal node is selected as the next hop node for data transmission, so that the probability of communication failure caused by the fact that the node moves out of the communication range in the data transmission process is reduced, and the stability of routing selection is improved.
Drawings
Other features, objects and advantages of the present application will become more apparent upon reading of the following detailed description of non-limiting embodiments thereof, made with reference to the accompanying drawings.
FIG. 1 is a flow chart of a V2V routing method based on online link duration prediction according to an embodiment of the present invention;
FIG. 2 is a block diagram of a V2V routing system based on online link duration prediction according to an embodiment of the present invention;
FIG. 3 is a flowchart illustrating a training data labeling method according to an embodiment of the present invention;
fig. 4 is an exemplary diagram of a link duration prediction model according to one embodiment of the invention.
Detailed Description
In order to make the objects, technical solutions and advantages 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 accompanying drawings, and it is apparent 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.
The present application will be described in further detail with reference to the following drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the relevant invention and not restrictive of the invention. It should be noted that, for convenience of description, only the portions related to the related invention are shown in the drawings.
It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict.
The invention discloses a V2V routing method based on online link duration prediction, which is applied to communication between vehicle nodes in a vehicle networking and comprises the following steps as shown in figure 1:
step S100, when a first node requests a data packet from a second node, if a neighbor table of the second node contains the first node, sending the corresponding data packet to the first node; otherwise, executing step S200; the data packet comprises a position, a speed, an acceleration, a motion direction and an MAC address;
step S200, respectively obtaining characteristic data of the second node and each neighbor node based on the difference value of corresponding data in the data packet of the second node and the data packet of each node in the neighbor node group; the neighbor node group comprises one or more neighbor nodes which are adjacent nodes of the second node; the characteristic data comprises distance difference, speed difference, motion direction difference, space similarity and relative displacement;
step S300, based on the characteristic data, obtaining link duration prediction values of the second node and each neighbor node through a link duration prediction model;
step S400, based on the link duration predicted value, obtaining the reachability coefficient between the second node and each neighbor node; selecting a neighbor node corresponding to the maximum reachability coefficient as a next hop node of the second node;
step S500, if the neighbor table of the next hop node contains the first node, the data packet is sent to the first node, otherwise, the steps S200 to S400 are executed in a circulating manner until the data packet is sent to the first node;
the link duration prediction model is constructed based on a width neural network and used for obtaining link duration prediction values of the two nodes according to the characteristic data.
In order to more clearly explain the V2V routing method based on online link duration prediction according to the present invention, the following describes steps in an embodiment of the method in detail with reference to the accompanying drawings.
In the following preferred embodiment, the link duration prediction model is detailed first, and then the on-line link duration prediction-based V2V routing method for obtaining the link duration prediction of two nodes by using the link duration prediction model is detailed.
1. Training of link duration prediction models
Step A100, labeling training data
In this embodiment, a vehicle is used as a node in the internet of vehicles, and a current node extracts a data packet and a routing table acquired from a neighboring node to form a training data set. The current node and the neighbor nodes are communicated once every 1s, and the data packet comprises data such as position, speed, movement direction and the like. And simultaneously, extracting the time from the beginning to the end of communication of the two nodes from the routing table as a label of the training data set. Then, features are extracted based on the data, and distance difference, speed difference and motion direction difference are calculated based on the positions, speeds, motion directions and the like of the two nodes respectively to form corresponding feature data.
And continuously and iteratively inquiring whether the communication record with the same neighbor node exists at the next moment or not aiming at each communication record, if so, adding 1 to the link duration until no such record exists, and at the moment, writing the difference between the current searched moment and the original moment into a corresponding data item to be used as the marking data.
The labeling of the training data of the link duration prediction model is shown in fig. 3, and the steps are as follows:
step A101, beginning to analyze a communication data packet and label data, and recording a communication Time of current labeled data as t;
step A102, inquiring a communication record item of a sink of a corresponding node according to a Time sequence, for example, a communication record item of the next Time Time +1 of the Time, recording the corresponding Time Time until no communication record of the corresponding node exists at the next Time, and marking the MAC address of the node as the ID of the node;
step A103, marking the difference value between the communication Time and t of the communication record of the corresponding node until the next moment as Time-t, and marking the difference value as duration of the link of the characteristic data;
step A104, based on the GPS position information (x) of the two nodes1,y1),(x2,y2) And calculating the distance d as shown in formula (1):
step A105: according to the speed v of two nodes1,v2And direction of motion dir1,dir2Velocity between calculationsDifference v and difference dir in direction of motion: dir ═ dir1-dir2|
If dir<90 degrees, then v ═ v1-v2|
If dir>90 degrees, then v ═ v1+v2|
Step A106: calculating the space similarity SLS between the two nodes according to the speed and the distance of the two nodes, as shown in a formula (2):
wherein R is the maximum communication distance of the on-board unit, vmaxThe maximum travel speed of the vehicle.
Step A107: according to the distance difference d between two nodes at the last momentoldAnd the current distance difference d, calculating the relative displacement RM of the two nodes, as shown in formula (3):
step A108: and combining the distance difference, the speed difference, the motion direction difference, the spatial similarity, the relative displacement and other items with the calculated link duration, and storing the combined link duration as a piece of training data in a database.
Step A200, training a link duration prediction model based on labeled data
Because the motion directions of the two nodes are the same or different, the link duration prediction models have larger difference, and the data of the two different modes are easy to interfere with the models, thereby influencing the accuracy of the models. Therefore, the data are divided into two parts according to the included angle of the vehicle moving direction, and two models are trained respectively. Wherein, the training same-direction link duration prediction model with the included angle smaller than 90 degrees and the training reverse link duration prediction model with the included angle larger than 90 degrees.
As shown in FIG. 4, the link duration prediction model is composed of spatial similarity, relative displacement, distance difference and velocity difference when training the modelThe two groups of characteristic data and labels thereof are respectively and sequentially input into each group of characteristic nodes F of two width neural networks1,…,FmAnd calculating the linear mapping output F of each group of characteristic nodesiAs shown in equation (4):
Fi＝f(XWfi+βi) (4)
wherein, WfiIs the connection weight of the input node and each characteristic node, which is the randomized weight, the input node is the first-stage node of the wide neural network for inputting the characteristic data, betaiThe method is characterized in that f () is a linear activation function, i is a subscript value, X is characteristic data and serves as an input node, and the activation function for operating the characteristic data in the wide neural network is a linear function.
Then, the output value E of each feature node is calculatedjSimultaneously as an enhanced node E1,E2…,EnAnd input to the output node according to formula Ej＝g(FWej+βj) Computing the output of a feature node, where WejJ is a subscript value, and the activation function of the enhanced node is a sigmoid nonlinear activation function. In the model training and predicting process, the connection weights between the input and the feature nodes and between the feature nodes and the enhancement nodes are always kept unchanged. Finally, the output F of the feature node and the output E of the enhancement node are combined into [ F | E]And calculates the output of the model by multiplication with the output weight matrix. The goal of the width learning training is to find the connection weight W to make the output of the model [ F | E]The error between W and the real link duration label is minimal. The connection weight W can be quickly found by the ridge regression method, which is calculated as shown in equation (5):
W＝(λI+[F|E]T[F|E])-1[F|E]TY (5)
wherein, Y is the real link duration label, λ is the random number matrix, I is the identity matrix with the main diagonal elements all being 1, and T is the transpose matrix.
When a node is to forward data, the trained model is first used to predict the link duration with a neighboring node. Extracting information such as speed, position and direction of neighbor nodes in a communication data packet, calculating distance difference, speed difference, spatial similarity and relative displacement of characteristic data by combining the information of the communication data packet and a historical record, selecting a prediction model according to the difference of the moving directions of the characteristic data and the historical record, inputting a characteristic matrix formed by data into a characteristic layer of a corresponding model, calculating outputs F ' and E ' of the characteristic node and an enhanced node respectively according to the matrix and an activation function in the model, and calculating the output Y ' of the model according to the trained output layer weight, namely [ F ' | E ' ] W, so that a link duration predicted value between the two nodes can be obtained.
And calculating and constructing a link value factor VF of the neighbor node based on a result LD of link duration prediction and the data volume transmission time TT, calculating a reachability coefficient AC of the neighbor node by combining the distance between the link value factor VF and the target node, ranking the nodes in the neighbor node table according to the reachability coefficient, and selecting the node with the highest reachability coefficient for transmission. Here, the link duration LD is a duration for predicting and outputting a corresponding link from the position data of the vehicle neighbor node and the like, and the data amount transmission time TT is a transmission time predicted to be required based on the data amount to be currently transmitted, and its calculation formula is:
where DS denotes the amount of data (packet size) to be transmitted by the node and Rb denotes the data transmission rate monitored by the node, for example, the data rate of short-range communication dedicated to general car networking communication technology is 1-3 Mbps. The time theoretically required to complete the data transmission can thus be obtained. Further, the link cost factor is defined as:
i.e. the ratio of the predicted link duration to the data volume transmission time. The reachability of the neighbor node is used for measuring the comprehensive property of the node which takes the stability of a link and the hop count into consideration when the node is used as a relay node (next hop node) to transmit data to a destination node, and the value of the reachability is defined as:
wherein α and 1- α represent a link value factor and a weight coefficient of a distance from a destination node, respectively, and the node distance is a ratio measure of a maximum distance difference max (d) from the destination node to a distance difference from the destination node. The larger the value of alpha is, the larger the proportion of the link stability in the routing is, and some scenes with sparse vehicle nodes, such as rural roads, are suitable for applying the larger alpha. And under the scenes of urban main roads with higher vehicle node density and the like, the size of alpha can be reduced, and the hop count of the transmission link is preferentially considered.
And constructing a forwarding selection strategy of each level of nodes according to a greedy strategy based on the reachability coefficient calculation result of the neighbor node, namely when each node selects the neighbor node for forwarding data, if the target node is in the communication range of the node, directly forwarding the data to the target node, and otherwise, taking the neighbor node with the highest reachability coefficient as a next hop node.
2. V2V routing method based on online link duration prediction
The invention discloses a V2V routing method based on online link duration prediction, which is applied to communication between vehicle nodes in a vehicle networking and comprises the following steps as shown in figure 1:
step S100, when a first node requests a data packet from a second node, if a neighbor table of the second node contains the first node, sending the corresponding data packet to the first node; otherwise, executing step S200; the data packet includes a position, a velocity, an acceleration, a direction of motion, a MAC address.
In this embodiment, each node broadcasts a route discovery packet at a fixed time interval, and inquires information such as its location from a neighboring node. After receiving the request message, the node replies basic information such as position, speed, acceleration, MAC address and the like to the sending node, so that each node normally maintains a neighbor table.
When the second node needs to send information, whether the MAC address of the first node is in the neighbor table of the second node is firstly inquired, if the first node is a friend node of the second node, the data is directly sent to the second node, and otherwise, the adjacent node of the second node is searched for transmitting the data packet.
Step S200, respectively obtaining characteristic data of the second node and each neighbor node based on the difference value of corresponding data in the data packet of the second node and the data packet of each node in the neighbor node group; the neighbor node group comprises one or more neighbor nodes which are adjacent nodes of the second node; the characteristic data comprises distance difference, speed difference, motion direction difference, space similarity and relative displacement.
In this embodiment, the characteristic data is obtained based on the difference between the second node and the data in the data packet of each adjacent node.
And step S300, obtaining link duration prediction values of the second node and each neighbor node through a link duration prediction model based on the characteristic data.
In this embodiment, a corresponding link duration prediction model is selected based on the motion direction difference between the second node and each adjacent node to obtain the link duration prediction value between the second node and each adjacent node. And the link duration prediction model is constructed based on the width neural network and used for acquiring link duration prediction values of the two nodes according to the characteristic data.
If the motion direction difference is smaller than 90 degrees, obtaining link duration prediction values of the second node and each neighbor node through a same-direction link duration prediction model according to the characteristic data; otherwise, obtaining the link duration prediction value of the second node and each adjacent node through a reverse link duration prediction model.
Step S400, based on the link duration predicted value, obtaining the reachability coefficient between the second node and each neighbor node; and selecting the neighbor node corresponding to the maximum reachability coefficient as the next hop node of the second node.
In this embodiment, the size of the second node data packet and the transmission rate of the second node data packet to each neighboring node are obtained, and the transmission time of the second node data packet to each neighboring node is obtained;
dividing the link duration predicted value with the transmission time to obtain a link value factor of the second node and each neighbor node;
and obtaining the reachability coefficients of the second node and each neighbor node according to the link value factor and the distance difference between the second node and each neighbor node.
And sequencing all the reachability coefficients, and selecting the neighbor node corresponding to the maximum reachability coefficient as the next hop node of the second node.
Step S500, if the neighbor table of the next hop node contains the first node, the data packet is sent to the first node, otherwise, the steps S200 to S400 are executed in a circulating manner until the data packet is sent to the first node.
In this embodiment, based on the found next hop node, a data packet is transmitted to the next hop node, it is determined that the neighbor table of the next hop node includes the first node, and if the neighbor table includes the first node, the data packet is sent to the first node. Otherwise, continuing to search for the next node until the data packet is sent to the first node.
A V2V routing system based on online link duration prediction according to a second embodiment of the present invention is applied to communication between vehicle nodes in an internet of vehicles, as shown in fig. 2, and includes: a request judgment module 100, a feature data acquisition module 200, a link duration prediction module 300, a reachability coefficient acquisition module 400, and a cycle output module 500;
the request determining module 100 is configured to, when a first node requests a data packet from a second node, send the data packet corresponding to a neighbor table of the second node to the first node if the neighbor table of the second node includes the first node; otherwise, executing the obtain feature data module 200; the data packet comprises a position, a speed, an acceleration, a motion direction and an MAC address;
the characteristic data obtaining module 200 is configured to obtain characteristic data of the second node and each neighboring node based on a difference value between the data packet of the second node and corresponding data in the data packet of each node in the neighboring node group; the neighbor node group comprises one or more neighbor nodes which are adjacent nodes of the second node; the characteristic data comprises distance difference, speed difference, motion direction difference, space similarity and relative displacement;
the predicted link duration module 300 is configured to obtain predicted link duration values of the second node and each neighboring node through a link duration prediction model based on the feature data;
the module 400 for obtaining reachability coefficient is configured to obtain reachability coefficients of the second node and neighboring nodes based on the predicted value of link duration; selecting a neighbor node corresponding to the maximum reachability coefficient as a next hop node of the second node;
the loop output module 500 is configured to send the data packet to the first node if the neighbor table of the next hop node includes the first node, and otherwise, to loop the obtain feature data module 200 — the obtain reachability coefficient module 400 until the data packet is sent to the first node;
the link duration prediction model is constructed based on a width neural network and used for obtaining link duration prediction values of the two nodes according to the characteristic data.
It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working process and related description of the system described above may refer to the corresponding process in the foregoing method embodiment, and details are not described herein again.
It should be noted that, the routing system of V2V based on online link duration prediction provided in the foregoing embodiment is only illustrated by the above-mentioned division of each functional module, and in practical applications, the above-mentioned function allocation may be completed by different functional modules according to needs, that is, modules or steps in the embodiments of the present invention are further decomposed or combined, for example, the modules in the foregoing embodiments may be combined into one module, or may be further split into multiple sub-modules, so as to complete all or part of the above-mentioned functions. The names of the modules and steps involved in the embodiments of the present invention are only for distinguishing the modules or steps, and are not to be construed as unduly limiting the present invention.
A storage device according to a third embodiment of the present invention stores therein a plurality of programs adapted to be loaded by a processor and to implement the above-described online link duration prediction-based V2V routing method.
A processing apparatus according to a fourth embodiment of the present invention includes a processor, a storage device; a processor adapted to execute various programs; a storage device adapted to store a plurality of programs; the program is adapted to be loaded and executed by a processor to implement the above-described online link duration prediction based V2V routing method.
It can be clearly understood by those skilled in the art that, for convenience and brevity, the specific working processes and related descriptions of the storage device and the processing device described above may refer to the corresponding processes in the foregoing method examples, and are not described herein again.
Those of skill in the art would appreciate that the various illustrative modules, method steps, and modules described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that programs corresponding to the software modules, method steps may be located in Random Access Memory (RAM), memory, Read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. To clearly illustrate this interchangeability of electronic hardware and software, various illustrative components and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as electronic hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
The terms "first," "second," and the like are used for distinguishing between similar elements and not necessarily for describing or implying a particular order or sequence.
The terms "comprises," "comprising," or any other similar term are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
So far, the technical solutions of the present invention have been described in connection with the preferred embodiments shown in the drawings, but it is easily understood by those skilled in the art that the scope of the present invention is obviously not limited to these specific embodiments. Equivalent changes or substitutions of related technical features can be made by those skilled in the art without departing from the principle of the invention, and the technical scheme after the changes or substitutions can fall into the protection scope of the invention.
Claims (8)
1. A V2V routing method based on online link duration prediction for communication between vehicle nodes in a vehicle networking, the method comprising the steps of:
step S100, when a first node requests a data packet from a second node, if a neighbor table of the second node contains the first node, sending the corresponding data packet to the first node; otherwise, executing step S200; the data packet comprises a position, a speed, an acceleration, a motion direction and an MAC address;
step S200, respectively obtaining characteristic data of the second node and each neighbor node based on the difference value of corresponding data in the data packet of the second node and the data packet of each node in the neighbor node group; the neighbor node group comprises one or more neighbor nodes which are adjacent nodes of the second node; the characteristic data comprises distance difference, speed difference, motion direction difference, space similarity and relative displacement;
step S300, based on the characteristic data, obtaining link duration prediction values of the second node and each neighbor node through a link duration prediction model;
step S400, based on the link duration predicted value, obtaining the reachability coefficient between the second node and each neighbor node; selecting a neighbor node corresponding to the maximum reachability coefficient as a next hop node of the second node;
step S500, if the neighbor table of the next hop node contains the first node, the data packet is sent to the first node, otherwise, the steps S200 to S400 are executed in a circulating manner until the data packet is sent to the first node;
the link duration prediction model is constructed on the basis of a width neural network and comprises a syntropy link duration prediction model and a reverse link duration prediction model; the link duration prediction model is used for acquiring link duration prediction values of the two nodes according to the characteristic data;
the calculation method of the reachability coefficient comprises the following steps:
wherein AC is a reachability coefficient, alpha is a weight coefficient, VF is a link cost factor, max (d) is the maximum distance difference between the second node and each neighbor node, diAnd the distance difference between the second node and each neighbor node is represented by i, the index value of the neighbor node is represented by LD, the link duration predicted value is represented by TT, and the data volume transmission time is represented by TT.
2. The method for routing V2V based on online link duration prediction according to claim 1, wherein in step S300, "obtaining the link duration prediction values of the second node and each neighboring node through a link duration prediction model based on the feature data" is performed by:
if the motion direction difference is smaller than 90 degrees, obtaining link duration prediction values of the second node and each neighbor node through a same-direction link duration prediction model according to the characteristic data; otherwise, link duration prediction values of the second node and each neighbor node are obtained through a reverse link duration prediction model.
3. The method for routing V2V based on online link duration prediction according to claim 1, wherein in step S400, "obtaining the reachability coefficients of the second node and each neighboring node based on the link duration prediction value" is performed by:
acquiring the size of the second node data packet and the transmission rate of the second node data packet to each neighbor node to obtain the transmission time of the second node data packet to each neighbor node;
dividing the link duration predicted value with the transmission time to obtain a link value factor of the second node and each neighbor node;
and obtaining the reachability coefficients of the second node and each neighbor node according to the link value factor and the distance difference between the second node and each neighbor node.
4. The method of claim 1, wherein the step S100 of sending the corresponding data packet to the first node if the neighbor table of the second node contains the first node comprises: and if the neighbor table of the second node contains the MAC address of the first node, sending a corresponding data packet to the first node.
5. The online link duration prediction-based V2V routing method according to claim 1, wherein the link duration prediction model is trained by:
respectively acquiring link duration and characteristic data of every two adjacent nodes as training data;
dividing the training data into homodromous training data and reverse training data according to the motion direction difference;
and respectively inputting the two groups of training data into the link duration prediction model to obtain a syntropy link duration prediction model and a reverse link duration prediction model.
6. A V2V routing system based on online link duration prediction is applied to communication among vehicle nodes in a vehicle networking and is characterized by comprising a request judging module, a characteristic data acquiring module, a link duration prediction module, an accessibility coefficient acquiring module and a cycle output module;
the request judging module is configured to send a corresponding data packet to a first node if a neighbor table of a second node contains the first node when the first node requests the data packet from the second node; otherwise, executing the characteristic data acquisition module; the data packet comprises a position, a speed, an acceleration, a motion direction and an MAC address;
the characteristic data acquisition module is configured to acquire characteristic data of the second node and each neighbor node based on a difference value between the data packet of the second node and corresponding data in the data packet of each node in the neighbor node group; the neighbor node group comprises one or more neighbor nodes which are adjacent nodes of the second node; the characteristic data comprises distance difference, speed difference, motion direction difference, space similarity and relative displacement;
the link duration prediction module is configured to obtain link duration prediction values of the second node and each neighbor node through a link duration prediction model based on the characteristic data;
the reachability coefficient acquisition module is configured to acquire reachability coefficients of the second node and each neighbor node based on the link duration predicted value; selecting a neighbor node corresponding to the maximum reachability coefficient as a next hop node of the second node;
the loop output module is configured to send the data packet to the first node if the neighbor table of the next hop node includes the first node, and otherwise, to loop the characteristic data obtaining module and the reachability coefficient obtaining module until the data packet is sent to the first node;
the link duration prediction model is constructed on the basis of a width neural network and comprises a syntropy link duration prediction model and a reverse link duration prediction model; the link duration prediction model is used for acquiring link duration prediction values of the two nodes according to the characteristic data;
the calculation method of the reachability coefficient comprises the following steps:
wherein AC is a reachability coefficient, alpha is a weight coefficient, VF is a link cost factor, max (d) is the maximum distance difference between the second node and each neighbor node, diAnd the distance difference between the second node and each neighbor node is represented by i, the index value of the neighbor node is represented by LD, the link duration predicted value is represented by TT, and the data volume transmission time is represented by TT.
7. A storage device having stored therein a plurality of programs, wherein said program applications are loaded and executed by a processor to implement the online link duration prediction based V2V routing method of any of claims 1-5.
8. A processing device comprising a processor, a storage device; a processor adapted to execute various programs; a storage device adapted to store a plurality of programs; characterized in that the program is adapted to be loaded and executed by a processor to implement the online link duration prediction based V2V routing method of any of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910978117.XA CN110784852B (en) | 2019-10-15 | 2019-10-15 | V2V routing method based on online link duration prediction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910978117.XA CN110784852B (en) | 2019-10-15 | 2019-10-15 | V2V routing method based on online link duration prediction |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110784852A CN110784852A (en) | 2020-02-11 |
| CN110784852B true CN110784852B (en) | 2021-03-02 |
Family
ID=69385677
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910978117.XA Active CN110784852B (en) | 2019-10-15 | 2019-10-15 | V2V routing method based on online link duration prediction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110784852B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113709701B (en) * | 2021-08-27 | 2022-06-17 | 西安电子科技大学 | Millimeter wave vehicle networking combined beam distribution and relay selection method, system and equipment |
| CN114666766B (en) * | 2022-05-24 | 2022-08-02 | 光谷技术有限公司 | Internet of things gateway communication load sharing method and system |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109003441A (en) * | 2017-06-06 | 2018-12-14 | 丰田自动车工程及制造北美公司 | According to the system and method for traffic dynamic control vehicle |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103476006A (en) * | 2013-08-29 | 2013-12-25 | 镇江青思网络科技有限公司 | Enhanced neighbor information-based multi-hop broadcast protocol |
| CN105228180B (en) * | 2015-09-29 | 2018-08-21 | 江苏大学 | A kind of vehicle-mounted Delay Tolerant Network method for routing based on the estimation of node transfer capability |
| CN107872807B (en) * | 2016-09-26 | 2021-07-09 | 富士通株式会社 | Routing node position determination method and device and terminal equipment |
| CN106713143A (en) * | 2016-12-06 | 2017-05-24 | 天津理工大学 | Adaptive reliable routing method for VANETs |
| CN107708173B (en) * | 2017-09-25 | 2020-06-19 | 京东方科技集团股份有限公司 | Selection method and device of fusion node |
-
2019
- 2019-10-15 CN CN201910978117.XA patent/CN110784852B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109003441A (en) * | 2017-06-06 | 2018-12-14 | 丰田自动车工程及制造北美公司 | According to the system and method for traffic dynamic control vehicle |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110784852A (en) | 2020-02-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ahmed et al. | Optimization procedure for intelligent internet of things applications | |
| Cerotti et al. | A crowd-cooperative approach for intelligent transportation systems | |
| CN110784852B (en) | V2V routing method based on online link duration prediction | |
| CN113409598B (en) | Cooperative linkage self-adaptive timing method and device for regional road traffic signal lamps | |
| Zou et al. | Energy-efficient roadside unit scheduling for maintaining connectivity in vehicle ad-hoc network | |
| Tasseron et al. | The potential impact of vehicle-to-vehicle communication on on-street parking under heterogeneous conditions | |
| Joshua et al. | An optimization framework for routing protocols in VANETs: A multi-objective firefly algorithm approach | |
| Maakar et al. | Performance evaluation of AODV and DSR routing protocols for flying ad hoc network using highway mobility model | |
| Kadadha et al. | A blockchain-enabled relay selection for QoS-OLSR in urban VANET: A Stackelberg game model | |
| Naeem et al. | Enhanced clustering based routing protocol in vehicular ad‐hoc networks | |
| Gomides et al. | Predictive congestion control based on collaborative information sharing for vehicular ad hoc networks | |
| Inam et al. | A novel protocol for vehicle cluster formation and vehicle head selection in vehicular ad-hoc networks | |
| Bintoro | A study of V2V communication on VANET: characteristic, challenges and research trends | |
| Liu et al. | Cpa-mac: A collision prediction and avoidance mac for safety message dissemination in mec-assisted vanets | |
| Rajalakshmi et al. | Investigation and Analysis of Path Evaluation for Sustainable Communication Using VANET | |
| Chang et al. | Design and implementation of the travelling time-and energy-efficient android gps navigation app with the vanet-based a* route planning algorithm | |
| Datta et al. | Ant-AODV-VANET: A bio-inspired routing protocol for VANET | |
| Li et al. | A traffic congestion aware vehicle-to-vehicle communication framework based on Voronoi diagram and information granularity | |
| Maharjan et al. | Smart parking utilizing iot embedding fog computing based on smart parking architecture | |
| Balzano et al. | A smart compact traffic network vision based on wave representation | |
| Bozdog et al. | RideMatcher: peer-to-peer matching of passengers for efficient ridesharing | |
| Dappuri et al. | A stable routing algorithm based on link prediction method for clustered VANET | |
| Cai et al. | Cooperative content caching and delivery in vehicular networks: A deep neural network approach | |
| CN114827946A (en) | Task similarity-based edge calculation method and system in Internet of vehicles scene | |
| Christakos | Sensor networks applied to the problem of building evacuation: An evaluation in simulation |
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 |