CN111464620A - Edge-assisted mobile crowd sensing truth value discovery system and excitation method thereof - Google Patents

Edge-assisted mobile crowd sensing truth value discovery system and excitation method thereof Download PDF

Info

Publication number
CN111464620A
CN111464620A CN202010236632.3A CN202010236632A CN111464620A CN 111464620 A CN111464620 A CN 111464620A CN 202010236632 A CN202010236632 A CN 202010236632A CN 111464620 A CN111464620 A CN 111464620A
Authority
CN
China
Prior art keywords
edge
user
cloud
truth
budget
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202010236632.3A
Other languages
Chinese (zh)
Other versions
CN111464620B (en
Inventor
徐佳
杨尚书
丁玉青
周远航
钱一航
徐力杰
鲁蔚锋
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.)
Nanjing University of Posts and Telecommunications
Original Assignee
Nanjing University of Posts 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 Nanjing University of Posts and Telecommunications filed Critical Nanjing University of Posts and Telecommunications
Priority to CN202010236632.3A priority Critical patent/CN111464620B/en
Publication of CN111464620A publication Critical patent/CN111464620A/en
Application granted granted Critical
Publication of CN111464620B publication Critical patent/CN111464620B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • 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
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • 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
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses an edge-assisted mobile crowd sensing truth value discovery system and an incentive method thereof, which particularly comprise a truth value discovery phase and a budget feasible reverse auction phase. In a truth finding stage, the method carries out truth finding on a deep cloud layer and an edge cloud layer. In the budget feasible reverse auction stage, the invention provides a greedy method for selecting users under the budget constraint to maximize the quality function. The method provided by the invention is effective and real in calculation, feasible in budget and capable of guaranteeing the approximation degree of the constant, and is superior to the similar method in the aspects of true value discovery precision and system expandability.

Description

Edge-assisted mobile crowd sensing truth value discovery system and excitation method thereof
Technical Field
The invention relates to an edge-assisted mobile crowd sensing truth value discovery system and an excitation method thereof.
Background
Mobile crowd-sourcing perception is a human-driven activity that exploits the ubiquity of wireless connections to enable various mobile devices with built-in sensing capabilities and inherent user mobility to create dense and dynamic data sets that can effectively characterize our environmental information. Mobile crowd sensing has become an effective method for data acquisition in sensing applications, such as photo selection, public bicycle travel selection, indoor positioning systems, and the like.
Most existing mobile crowd-sourcing sensing systems rely on cloud services to accomplish tasks of collecting/aggregating sensory data, assigning tasks, estimating truth values, and motivating mobile users. Cloud-based mobile crowd-sourcing aware systems have some significant drawbacks, for example, due to computing and communication congestion on cloud servers, which can lead to poor scalability of crowd-sourcing awareness, difficulty in identifying false locations of awareness data, and attendant risks of high data security and user privacy exposure.
The mobile crowd-sourcing perception system based on the edge computing architecture has the following main advantages: the computational complexity is reduced: the edge computing-based mobile crowd-sourcing aware architecture can parallelize computing by offloading the computing from the cloud to multiple edge servers; and (3) reducing delay: little or no communication between the cloud and the mobile user; position sensing: most mobile crowd-sourcing aware tasks are location dependent. Edge computing resources (e.g., base stations, access points) are typically located at specific locations: since the edge server collects the sensing data only in the deployment area thereof, the location attribute of the sensing data is easily verified; flexible data processing: mobile crowd-sourcing awareness based on edge computing gives edge servers flexibility in local data processing (e.g., aggregation, truth discovery and inference of temperature, noise level, transportation, air conditions in a particular area); reducing privacy threats: the sensing data is distributed among a plurality of edge servers. The distributed storage of the sensing data in the edge servers not only enhances the security of the data, but also reduces the privacy threat of the user.
Related research on edge-computing-based mobile crowd-sourcing awareness has been conducted to achieve or solve certain objective problems, such as edge-computing-based data processing, privacy protection, reputation management, task offloading of vehicle crowdsourcing applications, and extracting context information, among others. However, there is no real value in the prior art to find and motivate the method design.
Disclosure of Invention
The purpose of the invention is as follows: the invention aims to provide an edge-assisted mobile crowd sensing truth value discovery system, wherein a mobile crowd sensing task with budget is distributed to each edge cloud; users interested in performing tasks submit their bids, along with sensory data, to respective edge clouds for reference, each edge cloud performing a truth discovery for each task.
It is another object of the present invention to provide an incentive method for an edge-assisted mobile crowd-sensing truth discovery system, each edge cloud selecting a subset of users as winners to maximize the quality of the winners under budget constraints and determining a payment to pay the winners.
The technical scheme is as follows: the utility model provides an edge assistance removes crowd's wisdom perception truth and discovers system, is applicable to a mobile crowd's wisdom perception system which characterized in that: the truth discovery system is located on a platform in a deep cloud, the platform comprising a set of r edge clouds, E ═ E1,e2,...,erAnd a set of n smartphone users, U ═ 1,2kAs edge clouds ekSet of users in a coverage area, where U ═ UUk1, 2.., r, any user belongs to a plurality of edge clouds;
the deep cloud sets m tasks T as T1,t2,...,tmDistributing the data to all edge clouds to perform mobile crowd sensing, determining budget for each edge cloud according to the importance of the sensing data in the coverage area of the edge cloud, and enabling G to be (G)1,G2,...,Gr) For budget profiles of all edge clouds, each task tj∈ T and a task type mujAssociated, the type representing a task tjThe importance of (c); let mu be (mu)1,μ2,...,μm) Is composed ofThe types of all tasks;
each edge cloud ek∈ E records and distributes tasks to a subset of users UkEach user i ∈ U submits a triple B to the edge cloud to which it belongsi＝(Ti,bi,Xi)，TiIs the set of tasks that user i is willing to perform, biIs a bid of user i, XiIs submitted by user i and task set TiCorresponding perception data, each task set TiAnd cost ciThe cost refers to the cost for the user i to submit the data for data perception, such as the spent computing power, storage space, time, flow and the like; c. CiKnown only to user i, user submits task TiLet X be (X)1,X2,...,Xn) For all the perception data submitted by the user,
Figure BDA0002431213250000021
let TkAnd XkRespectively towards the edge cloud ekSubmitted tasks and sensory data;
upon receiving the perception data, each edge cloud ek∈ E calculate each i ∈ UkWeight w of the user(s)iEstimating truth values for all perceptual tasks through truth discovery
Figure BDA0002431213250000022
Let wkIs ekAll users U inkBy weight of (A), let X*Estimated true values for all final edge clouds.
Further, the cloud e at any edgekThe steps of the above truth finding phase are as follows:
step 201: the method comprises the steps that a user collects sensing data and submits the sensing data to an edge cloud server where the user is located;
step 202: real value of random initialization task of edge cloud server
Figure BDA0002431213250000023
And an iteration upper limit
Figure BDA0002431213250000024
Step 203: initializing a boundary cloud iteration counter, wherein K is 0;
step 204: on the edge cloud
Figure BDA0002431213250000031
Is given to
Figure BDA0002431213250000032
Step 205, calculate all users i ∈ UkWeight of (2)
Figure BDA0002431213250000033
Wherein
Figure BDA0002431213250000034
stdjIs task tjStandard deviation of all perceptual data of (a);
step 206: computing all tasks tj∈TkTrue value of
Figure BDA0002431213250000035
Step 207: updating the current iteration times K to K + 1;
step 208: if it is
Figure BDA0002431213250000036
And is
Figure BDA0002431213250000037
Step 204 is executed, otherwise step 209 is executed;
step 209: outputting true values for estimating all tasks
Figure BDA0002431213250000038
And weight w of all usersk。
Further, the edge cloud performs budget reverse auctions, giving a task set TkUser set UkBudget GkTask type μ and bid strategy
Figure BDA0002431213250000039
Each edge cloud ekCalculating a set of winners
Figure BDA00024312132500000310
And each winner i ∈ SkPayment of (p)iLet us order
Figure BDA00024312132500000311
Let p bekAnd p is each SkAnd the payment policy of S,
the utility of any user i is defined as the difference between the payment and its actual cost:
ui＝pi-ci(1)
for any task tj∈TiFrom edge clouds ekWinner set S inkThe obtained quality function is defined as:
Figure BDA00024312132500000312
wherein the log reflects the diminishing returns of the platform in terms of quality brought by participating users;
the quality function is maximized so that the total payment does not exceed the budget, which the present invention calls as the budget feasible quality optimization problem, expressed as:
target Maximize V (S)k) (3)
And (3) constraint:
Figure BDA0002431213250000041
further, the budget feasible reverse auction phase comprises the following steps:
step 301: initializing winner sets
Figure BDA0002431213250000042
Initializing a payment policy pk＝0；
Step 302: initializing a set of users whose bids do not exceed a budget
Figure BDA0002431213250000043
Step 303: let i*Is composed of
Figure BDA0002431213250000044
The most valuable users in the set:
Figure BDA0002431213250000045
step 304: performing step 305 with an 2/5 probability and performing step 306 with a 3/5 probability;
step 305: will i*Joining to a winner set SkAnd set user i*Is paid as
Figure BDA0002431213250000046
Step 317 is executed;
step 306: in that
Figure BDA0002431213250000047
Is found in
Figure BDA0002431213250000048
The user i with the largest value, wherein
Figure BDA0002431213250000049
Vi'(Sk)＝V(Sk∪{i'})＝V(Sk)；
Step 307: if it satisfies
Figure BDA00024312132500000410
Step 308 is executed, otherwise step 310 is executed;
step 308: adding user i to the set of winners SkPerforming the following steps;
step 309: in that
Figure BDA00024312132500000411
Is found in
Figure BDA00024312132500000412
The user i with the largest value, wherein
Figure BDA00024312132500000413
Is a set
Figure BDA00024312132500000414
And set SkStep 307 is performed;
step 310: judging the pair set SkWhether each user i has performed steps 311 to 316; if so, go to step 317;
step 311: order to
Figure BDA00024312132500000415
Order to
Figure BDA00024312132500000416
Step 312: finding collections
Figure BDA00024312132500000417
In
Figure BDA00024312132500000418
The user with the largest value i' where
Figure BDA00024312132500000419
Step 313: if user i' satisfies
Figure BDA00024312132500000420
Step 314 is executed, otherwise step 310 is executed;
step 314: finding collections
Figure BDA00024312132500000421
In
Figure BDA00024312132500000422
The user with the largest value i' where
Figure BDA00024312132500000423
Step 315: computing
Figure BDA00024312132500000424
Step 316: and adding i 'to S'kThe method comprises the following steps: s'k＝S'k∪ { i' }, go to step 313;
step 317: outputting the winner set SkPayment policy pk。
Further, the edge clouds submit estimated truth values to the deep clouds, and each edge cloud e is calculated by the deep cloudsk∈ E weight gkTime, edge cloud ekThe formula for weight update of (c) is:
Figure BDA0002431213250000051
wherein
Figure BDA0002431213250000052
α is two constants, α is a reliability coefficient, β is an importance coefficient, α is the influence of the number of people determining the edge cloud on the last truth discovery, the larger α is, the more edge cloud will have larger weight and play larger influence in the truth discovery process of the deep cloud, the larger β is, the more budget edge cloud will have larger weight and play larger influence in the truth discovery process of the deep cloud, wherein α∈ [0,1 ]],β∈[0,1]α + β is 1, finally according to X*Estimating final true value X of all perception tasks by a truth value discovery method**。
Furthermore, the method for discovering and exciting the true value of the edge-assisted mobile crowd sensing is feasible in budget.
And (3) proving that: is provided with
Figure BDA0002431213250000053
And is
Figure BDA0002431213250000054
For all i ∈ Sk'\SkSuppose there is
Figure BDA0002431213250000055
Then add all inequalities together, have
Figure BDA0002431213250000056
Is equivalent to
Figure BDA0002431213250000057
So that initially the assumption is false, there
Figure BDA0002431213250000058
Now let S0Is an empty set, S1There is only one user, and so on. Assuming presence of a user
Figure BDA0002431213250000059
Can offer a bid
Figure BDA00024312132500000510
Still become the winner (user j originally bids b)j) At this time j increments bid to b'jOthers remain unchanged.
Can find that
Figure BDA00024312132500000511
So j remains in the winner set Sj-1In (1).
The combination selected before j is included in the winning combination is denoted by S. Thus, there are
Figure BDA0002431213250000061
Figure BDA0002431213250000062
Can assume that
Figure BDA0002431213250000063
This is true in fact, otherwise S ∪ { j } ═ Sk∪ S and
Figure BDA0002431213250000064
thus is provided with
Figure BDA0002431213250000065
If R is SkS, applying equation (5) and the above equation, one can obtain:
for user r0∈ R \ j, there is
Figure BDA0002431213250000066
It is known that
Figure BDA0002431213250000067
Further obtain the
Figure BDA0002431213250000068
Combining the inequalities to obtain
Figure BDA0002431213250000069
Is ready to obtain
b'(Sk∪S)-b'(S∪{j})＝b'(R\{j})＝b(R\{j})≤b(Sk)。
Because of the fact that
Figure BDA00024312132500000610
To pair
Figure BDA00024312132500000611
Are all true, therefore have
Figure BDA00024312132500000612
And
Figure BDA00024312132500000613
can obtain
Figure BDA00024312132500000614
Thus V (S)k) < 2V (S ∪ { j }), and combining inequality (6), finally obtaining
To:
Figure BDA00024312132500000615
namely, it is
Figure BDA00024312132500000616
Therefore, the quotation of all users is less than or equal to G when being accumulatedk。
Furthermore, the method for discovering and exciting the true value of the edge-assisted mobile crowd sensing is computationally efficient. And (3) proving that: the truth on the edge cloud is first analyzed to find the phase time complexity. It takes time o (nm) to update the weights of all users in each edge cloud (step 205). It takes time o (nm) to update the true facies of all tasks in each edge cloud (step 206). The number of iterations (step 203-step 208) is at most
Figure BDA00024312132500000617
Thus, the runtime of the truth discovery phase is o (nm). The time complexity found by the truth in the deep cloud is the same as the time complexity found by the truth on the edge cloud. Since, the entire phase discovery takes time O (nm).
Next, the temporal complexity of budgeting a viable reverse auction is analyzed. Only the time complexity of the probabilistic branch (steps 306-317) of the random method 3/5 needs to be analyzed because it governs the runtime of the budget viable reverse auction. It takes time O (nm) to find the user with the greatest marginal benefit, and calculate Vi(Sk) The time spent takes the time o (m). Since there are m tasks, each winner should contribute at least one new task, so the number of winners is at most m. Therefore, the cycle of step 307 to step 309 requires time O (nm)2). In each iteration of the loop of steps 311-316, a process similar to steps 307-309 is performed. Thus, payment determinationTakes time O (nm)3). The runtime of the budget feasible reverse auction phase is controlled by the payment determination phase, so the time complexity of the budget feasible reverse auction phase is O (nm)3)。
Furthermore, the method for discovering excitation by the edge-assisted mobile crowd sensing truth value is real.
And (3) proving that: to prove that the excitation method of the present invention is realistic, the method should be proved to conform to the meisen theorem. First, to demonstrate monotonicity, at a probability of 2/5, the user who directly picks the greatest contribution becomes the winner, and no matter what bid he is, he is the winner, at a probability of 3/5, the idea of greedy algorithm is adopted to pick the winner, and the winner submits a lower bid, which will make their bid
Figure BDA0002431213250000076
Larger, ranked further forward, still becomes the winner, so the method satisfies the monotonicity of the melson theorem.
Next, the threshold payout is proved, and the user with the largest contribution is directly selected as the winner under the probability of 2/5, and the whole budget G of the user is directly paidkAs the payment, the winning user can not bring more profits to the winning user even if the winning user changes the price, and under the probability of 3/5, the winner is selected by adopting the idea of greedy algorithm, and the winning user can pay in the payment stage
Figure BDA0002431213250000072
To determine the compensation to be paid to user i when winning bid bi≤piAt that time, a lower bid may still make i the winner, when bi＞piWhen, there are two cases:
(1) when in use
Figure BDA0002431213250000073
The winner will not be reached because the condition of step 307 is violated.
(2) When in use
Figure BDA0002431213250000074
Due to the existence of
Figure BDA0002431213250000075
I' will replace the position of i, so that the profit of i is reduced and even cannot be a winner, and therefore the user cannot increase the profit by changing the price.
So that the threshold payment of mellson's theorem is met.
In conclusion, the excitation method of the present invention is true.
Further, the approximation ratio of the edge-assisted mobile crowd sensing true value discovery excitation method is 1/5.
And (3) proving that: let S*For the optimal solution, due to the submodular and monotonous properties of the cost function V (·), there are:
Figure BDA0002431213250000081
at this time, get
Figure BDA0002431213250000082
Can have
Figure BDA0002431213250000083
Since k is not in the winning set, there are
Figure BDA0002431213250000084
As can be seen from the above formula,
Figure BDA0002431213250000085
so that 2V (S) can be obtainedk)≤2(V(Sk-1)+V({k}))≤2(V(Sk-1)+V({i*})). To prove the approximation ratio, let S be the final result of the method. Method has probability of 3/5 and S is obtained by greedy algorithmkThe probability of 2/5 is directly obtained by the optimal single-user scheme*This results. Therefore, there are:
Figure BDA0002431213250000086
wherein E (V (S)) is V (S), the above results
Figure BDA0002431213250000087
Has the advantages that:
compared with the prior art, the method for discovering and exciting the true value of the edge-assisted mobile crowd sensing has the following advantages that:
1. the method distributes the mobile crowd sensing task with budget to each edge cloud; the truth value estimation can be carried out on the regional crowd sensing data, and the pertinence of the truth value estimation is improved.
2. The invention selects a user subset as a winner for each edge cloud under the condition of improving the accuracy of the true value estimation precision, so as to maximize the quality of the winner under the budget constraint and determine the payment paid to the winner. The calculation cost is low, and the value obtained by the platform is at least not lower than 1/5 of the optimal solution.
3. The truth value estimation is used as an evaluation method of the quality of the crowd sensing data, and the quality of the crowd sensing data is improved.
Drawings
FIG. 1 is a schematic diagram of an edge-assisted mobile crowd sensing truth discovery system;
FIG. 2 is a truth discovery flow diagram;
FIG. 3 budget feasible reverse auction flow diagram.
Detailed Description
The invention relates to an edge-assisted mobile crowd sensing truth finding system and an excitation method thereof, which comprise the following edge-assisted mobile crowd sensing temperature detection system, and FIG. 1 is a schematic diagram of the edge-assisted mobile crowd sensing truth finding system.
In a mobile crowd-sourcing temperature detection system, the system is located on a platform in a deep cloud, the platform comprising a set of r edge clouds, E ═ E1,e2,...,erAnd a set ofA set of n smartphone users interested in performing a task, U ═ 1, 2. Is provided with a UkAs edge clouds ekSet of users in a coverage area, where U ═ UUk,k＝1,2,...,r。
The platform firstly sets m temperature detection tasks T as T1,t2,...,tmAnd allocating the cloud to all edge clouds to perform mobile crowd sensing. Each edge cloud has a budget that is determined by the importance of the sensed data in the edge cloud coverage area. Let G be (G)1,G2,...,Gr) Budget profiles for all edge clouds. Each task tj∈ T are associated with a task type mujAnd (4) associating. Let mu be (mu)1,μ2,...,μm) All task types.
Each edge cloud ek∈ E all record and distribute tasks to the subset of users UkEach user i ∈ U submits a triple B to the edge cloud to which it belongsi＝(Ti,bi,Xi)，XiThe perceived temperature of user i. Let X ═ X1,X2,...,Xn) Perception data submitted for all users, wherein
Figure BDA0002431213250000091
Let TkAnd XkRespectively towards the edge cloud ekSubmitted tasks and sensory data.
Upon receiving the perception data, each edge cloud ek∈ E calculate each i ∈ UkWeight w of the user(s)iAnd by truth discovery, estimates truth values for all perceptual tasks
Figure BDA0002431213250000092
Let wkIs ekAll users U inkThe weight of (c). Let X*Estimated true values for all final edge clouds.
Optionally, the edge cloud may submit estimated truth values to the deep cloud. Deep cloud computing each edge cloud ek∈ E weight gkAnd according to X*By passingMethod for finding true value to estimate final true value X of all perception tasks**. Finally, each edge cloud conducts a budget-viable reverse auction. Each edge cloud ekCalculating a set of winners
Figure BDA0002431213250000093
And each winner i ∈ SkPayment of (p)i. Order to
Figure BDA0002431213250000094
Let p bekAnd p is each SkAnd the payment policy of S.
For any task tj∈TiThe invention will be from the edge cloud ekWinner set S inkThe obtained quality function is defined as:
Figure BDA0002431213250000095
further, the method for discovering and exciting edge-assisted mobile crowd sensing truth value comprises two stages, namely a truth value discovering stage and a budget feasible reverse auction stage, wherein any edge cloud e is in the following statekThe flow of the above truth finding stage is shown in fig. 2, and the steps are as follows:
simply simulating the execution situation on one edge cloud, and assuming a certain edge cloud ekThere are 4 users ui(i 1,2,3,4) to complete two tasks t1,t2。
Step 201: two temperature data collected by 4 users through sensors
Figure BDA0002431213250000101
And submitting the sensing data to the edge cloud e where the user is locatedkThe server of (2);
step 202: real value of random initialization task of edge cloud server
Figure BDA0002431213250000102
And an iteration upper limit
Figure BDA0002431213250000103
Step 203: initializing a boundary cloud iteration counter, wherein K is 0;
step 204: on the edge cloud
Figure BDA0002431213250000104
Is given to
Figure BDA0002431213250000105
Step 205, calculate all users i ∈ UkWeight of (2)
Figure BDA0002431213250000106
Wherein
Figure BDA0002431213250000107
stdjIs task tjStandard deviation of all perceptual data.
From the formula w1＝4.3982，w2＝3.3692，w3＝2.5392，w4＝0.5392。
Step 206: computing all tasks tj∈TkTrue value of
Figure BDA0002431213250000108
Step 207: updating the current iteration times K to K + 1;
step 208: if it is
Figure BDA0002431213250000109
And is
Figure BDA00024312132500001010
Step 204 is executed, otherwise step 209 is executed;
step 209: outputting true values for estimating all tasks
Figure BDA00024312132500001011
And weight w of all usersk；
Wherein
Figure BDA00024312132500001012
w1＝3.3986，w2＝10.7871，w3＝8.5912，w43.1094. And (4) the users on other edge clouds have the same reason, and finally all the edge clouds are summarized by using the same algorithm to obtain a final temperature true value.
Furthermore, the method for discovering and exciting the true value of the edge-assisted mobile crowd sensing further comprises a budget feasible reverse auction stage for simulating a bid price b for each user1＝5,b2＝4,b3＝4,b46, edge cloud ekThe budget obtained is 10, set mu1＝1，μ2＝2。
The process is shown in FIG. 3, and the steps are as follows:
step 301: initializing winner sets
Figure BDA00024312132500001013
Initializing a payment policy pk＝0；
Step 302: initializing a set of users whose bids do not exceed a budget
Figure BDA00024312132500001014
Therefore, it is not only easy to use
Figure BDA00024312132500001015
Step 303: let i*Is composed of
Figure BDA00024312132500001016
The most valuable users in the set:
Figure BDA00024312132500001017
the V values were calculated by w for 4 users as 4.4326, 8.7678, 7.5219, 5.9214, respectively. So the largest user is u2；
Step 304: step 305 is performed with a probability of 2/5, step 306 is performed with a probability of 3/5, and in order to demonstrate the integrity of the algorithm, it is assumed here that step 306 is performed with a probability of 3/5;
step 305: will i*Joining to a winner set SkAnd set user i*Is paid as
Figure BDA0002431213250000111
Step 317 is executed;
step 306: in that
Figure BDA0002431213250000112
Is found in
Figure BDA0002431213250000113
The user i with the largest value, wherein
Figure BDA0002431213250000114
Vi'(Sk)＝V(Sk∪{i'})-V(Sk) Of the current 4 users, u2Is/are as follows
Figure BDA0002431213250000115
Is composed of
Figure BDA0002431213250000116
Is the largest.
Step 307: if it satisfies
Figure BDA0002431213250000117
Step 308 is performed, otherwise step 310 is performed, obviously user u2Compliance with this constraint;
step 308: add user 2 to winner set SkPerforming the following steps;
step 309: in that
Figure BDA0002431213250000118
Is found in
Figure BDA0002431213250000119
The user i with the largest value, wherein
Figure BDA00024312132500001110
Is a set
Figure BDA00024312132500001111
And set SkStep 307 is executed to find the final set Sk＝{2,3}
Step 310: judging the pair set SkWhether each user i has performed steps 311 to 316; if so, go to step 317;
step 311: first calculating the reward of user 2
Figure BDA00024312132500001112
Order to
Figure BDA00024312132500001113
At this time
Figure BDA00024312132500001114
Step 312: finding collections
Figure BDA00024312132500001115
In
Figure BDA00024312132500001116
The user with the largest value i' where
Figure BDA00024312132500001117
This time also user 3.
Step 313: if user i' satisfies
Figure BDA00024312132500001118
Step 314 is performed, otherwise step 310 is performed, when user 3 is in compliance with this constraint.
Step 314: computing
Figure BDA00024312132500001119
Step 315: and user 3 is added to S'kThe method comprises the following steps: s'k＝S'k∪ {3}, execution of step 313 again adds USER-1 to S'kIn (b) to obtain p2＝4.66。
Step 316: the reward p of the user 3 is calculated in a cycle like the above3＝4.19
Step 317: outputting the winner set Sk-2, 3, payment policy pk＝{p2＝4.66，p3＝4.19}。

Claims (5)

1. The utility model provides an edge assistance removes crowd's wisdom perception truth and discovers system, is applicable to a mobile crowd's wisdom perception system which characterized in that: the system is located on a platform in a deep cloud, and the platform comprises a set of r edge clouds, namely a set E ═ E1，e2，...，erAnd a set of n smartphone users, U ═ 1,2kAs edge clouds ekSet of users in a coverage area, where U ═ UUk1, 2.., r, any user belongs to a plurality of edge clouds;
the deep cloud sets m tasks T as T1，t2，...，tmDistributing the data to all edge clouds to perform mobile crowd sensing, determining budget for each edge cloud according to the importance of the sensing data in the coverage area of the edge cloud, and enabling G to be (G)1，G2，...，Gr) For budget profiles of all edge clouds, each task tj∈ T and a task type mujAssociated, the type representing a task tjThe importance of (c); let mu be (mu)1，μ2，...，μm) All task types;
each edge cloud ek∈ E records and distributes tasks to a subset of users UkEach user i ∈ U submits a triple B to the edge cloud to which it belongsi＝(Ti，bi，Xi)，TiIs the set of tasks that user i is willing to perform, biIs a bid of user i, XiIs submitted by user i and task set TiCorresponding perception data, each task set TiAnd cost ciIs associated with ciKnown only to user i, user submits task TiLet X be (X)1，X2，...，Xn) For all the perception data submitted by the user,
Figure FDA0002431213240000011
let TkAnd xkRespectively towards the edge cloud ekSubmitted tasks and sensory data;
upon receiving the perception data, each edge cloud ek∈ E calculate each i ∈ UkWeight w of the user(s)iEstimating truth values for all perceptual tasks through truth discovery
Figure FDA00024312132400000111
Let wkIs ekAll users U inkBy weight of (a), let x*Estimated true values for all final edge clouds.
2. An edge-assisted mobile wisdom perception truth discovery system as claimed in claim 1, wherein: the cloud e at any edgekThe steps of the above truth finding phase are as follows:
step 201: the method comprises the steps that a user collects sensing data and submits the sensing data to an edge cloud server where the user is located;
step 202: real value of random initialization task of edge cloud server
Figure FDA0002431213240000012
And an iteration upper limit
Figure FDA0002431213240000013
Step 203: initializing a boundary cloud iteration counter, wherein K is 0;
step 204: on the edge cloud
Figure FDA0002431213240000014
Is given to
Figure FDA0002431213240000015
Step 205, calculate all users i ∈ UkWeight of (2)
Figure FDA0002431213240000016
Wherein
Figure FDA0002431213240000017
stdjIs task tjStandard deviation of all perceptual data of (a);
step 206: computing all tasks tj∈TkTrue value of
Figure FDA0002431213240000018
Step 207: updating the current iteration times K to K + 1;
step 208: if it is
Figure FDA0002431213240000019
And is
Figure FDA00024312132400000110
Step 204 is executed, otherwise step 209 is executed;
step 209: outputting true values for estimating all tasks
Figure FDA0002431213240000021
And weight w of all usersk。
3. An excitation method of an edge-assisted mobile crowd sensing truth discovery system is characterized by comprising the following steps: budget reverse auctioning the edge cloud, giving a set of tasks TkUser set UkBudget GkTask type μ and bid strategy
Figure FDA0002431213240000022
Each edgeCloud ekCalculating a set of winners
Figure FDA0002431213240000023
And each winner i ∈ SkPayment of (p)iLet us order
Figure FDA0002431213240000024
Let p bekAnd p is each SkAnd the payment policy of S,
the utility of any user i is defined as the difference between the payment and its actual cost:
ui＝pi-ci(1)
for any task tj∈TiFrom edge clouds ekWinner set S inkThe obtained quality function is defined as:
Figure FDA0002431213240000025
wherein the log reflects the diminishing returns of the platform in terms of quality brought by participating users;
maximizing the quality function so that the total payment does not exceed the budget is expressed as:
the target is as follows: maximize V (S)k) (3)
And (3) constraint:
Figure FDA0002431213240000026
4. the excitation method of the edge-assisted mobile crowd sensing truth discovery system according to claim 3, wherein: the budget feasible reverse auction comprises the following steps:
step 301: initializing winner sets
Figure FDA0002431213240000027
Initializing a payment policy pk＝0；
Step 302: initial bid no more thanOver-budgeted set of users
Figure FDA0002431213240000028
Step 303: let i*Is composed of
Figure FDA0002431213240000029
The most valuable users in the set:
Figure FDA00024312132400000210
step 304: performing step 305 with an 2/5 probability and performing step 306 with a 3/5 probability;
step 305: will i*Joining to a winner set SkAnd set user i*Is paid as
Figure FDA00024312132400000211
Step 317 is executed;
step 306: in that
Figure FDA00024312132400000212
Is found in
Figure FDA00024312132400000213
The user i with the largest value, wherein
Figure FDA00024312132400000214
Vi′(Sk)＝V(Sk∪{i′})-V(Sk)；
Step 307: if it satisfies
Figure FDA00024312132400000215
Step 308 is executed, otherwise step 310 is executed;
step 308: adding user i to the set of winners SkPerforming the following steps;
step 309: in that
Figure FDA00024312132400000216
Is found in
Figure FDA00024312132400000217
The user i with the largest value, wherein
Figure FDA00024312132400000218
Figure FDA00024312132400000219
Is a set
Figure FDA00024312132400000220
And set SkStep 307 is performed;
step 310: judging the pair set SkWhether each user i has performed steps 311 to 316; if so, go to step 317;
step 311: order to
Figure FDA0002431213240000031
Order to
Figure FDA0002431213240000032
Step 312: finding collections
Figure FDA0002431213240000033
In
Figure FDA0002431213240000034
The user with the largest value i' where
Figure FDA0002431213240000035
Step 313: if user i' satisfies
Figure FDA0002431213240000036
Step 314 is executed, otherwise step 310 is executed;
step 314: finding collections
Figure FDA0002431213240000037
The user with the largest value i' where
Figure FDA0002431213240000038
Step 315: computing
Figure FDA0002431213240000039
Step 316: and adding i 'to S'kThe method comprises the following steps: s'k＝S′k∪ { i' }, go to step 313;
step 317: outputting the winner set SkPayment policy pk。
5. The excitation method of the edge-assisted mobile crowd sensing truth discovery system according to claim 3, wherein: the edge cloud submits the estimated truth value to the deep cloud, and the deep cloud calculates each edge cloud ek∈ E weight gkTime, edge cloud ekThe formula for weight update of (c) is:
Figure FDA00024312132400000310
wherein
Figure FDA00024312132400000311
α is two constants, α is a reliability coefficient, β is an importance coefficient, α is the influence of the number of people determining the edge cloud on the last truth discovery, the larger α is, the more edge cloud will have larger weight and play larger influence in the truth discovery process of the deep cloud, the larger β is, the more budget edge cloud will have larger weight and play larger influence in the truth discovery process of the deep cloud, wherein α∈ [0,1 ]]，β∈[0，1]α + β is 1 according to x*Discovery through truthMethod for estimating final true value x of all perception tasks**。
CN202010236632.3A 2020-03-30 2020-03-30 Edge-assisted mobile crowd sensing true value discovery system and excitation method thereof Active CN111464620B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010236632.3A CN111464620B (en) 2020-03-30 2020-03-30 Edge-assisted mobile crowd sensing true value discovery system and excitation method thereof

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010236632.3A CN111464620B (en) 2020-03-30 2020-03-30 Edge-assisted mobile crowd sensing true value discovery system and excitation method thereof

Publications (2)

Publication Number Publication Date
CN111464620A true CN111464620A (en) 2020-07-28
CN111464620B CN111464620B (en) 2023-05-23

Family

ID=71679263

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010236632.3A Active CN111464620B (en) 2020-03-30 2020-03-30 Edge-assisted mobile crowd sensing true value discovery system and excitation method thereof

Country Status (1)

Country Link
CN (1) CN111464620B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112787809A (en) * 2020-12-31 2021-05-11 华南理工大学 Efficient crowd sensing data stream privacy protection truth value discovery method
CN112950251A (en) * 2021-05-11 2021-06-11 北京航空航天大学 Reputation-based vehicle crowd sensing node reverse combination auction excitation optimization method

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108876567A (en) * 2018-05-25 2018-11-23 华南理工大学 A kind of intelligent perception motivational techniques based on perception maximization of utility
CN110189174A (en) * 2019-05-29 2019-08-30 南京工业大学 A kind of mobile intelligent perception motivational techniques based on quality of data perception

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108876567A (en) * 2018-05-25 2018-11-23 华南理工大学 A kind of intelligent perception motivational techniques based on perception maximization of utility
CN110189174A (en) * 2019-05-29 2019-08-30 南京工业大学 A kind of mobile intelligent perception motivational techniques based on quality of data perception

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112787809A (en) * 2020-12-31 2021-05-11 华南理工大学 Efficient crowd sensing data stream privacy protection truth value discovery method
CN112787809B (en) * 2020-12-31 2022-08-12 华南理工大学 Efficient crowd sensing data stream privacy protection truth value discovery method
CN112950251A (en) * 2021-05-11 2021-06-11 北京航空航天大学 Reputation-based vehicle crowd sensing node reverse combination auction excitation optimization method
CN112950251B (en) * 2021-05-11 2024-01-26 北京航空航天大学 Reputation-based vehicle crowd sensing node reverse combination auction excitation optimization method

Also Published As

Publication number Publication date
CN111464620B (en) 2023-05-23

Similar Documents

Publication Publication Date Title
CN111754000B (en) Quality-aware edge intelligent federal learning method and system
Liu et al. Gtg-shapley: Efficient and accurate participant contribution evaluation in federated learning
US10715962B2 (en) Systems and methods for predicting lookalike mobile devices
US8782045B1 (en) Evaluating techniques for clustering geographic entities
US9002774B2 (en) Systems and methods for generating a forecasting model and forecasting future values
Feng et al. Towards truthful mechanisms for mobile crowdsourcing with dynamic smartphones
JP7285521B2 (en) System and method for predicting similar mobile devices
CN114902273A (en) System and method for optimizing resource allocation using GPU
CN111464620A (en) Edge-assisted mobile crowd sensing truth value discovery system and excitation method thereof
Alswailim et al. A reputation system to evaluate participants for participatory sensing
CN111461812A (en) Object recommendation method and device, electronic equipment and readable storage medium
CN111932318A (en) Region division method and device, electronic equipment and computer readable storage medium
CN111626563B (en) Dual-target robust mobile crowd sensing system and excitation method thereof
KR20110096488A (en) Collaborative networking with optimized inter-domain information quality assessment
CN117454416A (en) Federal learning method and system for air computing network with privacy protection incentive
Zhang et al. A reputation-based and privacy-preserving incentive scheme for mobile crowd sensing: A deep reinforcement learning approach
Halabi et al. Reliability-driven task assignment in vehicular crowdsourcing: A matching game
US11997508B2 (en) Geospatial-based forecasting for access point deployments
Orpana et al. Building spatial choice models from aggregate data
Tham et al. A spatio-temporal incentive scheme with consumer demand awareness for participatory sensing
Yang et al. Good work deserves good pay: A quality-based surplus sharing method for participatory sensing
CN113011703B (en) Topic-sensitive crowdsourcing task diffusion method
Singh et al. Quad: A quality aware multi-unit double auction framework for iot-based mobile crowdsensing in strategic setting
CN114331540A (en) Method for training model, method and device for determining asset valuation
CN114841435A (en) Navigation path planning method and system

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
CB02 Change of applicant information

Address after: 210003 Gulou District, Jiangsu, Nanjing new model road, No. 66

Applicant after: NANJING University OF POSTS AND TELECOMMUNICATIONS

Address before: Yuen Road Qixia District of Nanjing City, Jiangsu Province, No. 9 210000

Applicant before: NANJING University OF POSTS AND TELECOMMUNICATIONS

CB02 Change of applicant information
GR01 Patent grant
GR01 Patent grant