CN109492887B - Mobile crowd sensing and exciting method based on potential game theory - Google Patents
Mobile crowd sensing and exciting method based on potential game theory Download PDFInfo
- Publication number
- CN109492887B CN109492887B CN201811254046.0A CN201811254046A CN109492887B CN 109492887 B CN109492887 B CN 109492887B CN 201811254046 A CN201811254046 A CN 201811254046A CN 109492887 B CN109492887 B CN 109492887B
- Authority
- CN
- China
- Prior art keywords
- user
- strategy
- income
- policy
- social
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
- G06Q10/06375—Prediction of business process outcome or impact based on a proposed change
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
Abstract
The invention discloses a mobile crowd sensing excitation method based on a potential game theory, which comprises the following steps: firstly, a server issues a crowd sensing task; step two, the equipment receives the task broadcast; initializing the system, firstly setting control parameter theta and strategy updating frequency tau for the equipmentnWhile each device randomly selects a policy an∈AnAs the initial strategy selection of the user; fourthly, calculating social group income; step five, updating strategy behaviors; step sixthly, informing the server device of the strategy selection. The invention promotes users to achieve a stable social group Nash equilibrium based on the Markov chain-based mobile swarm intelligence perception incentive algorithm, and the income obtained by each user in the state is the optimal income which can be obtained by the user under the current strategy combination.
Description
Technical Field
The invention belongs to the field of mobile crowd sensing, and particularly relates to a mobile crowd sensing excitation method based on a potential game theory.
Background
The mobile crowd sensing mainly refers to that mobile equipment (mobile phones, tablet computers and the like) of common users are used as basic sensing units, conscious or unconscious cooperation is carried out through the mobile internet, sensing task distribution and sensing data collection are achieved, and large-scale and complex social sensing tasks are completed. Compared with the traditional wireless sensor network, the mobile crowd sensing overcomes the defects of large deployment difficulty, high maintenance cost, single purpose and the like of the wireless sensor network, and has the characteristics of diverse sensing information, wider sensing coverage space range, and random and anywhere sensing. At present, mobile crowd sensing has become a new important internet application mode, covers the aspects of people's life, including environmental monitoring, intelligent transportation, social network, business, medical treatment, transportation and other fields, brings unprecedented opportunities for promoting social and city management innovation, but also has a lot of problems to be solved urgently.
Although the mobile device can gain income when performing the mobile crowd sensing task, the cost is generated due to privacy disclosure and energy and bandwidth consumption caused by collecting, processing and uploading data, and how to effectively stimulate the mobile device to participate in the mobile crowd sensing becomes a very important research hotspot. Game theory is an important theoretical approach to solving this problem. However, the existing methods based on the game theory only pay attention to two extreme cases. One is when designing the excitation algorithm, the optimization goal is the overall performance of the mobile crowd sensing Network (NUM). The game theory model for solving the problem assumes that all users in the network are privately and collaborating, and the optimization objectives are completely consistent, namely, the gains of the whole network are optimized. The other is to design an incentive algorithm, the optimization goal of which is to maximize individual interests (NCG) of individual users. The game theory model for solving this problem assumes that all users in the network are selfish, rational, uncooperative, with their optimization goal being to maximize their own revenue. However, the advent of mobile social networks has made these two assumptions no longer true. Because people in the mobile social network have rich social relationships and the intimacy degree of each relationship is different, users in the mobile social network are not completely privet or selfish and often show certain group sex, and the benefits of other users with different social relationships with the users are considered.
The social relationship in the mobile social network is abstracted into a social relationship network, and a network structure is formed by a plurality of nodes and the relationship among the nodes. Nodes generally refer to individuals or groups of organizations, and the social relationship network represents social relationships between individuals or organizations, and objects are concatenated according to the social relationships. Social relationship networks are formed by relying on various relationships, such as consanguinity, friendship, hobbies, working properties, value, ideality, conflict, and the like, thereby generating an intricate and complex network structure. The Social Group profit Maximization framework (SGUM) is a brand-new application framework of game theory research, and can be used for integrating Social relations into profit functions by introducing the Social relations among users and solving the Social Group profit Maximization so as to solve the problem of a continuous space between the two extreme problems of NUM and NCG.
Since the introduction of the SGUM framework, the SGUM framework has been widely applied to various application scenarios, including database-based spectrum channel selection, micro-cloud computing transfer, location privacy protection, and the like. The motivational mechanisms for crowd sensing have been extensively studied. Traditional crowd sensing incentives mechanisms involve mainly five aspects, reward payment incentives, entertainment game incentives, social relationship incentives, virtual point incentives, and hybrid incentives. Lee and the like apply reverse auction in the economic field to research of a crowd sensing incentive mechanism for the first time, and guarantee higher participation rate while minimizing payment cost. Feng et al incentivize participants using a combined auction mode in a reverse auction, the participants can bid for multiple perception tasks according to their own location and perception range, and the server platform selects a winner according to the summarized bidding conditions of the participants. Ueyama et al motivate participants by introducing existing social network information to detect dishonest selfish participants and penalizing them. Luo et al propose an incentive mechanism based on the skirt-based relationship to ensure credibility of participant data quality by establishing an endorsement relationship between participants. The virtual point incentive is different from a reward payment mode, a participant cannot directly obtain payment, but the virtual point can meet the psychological requirements of realization of the participant's own value and virtual honor, plays a guiding role on the participant and increases the user stickiness. In crowd-sourcing aware incentive schemes, two or more incentive modes are often integrated in order to better motivate participants.
For social relations, Chen et al propose an SGUM game model and use it in database-based spectrum channel selection. The SGUM game model considers the influence of a physical layer and the influence of social relations, and aims to maximize social group income. Tang et al further use the SGUM game model in the computation transfer of a micro-cloud device, and design a computation transfer incentive algorithm based on a Markov chain to decide whether the device uploads a computation task to the micro-cloud for processing. Gong et al use the SGUM model to solve pareto optima of social awareness for location privacy protection in mobile networks. Through literature search, research of the SGUM game model aiming at an excitation mechanism of mobile crowd sensing and a distributed mobile crowd sensing excitation algorithm is still blank at present. The invention designs a mobile swarm intelligence perception excitation mechanism through an SGUM optimization framework based on a potential game theory, and mainly designs a potential game function and a theoretical proof based on the function so as to prove the feasibility of a scheme, and designs a mobile swarm intelligence perception excitation algorithm based on a Markov chain to promote users in a game to reach social swarm nash balance.
Disclosure of Invention
A crowd sensing application architecture that considers reward payment incentives is shown in fig. 1. The crowd sensing cloud platform firstly releases the sensing task and the ideal reward amount of the sensing task, a sensing task participant registers on the platform, and whether to participate in the task is determined according to task return and participation cost. The platform selects the required number of users based on the decisions submitted by the users. And uploading the collected data content to the crowd sensing cloud platform by the user, and finally distributing the task reward by the crowd sensing cloud platform.
Although the mobile device can obtain the return of the profit when performing the mobile crowd sensing task, the cost is generated due to the energy and bandwidth consumption caused by privacy disclosure, data collection and data uploading, and how to effectively stimulate the mobile device to participate in the mobile crowd sensing becomes a very important research hotspot. In order to solve the problem, the SGUM game model is utilized to fully consider the states of the users in both physical network and social network, and a social group revenue function is defined to prompt the device users to participate in the mobile crowd sensing task. The physical relationship layer and the social relationship layer under the application of the mobile crowd sensing are shown in fig. 2, and in the game process, the user equipment not only focuses on the income obtained by participating in the crowd sensing, but also gives consideration to the income of other users with different social relationships with the user. Based on a potential game theory, SGUM is applied to the design of a mobile swarm intelligence perception excitation mechanism, and a mobile swarm intelligence perception excitation algorithm based on a Markov chain is designed to promote user equipment to finally reach a stable social swarm nash balance. The specific technical scheme is as follows:
a mobile crowd sensing excitation method based on a potential game theory comprises the following steps:
firstly, a server issues a crowd sensing task;
step two, the equipment receives the task broadcast;
initializing the system, firstly setting control parameter theta and strategy updating frequency tau for the equipmentnWhile each device randomly selects a policy an∈AnAs the initial strategy selection of the user;
fourthly, calculating social group income;
step five, updating strategy behaviors;
step sixthly, informing the server device of the strategy selection.
Further, the step (i) is specifically as follows: the crowd sensing server initializes the sensing task and simultaneously sends a notification to acquire basic information of equipment near a target area; the server sets the reward b corresponding to each device according to the property of the tasknAnd the competitive risk cost c of the current task in the areas(ii) a After initializing these messages, a task is sent to each device for broadcast.
Further, the step two is specifically as follows: device receiving crowd wisdomAfter the task broadcast sent by the application server, corresponding cost C which needs to be borne when data is collected, processed and transmitted is calculated according to the task property and the performance of the sensor or other modules of the application servernFinally, the acquired reward bnCalculating the cost CnAnd a competitive risk cost csAnd substituting into the distributed algorithm model.
Further, the step (iv) is specifically as follows: the social group revenue function for user n is defined as follows:
wherein N represents a set of users in the model, a represents a policy combination,
a user m representing a social relationship with the user n; policy revenue function Un(a) Corresponding expression modes exist in different application scenes; each device n generates the social group income of itself according to the formula (1) by combining the strategies determined by the current devices, and meanwhile, the average value is followed
Generates its own timer and starts the timer.
Further, the fifth step is as follows: entering a circulation process, and sequentially updating the strategies of all the devices according to the expiration of the timer; if the timer for device n expires, device n reselects the new policy action a'n∈An\anAs a strategy to be updated by the user, the social group income S of the user at the moment is calculatedn(a′n,a-n) (ii) a Each time the timer expires, there is one and only one device user updating policy behavior; meanwhile, according to the new social group income, the equipment carries out strategy updating according to the following probability:
when the new policy behavior of the device provides better yield, Sn(a′n,a-n)≥Sn(an,a-n) At this point the user will transition to the new policy a 'with probability 1'nThe above. New strategy a 'selected according to the nature of potential game'nSocial group income S of user n can be improved in potential gamen(a) Meanwhile, the benefit of the potential function phi (a) can be improved; better yield is provided when the device is old, i.e. Sn(a′n,a-n)＜Sn(an,a-n) When this is the case, user n will be at probability
Stay in old policy anBy probability
Update to New policy a'n(ii) a When the income generated by the new strategy selected by the user n is less than the income generated by the old strategy, the probability that the user n stays in the old strategy is as the income function S of the old and new social groupsn(an,a-n)-Sn(a′n,a-n) The increase in the difference increases.
Further, the steps are as follows: when the equipment in the system reaches social consciousness Nash equilibrium, namely under the current strategy combination, no participant can improve the social group income by singly changing the strategy action of the participant, and the process of jumping out and ending the game is completed; meanwhile, each device sends a message to inform the swarm intelligence perception server of the own strategy.
Based on the potential game theory, the SGUM is applied to the design of a mobile crowd sensing excitation mechanism so as to excite mobile crowd sensing users to actively participate in sensing tasks and improve the application effect of crowd sensing. The social relationship is integrated into the mobile crowd sensing, and the strategy decision of the user not only focuses on the income of the user, but also strives to maximize the income of the user group with the social relationship. The game model becomes a potential game by constructing a potential function, so that the designed incentive mechanism can reach the social group nash equilibrium of a pure strategy theoretically proved. The invention promotes users to achieve a stable social group Nash equilibrium based on the Markov chain-based mobile swarm intelligence perception incentive algorithm, and the income obtained by each user in the state is the optimal income which can be obtained by the user under the current strategy combination.
Drawings
FIG. 1 is a diagram of a crowd sensing application architecture in view of reward payment incentives;
FIG. 2 is a diagram of a physical relationship layer and a social relationship layer under a mobile crowd sensing application;
figure 3 is a flow chart of a mobile crowd sensing incentive algorithm based on a Markov chain.
Detailed Description
The invention will be further explained with reference to the drawings.
The invention discloses a mobile crowd sensing excitation method based on a potential game theory, which has the following theories:
the first step is as follows: social group income maximization game model: in order to integrate social relationships among users into application scenes, the income of a user n is defined to be mainly composed of two parts, wherein one part is own strategy income Un(a) And part is the weighted sum of the strategic gains of the user group with social relationship with the user group. According to the social connection weight w between users in the social relationship graphnmWe can define the social group revenue function for user n as follows:
wherein N represents a set of users in the model, a represents a policy combination,
representing user m having a social relationship with user n. Policy collectorBenefit function Un(a) Corresponding expression modes are provided in different application scenes, so that the social group income function has good universality. But in every application, redefining the policy revenue function of the user is a problem that needs to be addressed with emphasis.
The second step is that: mobile crowd sensing revenue function Un(a) Defining: in the mobile crowd sensing application environment, the device collects, calculates and uploads data to the crowd sensing server through data. If the data is accepted by the server, the user equipment can obtain a revenue report b from the service providern. Meanwhile, the device also needs to bear the corresponding cost C when collecting, processing and transmitting datanThe cost here is mainly in view of the consumption of device resources caused by collecting data. The collection of relevant data by the equipment is a continuous measurement process, and the sampling frequency of the equipment n is fnThe cost of the device per unit sampling frequency is cnCost C of crowd sensing servicen＝cnfn. Meanwhile, when device n chooses to participate in the crowd sensing task, there is a risk of competing with other devices, assuming that the risk is csAnd the competitive risk increases with the number of remaining participating users. At this time, if the policy behavior of other devices remains unchanged, the revenue function of the device in the current state is as follows
When the equipment refuses to participate in the crowd sensing task, the equipment neither obtains income and return from a service provider nor bears the cost of data collection, and the income function of the equipment is 0. Therefore, in the context of crowd sensing applications, the user's revenue function is expressed as follows:
the third step: for the potential game theory, the following definitions are first given:
definition 1: nash equilibrium: if and only if none of the participants can pass through the sheetOnce the policy action is changed to improve his profit, we call the combination of policies in this model
For Nash equilibrium in a game, it is expressed as follows:
where N represents a set of users in the model, anA policy representing a selection of a user n, a-nRepresenting combinations of policies, U, of users in the set other than user nn(an,a-n) Representing the functional gain of user n under the current policy combination, a*Represents the combination of strategies in the state of nash equilibrium,
representing the policy selection for user n in nash equilibrium.
Definition 2: full potential game: if game Γ has a function φ (a) so that for any user N ∈ N, when the policy of N is an,an′∈AnAnd the policy combination of the rest of users is a-n∈∏i≠nAiExistence of equation Un(an′,a-n)-Un(an,a-n)＝φ(a′n,a-n)-φ(an,a-n). At this time, the game Γ is called a full potential game, the function Φ (a) is a potential function corresponding to the full potential game, and the full potential game has nash equilibrium with a pure strategy.
Wherein A isnAll policy set spaces, φ (a), representing users nn,a-n) Representing the value of a for the potential function under the current policy combination.
The definitions of nash equilibrium and full potential game in potential game theory are introduced in definition 1 and definition 2, respectively. Game Γ ═ N, { An},{Sn} is SGUM game, where N is denoted bySet of households AnRepresenting the policy composition space of user n, SnRepresenting the social group income of the user n under the strategy, the following definitions are continuously given:
definition 3: socially conscious nash equilibrium: we then call the combination of policies in this model when and only when none of the participants can promote his social group profits by a single change of their own policy actions
For the socially conscious nash balance in the above SGUM game, it is expressed as follows:
wherein S isn(an,a-n) And representing the social group function income of the user n under the current strategy combination.
Suppose that during the time interval when a particular task is broadcast, each device corresponds to the competition risk c of that tasksEqual and the social connections are symmetrical, i.e. wnm＝wmn. Next, a potential function needs to be constructed to make the SGUM game a potential game, so as to obtain social-awareness nash equilibrium of pure policy.
The fourth step: pure strategy social awareness nash equilibrium proof: the structural potential function is expressed as follows:
wherein, I{E}Is a flag function, I when event E holds{E}1 is ═ 1; otherwise, when the event E is not established I{E}0. According to the property Sn(an′,a-n)-Sn(an,a-n)＝φ(an′,a-n)-φ(an,a-n) The SGUM game described above is a full potential game and the potential function is shown in equation (5). Therefore, the SGUM game mustThere is a socially conscious nash equilibrium of pure strategies.
The fifth step: the mobile crowd sensing incentive mechanism based on the Markov chain comprises the following steps: and establishing a Markov approximate optimization solution aiming at the potential function of the formula (5).
Where Ω represents the combined space of all policies, qaRepresenting the probability that the policy combination a is selected, theta represents the control parameter of the formula and passes
Ensuring progressive optimality. When θ → ∞, the optimal solution that maximizes the potential function φ (a) will be chosen with a probability of 1. Meanwhile, the optimal solution can be solved as follows:
the invention designs a mobile crowd sensing excitation method based on a Markov chain, which determines whether a crowd sensing user participates in a sensing task in a distributed mode. Because of the time-reversible nature of the Markov chain, it is always possible to achieve a unique smooth profile that is independent of the initial system state and independent of the update sequence of states. Therefore, for any given initial strategy selection and updating sequence, the strategy selection of the equipment under the mobile swarm intelligence perception incentive mechanism can finally reach a smooth distribution state, namely unique social awareness nash equilibrium.
And solving social Nash equilibrium of the perception task strategy by setting a control parameter theta large enough. Order timernA timer generated for each user by the system, the timer being in accordance with the exponential distribution and having a value τn. Timer for user nnTime of passage τnLate expiration, which will then attempt to update the user's n own perception task participationA policy is decided, and the updated policy is different from the original policy. The algorithm flow is shown in fig. 3, the scheme is designed into 6 steps, wherein firstly, a crowd sensing task is issued for the server, secondly, a distributed algorithm flow of a mobile crowd sensing excitation mechanism based on a Markov chain is adopted, and sixth, the strategy selection of the equipment informing the server is adopted. The method comprises the following specific steps:
the server issues a crowd sensing task. And the crowd sensing server initializes the sensing task and simultaneously sends a notification to acquire basic information of the equipment near the target area. The server sets the reward b corresponding to each device according to the property of the tasknAnd the competitive risk cost c of the current task in the areas. After initializing these messages, a task is sent to each device for broadcast.
② the equipment receives the task broadcast. After the device receives the task broadcast sent by the crowd sourcing application server, corresponding cost C which needs to be borne during data collection, processing and transmission is calculated according to the task property and the performance of the sensor or other modules of the devicenFinally, the acquired reward bnCalculating the cost CnAnd a competitive risk cost csAnd substituting into the distributed algorithm model.
And initializing the system. After entering the algorithm, firstly, the system sets a control parameter theta and a strategy updating frequency tau for the equipmentnWhile each device randomly selects a policy an∈AnAs its initial policy selection.
And fourthly, calculating social group income. Each device n generates the social group income of itself according to the formula (1) by combining the strategies determined by the current devices, and meanwhile, the average value is followed
Generates its own timer and starts the timer.
Updating strategy behavior. Entering a circulation process, and sequentially updating the strategies of each device according to the expiration of the timer. If the timer of device n expires, at this point, device n reselects the new policy action a'n∈An\anAs a strategy to be updated by the user, the social group income S of the user at the moment is calculatedn(a′n,a-n). Each time the timer expires, there is one and only one device user updating the policy behavior. Meanwhile, according to the new social group income, the equipment carries out strategy updating according to the following probability.
Analytically, the new policy behavior of the device provides better yield, i.e., Sn(a′n,a-n)≥Sn(an,a-n) At this point the user will transition to the new policy a 'with probability 1'nThe above. According to the nature of the potential game, the new strategy a 'selected by us'nSocial group income S of user n can be improved in potential gamen(a) And simultaneously, the benefit of the potential function phi (a) can be improved. Better yield is provided when the device is old, i.e. Sn(a′n,a-n)＜Sn(an,a-n) When this is the case, user n will be at probability
Stay in old policy anBy probability
Update to New policy a'n. Obviously, when the revenue generated by the new strategy selected by the user n is smaller than the revenue generated by the old strategy, the probability that the user n stays in the old strategy is as a function of the revenue of the old and new social groupsn(an,a-n)-Sn(a′n,a-n) The increase in the difference increases.
Sixthly, notifying the server equipment of the strategy selection. When the devices in the system reach social awareness nash equilibrium, namely under the current strategy combination, no participant can promote the social group income of the participant by singly changing the strategy action of the participant, and the exit cycle ends the game process. Meanwhile, each device sends a message to inform the swarm intelligence perception server of the own strategy.
Claims (4)
1. A mobile crowd sensing excitation method based on a potential game theory comprises the following steps:
firstly, a server issues a crowd sensing task;
step two, the equipment receives the task broadcast;
initializing the system, firstly setting control parameter theta and strategy updating frequency tau for the equipmentnWhile each device randomly selects a policy an∈AnAs the initial strategy selection of the user;
fourthly, calculating social group income;
step five, updating strategy behaviors;
step sixthly, informing the strategy selection of the server equipment;
the step IV is as follows: according to the social connection weight w between users in the social relationship graphnmDefining the social group revenue function for user n as follows:
wherein N represents a set of users in the model, a represents a policy combination,
a user m representing a social relationship with the user n; policy revenue function Un(a) Corresponding expression modes exist in different application scenes; each device n generates the social group income of itself according to the formula (1) by combining the strategies determined by the current devices, and meanwhile, the average value is followed
Generating a self timer by the exponential distribution of the data and starting the timer;
the fifth step is as follows: entering a circulation process, and sequentially updating the strategies of all the devices according to the expiration of the timer; if the timer for device n expires, device n reselects the new policy action a'n∈An\anAs a strategy to be updated by the user, the social group income S of the user at the moment is calculatedn(a′n,a-n) (ii) a Each time the timer expires, there is one and only one device user updating policy behavior; meanwhile, according to the new social group income, the equipment carries out strategy updating according to the following probability:
when the new policy behavior of the device provides better yield, Sn(a′n,a-n)≥Sn(an,a-n) At this point the user will transition to the new policy a 'with probability 1'nThe above step (1); new strategy a 'selected according to the nature of potential game'nSocial group income S of user n can be improved in potential gamen(a) Meanwhile, the benefit of the potential function phi (a) can be improved; better yield is provided when the device is old, i.e. Sn(a′n,a-n)＜Sn(an,a-n) When this is the case, user n will be at probability
Stay in old policy anBy probability
Update to New policy a'n(ii) a When the income generated by the new strategy selected by the user n is less than the income generated by the old strategy, the probability that the user n stays in the old strategy is as the income function S of the old and new social groupsn(an,a-n)-Sn(a′n,a-n) The increase in the difference increases.
2. The mobile crowd-sourcing perception incentive method based on potential game theory as claimed in claim 1, wherein: the steps are as follows: the crowd sensing server initializes the sensing task and simultaneously sends a notification to acquire basic information of equipment near a target area; the server sets the reward b corresponding to each device according to the property of the tasknAnd the competitive risk cost c of the current task in the areas(ii) a After initializing these messages, a task is sent to each device for broadcast.
3. The mobile crowd-sourcing perception incentive method based on potential game theory as claimed in claim 2, wherein: the step II comprises the following concrete steps: the equipment receives the task broadcast sent by the crowd-sourcing application server, and calculates the corresponding cost C required to be borne during data collection, processing and transmission according to the task property and the performance of the sensor or other modulesnFinally, the acquired reward bnCalculating the cost CnAnd a competitive risk cost csSubstituting the user income function into a distributed algorithm model to obtain the following income function expression:
wherein the sampling frequency of the device n is fnAnd a denotes a policy set space of the user.
4. The mobile crowd-sourcing perception incentive method based on potential game theory as claimed in claim 1, wherein: the steps are as follows: when the equipment in the system reaches social consciousness Nash equilibrium, namely under the current strategy combination, no participant can improve the social group income by singly changing the strategy action of the participant, and the process of jumping out and ending the game is completed; meanwhile, each device sends a message to inform the swarm intelligence perception server of the own strategy.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811254046.0A CN109492887B (en) | 2018-10-25 | 2018-10-25 | Mobile crowd sensing and exciting method based on potential game theory |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811254046.0A CN109492887B (en) | 2018-10-25 | 2018-10-25 | Mobile crowd sensing and exciting method based on potential game theory |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109492887A CN109492887A (en) | 2019-03-19 |
| CN109492887B true CN109492887B (en) | 2021-10-26 |
Family
ID=65691924
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811254046.0A Active CN109492887B (en) | 2018-10-25 | 2018-10-25 | Mobile crowd sensing and exciting method based on potential game theory |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109492887B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110034958B (en) * | 2019-04-02 | 2022-11-11 | 浙江工商大学 | Vehicle networking pseudonym change incentive algorithm and change method based on SGUM theory |
| CN110009233B (en) * | 2019-04-08 | 2020-12-25 | 清华大学深圳研究生院 | Game theory-based task allocation method in crowd sensing |
| CN110232517B (en) * | 2019-06-06 | 2023-08-01 | 哈尔滨工程大学 | Mobile crowd sensing user profit selection method |
| CN111083675A (en) * | 2019-12-03 | 2020-04-28 | 西北工业大学 | Resource allocation method based on social perception in industrial Internet of things |
| CN111432361B (en) * | 2020-03-19 | 2022-09-13 | 长春理工大学 | User incentive strategy determination method and system based on crowd sensing network |
| CN112016047A (en) * | 2020-07-24 | 2020-12-01 | 浙江工业大学 | Heuristic data acquisition method and device based on evolutionary game, computer equipment and application thereof |
| CN115102648B (en) * | 2022-05-30 | 2023-07-25 | 南京邮电大学 | Multi-frequency Duan Qunzhi spectrum sensing method based on Stackelberg game |
| CN114905527A (en) * | 2022-05-31 | 2022-08-16 | 江苏经贸职业技术学院 | Football robot interception method based on Markov chain and football robot |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105528248A (en) * | 2015-12-04 | 2016-04-27 | 北京邮电大学 | Crowd sensing incentive mechanisms for multitask cooperative application |
| CN106485415A (en) * | 2016-10-11 | 2017-03-08 | 安徽慧达通信网络科技股份有限公司 | A kind of based on the mobile intelligent perception motivational techniques with budget for the relation between supply and demand |
| US20170216034A1 (en) * | 2016-01-28 | 2017-08-03 | DePuy Synthes Products, Inc. | Helical Graft Containment Cage |
| CN107301509A (en) * | 2017-06-23 | 2017-10-27 | 武汉大学 | It is a kind of based on intelligent perception system towards the intelligent perception motivational techniques participated at random |
| CN107944947A (en) * | 2017-11-07 | 2018-04-20 | 南京邮电大学 | A kind of mobile intelligent perception motivational techniques using social networks |
| CN108668253A (en) * | 2018-04-09 | 2018-10-16 | 南京邮电大学 | A kind of gunz cooperative sensing motivational techniques based on evolutionary Game |
| CN108681921A (en) * | 2018-05-14 | 2018-10-19 | 北京信息科技大学 | A kind of method and device obtaining intelligent perception incentives strategy based on Stochastic Game |
-
2018
- 2018-10-25 CN CN201811254046.0A patent/CN109492887B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105528248A (en) * | 2015-12-04 | 2016-04-27 | 北京邮电大学 | Crowd sensing incentive mechanisms for multitask cooperative application |
| US20170216034A1 (en) * | 2016-01-28 | 2017-08-03 | DePuy Synthes Products, Inc. | Helical Graft Containment Cage |
| CN106485415A (en) * | 2016-10-11 | 2017-03-08 | 安徽慧达通信网络科技股份有限公司 | A kind of based on the mobile intelligent perception motivational techniques with budget for the relation between supply and demand |
| CN107301509A (en) * | 2017-06-23 | 2017-10-27 | 武汉大学 | It is a kind of based on intelligent perception system towards the intelligent perception motivational techniques participated at random |
| CN107944947A (en) * | 2017-11-07 | 2018-04-20 | 南京邮电大学 | A kind of mobile intelligent perception motivational techniques using social networks |
| CN108668253A (en) * | 2018-04-09 | 2018-10-16 | 南京邮电大学 | A kind of gunz cooperative sensing motivational techniques based on evolutionary Game |
| CN108681921A (en) * | 2018-05-14 | 2018-10-19 | 北京信息科技大学 | A kind of method and device obtaining intelligent perception incentives strategy based on Stochastic Game |
Non-Patent Citations (5)
| Title |
|---|
| Exploiting Social Tie Structure for Cooperative Wireless Networking:A Social Group Utility Maximization Framework;Xu Chen,et al;《IEEE/ACM Transactions on Networking》;20160311;第24卷(第6期);第3593-3606页 * |
| Markov Approximation for Combinatorial Network Optimization;Minghua Chen,et al;《IEEE Transactions on Information Theory》;20130614;第59卷(第10期);第6301-6327页 * |
| 基于博弈论的感知任务参与决策机制研究;王秀兰;《中国优秀硕士学位论文全文数据库 信息科技辑》;20171115(第11期);第I138-315页 * |
| 基于移动社交网络的群智计算任务分配算法研究;茆华林;《中国优秀硕士学位论文全文数据库 信息科技辑》;20170315(第3期);第I139-362页 * |
| 群智感知激励机制研究综述;吴垚等;《软件学报》;20160317;第27卷(第8期);第2025-2047页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109492887A (en) | 2019-03-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109492887B (en) | Mobile crowd sensing and exciting method based on potential game theory | |
| Gao et al. | Truthful incentive mechanism for nondeterministic crowdsensing with vehicles | |
| Nie et al. | A stackelberg game approach toward socially-aware incentive mechanisms for mobile crowdsensing | |
| Wang et al. | EIHDP: Edge-intelligent hierarchical dynamic pricing based on cloud-edge-client collaboration for IoT systems | |
| Yang et al. | Promoting cooperation by the social incentive mechanism in mobile crowdsensing | |
| Ning et al. | Social-oriented adaptive transmission in opportunistic Internet of smartphones | |
| Jiang et al. | Toward optimal participant decisions with voting-based incentive model for crowd sensing | |
| Liu et al. | An incentive mechanism for privacy-preserving crowdsensing via deep reinforcement learning | |
| Charilas et al. | A survey on game theory applications in wireless networks | |
| Liang et al. | A survey on game theoretical methods in human–machine networks | |
| Zhao et al. | Social-aware incentive mechanism for vehicular crowdsensing by deep reinforcement learning | |
| Zhao et al. | Coalition-based task assignment in spatial crowdsourcing | |
| Xie et al. | Bargain-based stimulation mechanism for selfish mobile nodes in participatory sensing network | |
| Xu et al. | Trust based incentive scheme to allocate big data tasks with mobile social cloud | |
| Yucel et al. | User satisfaction aware maximum utility task assignment in mobile crowdsensing | |
| Liu et al. | A truthful auction mechanism for mobile crowd sensing with budget constraint | |
| Huang et al. | Multi-agent reinforcement learning for cost-aware collaborative task execution in energy-harvesting D2D networks | |
| Huang et al. | Group buying based incentive mechanism for mobile crowd sensing | |
| Li et al. | Entropy-based reinforcement learning for computation offloading service in software-defined multi-access edge computing | |
| Yang et al. | Competition-congestion-aware stable worker-task matching in mobile crowd sensing | |
| CN110034958B (en) | Vehicle networking pseudonym change incentive algorithm and change method based on SGUM theory | |
| Pouryazdan et al. | Ta-crocs: Trustworthiness-aware coalitional recruitment of crowd-sensors | |
| Yang et al. | Edge–IoT computing and networking resource allocation for decomposable deep learning inference | |
| CN107330754B (en) | Mobile crowd sensing excitation method for cooperative task | |
| Bastani et al. | A social node model for realising information dissemination strategies in delay tolerant networks |
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 |