CN110113725A - A kind of motivational techniques of the car networking node based on betting model - Google Patents
A kind of motivational techniques of the car networking node based on betting model Download PDFInfo
- Publication number
- CN110113725A CN110113725A CN201910360153.XA CN201910360153A CN110113725A CN 110113725 A CN110113725 A CN 110113725A CN 201910360153 A CN201910360153 A CN 201910360153A CN 110113725 A CN110113725 A CN 110113725A
- Authority
- CN
- China
- Prior art keywords
- node
- data packet
- car networking
- stage
- value
- 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
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/02—Arrangements for optimising operational condition
-
- 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]
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Information Transfer Between Computers (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The invention discloses a kind of node activations methods of car networking based on betting model, a kind of new incentive mechanism is proposed based on betting model, based on the credit value of node, the node high to contribution degree is rewarded, the node low to contribution degree is punished, the inconsistent problem of contribution degree between node is improved, so that the enthusiasm of all nodes is improved.The data packet forwarding task for the participation network that vehicle net node can be motivated positive improves data packet delivery fraction on the basis of original AODV agreement, reduces average end-to-end time delay, improves the overall network performance of car networking.
Description
Technical field
The invention belongs to car networking vehicle node fields of communication technology, are related to a kind of car networking node based on betting model
Motivational techniques.
Background technique
With economic continuous development, people's lives level is being promoted steadily.In order to accommodating out, more and more families
Front yard uses automobile as walking-replacing tool.By to the first half of the year in 2018, Chinese vehicle guaranteeding organic quantity has been over 300,000,000,
Middle automobile accounts for 2.17 hundred million.While automobile offers convenience for people, many social concerns are also caused.For example, take place frequently
Traffic accident, traffic congestion and air pollution.Therefore, there is an urgent need to a kind of technologies of vehicle communication to alleviate the above problem.Closely
Several years, the appearance of car networking technology provided answer for these problems.
Car networking is developed from vehicular ad hoc network, and in car networking, each vehicle can be regarded as net
A vehicle node in network, the interaction between vehicle need to be in communication with each other between node.When vehicle node attempts and position
When vehicle node except oneself communication range communicates, it is necessary to the assistance of relay node.However in reality, vehicle
The energy and resource of node are limited, and memory, bandwidth and the energy of vehicle node itself can be consumed by participating in data packet forwarding
Deng.Since each vehicle node is rationality, that is, maximize the income of itself.Therefore, certain vehicle nodes are to save itself
Resource and refusing forwards other vehicle nodes to relay the data of coming, and shows the selfishness of vehicle node, these nodes are referred to as
Selfish node.Even if studies have shown that also resulting in internetworking there is the selfish node of fraction (10%-40%) in network
Energy is remarkably decreased (16%-32%).Therefore, it cooperates how effectively being motivated between node in car networking, Logistics networks
Availability and high-performance be significant challenge that current car networking research field faces.
Currently, motivating the mechanism of selfish node cooperative communication to be divided into two classes in self-organizing network.The first kind is based on payment
Incentive mechanism.The mechanism is mainly cooperated between node by the way of virtual electronic currency to motivate, and Typical Representative is
Nuglet Counter strategy.The strategy is based on Counter Value, and when node sends the data packet of oneself, Counter Value reduces,
And when the data packet of node forwarding neighbor node, Counter Value increases, but it is positive number that node, which must maintain the value of counter,
The data packet of itself can just be sent.Therefore, if node is wanted to send the data packet of oneself, must cooperate.The mechanism
Disadvantage is that usually requiring additional equipment carrys out the smooth implementation of pledge system, but is difficult reality in actual network environment
It is existing, therefore constrain the development of the mechanism.
Second class is the incentive mechanism based on prestige.The thought of the mechanism mainly assesses its letter by the prestige of node
Ren Du judges whether believable node is according to the height of degree of belief, and the decision-making foundation as Route Selection.The mechanism
The disadvantage is that, the calculating about prestige is complex, and is directed to different schemes, the calculated value of prestige may be inconsistent.
Summary of the invention
The motivational techniques of the object of the present invention is to provide a kind of car networking node based on betting model, solve existing skill
The poor problem for causing network performance bad of the interaction enthusiasm of each node of car networking in art.
The technical scheme adopted by the invention is that a kind of node activations method of the car networking based on betting model, including
Following steps:
Step 1: establishing the single phase betting model of the adjacent node of car networking network, rear extend establishes infinite repeated game
Model;
Step 2: on repeated game model in step 1, and the basis for strategy of giving tit for tat, using appearance between node
The strategy of mistake;
Step 3: based on the credit value of node, the node high to contribution degree is rewarded, the node low to contribution degree
It is punished, improves the inconsistent problem of contribution degree between node, so that the enthusiasm of all nodes is improved.
In step 1, entire car networking network G (V, E) is made of N number of rationality node, and G indicates that any connected graph, V indicate section
The set of point, E indicate link set;System time is made of discrete time slots t one by one, and in each time slot, node all may be used
To make the selection of rationality;The infinite repeated game model is specifically, make repeatedly to be interacted between each node, mutually
Cooperation carries out data packet forwarding, to improve the overall network performance of car networking.
Fault-tolerance strategy in step 2 specifically:
For there is a pair of of adjacent node A and B in car networking, take strategy as follows any node:
Initial stage: the attitude of node A and node B be all it is friendly, that is, take the attitude of cooperation, actively forward other side
Data packet;
The tolerance stage: when k-th of time slot, node A takes disoperative attitude, then within a certain period of time, being set as T, saves
Point B adopts a tolerant attitude, i.e., still forwards data packet for node A, be set as patient time;
The judgement stage: if after the tolerance stage, the attitude of node A is still unfriendly, that is, selecting uncooperative behavior, then
Node A can be broadcasted by node B, inform other nodes, and then node A enters the punishment stage, be set as P;
The punishment stage: in penalty period, other nodes can all refuse the data packet of forward node A transmission, and node A exists
Its neighbor node must be helped gratuitously to forward data packet in penalty period;
The forgetting stage: terminating when the punishment stage, and node A can reenter network, and uncooperative history can also pass into silence;
The value of patient time T and penalty period P are specifically determined according to following rule:
In patient time T determining first, patient time T, node B selects cooperation policy always, but neighbor node A takes
Disoperative attitude, according to the knowledge of game theory, in the tolerance stage, the income f of node BBSuch as following formula:
Wherein δ is discount factors, and value range is (0,1), it is considered as the patient degree of node cooperation, and δ value is bigger,
Then node performance is more patient, and β indicates the consumption that node for data forwarding generates
During this period, node B gratis forwards data always for node A, but the data of itself are not forwarded,
So that T is set up, then need to make the expected revenus U of node BBMore than or equal to zero, it may be assumed that
That is, the value of patient time T is exactly the value for setting up above formula.
Finally determine the value of penalty period P.Since node A is since k-th time slot, after experienced patient time, still
It is uncooperative, it is punished so applying to node A.If tolerating that the interests of stage acquisition are less than the consumption of penalty period, all due to node
It is rationality, so since next stage, in order to maximize itself interests, the attitude of node can switch to close from uncooperative
Make, see below formula:
Meet formula (7), corresponding is exactly the value of penalty period P.
Reward based on credit value and penalty mechanism in step 3 specifically:
Initial phase, each node itself include a weight table, and the initial value of weight is initialized as 0, works as section
The every participation forwarding of point is primary, and weight adds 1;
In each time slot t, table can all update once, and each node is sorted according to the size of weight;
When node B and node C require that node A forwards respective data packet, node A can be according to weight in weight table
Size, the data packet of the preferential node for forwarding weight big, the data packet for the node for forwarding weight small again later.
The beneficial effects of the invention are as follows the present invention is based on betting models to propose a kind of new incentive mechanism, vehicle can be motivated
The data packet of the positive participation network of net node forwards task, on the basis of original AODV agreement, improves data packet throwing
Rate is passed, average end-to-end time delay is reduced, improves the overall network performance of car networking.
Detailed description of the invention
Fig. 1 is a kind of node activations method single phase node forwarding game mould of car networking based on betting model of the present invention
Type figure;
Fig. 2 is a kind of node activations method data packet delivery fraction of the car networking based on betting model of the present invention and selfish section
The comparison diagram of point ratio;
Fig. 3 is that a kind of node activations method of the car networking based on betting model of the present invention be averaged end-to-end time delay and selfishness
The comparison diagram of node ratio;
Specific embodiment
The following describes the present invention in detail with reference to the accompanying drawings and specific embodiments.
A kind of node activations method of the car networking based on betting model of the present invention, comprising the following steps:
Step 1: establishing the single phase betting model of the adjacent node of car networking network, rear extend establishes infinite repeated game
Model;
Step 2: on repeated game model in step 1, and the basis for strategy of giving tit for tat, using appearance between node
The strategy of mistake;
Step 3: based on the credit value of node, the node high to contribution degree is rewarded, the node low to contribution degree
It is punished, improves the inconsistent problem of contribution degree between node, so that the enthusiasm of all nodes is improved.
In step 1, entire car networking network G (V, E) is made of N number of rationality node, and G indicates that any connected graph, V indicate section
The set of point, E indicate link set;System time is made of discrete time slots t one by one, and in each time slot, node all may be used
To make the selection of rationality;The infinite repeated game model is specifically, make repeatedly to be interacted between each node, mutually
Cooperation carries out data packet forwarding, to improve the overall network performance of car networking.
Fault-tolerance strategy in step 2 specifically:
For there is a pair of of adjacent node A and B in car networking, take strategy as follows any node:
Initial stage: the attitude of node A and node B be all it is friendly, that is, take the attitude of cooperation, actively forward other side
Data packet;
The tolerance stage: when k-th of time slot, node A takes disoperative attitude, then within a certain period of time, being set as T, saves
Point B adopts a tolerant attitude, i.e., still forwards data packet for node A, be set as patient time;
The judgement stage: if after the tolerance stage, the attitude of node A is still unfriendly, that is, selecting uncooperative behavior, then
Node A can be broadcasted by node B, inform other nodes, and then node A enters the punishment stage, be set as P;
The punishment stage: in penalty period, other nodes can all refuse the data packet of forward node A transmission, and node A exists
Its neighbor node must be helped gratuitously to forward data packet in penalty period;
The forgetting stage: terminating when the punishment stage, and node A can reenter network, and uncooperative history can also pass into silence;
The value of patient time T and penalty period P are specifically determined according to following rule:
In patient time T determining first, patient time T, node B selects cooperation policy always, but neighbor node A takes
Disoperative attitude, according to the knowledge of game theory, in the tolerance stage, the income f of node BBSuch as following formula:
Wherein δ is discount factors, and value range is (0,1), it is considered as the patient degree of node cooperation, and δ value is bigger,
Then node performance is more patient, and β indicates the consumption that node for data forwarding generates
During this period, node B gratis forwards data always for node A, but the data of itself are not forwarded,
So that T is set up, then need to make the expected revenus U of node BBMore than or equal to zero, it may be assumed that
That is, the value of patient time T is exactly the value for setting up above formula.
Finally determine the value of penalty period P.Since node A is since k-th time slot, after experienced patient time, still
It is uncooperative, it is punished so applying to node A.If tolerating that the interests of stage acquisition are less than the consumption of penalty period, all due to node
It is rationality, so since next stage, in order to maximize itself interests, the attitude of node can switch to close from uncooperative
Make, see below formula:
Meet formula (7), corresponding is exactly the value of penalty period P.
Reward based on credit value and penalty mechanism in step 3 specifically:
Initial phase, each node itself include a weight table, and the initial value of weight is initialized as 0, works as section
The every participation forwarding of point is primary, and weight adds 1;
In each time slot t, table can all update once, and each node is sorted according to the size of weight;
When node B and node C require that node A forwards respective data packet, node A can be according to weight in weight table
Size, the data packet of the preferential node for forwarding weight big, the data packet for the node for forwarding weight small again later.
Step 1: establishing the single phase betting model of adjacent node, the strategy that analysis node can be taken, and extension establishes nothing
Limit repeated game model;
1) entire car networking network G (V, E) is made of N number of rationality node, and G indicates that any connected graph, V indicate the collection of node
It closes, E indicates link set;
2) link between node is two-way, i.e., and if only if node w, when z is within transmission range each other, link
(w, z) is the subset of link E;
3) system time is made of discrete time slots t one by one, and in each time slot, node can make the choosing of rationality
It selects;
4) all nodes all show two kinds of behavioral strategies, and one kind is cooperation policy, that is, are ready to turn for other nodes relaying
Data are sent out, are indicated with C;Another kind is selfish strategy, i.e. refusal relay forwarding data, is indicated with D;
5) nodes revenue is generally made of two parts, and a part is that node sends data acquisition by neighbor node forwarding
Income indicates that another part is the consumption that node for data forwarding generates with α, is indicated with β.So the income of node can indicate
Are as follows: f=alpha-beta;
As shown in Figure 1, A, B, D1 and D2 are four nodes in car networking.D2 and A, A and B, B and D1 are adjacent segments
Point, dotted line represent the transmission range of vehicle node signal.In car networking, there are two types of modes for the intercommunication of node, if phase
Neighbors is within the transmission range of respective signal, then direct communication, such as adjacent node A and B in figure;Another kind is due to logical
Letter other side is in except the transmission range of signal, communicates then to be realized by the relay forwarding of neighbor node, such as A and D1 in figure
Between communication must pass through the forwarding of node B.Assuming that node A wants data packet issuing D1, since signal transmission ranges have
Limit, it is necessary to via the forwarding of adjacent node B;Similarly, the precondition of node B and D2 communication is the forwarding of node A.So obtaining
Adjacent node A and B only carry out the gain matrix of single game, as shown in table 1.
The gain matrix of table 1 single phase node forwarding game
Adjacent node A and B | C (cooperation) | D (uncooperative) |
C (cooperation) | (α-β,α-β) | (-β,α) |
D (uncooperative) | (α,-β) | (0,0) |
As shown in table 1, α represents node and sends the income that a data coating neighbor node forwarding obtains, and β represents node and turns
The consumption that data generate is sent out, and β is much smaller than α.From the above it can be seen that this is a typical single prisoners' dilemma game, because right
For each node, regardless of what strategy game other side takes, oneself taking uncooperative strategy is all optimal selection.Because
When taking uncooperative strategy, for node itself, the income of acquisition is the largest.Game is forwarded in entire single phase node
In the process, all nodes are all the individuals of rationality, and each node can pursue the maximum of number one during game
Change.
Therefore, the Nash Equilibrium Solution of available this time game is (uncooperative, uncooperative), i.e., game both sides refuse to turn
Send out data packet, node A and node D1 cannot be communicated at this time, be similarly also with node D2 for node B as.This be we not
It is ready to see, so introduce repeated game model, motivates to cooperate between node and forward data packet, to improve car networking
Overall network performance.
Step 2: on the basis of repeated game, based on strategy of giving tit for tat, a kind of plan of fault-tolerance is proposed
Slightly;
Cause disoperative reason between node to be that node takes disoperative strategy, not will lead to corresponding punishment,
This punishes the income that can reduce node.If a kind of penalty mechanism of design of node is directed to, so that the uncooperative acquisition of node is short
Phase interests are less than the loss being subject in the punishment stage.Since each node is rationality, that is, maximize the interests of itself.So save
Point is since next stage, it will considers the influence that itself behavior generates subsequent game carefully.At this point, it and neighbor node
Between multiple interaction be no longer a series of mutually independent single phase nodes forwarding games, but be extended to one it is unlimited
Duplicate message forwards game.
According to the principle that repeated game is discussed, the expection total revenue of node i are as follows:
In formula 1,Node i is represented in the income of time slot t single phase game, δ is discount factors, value range be (0,
1), it is considered as the patient degree of node cooperation, and δ value is bigger, then node performance is more patient, also more payes attention to long-term benefit;Instead
It, then node more focuses on interests at the moment;In general, the network discount factors temporarily constructed are relatively small, and stable
Network discount factors it is relatively large.
It is that people generally use there are two kinds of punishment strategies according to repeated game theory, that is, gives tit for tat tactful and cold
Cruel strategy.In repeated game, optimal strategy is give tit for tat tactful (TFT, Tit-For-Tat), that is, plan of returning like for like
Slightly.The strategy can be described as: the both sides for participating in game know the strategy that other side takes, if wherein a side is in the first stage
Selection cooperation, then next stage another party can also select cooperation policy, if a certain stage, wherein a side took uncooperative plan
Slightly, corresponding next stage, another party can directly take uncooperative strategy, circuit sequentially.Summarize the strategy it is found that game both sides
Mode is to return like for like.See formula (2):
Assuming that node i takes uncooperative strategy in second time slot, the strategy taken between node will as shown in formula (3),
It alternates, you come, and I am past, is easy to lead to the unstable of network, it is therefore desirable to improve this strategy.
If node i is on the basis of second time slot takes uncooperative strategy, the attitude of third time slot is remained as not
Cooperative attitude, then node j will not give the impression of weakness, can also take immediately it is uncooperative, to show oneself attitude, as shown in formula (4).
If things go on like this, the switching performance of network will sharply decline.
By analyzing step by step, if the strategy relatively that will fight for mastery is applied directly in car networking network, it is found that the strategy
It is excessively stringent.Therefore it is tactful (TTFT) that this paper presents a kind of TFT of fault-tolerance.Assuming that network exist a pair of adjacent node A with
B takes strategy as follows any node:
1) initial stage: the attitude of node A and node B be all it is friendly, that is, take the attitude of cooperation, actively forward other side
Data packet;
2) tolerate the stage: when k-th of time slot, node A takes disoperative attitude, then (being set as T) within a certain period of time
Node B adopts a tolerant attitude, i.e., still forwards data packet for node A, be set as patient time;
3) judge the stage: if after the tolerance stage, the attitude of node A is still unfriendly, that is, selecting uncooperative behavior, that
Node A can be broadcasted by node B, inform other nodes, and then node A enters punishment stage (being set as P);
4) punish the stage: in penalty period, other nodes can all refuse the data packet of forward node A transmission, and node A
Its neighbor node forwarding data packet (free that forwarding service is provided) must be helped in penalty period;
5) forget the stage: terminating when the punishment stage, node A can reenter network, and uncooperative history can also be lost
Forget;
According to the strategic thinking, it is thus necessary to determine that the value of patient time T and penalty period P.Patient time T, tolerance are determined first
In time T, node B selects cooperation policy always, but neighbor node A takes disoperative attitude, according to the knowledge of game theory,
In the tolerance stage, the income of node B is shown in formula (5):
During this period, node B gratis forwards data always for node A, but the data of itself are not forwarded,
So that T is set up, then need to make the expected revenus of node B to be more than or equal to zero, it may be assumed that
That is, the value of patient time T is exactly the value that formula (6) are set up.
Finally determine the value of penalty period P.Since node A is since k-th time slot, after experienced patient time, still
It is uncooperative, it is punished so applying to node A.If tolerating that the interests of stage acquisition are less than the consumption of penalty period, all due to node
It is rationality, so since next stage, in order to maximize itself interests, the attitude of node can switch to close from uncooperative
Make, see formula (7):
Meet formula (7), corresponding is exactly the value of penalty period P.
Step 3: based on the credit value of node, the inconsistent problem of contribution degree between node is improved;
It analyses in depth set forth above based on fault-tolerant TFT strategy, it is found that the strategy does not have the contribution degree of node
It distinguishes, i.e. it is equal that node B, which forwards 10 data packets and node C to forward the status of 1 data packet,.Due to based on fault-tolerant
TFT strategy excitation node limits the behavior of node using the means of punishment, so that node has to take into account that current behavior to not
Carry out the influence of income.Therefore on the basis of repeated game, the high node of the mechanism reward contribution degree based on credit value is proposed,
To improve the enthusiasm of node.
Basic thought: initial phase, each node itself include a weight table, which is (node name
Word, weight).Namespace node represents the nodename for helping node for data forwarding packet, and the initial value of weight is initialized as 0, forwarding
Once, weight adds 1.For example node B and node C help the forwarded over data packet of node A, but percentage contribution is different, such as:
Node B helps A to forwarded 10 times, and node C helps A to forwarded 1 time.In each time slot t, table can all update once, and meeting
It sorts according to the size of weight.When node B and node C require that node A forwards respective data packet, node A can be according to power
It is worth the size of weight in table, the data packet of the preferential node for forwarding weight big, the data packet for the node for forwarding weight small again later.
By above analysis, the enthusiasm of node can be greatlyd improve according to this credit mechanism, successfully to the contribution degree of node
It is distinguished.
In order to verify the correctness and feasibility of the mechanism, we utilize Network Simulation Software NS2 (Network
Simulation Version 2) emulation is realized on operating system Ubuntu 16.04.NS2 be a object-oriented,
Event driven, powerful network analog simulation software, is combined by C++ and OTcl language and writes completion.
Simulated environment is as shown in the table:
The major parameter of 2 node running track of table is arranged
Performance indicator parameter are as follows:
1) data packet delivery fraction (Packet Delivery Ratio): the data packet that the vehicle node of feeling the pulse with the finger-tip receives accounts for
The ratio for the data packet that source vehicle node is sent, directly reflects the reliability of Routing Protocol, the higher routing of data packet delivery fraction
The reliability of agreement is better.As shown in formula 3-1:
2) average time delay (Average End-to-End Delay) end to end: refer to that source vehicle node transmits packets to
The ratio of the data packet number sent up to time consumed by purpose vehicle node and time period.It is usually anti-with end-to-end time delay
Whether unobstructed reflect network, time delay is smaller, and the patency of network is better.As shown in formula 3-2:
Embodiment:
This experimental result is divided into two groups.Data packet delivery fraction peace the two parameters of end-to-end time delay are investigated respectively,
Original AODV agreement, be added AODV agreement (0.6AODV) and addition incentive mechanism that incentive mechanism and discount factors are 0.6 and
Under ADOV agreement (0.8AODV) both of these case that discount factors are 0.8, as the increase of selfish node ratio in network is each
The variation of parameter, and then analyze and obtain the performance of incentive mechanism.
Incentive mechanism AODV agreement, discount factors are not added as 0.6 AODV agreement and discount as shown in Fig. 2, giving
The AODV agreement of the factor is in the case of these three, when the ratio of selfish node in network increases, the variation of data packet delivery fraction.From
Analogous diagram observes, these three situations have a common ground, data packet delivery fraction with selfish node ratio in network increase
And reduce.When the quantity of selfish node in network is more and more, selfish node all refuses relay forwarding data packet and results in data
The decline of packet delivery fraction.But when incentive mechanism is added to AODV agreement, it can be found that data packet delivery fraction downward trend subtracts
It is slow, and it is 0.6 that data packet delivery fraction when discount factors are 0.8, which is higher than discount factors,.Reason is: excitation set is added
System improves the enthusiasm cooperated between node, and the state of selfish node switchs to cooperate from uncooperative, improves network entirety
Cooperative, and discount factors are higher, and the patient degree of node is higher, so data packet delivery fraction is higher.
Incentive mechanism AODV agreement, discount factors are not added as 0.6 AODV agreement and discount as shown in figure 3, giving
The AODV agreement of the factor is in the case of these three, when the ratio of selfish node in network increases, the variation of average end-to-end time delay.
From the figure, it can be seen that the overall trend of average end-to-end time delay is all incremental with the increase of selfish node ratio, and
The time delay value that joined incentive mechanism AODV agreement is higher than the value of original AODV agreement.This is because there is section in incentive mechanism
The process of repeated game between point, spends to the more time, improves the value of time delay.It could be observed that discount
The time delay value of the factor 0.8 is higher than the time delay of discount factors 0.6 on the whole, because discount factors embody the patient degree of node,
Discount factors are higher, and the patient degree of node is higher, therefore node more focuses on following income, so time delay value is higher.
Claims (5)
1. a kind of node activations method of the car networking based on betting model, which comprises the following steps:
Step 1: establishing the single phase betting model of the adjacent node of car networking network, rear extend establishes infinite repeated game mould
Type;
Step 2: on repeated game model in step 1, and the basis for strategy of giving tit for tat, fault-tolerance is used between node
Strategy;
Step 3: based on the credit value of node, the node high to contribution degree is rewarded, and the node low to contribution degree carries out
Punishment, improves the inconsistent problem of contribution degree between node, so that the enthusiasm of all nodes is improved.
2. a kind of node activations method of car networking based on betting model according to claim 1, which is characterized in that institute
It states in step 1, entire car networking network G (V, E) is made of N number of rationality node, and G indicates that any connected graph, V indicate the collection of node
It closes, E indicates link set;System time is made of discrete time slots t one by one, and in each time slot, node can be made
The selection of rationality;The infinite repeated game model specifically, make repeatedly to be interacted between each node, work in coordination into
Row data packet forwarding, to improve the overall network performance of car networking.
3. a kind of node activations method of car networking based on betting model according to claim 1, which is characterized in that institute
State fault-tolerance strategy in step 2 specifically:
For there is a pair of of adjacent node A and B in car networking, take strategy as follows any node:
Initial stage: the attitude of node A and node B be all it is friendly, that is, take the attitude of cooperation, actively forward other side data
Packet;
The tolerance stage: when k-th of time slot, node A takes disoperative attitude, then within a certain period of time, being set as T, node B is adopted
It tries to please the attitude born, i.e., still forwards data packet for node A, be set as patient time;
The judgement stage: if after the tolerance stage, the attitude of node A is still unfriendly, that is, uncooperative behavior is selected, then node
Node A can be broadcasted by B, inform other nodes, and then node A enters the punishment stage, be set as P;
The punishment stage: in penalty period, other nodes can all refuse the data packet of forward node A transmission, and node A is being punished
Its neighbor node must be helped gratuitously to forward data packet in phase;
The forgetting stage: terminating when the punishment stage, and node A can reenter network, and uncooperative history can also pass into silence.
4. a kind of node activations method of car networking based on betting model according to claim 3, which is characterized in that institute
The value for stating patient time T and penalty period P is specifically determined according to following rule:
In patient time T determining first, patient time T, node B selects cooperation policy always, but neighbor node A takes and do not conform to
The attitude of work, according to the knowledge of game theory, in the tolerance stage, the income f of node BBSuch as following formula:
Wherein δ is discount factors, and value range is (0,1), it is considered as the patient degree of node cooperation, and δ value is bigger, then saves
More patient, the consumption that β expression node for data forwarding generates of point performance
During this period, node B gratis forwards data always for node A, but the data of itself are not forwarded, to be made
T is set up, then needs to make the expected revenus U of node BBMore than or equal to zero, it may be assumed that
That is, the value of patient time T is exactly the value for setting up above formula.
Finally determine the value of penalty period P.Since node A is since k-th time slot, after experienced patient time, still do not conform to
Make, is punished so applying to node A.If tolerating that the interests of stage acquisition are less than the consumption of penalty period, since node is all reason
Property, so since next stage, in order to maximize itself interests, the attitude of node can switch to cooperate from uncooperative,
See below formula:
Meet formula (7), corresponding is exactly the value of penalty period P.
5. a kind of node activations method of car networking based on betting model according to claim 1, which is characterized in that institute
State the reward based on credit value in step 3 and penalty mechanism specifically:
Initial phase, each node itself include a weight table, the initial value of weight are initialized as 0, when node is every
It is primary to participate in forwarding, weight adds 1;
In each time slot t, table can all update once, and each node is sorted according to the size of weight;
When node B and node C require node A forward respective data packet when, node A can according in weight table weight it is big
It is small, the data packet of the preferential node for forwarding weight big, the data packet for the node for forwarding weight small again later.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910360153.XA CN110113725B (en) | 2019-04-30 | 2019-04-30 | Internet of vehicles node excitation method based on game model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910360153.XA CN110113725B (en) | 2019-04-30 | 2019-04-30 | Internet of vehicles node excitation method based on game model |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110113725A true CN110113725A (en) | 2019-08-09 |
CN110113725B CN110113725B (en) | 2022-11-11 |
Family
ID=67487731
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910360153.XA Active CN110113725B (en) | 2019-04-30 | 2019-04-30 | Internet of vehicles node excitation method based on game model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110113725B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115230433A (en) * | 2022-08-23 | 2022-10-25 | 重庆大学 | Cooperative heating control method and device for electric vehicle passenger compartment and power battery |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20180173538A1 (en) * | 2016-03-29 | 2018-06-21 | Intel Corporation | Systems, methods and devices for dynamic power management of devices using game theory |
CN109040075A (en) * | 2018-08-08 | 2018-12-18 | 中国联合网络通信集团有限公司 | Management method, server and the system of wireless mobile sensor network interior joint |
-
2019
- 2019-04-30 CN CN201910360153.XA patent/CN110113725B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20180173538A1 (en) * | 2016-03-29 | 2018-06-21 | Intel Corporation | Systems, methods and devices for dynamic power management of devices using game theory |
CN109040075A (en) * | 2018-08-08 | 2018-12-18 | 中国联合网络通信集团有限公司 | Management method, server and the system of wireless mobile sensor network interior joint |
Non-Patent Citations (4)
Title |
---|
MEIJUAN KOU; YANLIN ZHAO; HANYU CAI; XIUMEI FAN: "Study of a Routing Algorithm of Internet of Vehicles Based on Selfishness", 《2018 IEEE INTERNATIONAL CONFERENCE ON SMART INTERNET OF THINGS (SMARTIOT)》 * |
于季弘: "机会网络中的节点激励策略研究", 《信息科技辑》 * |
张健: "基于博弈论的移动Ad_Hoc网络节点合作策略研究", 《信息科技辑》 * |
谭永银: "基于重复博弈的机会网络路由算法研究", 《信息科技辑》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115230433A (en) * | 2022-08-23 | 2022-10-25 | 重庆大学 | Cooperative heating control method and device for electric vehicle passenger compartment and power battery |
CN115230433B (en) * | 2022-08-23 | 2024-05-14 | 重庆大学 | Electric vehicle passenger cabin and power battery cooperative heating control method and device |
Also Published As
Publication number | Publication date |
---|---|
CN110113725B (en) | 2022-11-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Wang et al. | Minimizing the age-of-critical-information: an imitation learning-based scheduling approach under partial observations | |
Morris et al. | Sexual networks, concurrency, and STD/HIV | |
Barbera et al. | Data offloading in social mobile networks through VIP delegation | |
CN102118450B (en) | Betweenness centrality-based opportunistic network P2P (peer-to-peer) information query method | |
CN103647671A (en) | Gur Game based crowd sensing network management method and system | |
Xu et al. | Data management in mobile peer-to-peer networks | |
CN104579957A (en) | Routing method of delay tolerant network based on degree of intimacy and time constraint forwarding | |
Zhu et al. | Information dissemination model for social media with constant updates | |
CN110134865A (en) | A kind of commuting passenger's social recommendation method and platform based on urban public transport trip big data | |
CN104408306A (en) | Method for stimulating parked vehicles for cooperative retransmission to perform analysis modeling by using evolutionary game theory in vehicle ad hoc network (VANET) | |
Cai et al. | A survey on routing algorithms for opportunistic mobile social networks | |
CN107181793A (en) | The transportation service information forwarding mechanism discussed based on dynamic game | |
CN103198432B (en) | Detection method and detection system of network groups in online social network | |
CN110113725A (en) | A kind of motivational techniques of the car networking node based on betting model | |
CN108460628A (en) | Information-pushing method, device, computer equipment and storage medium | |
Sanchez et al. | Green Bear-A LoRaWAN-based Human-in-the-Loop case-study for sustainable cities | |
Kong et al. | Disseminating authorized content via data analysis in opportunistic social networks | |
CN107357858A (en) | A kind of network reconstruction method based on geographical position | |
CN108985631A (en) | A kind of private car share-car action selection method based on game theory | |
CN109284932A (en) | A kind of stranger's social activity user's evaluation method and system based on big data | |
Yoneki et al. | Distinct types of hubs in human dynamic networks | |
Layuan et al. | A routing protocol for dynamic and large computer networks with clustering topology | |
CN109634725B (en) | Method and device for dispatching crowd sensing task | |
Agrawal et al. | Efficient taxi and passenger searching in smart city using distributed coordination | |
CN105141544A (en) | Data distribution method in Internet of vehicles |
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 | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20221009 Address after: No. 30728 Xi'an Square, Yanta District, Xi'an City, Shaanxi Province, 710000 Applicant after: Xi'an Zhixing Changjia Network Technology Co.,Ltd. Address before: 710048 No. 5 Jinhua South Road, Shaanxi, Xi'an Applicant before: XI'AN University OF TECHNOLOGY |
|
TA01 | Transfer of patent application right | ||
GR01 | Patent grant | ||
GR01 | Patent grant |