CN108269129B - User incentive method in mobile crowd sensing network based on reverse auction - Google Patents

User incentive method in mobile crowd sensing network based on reverse auction Download PDF

Info

Publication number
CN108269129B
CN108269129B CN201810061996.5A CN201810061996A CN108269129B CN 108269129 B CN108269129 B CN 108269129B CN 201810061996 A CN201810061996 A CN 201810061996A CN 108269129 B CN108269129 B CN 108269129B
Authority
CN
China
Prior art keywords
task
user
winning
tasks
users
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201810061996.5A
Other languages
Chinese (zh)
Other versions
CN108269129A (en
Inventor
李慧聪
李万林
刘媛妮
赵国锋
唐红
段洁
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chongqing University of Post and Telecommunications
Original Assignee
Chongqing University of Post and Telecommunications
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Chongqing University of Post and Telecommunications filed Critical Chongqing University of Post and Telecommunications
Priority to CN201810061996.5A priority Critical patent/CN108269129B/en
Publication of CN108269129A publication Critical patent/CN108269129A/en
Application granted granted Critical
Publication of CN108269129B publication Critical patent/CN108269129B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0207—Discounts or incentives, e.g. coupons or rebates
    • 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
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Strategic Management (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Accounting & Taxation (AREA)
  • Development Economics (AREA)
  • Finance (AREA)
  • Economics (AREA)
  • Game Theory and Decision Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Marketing (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention relates to a reverse auction-based user incentive method in a mobile crowd sensing network, which utilizes a reverse auction framework to motivate users to participate in position-related sensing activities and considers the random exit condition in the user sensing process. The scheme aims to maximize the utility of the user on the premise that the budget is feasible, and improves the enthusiasm of the user for participating in the perception activity. Firstly, a task-centric approach is adopted to select users as winner of the perception task to guarantee a higher task coverage. And secondly, according to the task completion condition of the winning bidder, paying the reward to the user by adopting a time proportion sharing rule so as to ensure the authenticity of the incentive. The user incentive method in the mobile crowd sensing network based on the reverse auction has the characteristics of computational effectiveness, individuality, budget balance, authenticity and the like. The invention can enable the user to obtain maximum utility, stimulate the user to participate in the task and improve the participation enthusiasm of the user.

Description

User incentive method in mobile crowd sensing network based on reverse auction
Technical Field
The invention belongs to the technical field of crowd sensing, and particularly relates to a user incentive method in a mobile crowd sensing network based on reverse auction.
Background
Currently, smart phones have integrated many sensors, such as GPS, accelerometer, gyroscope, microphone, camera, etc., which can monitor human activities and the surrounding environment together, so that people can sense and acquire surrounding environment information anytime and anywhere. Sensing and collecting large-scale data by using ubiquitous smart phone users has become a new sensing mode. At present, different applications in the fields of environment monitoring, intelligent transportation, behavior monitoring, indoor positioning and the like are realized by some projects based on mobile crowd sensing.
Data acquisition of the mobile crowd sensing application needs to collect data by means of a large number of terminals of mobile users, however, the mobile terminals consume resources such as time, electric quantity, flow and the like while collecting data, and in the sensing application related to the position, the users may face privacy disclosure threat while sharing own position information. Typically, a normal rational user will only provide perception or computational services in return for incentives. Therefore, in order to recruit enough users, a reasonable incentive mechanism needs to be designed, so that the enthusiasm of the users for participating in the sensing activities is improved, and the users are prompted to participate in the crowd sensing application.
At present, the research of the mobile crowd sensing incentive mechanism mainly stimulates users to participate in the sensing task through different incentive modes, and the incentive modes can be mainly divided into three categories: entertainment games, reputation value and payment of rewards. The entertainment game and the reputation value belong to a non-monetary incentive mode, the entertainment game means that users are encouraged to participate in mobile crowd sensing tasks through position-based games, and research of the mechanism focuses on enriching user experience by setting position-related entertainment games suitable for the sensing tasks, such as mobile phone games Ostereiersuche, Treasure and the like. The reputation value refers to that a user obtains a certain reputation value by executing a perception task, the user can obtain satisfaction (social status and the like), and the platform can also select a user with high quality according to the reputation value of the user to execute the perception task. Such mechanisms focus on improving the quality of the perception data through the enthusiasm of users to participate in perception tasks for maintaining their own social status, interests, and the like. Such as DWI mechanisms, IRONMAN mechanisms, etc. Reddy S and Estrin D et al indicate that incentive mechanisms for monetary adoption, i.e., incentive mechanisms based on payment of a reward, tend to increase the interest of the user in participating more than incentive approaches that are not monetary. Reverse auction is a common payment mode and is used as a basic model in different research works to solve the problems in the mobile crowd sensing incentive mechanism. The crowd sensing platform is a service requester, namely a buyer, and the user is a service provider, namely a seller.
However, due to the selfishness and randomness of the human, the design of the mobile crowd-sourcing perception incentive method based on the reverse auction model faces certain challenges.
Problems caused by selfishness: most participants are stimulated by payment of the crowd sensing platform to participate in the sensing task, and in the crowd sensing system, user income is a main factor influencing user participation, namely, the higher the user utility is, the higher the participation sensing enthusiasm is. Currently, researchers have proposed many motivational models to attract more user participation. For example, Yang Dejun et al proposed a user-centric incentive model Msensing at the mobicom conference of 2012, which mainly aims to seek an optimal set of users to achieve the purpose of maximizing platform utility while compensating for user costs, and users have low utility, resulting in low user participation enthusiasm. Feng Zhenni et al propose a TRAC mechanism that considers the relevance of tasks to location and selects the user with the least payment cost as the winner with the goal of minimizing platform cost. Therefore, in the design of the current mechanism, the maximum benefit of the crowd sensing platform is considered, which is also the reason that the enthusiasm of the participants is not high. The platform utilizes the competitive relationship among the participants to select the participant subset with the lowest price and the lowest payment cost as the winning bidder. Thus, platform-centric incentive mechanisms do not prioritize the interests of participants and do not more efficiently incentivize participants to join a sensory task. Designing a user-centric incentive scheme, aiming at maximizing user utility, will serve a better incentive.
Problems caused by randomness: random movement of people or sudden situations cause the user to leave the perception place during the performance of a task, resulting in the task not being completed. Lee Juongsik and the like artificially ensure higher participation rate while minimizing platform payment cost, adopt a reverse auction mechanism, select the participant with the lowest bid as a winner to pay, and introduce a concept of virtual participation points to avoid that the participant who frequently fails in bidding quits participation. Zhang Xiang et al proposed three excitation models, namely, SS-Model (Single-request Single-bid), SM-Model (Single-request Multiple-bid) and MM-Model (Multiple-request Multiple-bid), respectively according to the number of crowd-sourcing sensing platforms and the number of users bidding, wherein SM-Model is a general form of SS-Model, and MM-Model considers two competition modes of competition between a plurality of crowd-sourcing sensing platforms and competition between a plurality of users, but the three excitation models do not fully consider the problem that a winner quits sensing with random probability in the process of executing a task, and all assume that the winner can complete a won sensing task, however, the above assumptions are not practical.
Disclosure of Invention
The present invention is directed to solving the above problems of the prior art. A user Incentive Method IMRAL (inclusive Method based on reversed experience interaction for Location-aware Sensing) in a mobile crowd Sensing network based on reverse auction is provided, under a reverse auction framework, the selfishness and the randomness of a user are comprehensively considered, the user efficiency is maximized on the premise that the budget of a crowd Sensing platform is feasible to improve the participation enthusiasm of the user, and the user is paid under the condition that the user exits midway in the task execution process to ensure the Incentive authenticity. The invention aims to provide a method for stimulating users to participate in tasks and improving the enthusiasm of the users to participate in perception and the task coverage rate by a task-centered winner selection method and a time proportion sharing rule payment method. The technical scheme of the invention is as follows:
a user incentive method in a mobile crowd sensing network based on reverse auction comprises the following steps:
the crowd sensing platform issues a plurality of sensing tasks, the sensing tasks form a set gamma {1,2, …, m }, m represents the number of the tasks, each task k epsilon gamma has corresponding attributes, and a quadruple is used for representing the attributes<sk,dk,tk,Vk>Wherein s iskIndicating the start time of the task, dkIndicating the task deadline, the time period from the task start time to the task deadline being the task effective time, tkTime required for the user to complete the perception task, VkRepresenting the value of the task;
and if the set of users interested in the task is U ═ 1,2, …, n represents the number of users interested in the task, and the user i reports the bid price of the task to the user B according to the position of the user ii＝(Γi,bi) Wherein a subset of tasks is reported by a user
Figure BDA0001555592440000031
biBidding for the user, namely, the user i is willing to provide the reverse bidding price of the data service;
according to task bidding pairs submitted by users, the crowd sensing platform selects a user subset by adopting a task-centered user selection method
Figure BDA0001555592440000041
As a winner of the task to maximize task coverage;
the winner executes the perception task and submits perception data to the crowd sensing platform;
according to the task completion condition, the crowd sensing platformPayment piGiving the winning bidder i.
Further, in the step 3), the crowd-sourcing sensing platform selects a subset of users that maximize task coverage
Figure BDA0001555592440000042
The method for selecting the win-win target person by taking the task as the center is designed as the win-win target person of the task, the task coverage rate is maximized, and the method specifically comprises the following steps:
31) according to the reported condition of user B ═ U-i∈UBiCounting a task set gamma' of the user participating in bidding, and counting a bidder set U of each task k epsilon gammakAnd the number n of biddersk；
32) Initializing a task set covered by the winner, wherein S is phi, and Γ is phi and represents a task set covered by the winner;
33) for each task k e Γ', perform step 34);
34) if n iskIf 1, step 35) is executed, if n is equal to 1)kIf > 1, executing step 36);
35) calculation of bi/viWherein b isiFor the bidding of user i, viReporting the total value of the task for the user i, if bi/viAnd (5) if the number of the users i is less than or equal to 1, adding the winning bid set S as an executor of the task k to the user i: s ═ S utou { i }, step 37) is performed, otherwise 33) is performed;
36) for all users i ∈ UkB of (a)i/viAnd (3) performing ascending sequencing: b1/v1≤b2/v2≤…≤bL/vLWherein b isL/vLMaximum value representing the ratio of bids among all bidders for task k to the total value of tasks reported, b1/v1B is the minimum value of the ratio of the bid of all bidders of the task k to the total value of the task reported by the bidders1/v1If the result is less than or equal to 1, adding the user into the winning bid set S as an executor of the task k, and executing the step 37), otherwise, executing the step 33);
37) add task k to set Γ ": Γ "═ Γ" { u { k };
38) return winning bid set S, quiltThe task set Γ' covered by the winning set, and b of each task k ∈ Γ1/v1And bL/vLAnd then, the process is ended.
Further, the task coverage is defined as: if number of winning targets num of task kkEqual to or greater than 1, the task is covered, numkNum represents task unmanned executionkNum represents one person performing the taskkMore than two persons execute the task when the number is more than or equal to 2;
further, in the step 5), the reward payment is performed on the winning bidder according to the task completion condition of the winning bidder, the payment is performed on the user by adopting a critical value payment method based on a time proportion sharing rule, when the user executes a certain task, only two possibilities of normally completing the task and quitting midway in the task execution process are provided, the probability of normally completing the task is p, the probability of quitting midway is q 1-p, the Bernoulli distribution is obeyed, and the payment comprises user bidding biWith task awards TkTwo parts.
Further, the payment includes a user bid biWith task awards TkThe method comprises the following two steps:
51) initialization of p i0; Γ '"Φ, Γ'" represents a set of tasks completed by the user;
52) for each task k ∈ Γ ", perform step 53);
53) let TkRewarding the task for task k, if task k is completed then there is Tk＝Vk(bL/vL-b1/v1) Γ' "═ u { k }, otherwise
Figure BDA0001555592440000051
Where Δ tkRepresenting a time at which the user performed the task;
54) computing
Figure BDA0001555592440000052
vΓ”'Representing the total value of the tasks completed by the winner;
55) for each user i e S, perform step 56);
56) calculating a reward for each winning bidder
Figure BDA0001555592440000053
Wherein x represents the number of tasks performed by user i;
57) computing
Figure BDA0001555592440000054
If P > vΓ”'Then step 58) is performed, otherwise step 59) is performed;
58)S＝Φ,pi＝0；
59) return piAnd then, the process is ended.
Further, the method also comprises the step of evaluating the four aspects of the average utility of the user, the task coverage rate, the task completion rate and the platform utility.
Further, the user average utility is: the user average utility is defined as the ratio of the total utility of all winning bidders to the number of winning bidders, and is calculated as follows:
Figure BDA0001555592440000055
wherein | S | represents the number of winning bidders;
task coverage rate: task coverage rate
Figure BDA0001555592440000056
Where cov represents the number of tasks covered by all winning users and m represents the total number of tasks;
task completion rate: the task completion rate γ is defined as the ratio of the number of tasks com completed by all winning bidders to the total number of tasks m, and is calculated as follows:
Figure BDA0001555592440000061
the utility of the platform: the platform utility is an important evaluation index for estimating the feasibility of the incentive method, and can be calculated according to formula 2, and is specifically defined as follows:
the utility function of the crowd-sourcing perception platform is defined as the total value v(s) of the tasks completed by all winning biddersThe difference from the total payout to all winning bidders,
Figure BDA0001555592440000062
the invention has the following advantages and beneficial effects:
compared with the prior art, the invention provides a user incentive method in a mobile crowd sensing network based on reverse auction. The selfishness and the randomness of the user are comprehensively considered, the user participation perception enthusiasm is improved by taking the maximized user utility as a target, and the situation that the user quits perception with random probability in the task execution process due to the randomness of the user is considered. Firstly, a user selection method taking a task as a center is adopted to maximize the task coverage rate, and secondly, under the condition that a winning bid person quits perception with random probability, the winning bid user is paid according to a time proportion sharing rule so as to ensure the authenticity of incentive. Simultaneously, the method also has the following advantages:
the computation time complexity is low, and the time complexity of the user selection method taking the task as the center in the incentive method is O (mn)2) The time complexity of the payment method based on the time proportion sharing rule is O (mn), wherein n is the number of users, and m is the number of tasks. Is a complete polynomial time method and has practical application value.
The incentive method is personal, i.e. the utility u of the user iiNot less than zero. When in use
Figure BDA0001555592440000063
When u is turned oni0; when i ∈ S, ui＝pi-ciWherein
Figure BDA0001555592440000064
And ci≤bi，TkIf u is greater than or equal to 0, u isiIs more than or equal to 0. Thus, the stimulation method is personal.
The incentive method is budget balanced for the platform, namely the platform utility u0Not less than zero. When there is no user to win the set, u 00; when the winning bid set is not nullWhen it is collected, v) is known from step 57)Γ”'P is not less than 0. The incentive method is thus budget balanced.
This excitation method is true. According to Myerson, the authenticity of the stimulation method is explained both in terms of monotonicity and key value.
Monotonicity: due to the fact thati/viSorting from small to large, if user i takes biBecomes the winning bid when the user wins the bid with bi'≤biWhen bidding, because viThe user i can also become the winner.
Key value: assuming that the number of users participating in the bidding task k is greater than or equal to 2, the user i bids biIf it becomes the winning bid, it pays pi＝bi+TkIf user i is greater than piAs a bid, then bi＞bi+TkThen T isk< 0, because
Figure BDA0001555592440000071
B is availableL/vL＜b1/v1Then user i will not win task k, so if user i is greater than piThe value of (c) will not be a winner as a bid price.
Therefore, the incentive method satisfies the reality which has an important role in preventing market monopoly.
Drawings
FIG. 1 is a flowchart illustrating the implementation of the interaction method between the crowd sensing platform and the user according to the preferred embodiment of the present invention;
FIG. 2 is a flow of execution of a task-centric user selection algorithm;
FIG. 3 is a flow of execution of a reward payment method of the proportion-by-time sharing rule;
FIG. 4 is a graph comparing average utility of users;
FIG. 5 is a comparison graph of perceived task coverage;
FIG. 6 is a comparison graph of perceived task completion rates;
FIG. 7 is a graph comparing utility of crowd sensing platform.
Detailed Description
The technical solutions in the embodiments of the present invention will be described in detail and clearly with reference to the accompanying drawings. The described embodiments are only some of the embodiments of the present invention.
The technical scheme for solving the technical problems is as follows: a user incentive method in a reverse auction based mobile crowd-sourcing aware network, the method comprising: the interaction method of the mobile crowd sensing platform and the user comprises the following steps:
1) the crowd sensing platform publishes a sensing task set gamma {1,2, …, m }, each task k epsilon gamma has corresponding attributes, and a quadruple is used for representing<sk,dk,tk,Vk>. Wherein s iskAnd dkThe task start time and the task deadline are respectively shown, and the time period from the task start time to the task deadline is the task effective time. t is tkThe time required for the user to complete the perception task is not greater than the effective time of the task. VkRepresents the value of the task, which is private information of the platform, and the perceptual task is location dependent, i.e. each task needs to be done at a specific location.
2) And setting the set of users interested in the task as U ═ 1,2, …, n }, and reporting the bid pair B of the task by the user i according to the position of the user ii＝(Γi,bi) Wherein a subset of tasks is reported by a user
Figure BDA0001555592440000081
biUsers are bid, i.e. the reverse bid price that user i is willing to provide data services.
3) According to task bidding pairs submitted by users, the crowd sensing platform selects a user subset capable of maximizing task coverage
Figure BDA0001555592440000082
As a winning bid for the task.
4) And the winner executes the perception task and submits perception data to the crowd sensing platform.
5) According to taskCompletion, crowd sensing platform Payment piGiving the winning bidder i.
The cost of user participation in a task is determined by factors such as energy consumption caused by providing services, network bandwidth resource consumption and potential privacy threats. The cost of user i participating in the perception task is ci(ci≤bi) Which is private information of the user.
The user i utility function is defined as follows:
Figure BDA0001555592440000083
pithe platform is paid a reward to winning bidder i.
The utility function of the crowd sensing platform is defined as the difference value between the total value v(s) of tasks completed by all winning bidders and the total payment to all winning bidders, and is specifically defined as follows:
Figure BDA0001555592440000084
the invention aims to preferentially maximize the utility of a user on the premise of feasible platform budget through a user selection and reward payment method, stimulate user participation and improve the perception enthusiasm of user participation, which can be expressed as:
the target is as follows:
Figure BDA0001555592440000085
conditions are as follows:
Figure BDA0001555592440000086
where B represents the platform's payment budget, assuming that its task budget B does not exceed the total value v(s) of the tasks completed by all winning bidders. One feasible and effective excitation method needs to satisfy the following requirements:
the calculation is valid: computationally efficient means that its operating results can be output within polynomial time.
Individuality: the utility of each user is non-negative.
Budget balance: the utility of the crowd-sourcing perception platform is non-negative, i.e. the total payment is not greater than the total value of the tasks completed by the winning bid set.
Authenticity: reality means that, no matter the bids of other bidders, the real cost bids of any user are taken as the dominant strategy, namely, the user can not obtain more utility from the bid value deviating from the real valuation of the task.
The first three characteristics are the basic conditions that ensure that an auction is feasible, authenticity may eliminate user concerns about market manipulation, and Myerson et al demonstrate that an auction mechanism is real and must satisfy both the condition that selection rules are monotonic and that the prize value of winning bidders is critical. Monotonicity, i.e. if the user bids bjIf it becomes the winning target, then use bj'＜bjCan still be the winning target. Key value means if bid price bjHigher than reward pjIt will not be the winner.
The invention provides a realization method of a user incentive method in a mobile crowd sensing network based on reverse auction, which relates to the problems of user selection and reward payment.
In step 3), a user selection method taking the task as the center is designed, and a winning bidder set S is returned to maximize the task coverage rate. Firstly, it proves that the user selection problem meeting the maximized task coverage rate is NP-hard, and secondly, an effective user selection method with low computational complexity, namely a user selection method taking a task as a center, is designed by combining the marginal utility decrement existing in the collection of perception data.
Definition 1 (task overlay): if number of winning targets num of task kkEqual to or greater than 1, the task is covered, numkNum represents task unmanned executionkNum represents one person performing the taskkMore than or equal to 2 represents that more than two persons execute the task.
Theorem 1: the user selection problem is NP-hard.
And (3) proving that: the Weighted Multiple Set Coverage Problem (WMSCP) has been proven by Yang Jian et al to be an NP-hard problem. While the weighted multi-tasking coverage problem may be reduced to a user-selected problem in linear time. Thus, the user selection problem is NP-hard.
The diminishing marginal utility caused by data redundancy is increasingly severe as the number of users participating in the same perceptual task increases. For example, when several mobile phones simultaneously collect noise in a certain area, the data collected by one or two mobile phones is enough to estimate the noise level in the area, and collecting more mobile phones cannot effectively improve the accuracy of the estimated noise, but increases data redundancy and social cost. The perception task value is certain, and the more users participating in the same task, the less reward each user can obtain. Therefore, in order to improve the user utility and the task coverage rate, in the user selection stage, a winner selection algorithm taking the task as the center is provided, and the user with low bid price and high total reported task value is selected to execute the perception task according to the principle that any one task is executed by one person, so that the user utility is improved, and the user participation is promoted.
In step 3), a user selection method taking the task as the center is designed, and a winning bidder set S is returned to maximize the task coverage rate. Firstly, it proves that the user selection problem meeting the maximized task coverage rate is NP-hard, and secondly, an effective user selection method with low computational complexity, namely a user selection method taking a task as a center, is designed by combining the marginal utility decrement existing in the collection of perception data. The execution flow of the task-centric user selection algorithm is shown in fig. 2, and comprises the following steps:
31) according to the reported condition of user B ═ U-i∈UBiCounting a task set gamma' of the user participating in bidding, and counting a bidder set U of each task k epsilon gammakAnd the number n of biddersk；
32) Initializing a task set covered by the winner, wherein S is phi, and Γ is phi and represents a task set covered by the winner;
33) for each task k e Γ', perform step 34);
34) if n iskIf 1, step 35) is executed, if n is equal to 1)kIf > 1, executing step 36);
35) calculation of bi/viWherein b isiFor the bidding of user i, viReporting the total value of the task for the user i, if bi/viAnd (5) if the number of the users i is less than or equal to 1, adding the winning bid set S as an executor of the task k to the user i: s ═ S utou { i }, step 37) is performed, otherwise 33) is performed;
36) for all users i ∈ UkB of (a)i/viAnd (3) performing ascending sequencing: b1/v1≤b2/v2≤…≤bL/vLIf b is1/v1If the result is less than or equal to 1, adding the user into the winning bid set S as an executor of the task k, and executing the step 37), otherwise, executing the step 33);
37) add task k to set Γ ": Γ "═ Γ" { u { k };
38) returning a winning bid set S, a task set Γ' covered by the winning bid set, and b of each task k ∈ Γ1/v1And bL/vLAnd then, the process is ended.
In step 5), payment is made for the reward winning person according to the task completion condition of the winning person. In order to ensure the authenticity of the incentive method and simultaneously consider the random exit condition in the user perception process, a critical value payment method sharing the rule according to the time proportion is adopted to pay for the user. When a user executes a certain task, only two possibilities of normally completing the task and quitting halfway in the process of executing the task are available, the probability of normally completing the task is p, and the probability of quitting halfway is q 1-p, so that the Bernoulli distribution is obeyed. Payment includes user bids biWith task awards TkThe method comprises the following two steps:
51) initialization of p i0; Γ '"Φ, Γ'" represents a set of tasks completed by the user;
52) for each task k ∈ Γ ", perform step 53);
53) let TkRewarding the task for task k, if task k is completed then there is Tk＝Vk(bL/vL-b1/v1) Γ' "═ u { k }, otherwise
Figure BDA0001555592440000111
Where Δ tkRepresenting a time at which the user performed the task;
54) computing
Figure BDA0001555592440000112
vΓ”'Representing the total value of the tasks completed by the winner;
55) for each user i e S, perform step 56);
56) calculating a reward for each winning bidder
Figure BDA0001555592440000113
Wherein x represents the number of tasks performed by user i;
57) computing
Figure BDA0001555592440000114
If P > vΓ”'Then step 58) is performed, otherwise step 59) is performed;
58)S＝Φ,pi＝0；
59) return piAnd then, the process is ended.
In order to illustrate the feasibility and the effectiveness of the excitation method, the performance of the excitation method IMRAL and TRAC (transmissive interaction for localization-aware chromatography sensing in mobile throughput) and IMC-SS (inductive mechanism for throughput chromatography in the single-request single-bit-model) in the four aspects of user average utility, task coverage rate, task completion rate and platform utility is researched under the same experimental conditions.
Average utility of users: the user average utility is defined as the ratio of the total utility of all winning bidders to the number of winning bidders, and is calculated as follows:
Figure BDA0001555592440000115
where | S | represents the number of winning bidders.
Task coverage rate: task coverage rate
Figure BDA0001555592440000121
Wherein cov indicates that the winning bid is covered by all usersThe number of tasks of the lid, m, represents the total number of tasks.
Task completion rate: the task completion rate γ is defined as the ratio of the number of tasks com completed by all winning bidders to the total number of tasks m, and is calculated as follows:
Figure BDA0001555592440000122
the utility of the platform: the platform utility is an important evaluation index for estimating the feasibility of the incentive method, and can be calculated according to the formula 2.
Firstly, initializing relevant parameters, assuming that the probability of the user exiting from perception in the process of executing the task is 0.2, and if the user exits from perception task k, assuming that delta t is delta tk/tkUniform distribution is obeyed between 0 and 1, and the number of tasks that the user is allowed to report is 5 at most. Specific parameter settings are shown in table 1.
Table 1 experimental parameter settings
Figure BDA0001555592440000123
As most participants are stimulated by payment of the crowd sensing platform to participate in the sensing task, the user income is an important factor influencing the participation of the user in the crowd sensing system, and the higher the user utility is, the higher the enthusiasm of the user to participate in the sensing task is. On the other hand, the higher the user's enthusiasm, the greater the likelihood that the task will be accepted, i.e., the higher the task coverage, which is also related to factors such as the user selection method and the payment method. Therefore, for comparison, the enthusiasm of the user for participating in the perception task can be illustrated by two indexes, namely the average utility of the user and the task coverage rate.
The average utility of the user is shown in fig. 4. Fig. 4 shows a comparison of the user average utility of the excitation methods IMRAL and TRAC, IMC-SS proposed by the present description. In contrast to TRAC and IMC-SS, payment of reward by TRAC and IMC-SS mechanisms is only related to their bid and number of tasks, since the incentive method proposed by the present description pays for T taskskThe user can get relatively highThe task compensation improves the willingness degree of the user to participate in the perception task. Fig. 5 shows the task coverage for three excitation methods. As can be seen, the task coverage under IMC-SS is slightly lower than that under IMRAL. The user effectiveness and the task coverage condition are combined, comprehensive obtaining is achieved, and compared with a TRAC mechanism and an IMC-SS mechanism, the user enthusiasm under IMRAL is higher.
As shown in fig. 6, fig. 6 is a comparison graph of perceived task completion rates. Both TRAC and IMC-SS mechanisms assume that task coverage is realized, so that the task completion rate and the task coverage rate have the same change trend, and IMRAL considers that a user exits midway in the task execution process due to the randomness of the user, so that the situation that the task is not completed is sensed. It can be seen that the randomness of the user will reduce the perceived task completion rate.
FIG. 7 is a comparison of crowd sensing platform utility. As can be seen from fig. 7(a), at a certain number of users, as the number of tasks increases, the platform utility under IMRAL and IMC-SS increases, and the IMRAL platform utility is lower compared to IMC-SS. In fig. 7(b), the platform utilities of the three excitation methods tend to be stable after increasing, because the task completion rate increases with the increase of the number of users, and when the task set is saturated, the platform utilities do not increase. It can be seen that the randomness of the user will reduce the utility of the crowd sensing platform.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that various changes and modifications may be made without departing from the spirit or essential attributes thereof, and all such equivalent changes and modifications as fall within the scope of the appended claims.
The above examples are to be construed as merely illustrative and not limitative of the remainder of the disclosure. After reading the description of the invention, the skilled person can make various changes or modifications to the invention, and these equivalent changes and modifications also fall into the scope of the invention defined by the claims.

Claims (6)

1. A user incentive method in a mobile crowd sensing network based on reverse auction is characterized by comprising the following steps:
1) the crowd sensing platform issues a plurality of sensing tasks, the sensing tasks form a set gamma {1,2, …, m }, m represents the number of the tasks, each task k epsilon gamma has corresponding attributes, and a quadruple is used for representing the attributes<sk,dk,tk,Vk>Wherein s iskIndicating the start time of the task, dkIndicating the task deadline, the time period from the task start time to the task deadline being the task effective time, tkTime required for the user to complete the perception task, VkRepresenting the value of the task;
2) and if the set of users interested in the task is U ═ 1,2, …, n represents the number of users interested in the task, and the user i reports the bid price of the task to the user B according to the position of the user ii＝(Γi,bi) Wherein a subset of tasks is reported by a user
Figure FDA0003310141500000011
biBidding for the user, namely, the user i is willing to provide the reverse bidding price of the data service;
3) according to task bidding pairs submitted by users, the crowd sensing platform selects a user subset by adopting a task-centered user selection method
Figure FDA0003310141500000012
As a winner of the task to maximize task coverage;
4) the winner executes the perception task and submits perception data to the crowd sensing platform;
5) according to the task completion condition, paying by the crowd sensing platformiWinning the tender marki；
In the step 3), the crowd-sourcing platform selects a subset of users that can maximize task coverage
Figure FDA0003310141500000013
As a winning bid for the task,a method for selecting a winner with a task as a center is designed, and the task coverage rate is maximized, and specifically comprises the following steps:
31) according to the reported condition of user B ═ U-i∈UBiCounting a task set gamma' of the user participating in bidding, and counting a bidder set U of each task k epsilon gammakAnd the number n of biddersk；
32) Initializing a task set covered by the winner, wherein S is phi, and Γ is phi and represents a task set covered by the winner;
33) for each task k e Γ', perform step 34);
34) if n iskIf 1, step 35) is executed, if n is equal to 1)kIf > 1, executing step 36);
35) calculation of bi/viWherein b isiFor the bidding of user i, viReporting the total value of the task for the user i, if bi/viAnd (5) if the number of the users i is less than or equal to 1, adding the winning bid set S as an executor of the task k to the user i: s ═ S utou { i }, step 37) is performed, otherwise 33) is performed;
36) for all users i ∈ UkB of (a)i/viAnd (3) performing ascending sequencing: b1/v1≤b2/v2≤…≤bL/vLWherein b isL/vLMaximum value representing the ratio of bids among all bidders for task k to the total value of tasks reported, b1/v1Then the minimum value of the ratio of the bid of all bidders for task k to the total value of the task it reports, if b1/v1If the result is less than or equal to 1, adding the user into the winning bid set S as an executor of the task k, and executing the step 37), otherwise, executing the step 33);
37) add task k to set Γ ": Γ "═ Γ" { u { k };
38) returning a winning bid set S, a task set Γ' covered by the winning bid set, and b of each task k ∈ Γ1/v1And bL/vLAnd then, the process is ended.
2. The reverse auction based mobile crowd-sourcing aware network user incentive method of claim 1, wherein the method comprisesThe definition of task coverage is: if number of winning targets num of task kkEqual to or greater than 1, the task is covered, numkNum represents task unmanned executionkNum represents one person performing the taskkMore than or equal to 2 represents that more than two persons execute the task.
3. The reverse auction based user incentive method in mobile crowd sensing network of claim 2, wherein in the step 5), the reward payment is made to the winning bidder according to the task completion condition of the winning bidder, the payment is made to the user by a critical value payment method based on the time proportion sharing rule, when the user executes a certain task, the user only has two possibilities of normally completing the task and quitting midway in the process of executing the task, the probability of normally completing the task is p, the probability of quitting midway is q-1-p, and obeys Bernoulli distribution, the payment includes the user bidding biWith task awards TkTwo parts.
4. A reverse auction based mobile crowd-sourcing aware network user incentive method according to claim 3, wherein the payment comprises a user bid biWith task awards TkThe method comprises the following two steps:
51) initialization of pi0; Γ '"Φ, Γ'" represents a set of tasks completed by the user;
52) for each task k ∈ Γ ", perform step 53);
53) let TkRewarding the task for task k, if task k is completed then there is Tk＝Vk(bL/vL-b1/v1) Γ' "═ u { k }, otherwise
Figure FDA0003310141500000021
Where Δ tkRepresenting a time at which the user performed the task;
54) computing
Figure FDA0003310141500000022
vΓ”'Representing the total value of the tasks completed by the winner;
55) for each user i e S, perform step 56);
56) calculating a reward for each winning bidder
Figure FDA0003310141500000031
Wherein x represents the number of tasks performed by user i;
57) computing
Figure FDA0003310141500000032
If P > vΓ”'Then step 58) is performed, otherwise step 59) is performed;
58)S＝Φ,pi＝0；
59) return piAnd then, the process is ended.
5. The reverse auction based mobile crowd-sourcing awareness network user incentive method of claim 1, further comprising the step of evaluating the user average utility, task coverage, task completion and platform utility.
6. The reverse auction based mobile crowd-sourcing aware network user incentive method of claim 5, wherein the user average utility is: the user average utility is defined as the ratio of the total utility of all winning bidders to the number of winning bidders, and is calculated as follows:
Figure FDA0003310141500000033
wherein | S | represents the number of winning bidders; ci represents the cost of user i to participate in the perception task;
task coverage rate: task coverage rate
Figure FDA0003310141500000034
Where cov represents the number of tasks covered by all winning users and m represents the total number of tasks;
task completion rate: the task completion rate γ is definedThe ratio of the number of tasks com to the total number of tasks m completed by all winning bidders is calculated as follows:
Figure FDA0003310141500000035
the utility of the platform: the platform utility is an important evaluation index for estimating the feasibility of the incentive method, and can be calculated according to formula 2, and is specifically defined as follows:
the utility function of the crowd-sourcing perception platform is defined as the difference between the total value v(s) of the tasks completed by all the winning bidders and the total payout to all the winning bidders,
Figure FDA0003310141500000036
CN201810061996.5A 2018-01-23 2018-01-23 User incentive method in mobile crowd sensing network based on reverse auction Active CN108269129B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810061996.5A CN108269129B (en) 2018-01-23 2018-01-23 User incentive method in mobile crowd sensing network based on reverse auction

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810061996.5A CN108269129B (en) 2018-01-23 2018-01-23 User incentive method in mobile crowd sensing network based on reverse auction

Publications (2)

Publication Number Publication Date
CN108269129A CN108269129A (en) 2018-07-10
CN108269129B true CN108269129B (en) 2022-02-08

Family

ID=62776490

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810061996.5A Active CN108269129B (en) 2018-01-23 2018-01-23 User incentive method in mobile crowd sensing network based on reverse auction

Country Status (1)

Country Link
CN (1) CN108269129B (en)

Families Citing this family (20)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109408228B (en) * 2018-09-30 2019-10-15 陕西师范大学 Intelligent perception method for allocating tasks based on allocation of budget
CN109447489B (en) * 2018-11-02 2020-10-30 华东交通大学 Selection method and device of crowd sensing user, electronic equipment and storage medium
CN109460929B (en) * 2018-11-15 2023-04-07 平安科技（深圳）有限公司 Crowd-sourced task pushing method based on geographic position operation and related equipment
CN109636484B (en) * 2018-12-25 2021-12-07 中南大学 Motivation method of crowd sensing system based on behavior economics preference theory
CN109872058B (en) * 2019-01-31 2022-12-13 南京工业大学 Multimedia crowd sensing excitation method for machine learning system
CN109858831A (en) * 2019-02-25 2019-06-07 重庆大学 Task crowdsourcing allocation processing method based on personalized competitive bidding Task-decomposing
CN109615285A (en) * 2019-02-25 2019-04-12 重庆大学 A kind of task crowdsourcing allocation processing method based on the personalized competitive bidding relations of distribution
CN109919742A (en) * 2019-02-25 2019-06-21 重庆大学 Task crowdsourcing allocation processing method based on personalized competitive bidding excitation
CN109886596B (en) * 2019-03-01 2023-01-13 中南大学 Method for improving cooperative rate of crowd sensing system based on psychological account theory
CN109978352A (en) * 2019-03-14 2019-07-05 南京邮电大学 A kind of motivational techniques towards space-time big data intelligent perception task
CN109978353B (en) * 2019-03-14 2021-07-27 南京邮电大学 Big data crowd sensing incentive method facing position-adjustable user
CN110366125A (en) * 2019-06-19 2019-10-22 武汉科技大学 Mobile intelligent perception motivational techniques under the conditions of space-time covering and budget
CN110992121B (en) * 2019-10-22 2024-03-22 西安电子科技大学 Perception task information distribution system and method based on perception error in crowd sensing
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
CN111507757B (en) * 2020-04-09 2024-03-15 中南大学 Crowd sensing excitation method for improving task completion rate in remote areas
CN111626563B (en) * 2020-04-27 2022-09-23 南京邮电大学 Dual-target robust mobile crowd sensing system and excitation method thereof
CN113592610A (en) * 2021-05-14 2021-11-02 南京航空航天大学 Reputation updating mobile crowd sensing excitation method based on fuzzy control
CN114760585B (en) * 2022-04-18 2024-04-16 中南大学 Method, system and equipment for intelligent perception excitation of vehicle group
CN115002713B (en) * 2022-08-03 2022-10-18 中南大学 Method, system, medium and device for improving crowd sensing coverage rate
CN115099535B (en) * 2022-08-24 2022-12-20 东南大学 Dual-target crowd sensing excitation mechanism method

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104850935A (en) * 2015-04-15 2015-08-19 南京邮电大学 Mobile group intelligent perception excitation method with minimized payment as object
CN105550020A (en) * 2014-10-28 2016-05-04 北京邮电大学 Object selection method and apparatus
CN105787788A (en) * 2016-03-11 2016-07-20 南京邮电大学盐城大数据研究院有限公司 Budget-based mobile crowd sensing incentive frame with continuous time interval coverage
CN107563616A (en) * 2017-08-17 2018-01-09 上海大学 A kind of user task distribution and the quorum-sensing system system and method for incentives strategy

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8219432B1 (en) * 2008-06-10 2012-07-10 Amazon Technologies, Inc. Automatically controlling availability of tasks for performance by human users
WO2016038412A1 (en) * 2014-09-10 2016-03-17 Umm Al-Qura University A spatio-temporal method and system to implement boundary regulation
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

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105550020A (en) * 2014-10-28 2016-05-04 北京邮电大学 Object selection method and apparatus
CN104850935A (en) * 2015-04-15 2015-08-19 南京邮电大学 Mobile group intelligent perception excitation method with minimized payment as object
CN105787788A (en) * 2016-03-11 2016-07-20 南京邮电大学盐城大数据研究院有限公司 Budget-based mobile crowd sensing incentive frame with continuous time interval coverage
CN107563616A (en) * 2017-08-17 2018-01-09 上海大学 A kind of user task distribution and the quorum-sensing system system and method for incentives strategy

Non-Patent Citations (6)

* Cited by examiner, † Cited by third party
Title
Reverse Auction Based Incentive Mechanism for Location-Aware Sensing in Mobile Crowd Sensing;Yuanni Liu 等;《Conference: 2018 IEEE International Conference on Communications (ICC 2018)》;20180731;1-6 *
Truthful Incentive Mechanisms for Crowdsourcing;Xiang Zhang 等;《2015 IEEE Conference on Computer Communications》;20150824;2830-2838 *
基于不完全信息博弈的移动群智感知网络激励机制研究;李慧聪;《中国优秀硕士学位论文全文数据库 信息科技辑》;20200115(第01期);I136-1209 *
基于区域覆盖的群智感知激励机制;邢春晓 等;《南京工程学院学报 (自然科学版)》;20171231;第15卷(第4期);14-20 *
基于拍卖模型的移动群智感知网络激励机制;刘媛妮 等;《通信学报》;20190725;第40卷(第7期);208-222 *
邢春晓 等.基于区域覆盖的群智感知激励机制.《南京工程学院学报 (自然科学版)》.2017,第15卷(第4期),14-20. *

Also Published As

Publication number Publication date
CN108269129A (en) 2018-07-10

Similar Documents

Publication Publication Date Title
CN108269129B (en) User incentive method in mobile crowd sensing network based on reverse auction
Luo et al. Sustainable incentives for mobile crowdsensing: Auctions, lotteries, and trust and reputation systems
Jiang et al. Toward optimal participant decisions with voting-based incentive model for crowd sensing
Guo et al. TaskMe: Toward a dynamic and quality-enhanced incentive mechanism for mobile crowd sensing
CN110189174A (en) A kind of mobile intelligent perception motivational techniques based on quality of data perception
Syrgkanis et al. Composable and efficient mechanisms
Krontiris et al. Monetary incentives in participatory sensing using multi-attributive auctions
US8195498B2 (en) Modeling a plurality of contests at a crowdsourcing node
Krause et al. The demand for water: Consumer response to scarcity
Zhu et al. A fair incentive mechanism for crowdsourcing in crowd sensing
Wang et al. QuaCentive: a quality-aware incentive mechanism in mobile crowdsourced sensing (MCS)
CN108681921B (en) Method and device for acquiring crowd sensing incentive strategy based on random game
CN110390560A (en) A kind of mobile intelligent perception multitask pricing method based on Stackelberg game
CN108171593B (en) Mobile crowd sensing incentive method with budget based on site coverage
Liu et al. A truthful auction mechanism for mobile crowd sensing with budget constraint
Liu et al. Reverse auction based incentive mechanism for location-aware sensing in mobile crowd sensing
CN106973392A (en) A kind of motivational techniques based on region overlay in intelligent perception
Huang et al. Group buying based incentive mechanism for mobile crowd sensing
CN105787788A (en) Budget-based mobile crowd sensing incentive frame with continuous time interval coverage
Xiong et al. MAIM: A novel incentive mechanism based on multi-attribute user selection in mobile crowdsensing
Liu et al. TGBA: A two-phase group buying based auction mechanism for recruiting workers in mobile crowd sensing
Suliman et al. A greedy-proof incentive-compatible mechanism for group recruitment in mobile crowd sensing
Sun et al. Collection-behavior based multi-parameter posted pricing mechanism for crowd sensing
Liu et al. Incentive mechanisms for crowdsensing: Motivating users to preprocess data for the crowdsourcer
CN111507757B (en) Crowd sensing excitation method for improving task completion rate in remote areas

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