CN110061862A - A kind of distributed multi-task intelligent perception method based on fairness in dense network - Google Patents
A kind of distributed multi-task intelligent perception method based on fairness in dense network Download PDFInfo
- Publication number
- CN110061862A CN110061862A CN201910226832.8A CN201910226832A CN110061862A CN 110061862 A CN110061862 A CN 110061862A CN 201910226832 A CN201910226832 A CN 201910226832A CN 110061862 A CN110061862 A CN 110061862A
- Authority
- CN
- China
- Prior art keywords
- user
- task
- ror
- remuneration
- 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.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/12—Discovery or management of network topologies
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a 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
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
- H04L67/1004—Server selection for load balancing
-
- 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]
Abstract
The present invention provides a kind of distributed multi-task intelligent perception methods in dense network based on fairness, by introducing rate of returns RoR, to describe the relationship between each user remuneration obtained and cost, and using a kind of distribution algorithm CRA based on consistency that can be applied to the density network with topology limitation, according to the remuneration of each task of corresponding absorption of costs of each user, so that all rate of returns RoR of user reach unanimity, each user can use the neighbor user information of participation task to update its own to the cost of task and the resource allocation obtained from server, so that each user obtains corresponding remuneration according to the size of its contribution, so that it is guaranteed that the justice of system is effectively.The method of the invention can ensure that fair effective and system fast convergence of each user's profit, realize the fairness of distribution in distributed multi-task system.
Description
Technical field
The present invention relates to points in wireless communication and mobile awareness field more particularly to a kind of dense network based on fairness
Cloth multitask intelligent perception method.
Background technique
In recent years, with the continuous development of technology of Internet of things, especially individual mobile terminal, such as smart phone, plate electricity
Brain, vehicle-mounted awareness apparatus etc. are popularized, and are promoted those and are only relied on individual extensive, complicated society's perception times difficult to realize
The development of business.Intelligent perception (Crowd Sensing) will be commonly used using the thought for combining crowdsourcing with mobile awareness
The mobile device at family forms intelligent perception network as basic sension unit, by network communication, to realize perception task point
Hair is collected with perception data, completes extensive, complicated social perception task.Therefore mobile intelligent perception also has wide answer
With prospect, the numerous areas such as environmental monitoring, Vehicle Detection, city management, social networks can be used in.
With traditional data collection compared with cognitive method, the advantage of intelligent perception is that not needing additional hardware sets
Standby, the mobile terminal for only relying on ordinary user can carry out data acquisition, therefore at low cost, easy to maintain with network deployment,
The advantages that system expandability is good.However, also consuming itself resource while user sends acquisition data, as electric energy,
Bandwidth, calculating and storage resource etc. will also undertake the risk of such as exposure self-position privacy information.Therefore, it is necessary to give user
Reasonable remuneration or compensation are provided and sending loss when acquiring data to make up it.How to design reasonable remuneration mechanism is group
Intelligence sensory perceptual system runs the matter of utmost importance to be faced.
Current most incentive mechanism is broadly divided into following three classes: amusement, service and money.
Intelligent perception task is converted perception game by some systems, by designing interesting game, participates in user simultaneously
Contribute its mobile computing or sensing capability.Some specified area is drawn as Barkhuus et al. devises Mobile phone treasure hunt game
The WiFi in domain covers map.Player participates in game by carrying the mobile device with GPS or WiFi.They need in gaming
The virtual gold coin for being dispersed in different zones is picked up, gold coin is then uploaded to server to obtain corresponding game points.Due to
Preferably network connection will have more maximum probability successfully upload the gold coin being collected into, thus encourage player look for having it is stronger
The region of WiFi covering.It is same to solve the problems, such as, Bell et al. devise an entitled Feeding Yoshi based on position
Game, it is desirable that each opening of players removal search and closing Wi-Fi hotspot.Wherein, virtual fruit represents open heat
Point, the referred to as virtual pet of Yoshi then indicate the hot spot closed.In order to get more points, players need as far as possible
It searches more fruit and the location Yoshi is gone to carry out feeding.This process for finding hot spot then can clearly be drawn out
It specifies the WiFi coverage condition in region and is uploaded onto the server.
It is then based on mutually beneficial principle that second class, which services incentive mechanism,.The consumer of service is also mentioning for service simultaneously
Donor.In other words, if user wants to be benefited from the service that system provides, he is also necessary for system and contributes.Traffic prison
Guan being exactly a typical example.When user drives vehicle driving in road, if real time traffic data passed through on network
It is transmitted to cloud server, then the role of user at this time is a supplier.The information that server provides all suppliers carries out
Processing, and a real-time traffic states are generated, so as to preferably be that each user (supplier) services.
Third class, is also presently the most common one kind, then is the incentive mechanism based on money.System is by being supplied to letter
Uploader pecuniary reward is ceased, so that user be stimulated to upload more better informations.Rivest and Shamir is attempted by giving use
The a certain number of money in family come encourage user access website, to obtain the use information of the web page contents.With Online Music
With the prevalence of application on site, remuneration mechanism is also applied to these fields.Company, robot, Amazon Turkey is equally by money
Incentive mechanism is fulfiled applied to task.For the mankind, posting request is being easily accomplished for task, but for computer
Highly difficult, the worker for then undertaking the task will obtain additional remuneration.
For at present, mainly by the method for game theory come the relationship between processing server and user, in particular by
Stackelberg model.The work of the overwhelming majority is all built upon on the basis of centralized policy.Since server must be collected
To the information of all users, thus it is easy to the load problem in terms of generation calculating/communication, and may cause Single Point of Faliure.
Therefore, it is badly in need of a kind of distributed strategy to handle intelligent perception problem.In addition, fair property, that is, distribute to each participant's
Remuneration matches with its cost, it is made to experience even treatment, is another pass for motivating user that cluster perception task is added
Key factor.
Summary of the invention
Based on problem of the existing technology, the present invention provides the distribution in a kind of dense network based on fairness is more
Task intelligent perception method, it is intended to realize the fairness of user's distribution in the case of multitask, and consistent using multiple agent
The principle of property, it is ensured that each user is based only upon the information of neighbor user, is just able to achieve the remuneration target directly proportional to cost.The plan
A kind of distribution scheme based on consistency is slightly proposed, the density network with topology limitation is mainly used in.
In order to reach the purpose of the present invention, the present invention is adopted the following technical scheme that:
The present invention provides a kind of distributed multi-task intelligent perception methods in dense network based on fairness, main
Thought is to propose the concept of fairness, and by introducing rate of returns (Return of Rate), referred to as RoR is each to describe
Relationship between user's remuneration obtained and cost, and propose a kind of excitation that can be worked in distributed multi-task system
Mechanism.The invention proposes a kind of distribution algorithms based on consistency that can be applied to the density network with topology limitation
(Consensus based Reward Allocation), referred to as CRA is each according to the corresponding absorption of costs of each user
The remuneration of task, so that all rate of returns RoR of user reach unanimity.
Described method includes following steps:
S1: mobile crowdsourcing system model is established:
Server distributes task-set Γ to n user, and each user can simultaneously participate in and execute multiple tasks to obtain
Remuneration, and each user i has the cost C generated due to perceptioni, i=1 ..., n;
The task-set Γ has m task, and each task l has total remuneration Rl, l=1 ..., m;
DefinitionIt indicates to complete the obtained rate of returns RoR of task l in k moment user i,
WhereinWithCorresponding remuneration and cost are respectively indicated, whenWhen, i-th of user cannot execute first
Corresponding RoR value is not present in business;
S2: it is interacted between server and user:
Task-set Γ is sent to all user nodes by S21, server, while obtaining participating in use corresponding to each task
Amount Nl;
S22, server is by remuneration Rl/Nl, l=1 ..., m is sent to all user nodes;
S23, each user are communicated by local AC and carry out information exchange;
S24, after completing local AC communication, the data being collected into are uploaded to server by all users;
S3: it is interacted between user:
S31 is iterated using CRA algorithm, and when each iteration, a user obtains the RoR information of its neighbor user;
S32, according to neighbor user and the RoR information update of its own, it participates in task remuneration obtained to each user
Value, so that the RoR value of each task is equal;
The cost of oneself is reassigned to all tasks of their participations by S33, all users, appoints each of its participation
The RoR value of business is equal;
S34 repeats step S31 to S33, so that the RoR value for each task that all users participate in is consistent.
Preferably, the neighbor user of user i is the use that can be obtained the information of user i and share same task l with user j
Family;The same user i has the task neighborhood equal in number participated in it, between any two user i and user j
There is a connection path.
Preferably, the connection relationship between user node meets the following conditions:
1) task given for one, when in-degree and equal out-degree, a user reaches balance in this task, when
When all users in the task realize justice, topology reaches balance;
2) CRA algorithm is executed in different tasks as all users, and when the topology of each task reaches balance, they
ROR value Asymptotic Synchronization.
Compared with the prior art, the invention has the following beneficial effects:
The present invention proposes to use the concept of rate of returns (RoR), to describe between each user remuneration obtained and cost
Relationship, realize the fairness of distribution in distributed multi-task system;Meanwhile incentive mechanism proposed by the present invention can be with
It works in distributed multi-task system, constructs intelligent perception network using the mobile device of user, complete extensive, complexity
Social perception task;On the basis of network topology, the Resource Allocation Formula of proposition is applicable to the intensive of topology limitation
Network.
Detailed description of the invention
Fig. 1 is the system model structure in the present invention.
Fig. 2 is the specific flow chart of CRA algorithm in the present invention;
Fig. 3 is CRA algorithm implementing result analogous diagram in the embodiment of the present invention.
Fig. 4 is in the embodiment of the present invention as user and task number of variations, and the convergence in mean obtained using CRA algorithm is secondary
Number figure.
Specific embodiment
It elaborates with reference to the accompanying drawing to the embodiment of the present invention.
Distributed multi-task intelligent perception method in dense network of the present invention based on fairness, first against distribution
The characteristics of formula multitask system, establishes following crowdsourcing system model, as shown in Figure 1:
Mobile crowdsourcing system is made of server and user's two parts.The function of server be aggregation environmental monitoring data,
The demands such as network coverage data, and collect the data from user mobile phone.Assuming that task-set Γ shares m task, each task
All there is total remuneration Rl, l=1 ... m.The task-set is distributed to n user by server, and each user can simultaneously participate in and execute
Multiple tasks are to obtain remuneration.And each user has the cost C generated due to perceptioni.In order to facilitate analysis, it is assumed that deposit
In adequate remuneration, so that all users are ready to undertake Data Collection task.In order to describe the network topology of user and task
Structure, the neighbours for defining user i are the user that can be obtained the information of user i and share same task l with user j.It should infuse
Meaning, for a user i, it has the task neighborhood equal in number participated in it.Two users i and j it
Between path represent different catenation sequences, for simplifying the analysis, present invention assumes that for any two user i and j, each other it
Between exist a connection path.
In TDMA MAC protocol, the time is divided into several isometric time slots, and the length of each time slot is enough to ensure that information
It can be successfully transmitted.Each user i has given cost Ci, i=1 ..., n may participate in several tasks in task-set.According to
The rule of fairness, contribution is more, and remuneration is bigger.Therefore, it is proposed thatIt indicates to complete to appoint in k moment user i
Be engaged in the obtained rate of returns RoR of l, i.e.,WhereinWithRespectively indicate corresponding remuneration and cost.?
In special circumstancesI.e. i-th of user cannot execute first of task, and corresponding RoR value is not present.
Secondly, interacting between server and user, interactive process is as follows:
Task-set Γ is sent to all user nodes by S21, server, while obtaining participating in corresponding to each task l
Number of users Nl。
S22, server is by remuneration Rl/Nl, l=1 ..., m is sent to all user nodes.
S23, each user carry out information exchange by the exchange communication of part.
S24, after completing local communication, the data of collection are uploaded to server by all users.
Again, it is interacted between user, for the density network with topology limitation, the present invention proposes to use a kind of base
In the distribution scheme CRA algorithm of consistency, specific steps are as shown in Figure 2, comprising the following steps:
S31, in each update, each user can only obtain the RoR information of its neighbor user.
S32, each user update its participation according to neighbor user and the RoR information of its own
Task consideration value obtained, so that the RoR value of each task is equal.
The cost of oneself is reassigned to all tasks that they are participated in by S33, all users,
To which the RoR value for each task for participating in it is equal.
S31-S33 is repeated several times, to reach all consistent targets of user RoR value in S34.
In order to make the fairness of all users be protected, need to guarantee that the connection relationship between user node meets centainly
Condition:
1) task given for one, when its in-degree and equal out-degree, a user is in this task
On be only balance.Only when all users in the task realize justice, topology, which is just referred to as, to be balanced.
2) CRA algorithm is executed in different tasks as all users, and when the topology of each task reaches balance, they
ROR value be Asymptotic Synchronization.
In order to assess proposed algorithm, the present invention verifies the convergence and convergence of the theoretical algorithm by emulation experiment
Time.
Assuming that shared n=10 user and m=4 task.Total remuneration of each task is respectively R1=1000, R2=
500,R3=700, R4The cost of=400,10 users be respectively 12,5,6,15,6,11,4,10,8 and 4 (analyze for convenience,
Unit is omitted in we).In the simulation, shown in the relation table 1 between task and user, wherein symbol * indicates each user
The corresponding task participated in.
The emulation of table 1 setting (*, which is represented, to be participated in)
Task 1 | Task 2 | Task 3 | Task 4 | |
User 1 | * | * | ||
User 2 | * | * | * | |
User 3 | * | * | ||
User 4 | * | * | * | * |
User 5 | * | * | ||
User 6 | * | * | * | |
User 7 | * | * | ||
User 8 | * | * | * | |
User 9 | * | * | * | |
User 10 | * | * |
The assessment of experimental result is as follows:
1, convergence
In the case where having the dense network of topology limitation, the simulation result using CRA algorithm is as shown in Figure 3.By tying
Fruit is synchronized it is found that in the case where having the dense network of topology limitation using the RoR value in the allocation plan, is realized
The fairness of distribution in distributed multi-task system.When application CRA algorithm, convergence rate is more rapid, when convergence
Between it is shorter.This is because in CRA algorithm, it is all nodes every time while updates, and all receives to reaching consistent direction
It holds back, all parameters all update at the same time, it can faster reach convergence, and all RoR values in each emulation gradually restrain unanimously,
So that the fairness in system is all achieved.
2, convergence time
It can be seen that from above-mentioned simulation result when application CRA algorithm, convergence rate is more rapid, and convergence time is shorter.
This is because be all nodes every time while updating in CRA algorithm, and all own to consistent direction convergence is reached
Parameter all updates at the same time, and convergence is very fast.
Meanwhile to probe into the relationship between convergence time and number of tasks and number of users, the present invention discuss task number by 2 to
5, when the number of users of participation is by 2 to 15 variation, using the convergence time situation of change of CRA algorithm.Assuming that each task is total
Remuneration is respectively 1~1000 random value, and the value at cost of each user is respectively 2~12 random value, and each user and institute
Have and there is a connection path between task.For number of users and number of tasks purpose each case, distinguish iteration 1000 times,
Average time needed for convergence must be reached is as shown in Figure 4.As can be seen from the results, quickly using CRA algorithmic statement, average about 5 to 6
Secondary to can reach convergence, and number of users and task number are more, and it is faster that RoR value reaches convergent speed, can satisfy larger
The demand of type task distribution system.
The preferred embodiment of the present invention and principle are described in detail above, to those skilled in the art
Speech, the thought provided according to the present invention will change in specific embodiment, and these changes also should be regarded as the present invention
Protection scope.
Claims (3)
1. a kind of distributed multi-task intelligent perception method in dense network based on fairness, it is characterised in that: including as follows
Step:
S1: mobile crowdsourcing system model is established:
Server distributes task-set Γ to n user, and each user can simultaneously participate in and execute multiple tasks to obtain remuneration,
And each user i has the cost C generated due to perceptioni, i=1 ..., n;
The task-set Γ has m task, and each task l has total remuneration Rl, l=1 ..., m;
DefinitionIt indicates to complete the obtained rate of returns RoR of task l in k moment user i,
WhereinWithCorresponding remuneration and cost are respectively indicated, whenWhen, i-th of user cannot execute first of task, no
There are corresponding RoR values;
S2: it is interacted between server and user:
Task-set Γ is sent to all user nodes by S21, server, while obtaining participating in number of users corresponding to each task
Nl;
S22, server is by remuneration Rl/Nl, l=1 ..., m is sent to all user nodes;
S23, each user are communicated by local AC and carry out information exchange;
S24, after completing local AC communication, the data being collected into are uploaded to server by all users;
S3: it is interacted between user:
S31 is iterated using CRA algorithm, and when each iteration, a user obtains the RoR information of its neighbor user;
S32, according to neighbor user and the RoR information update of its own, it participates in task consideration value obtained to each user, with
Keep the RoR value of each task equal;
The cost of oneself is reassigned to all tasks of their participations by S33, all users, each task for participating in it
RoR value is equal;
S34 repeats step S31 to S33, so that the RoR value for each task that all users participate in is consistent.
2. according to the method described in claim 1, it is characterized by:
The neighbor user of user i is the user that can be obtained the information of user i and share same task l with user j;The same use
Family i has the task neighborhood equal in number participated in it, there is one between any two user i and user j
Connection path.
3. according to the method described in claim 2, it is characterized by:
Connection relationship between user node meets the following conditions:
1) task given for one, when in-degree and equal out-degree, a user reaches balance in this task, when this
When all users in business realize justice, topology reaches balance;
2) CRA algorithm is executed in different tasks as all users, and when the topology of each task reaches balance, they
ROR value Asymptotic Synchronization.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910226832.8A CN110061862B (en) | 2019-03-25 | 2019-03-25 | Fairness-based distributed multi-task crowd sensing method in dense network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910226832.8A CN110061862B (en) | 2019-03-25 | 2019-03-25 | Fairness-based distributed multi-task crowd sensing method in dense network |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110061862A true CN110061862A (en) | 2019-07-26 |
CN110061862B CN110061862B (en) | 2022-02-18 |
Family
ID=67317397
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910226832.8A Active CN110061862B (en) | 2019-03-25 | 2019-03-25 | Fairness-based distributed multi-task crowd sensing method in dense network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110061862B (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103514048A (en) * | 2013-10-15 | 2014-01-15 | 上海交通大学 | Sensing participation system and task distribution method of sensing participation system |
US20140208331A1 (en) * | 2013-01-18 | 2014-07-24 | Nec Laboratories America, Inc. | Methods of processing core selection for applications on manycore processors |
CN105528248A (en) * | 2015-12-04 | 2016-04-27 | 北京邮电大学 | Crowd sensing incentive mechanisms for multitask cooperative application |
CN107103405A (en) * | 2017-03-22 | 2017-08-29 | 暨南大学 | A kind of mass-rent system and its building method based on block chain technology |
CN107688964A (en) * | 2017-08-07 | 2018-02-13 | 南京邮电大学 | A kind of motivational techniques for the diffusion of extensive mass-rent task |
CN107784561A (en) * | 2017-11-10 | 2018-03-09 | 东南大学 | The implementation method of online incentive mechanism in a kind of mobile mass-rent system |
-
2019
- 2019-03-25 CN CN201910226832.8A patent/CN110061862B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140208331A1 (en) * | 2013-01-18 | 2014-07-24 | Nec Laboratories America, Inc. | Methods of processing core selection for applications on manycore processors |
CN103514048A (en) * | 2013-10-15 | 2014-01-15 | 上海交通大学 | Sensing participation system and task distribution method of sensing participation system |
CN105528248A (en) * | 2015-12-04 | 2016-04-27 | 北京邮电大学 | Crowd sensing incentive mechanisms for multitask cooperative application |
CN107103405A (en) * | 2017-03-22 | 2017-08-29 | 暨南大学 | A kind of mass-rent system and its building method based on block chain technology |
CN107688964A (en) * | 2017-08-07 | 2018-02-13 | 南京邮电大学 | A kind of motivational techniques for the diffusion of extensive mass-rent task |
CN107784561A (en) * | 2017-11-10 | 2018-03-09 | 东南大学 | The implementation method of online incentive mechanism in a kind of mobile mass-rent system |
Non-Patent Citations (2)
Title |
---|
JIAN HOU ET AL.: "Hierarchical Consensus Problem via Group Information Exchange", 《IEEE》 * |
方文凤等: "移动群智感知中任务分配的研究", 《计算机应用研究》 * |
Also Published As
Publication number | Publication date |
---|---|
CN110061862B (en) | 2022-02-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Zhang et al. | Energy-latency tradeoff for energy-aware offloading in mobile edge computing networks | |
Wang et al. | Dynamic UAV deployment for differentiated services: A multi-agent imitation learning based approach | |
Farris et al. | MIFaaS: A mobile-IoT-federation-as-a-service model for dynamic cooperation of IoT cloud providers | |
Chen et al. | From technological networks to social networks | |
Sun et al. | Task replication for vehicular edge computing: A combinatorial multi-armed bandit based approach | |
Zhao et al. | Social-aware data dissemination via device-to-device communications: Fusing social and mobile networks with incentive constraints | |
Ma et al. | Economic analysis of crowdsourced wireless community networks | |
Chen et al. | Cluster-group based trusted computing for mobile social networks using implicit social behavioral graph | |
Fu et al. | Performance optimization for blockchain-enabled distributed network function virtualization management and orchestration | |
Kakhbod et al. | An efficient game form for unicast service provisioning | |
Li et al. | Dynamic routing for social information sharing | |
Qian et al. | Spectrum trading in cognitive radio networks: An agent-based model under demand uncertainty | |
Wang et al. | Service placement for collaborative edge applications | |
Dong et al. | Collaborative edge computing for social internet of things: Applications, solutions, and challenges | |
Wang et al. | A secure and efficient wireless charging scheme for electric vehicles in vehicular energy networks | |
Li et al. | Designing truthful spectrum auctions for multi-hop secondary networks | |
Liu et al. | COMP: Online control mechanism for profit maximization in privacy-preserving crowdsensing | |
Seredynski et al. | Analysing the development of cooperation in MANETs using evolutionary game theory | |
Li et al. | Entropy-based reinforcement learning for computation offloading service in software-defined multi-access edge computing | |
Hu et al. | Leveraging blockchain for multi-operator access sharing management in internet of vehicles | |
Bilò et al. | When ignorance helps: Graphical multicast cost sharing games | |
Nguyen et al. | Intelligent blockchain-based edge computing via deep reinforcement learning: solutions and challenges | |
Li et al. | Fedrelay: Federated relay learning for 6g mobile edge intelligence | |
Li et al. | Analysis for behavioral economics in social networks: An altruism-based dynamic cooperation model | |
Zhang et al. | Cellular traffic offloading via link prediction in opportunistic networks |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |