CN112132202B - Edge computing collaboration alliance discovery method based on comprehensive trust evaluation - Google Patents
Edge computing collaboration alliance discovery method based on comprehensive trust evaluation Download PDFInfo
- Publication number
- CN112132202B CN112132202B CN202010985519.5A CN202010985519A CN112132202B CN 112132202 B CN112132202 B CN 112132202B CN 202010985519 A CN202010985519 A CN 202010985519A CN 112132202 B CN112132202 B CN 112132202B
- Authority
- CN
- China
- Prior art keywords
- node
- cooperative
- nodes
- service
- trust
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000011156 evaluation Methods 0.000 title claims abstract description 29
- 238000000034 method Methods 0.000 title claims abstract description 21
- 238000003860 storage Methods 0.000 claims abstract description 15
- 238000004364 calculation method Methods 0.000 claims description 25
- 230000002776 aggregation Effects 0.000 claims description 16
- 238000004422 calculation algorithm Methods 0.000 claims description 16
- 230000008901 benefit Effects 0.000 claims description 12
- 238000004220 aggregation Methods 0.000 claims description 10
- 230000003993 interaction Effects 0.000 claims description 9
- 230000006978 adaptation Effects 0.000 claims description 7
- 238000004891 communication Methods 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 6
- 230000008859 change Effects 0.000 claims description 4
- 230000003044 adaptive effect Effects 0.000 claims description 3
- 230000032683 aging Effects 0.000 claims description 3
- 230000015572 biosynthetic process Effects 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000013139 quantization Methods 0.000 claims description 3
- 230000004931 aggregating effect Effects 0.000 claims description 2
- 230000008569 process Effects 0.000 claims description 2
- 239000000872 buffer Substances 0.000 claims 1
- 238000011010 flushing procedure Methods 0.000 claims 1
- 230000000694 effects Effects 0.000 abstract description 6
- 230000004927 fusion Effects 0.000 abstract description 5
- 230000006399 behavior Effects 0.000 description 16
- 238000010586 diagram Methods 0.000 description 11
- 230000004044 response Effects 0.000 description 7
- 230000007246 mechanism Effects 0.000 description 6
- 238000004088 simulation Methods 0.000 description 6
- 238000011161 development Methods 0.000 description 5
- 230000018109 developmental process Effects 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 239000006185 dispersion Substances 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 238000005304 joining Methods 0.000 description 2
- 238000013508 migration Methods 0.000 description 2
- 230000005012 migration Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000000739 chaotic effect Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002401 inhibitory effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000006116 polymerization reaction Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
- G06F2009/4557—Distribution of virtual machine instances; Migration and load balancing
-
- 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)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Bioinformatics & Computational Biology (AREA)
- Economics (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Evolutionary Computation (AREA)
- Entrepreneurship & Innovation (AREA)
- Strategic Management (AREA)
- Game Theory and Decision Science (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Probability & Statistics with Applications (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses an edge computing cooperative alliance member discovery method based on comprehensive trust evaluation, which carries out self-adaptive fusion on computing power, storage, bandwidth and behavior difference edge nodes by constructing a task driving cooperative virtual service pool, evaluates the reliability of resources by the trust degree of the nodes, constructs a cooperative service alliance member cluster by the availability, reliability, robustness and resource sharing of the nodes, measures the performance of the constructed cooperative alliance member cluster by taking load balancing capability, packet loss rate, delay, task completion rate and the like as indexes, and proves that node cooperative game has Nash balancing steady state by establishing a node cooperative service utility model, realizes load balancing smoothness, inhibits hot zone effect and improves the performance of a selected alliance member and the efficiency of cooperative service.
Description
Technical Field
The invention relates to the technical field of edge calculation, in particular to an edge calculation collaboration alliance discovery method based on comprehensive trust evaluation in edge calculation.
Background
With the rapid development of the Internet and the wide use of mobile intelligent Internet of things equipment, the application and service demands of the Internet of things, ultra-high definition video, real-time video streaming, unmanned, industrial Internet of things, air-sky-sea integrated comprehensive communication and other emerging networks are greatly emerging. These emerging applications all have the characteristics of large data volume, high bandwidth and delay sensitivity. Due to rapid development and mutual fusion of edge computing and artificial intelligence technologies, the trend of network intellectualization, digitization, ubiquitous integration and heterogeneous integration in the future is increasingly obvious. IDC reports indicate that there will be over 500 million terminals and devices accessing the network in 2020 and up to 1000 million in 2030. The development of the network 'three-in-one' makes the number and variety of devices, service and computing modes, protocol clusters and the like carried by the network become more and more complex, and finally, the network can evolve into a stable, open and complex huge chaotic information and data stream type service system. Network traffic will grow exponentially for live webcast, short video, etc. applications. Currently, bandwidth, capacity, delay, resources, computing power and the like of a cloud computing center have become bottlenecks for restricting development of the cloud computing center.
Heterogeneous fusion ubiquity and mass data traffic processing are one of the main challenges faced by the next generation network development, and high bandwidth and low latency are key indicators thereof. The edge computing is one of effective solutions, and its core idea is to push down computing tasks and services from cloud computing and network core to a novel mode of network edge, implement computing and analysis processing on the network edge side generated by data, reduce bandwidth consumption and delay, and implement on-site and nearby services so as to meet real-time service requirements. The industry and academia sequentially put forward new network computing modes which can provide computing resources on the intelligent terminal side and process data nearby according to application requirements. The edge computing promotes the cloud computing to evolve towards the layered architecture, and the real-time and reliable high-quality service is provided for users through the three dimensions of time, space and distance by fusing the resources such as a network edge server, a distributed intelligent terminal and the like.
With the increase of computing power, storage capacity and other resources of the edge intelligent terminal and the change of computing modes, network edge resources begin to evolve towards AI computing, servers, micro-server clusters and micro-data centers. The intelligent mobile phone has various application scenes with different complexity to resource requirements, such as social, shopping, entertainment and other application, has many and idle resources such as calculation power, storage, bandwidth and the like, but has the characteristics of isomerism, sensitive time delay, real-time response and the like for the application fields such as ocean monitoring, unmanned driving, video monitoring, industrial control and the like, and the intelligent terminal cannot meet the functional service requirements of the intelligent terminal. Edge computation can effectively solve this problem.
In summary, the prior art mainly has the following problems: 1) Because the edge computing task is generally uneven, and the limit of the self capacity of the edge node is added, the problems of network congestion, jitter, failure, downtime and the like are generated under the conditions of surge, hot zone and high load due to the overload of the task of the edge node. Through the interaction and collaboration of edges, the advantages of the cloud, edge and end nodes can be fully brought into play, the resource use is balanced, and the comprehensive optimization of resource utilization rate, energy consumption, bandwidth, storage and the like is realized; 2) Due to the differences in computing power, storage, bandwidth and behavior of the edge nodes, the selection of the cooperative edge nodes directly affects the QoS of the edge computation.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide an edge computing cooperative alliance member discovery method based on comprehensive trust evaluation, which adaptively fuses different edge nodes such as calculation power, storage, bandwidth, behaviors and the like in a mode of constructing a task driven cooperative virtual service pool, evaluates the reliability of resources by the trust degree of the nodes, constructs a cooperative service alliance member cluster by the availability, reliability, robustness, resource sharing and the like of the nodes, measures the performance of the constructed cooperative alliance member cluster by taking load balancing capability, packet loss rate, delay, task completion rate and the like as indexes, establishes a node bad behavior punishment mechanism, inhibits the cheating behaviors of nodes such as selfish, rationality, strategies and the like, effectively realizes load balancing smoothness, inhibits hot zone effects, improves the performance of selection alliance members and the efficiency of cooperative service, and realizes efficient and reliable resource cooperative service.
The technical scheme for realizing the aim of the invention is as follows:
an edge computing cooperative alliance discovery method based on comprehensive trust evaluation assumes that n nodes m sides exist in an edge computing system at a certain moment, a formed network topology formalized description is G (V, E), E is an adjacent matrix of a network, and if a node i is connected with a node j, E is formed ij =1, otherwise e ij =0, v= {1,2,3,., n } represents a set of all server nodes, and assuming that the nodes autonomously broadcast their calculation power and storage resource information after joining the network, the master node automatically washes and caches the resource related information in the information life cycle, and stores the information in the local database and continuously updates the information; the method specifically comprises the following steps:
1) Modeling service probability among nodes through node connectivity, analyzing stability of a network topological structure according to average service duration of the nodes, and constructing a cooperative communication probability model among the nodes; evaluating the node direct trust degree by using the characteristics of the interaction times and the interaction success rate among the nodes, recommending trust by using neighbors, filtering cheating nodes by adopting improved mean square error, and fusing the direct trust degree and the indirect trust degree to construct an aging node comprehensive trust degree quantization model;
2) Evaluating the comprehensive performance of the nodes by using node computing power, storage, bandwidth and comprehensive trust, selecting cooperative service nodes meeting requirements according to a cooperative member discovery strategy, and constructing a trusted cooperative service member set;
3) Constructing a node income utility model by using node selfish and rational behavior characteristics and using node trust and available service resources, and verifying that the discovered collaborative service set has Nash equilibrium steady state by evolution repeated game to construct a trusted collaborative service cluster;
4) Designing an improved greedy algorithm to optimize aggregation clustering, constructing a virtual cooperative service pool, cooperatively executing computing tasks, measuring the performance of the constructed cooperative alliance cluster by taking load balancing capability, packet loss rate, delay and task completion rate as indexes, and verifying and finding out the policy efficiency of the cooperative service alliance;
5) The cooperative service is completed, the quality of the cooperative service is evaluated, and the trust information of the pushing cooperative service node is updated in the local network; periodically updating the cooperative service node characteristic database.
The step 1) specifically comprises the following steps:
1-1) the formula of the probability p (i) that an arbitrary node k reaches a node i is:
wherein ζ (i) is the connectivity of node i, N (i) is the set of neighboring nodes of node i, and j is the neighboring nodes of node i;
stability probability p of node i during arbitrary period t i online The formula of (2) is:
wherein T represents an observation period, T k,out 、t k,in Respectively representing the time of node down and on line in the observation period T;
1-2) constructing a connectivity probability P (i) that an edge network arbitrary node k can cooperate with to reach a alliance master node i by formulas (1), (2):
1-3) direct trust T of node j to node i i ：
s.t.k 1 ＜n,f(ε 0 )＝0
Wherein T is i,j (t) Representing the direct trust degree of t-period node j to node i, s ij Representing the times of successful service of node j for node i in t period, f ij Representing the number of times node j fails to serve node i, f (ε) i ) As a penalty function, k 1 Representing the number of nodes with direct trust with the node i;
1-4) according to different recommendation situations and recommender behavior credibility, constructing recommendation credibility r of the node i i ：
Wherein r is i,j Representing recommendation degree of node j to node i, T i,m Representing the direct trust degree of the node i and the recommended node m, T m,j Representing the direct trust degree, k, of a node m and a recommended node j 2 The number of recommenders;
1-5) judging whether the recommended trust level is a malicious node for collaborative cheating or not by calculating the root mean square of the recommended trust level, if the root mean square isIf the node is smaller than the lower limit theta, the node is considered to be a malicious node, the recommendation is received with small probability p, otherwise, the recommendation is received with probability 1-p, and r of a recommendation trust degree formula (5) i Correcting, and recommending trust degree R after correction i The method comprises the following steps:
s.t.abandoning the recommendation
Wherein,representing the root mean square, and the cooperative communication probability of any node k of P (i) reaching the alliance node i;
1-6) the dynamic trust value of node i is expressed as follows:
ω l for a period of observation T l The corresponding weights:
wherein μ is a time decay coefficient, t 0 Is the initial time.
The step 2) specifically comprises the following steps:
2-1) assuming that the attributes are independent of each other, each node feature vector is denoted as X i ＝(c i ,s i ,Trust new (i) The overall performance of the node)The expression is:
wherein c i * 、s i * Trust (i) respectively represents the computing power, shared storage and lambda as weight factors after normalization processing;
2-2) comprehensive evaluation of nodesGreater than or equal to a threshold value psi, the node joins the trusted collaboration service set; when node comprehensive evaluation->Below the threshold value ψ, the trusted collaborative service set is not participated in construction, but is still reserved in the network, and the dynamic stability of the network is maintained.
The step 3) specifically comprises the following steps:
3-1) assuming in any game that i represents the node itself, -i refers to other adjacent game nodes not containing node i, the benefit obtained at the end of one stage is B s The resources consumed at this stage are C s Thus, the node is at each time slot t n Utility function of (2)The expression is:
utility function U of the node during observation period T i The expression is:
wherein mu 1 Discount rate for node benefit at each time period;
3-2) assuming that the nodes start gaming by a cooperative strategy, adopting a 'past and future' strategy in the later gaming stage, and enabling the node i to imitate the behavior of the opponent-i in the previous stage; let stage 1 be considered, all node cooperation remain approximateThe rate is 1, so that cooperation is maintained all the time; stage 2, node i changes its own policy due to its attribute characteristics and stability, i.e. the collaboration probability changes toIts cooperative probability of opponent-i in stage 3 is changed to +.>Node i mimics the cooperative probability of an adversary in phase 2 in phase 3, i.e. +.>After repeated game, the cooperative probability sequence of the final generated node is as follows:
wherein the node is in time slot t n The cooperative probability rho of the self strategy is changed to be related to the attribute and the stability of the node, the higher the comprehensive evaluation is, the larger the cooperative probability after the change is, and the expression is as follows;
3-3) substituting the formula (12) into the formula (11) to obtain the final benefit expression of the node i, wherein the final benefit expression is as follows:
only U i The policy is adopted by the 0 node, so whenAnd the node takes the strategy to reach Nash equilibrium steady state, and builds a trusted collaborative service cluster.
The step 4) specifically comprises the following steps:
4-1) building a priority function g for member node j j The expression is:
wherein,rtt (i, j) is the network delay from the federation master node i to the federation member node j, which is the comprehensive capability of node j;
4-2) constructing a load balancing function f (i) expression of the alliance master node according to the load of the alliance master node and the average connectivity probability, wherein the expression is as follows:
wherein Q is i For the load threshold of the alliance master node, xi receives the concurrent control coefficient of the task request of the node i, delta Q i Dynamically adjusting the increment for load adaptation;
4-3) constructing and evaluating an adaptive function eta (i) expression of the cluster formed by the alliance master node i according to the alliance master node load balance and the alliance master node priority, wherein the expression is as follows:
solving through an improved greedy algorithm, and optimizing the aggregation clustering;
4-4) forming a resource pool X= { S by completing cluster formation 1 ,S 2 ,…,S k |j∈S i J=1, 2, &..n }, to ensure the quality of collaborative service, dynamically aggregating optimal resources according to the performance of a resource pool, constructing an evaluation function eta (X) expression of the performance of the resource pool as follows:
wherein S is i And k is the number of constructed clusters, namely the service cluster formed by the alliance master node i.
In the step 4-3), the improved greedy algorithm is described as follows:
4-3-1) setting the number of expected clusters as k, and taking the first k nodes with the optimal comprehensive performance evaluation as initial aggregation centers;
4-3-2) for each cluster center, calculating the priority and the load increment of adjacent nodes according to a formula (15) and a formula (16), calculating f (i) according to the formula (16), and obtaining greedy under the condition that f (i) is more than 0 to be classified into the same collaborative service cluster; if no more nodes are added, selecting the node with the highest priority in the collaborative service cluster to be a new aggregation center for repeating 4-3-2), otherwise stopping adding;
4-3-3) if the system still has nodes which do not join the collaborative service cluster, selecting according to the greedy principleThe optimal node becomes a new aggregation center, and the steps of 4-3-2 are repeated until all nodes join the collaborative service cluster;
4-3-4) note the current solution x= { S 1 ,S 2 ,...,S k -adapting the value η (X) according to equation (18);
4-3-5) let t=0, t < t max Performing:
randomly selecting one node in the system, exiting all adjacent nodes of the node from the current collaborative service cluster, and repeating 4-3-2) to form a new collaborative service cluster; recalculating the current solution X' = { S 1 ′,S 2 ′,...,S k 'computing a system adaptation value η (X');
if η (X ') > η (X), then X' = { S is recorded 1 ′,S 2 ′,...,S k ' is the optimal solution; otherwise t++;
4-3-6) constructing an edge calculation collaborative service cluster according to the optimal solution, mapping collaborative service resources, and initializing a collaborative service pool.
According to the edge computing co-member discovery method based on comprehensive trust evaluation, by means of constructing a task-driven co-virtual service pool, different edge nodes such as computing power, storage, bandwidth and behavior are adaptively fused, reliability of resources is evaluated according to trust degree of the nodes, a co-member cluster is constructed according to availability, reliability, robustness and resource sharing of the nodes, performance of the constructed co-member cluster is measured by taking load balancing capacity, packet loss rate, delay, task completion rate and the like as indexes, and a node co-service utility model is established, so that node co-game has Nash balancing steady state, load balancing smoothness is achieved, hot zone effect is restrained, and performance of selected members and efficiency of co-service are improved.
Drawings
FIG. 1 is a diagram of a trusted allier discovery policy framework;
FIG. 2 is a different case recommendation trust;
FIG. 3 is a diagram of a collaborative services topology;
FIG. 4 is a graph of average load rate of a base station, a federation co-cluster;
FIG. 5 is a flow diagram of a collaborative services system;
FIG. 6 is a concurrent number diagram of a collaborative services system;
FIG. 7 is a graph comparing average trust levels of base stations and members;
fig. 8 is a graph of task reception rate, cooperative service success rate, packet loss rate, and cooperative service response delay;
fig. 9 is a base station load factor graph;
FIG. 10 is a collaborative service pool traffic diagram;
FIG. 11 is a concurrent number graph of a collaboration service pool;
FIG. 12 is a diagram of co-service member reputation;
fig. 13 is a graph showing the cooperative task reception rate, the cooperative service success rate, the cooperative service packet loss rate, and the cooperative service response delay.
Detailed Description
The present invention will now be further illustrated with reference to the drawings and examples, but is not limited thereto.
An edge computing collaboration alliance discovery method based on comprehensive trust evaluation comprises the following steps:
assuming that there are n nodes m edges in the edge computing system at a time, the formed network topology is formally described as G (V, E). E is the adjacency matrix of the network, if node i is connected with node j, thene ij =1, otherwise e ij =0. V= {1,2,3,..n } represents all server node sets. And assuming that the node autonomously broadcasts resource information such as calculation power, storage and the like after joining the network, the main node automatically washes and caches the related information of the resource in the information life cycle, and stores the information in a local database and continuously updates the information.
1) Modeling service probability among nodes through node connectivity, analyzing stability of a network topological structure according to average service duration of the nodes, and constructing a cooperative communication probability model among the nodes; evaluating the direct trust degree of the nodes by using characteristics such as the interaction times and the interaction success rate among the nodes, recommending trust by using neighbors, filtering cheating nodes by adopting improved mean square error, and fusing the direct trust degree and the indirect trust degree to construct an aging node comprehensive trust degree quantization model;
2) Evaluating the comprehensive performance of the nodes by using node computing power, storage, bandwidth, comprehensive trust and the like, selecting cooperative service nodes meeting requirements according to a cooperative member discovery strategy, and constructing a trusted cooperative service member set;
3) Constructing a node income utility model by using node selfish and rational behavior characteristics and using node trust and available service resources, and verifying that the discovered collaborative service set has Nash equilibrium steady state by evolution repeated game to construct a trusted collaborative service cluster;
4) Designing an improved greedy algorithm to optimize aggregation clustering, constructing a virtual cooperative service pool, cooperatively executing computing tasks, measuring the performance of the constructed cooperative alliance cluster by taking load balancing capability, packet loss rate, delay, task completion rate and the like as indexes, and verifying and finding out the policy efficiency of the cooperative service alliance;
5) The cooperative service is completed, the quality of the cooperative service is evaluated, and the trust information of the pushing cooperative service node is updated in the local network; periodically updating the cooperative service node characteristic database.
The step 1) specifically comprises the following steps:
1-1) the formula of the probability p (i) that an arbitrary node k reaches a node i is:
wherein ζ (i) is the connectivity of node i, N (i) is the set of neighboring nodes of node i, and j is the neighboring nodes of node i;
stability probability p of node i during arbitrary period t i o nline The formula of (2) is:
wherein T represents an observation period, T k ,o ut 、t k,in Respectively representing the time of node down and node up in the observation period T.
1-2) the expression of the connectivity probability P (i) that any node k of the edge network can be constructed to reach the alliance master node i by the formulas (1) and (2) is as follows:
1-3) direct trust T of node j to node i i The expression is:
s.t.k 1 ＜n,f(ε 0 )＝0
wherein T is i,j (t) Representing the direct trust degree of t-period node j to node i, s ij Representing the times of successful service of node j for node i in t period, f ij Representing the number of times node j fails to serve node i, f (ε) i ) As a penalty function, k 1 Representing the number of nodes that have direct trust with node i.
1-4) according to the figures2, constructing recommendation trust r of the node i according to different recommendation situations and recommender behavior credibility shown in 2 i The expression is:
wherein r is i,j Representing recommendation degree of node j to node i, T i,m Representing the direct trust degree of the node i and the recommended node m, T m,j Representing the direct trust degree, k, of a node m and a recommended node j 2 Is the number of recommenders.
1-5) judging whether the recommended trust level is a malicious node for collaborative cheating or not by calculating the root mean square of the recommended trust level, if the root mean square isIf the node is smaller than the lower limit theta, the node is considered to be a malicious node, the recommendation is received with small probability p, otherwise, the recommendation is received with probability 1-p, and r of a recommendation trust degree formula (5) i Correcting, and recommending trust degree R after correction i The method comprises the following steps:
s.t.abandoning the recommendation
Wherein,the root mean square, the probability of connectivity that any node k of P (i) can cooperate to reach the federation master node i, is expressed.
1-6) the dynamic trust value of node i is expressed as follows:
ω l for a period of observation T l The corresponding weight expression is:
wherein μ is a time decay coefficient, t 0 Is the initial time;
the step 2) specifically comprises the following steps:
2-1) assuming that the attributes are independent of each other, each node feature vector is denoted as X i ＝(c i ,s i ,Trust new (i) The overall performance of the node)The expression is:
wherein c i * ，s i * Trust (i) represents the computing power and shared storage after normalization processing, and λ is a weight factor.
2-2) comprehensive evaluation of nodesGreater than or equal to a threshold value psi, the node joins the trusted collaboration service set; when node comprehensive evaluation->Below the threshold value ψ, the trusted collaborative service set is not participated in construction, but is still reserved in the network, and the dynamic stability of the network is maintained.
The step 3) specifically comprises the following steps:
3-1) assuming that in any game i represents the node itself, -i refers to other adjacent game nodes that do not contain node i. The gain obtained at the end of a phase is B s The resources consumed at this stage are C s Thus, the node is at each time slot t n Utility function of (2)The expression is:
utility function U of the node during observation period T i The expression is:
wherein mu 1 The discount rate at each time period for the node benefit.
3-2) assuming that the nodes start gaming with a collaborative strategy, in the later gaming stage, a "past and present" strategy is adopted, and node i mimics the behavior of its opponent-i in the previous stage. Let the 1 st stage, all nodes cooperate to maintain the probability as 1, so that cooperation will be maintained all the time. Stage 2, node i changes its own policy due to its attribute characteristics and stability, i.e. the collaboration probability changes toIts cooperative probability of opponent-i in stage 3 is changed to +.>Node i mimics the cooperative probability of an adversary in phase 2 in phase 3, i.e. +.>After repeated game, the cooperative probability sequence expression of the final generated node is as follows:
wherein the node is in time slot t n The cooperative probability rho of the self strategy is changed to be related to the attribute and the stability of the node, the higher the comprehensive evaluation is, the larger the cooperative probability rho after the change is, and the expression of the cooperative probability rho is as follows:
3-3) substituting the formula (12) into the formula (11) to obtain the final benefit expression of the node i, wherein the final benefit expression is as follows:
only U i The 0 node will take the policy. Therefore, whenAnd the node takes the strategy to reach Nash equilibrium steady state, and builds a trusted collaborative service cluster.
The step 4) specifically comprises the following steps:
4-1) building a priority function g for member node j j The expression is:
wherein,rtt (i, j) is the network delay from the federation master node i to the federation member node j, which is the comprehensive capability of node j.
4-2) constructing a load balancing function f (i) expression of the alliance master node according to the load of the alliance master node and the average connectivity probability, wherein the expression is as follows:
wherein Q is i For the load threshold of the alliance master node, xi receives the concurrent control coefficient of the task request of the node i, delta Q i The increment is dynamically adjusted for load adaptation.
4-3) constructing and evaluating an adaptive function eta (i) expression of the cluster formed by the alliance master node i according to the alliance master node load balance and the alliance master node priority, wherein the expression is as follows:
and solving by an improved greedy algorithm to optimize the aggregation clustering.
4-4) forming a resource pool X= { S by completing cluster formation 1 ,S 2 ,…,S k |j∈S i To ensure the quality of collaborative services, dynamically aggregate optimal resources according to the performance of a resource pool:
wherein S is i And k is the number of constructed clusters, namely the service cluster formed by the alliance master node i.
The specific steps of the step 4-3) improved greedy algorithm are described as follows:
4-3-1) setting the number of expected clusters as k, and taking the first k nodes with the optimal comprehensive performance evaluation as initial aggregation centers;
4-3-2) for each cluster center, calculating the priority and the load increment of adjacent nodes according to a formula (15) and a formula (16), calculating f (i) according to the formula (16), and obtaining greedy under the condition that f (i) is more than 0 to be classified into the same collaborative service cluster; if no more nodes are added, selecting the node with the highest priority in the collaborative service cluster to be a new aggregation center for repeating 4-3-2), otherwise stopping adding;
4-3-3) if the system still has nodes which do not join the collaborative service cluster, selecting according to the greedy principleThe optimal node becomes a new aggregation center, and the steps of 4-3-2 are repeated until all nodes join the collaborative service cluster;
4-3-4) note the current solution x= { S 1 ,S 2 ,...,S k -calculating a system adaptation value η (X) according to equation (18);
4-3-5) let t=0, t < t max Performing:
randomly selecting one node in the system, exiting all adjacent nodes of the node from the current collaborative service cluster, and repeating 4-3-2) to form a new collaborative service cluster; re-establishmentCalculate the current solution X' = { S 1 ′,S 2 ′,...,S k 'computing a system adaptation value η (X');
if η (X ') > η (X), then X' = { S is recorded 1 ′,S 2 ′,...,S k ' is the optimal solution; otherwise t++;
4-3-6) constructing an edge calculation collaborative service cluster according to the optimal solution, mapping collaborative service resources, and initializing a collaborative service pool.
Examples:
in order to verify and analyze the performance of the algorithm model, a coastal sea area edge calculation simulation system is designed and realized, and collaborative services, calculation migration, task collaboration, interaction behavior, reputation evaluation and the like in edge calculation are simulated. And constructing a collaborative service system based on the Route Views public data set, simulating the uploading and downloading simulation experiment of the Web browser service for simulating the offshore area edge calculation, and simulating the Web browsing application scene. The edge server establishes a collaborative service pool based on trust evaluation autonomous fusion, sets a load balancing trigger in a task driving mode, triggers collaborative operation when the load of the edge server reaches a threshold value, and transfers part of tasks to the collaborative service pool for execution. The trusted federation discovery policy is shown in figure 1.
Experiment one: SCMECT model collaborative fusion performance simulation analysis
The data center server cluster is provided with 200 calculation tasks to be migrated, the quantity of each calculation task to be migrated is 256-1024, and the calculation complexity is O (n 2 ) Every time 5 continuous calculation tasks are requested, the experimental test time is 4h, the flow sampling period is 30s, and the sampling period of other indexes is 5s.
1) And (3) taking similarity coupling degree, creditworthiness, service capability and the like as evaluation indexes, and comparing and analyzing SCMECT and k-means and KNN proposed herein. The collaborative service topology structure diagram is shown in fig. 3, and the specific service system performance characteristic table constructed by different algorithms is shown in table 1. The polymerization results showed that:
a) Compared with KNN, k-means algorithm, the SCMECT algorithm reduces the base station cluster alliance dispersion by 34.39% and 5.54%, reduces the alliance main cluster alliance dispersion by 34.93% and 39.31%, improves the system similarity coupling by 64.40% and 51.56%, and improves the average aggregation capability of the system by 34.5% and 15.67%;
b) The nodes with better performance can aggregate more nodes to form a collaborative service cluster, and the martai effect is achieved.
2) And starting an experiment at the 5 th minute, generating a surge service request every 30 minutes, wherein the load rate, the flow diagram, the concurrency number diagram and the average trust degree comparison diagram of the base station and the allied member are respectively shown in fig. 4-7, and the task receiving rate, the collaborative service success rate, the packet loss rate and the collaborative service response delay diagram are respectively shown in fig. 8. The experimental results show that:
a) Compared with KNN, K-means, the SCMECT reduces the load of the base station cluster by 14.1%, improves the load balancing rate of the allied main cooperative service cluster by 58.65%, and improves the balancing capacity of the constructed cooperative service system by 36.38%;
b) Compared with KNN, k-means, the SCMECT improves the service flow of the base station cluster by 25.53 percent and 44.06 percent, improves the service flow of the allied cooperative cluster by 19.80 percent and 28.48 percent, improves the overall service capability of the system by 21.23 percent and 31.26 percent, and improves the concurrency performance by 22.00 percent and 32.61 percent;
c) The trust mechanism of SCMECT stimulates the enthusiasm of cooperative service alliers to participate in cooperation, and effectively selects trusted cooperative service nodes to finish tasks cooperatively;
d) The collaborative service cluster constructed by SCMECT is superior to KNN, k-means in response delay, packet loss rate, collaborative success rate and task receiving rate; the overall service efficiency of the SCMECT system is improved by 36.14 percent and 33.26 percent compared with KNN and k-means;
3) The comprehensive performance of the collaborative service cluster constructed by SCMECT is superior to that of KNN and k-means algorithm, and the trust mechanism constructed by SCMECT effectively selects trusted collaborative service nodes to efficiently finish collaborative service, so that the service quality is improved, the load balance is realized, and the guarantee is provided for large-scale trusted edge computing service.
Experiment II: SCMECT model collaborative service pool performance analysis
The cloud server is provided with 100 calculation tasks to be migrated, the quantity of each calculation task to be migrated is more than or equal to 500 and less than or equal to 1024, surge service request testing is carried out on a cloud server where a target calculation task is located, a user requests 5 continuous stream data files each time, the number of times of link re-searching is not more than 3, and the experimental testing time is 4 hours.
4) And (3) taking flow, load, credibility, task receiving rate, cooperative success rate and the like as evaluation indexes, and comparing and analyzing the performance of the cooperative service pool with random nomadism (stochastic routing, SR) and on-demand cooperative routing (ad hoc on-demand distance vector routing, AODV). The base station load rate, the cooperative service pool flow rate, the cooperative service pool concurrency number and the cooperative service member credibility are respectively shown in fig. 9 to 12, and the cooperative task receiving rate, the cooperative service success rate, the cooperative service packet loss rate and the cooperative service response delay are respectively shown in fig. 13. The experimental results show that.
a) Compared with AODV and SR, the SCMECT improves the load unloading rate of the base station by 30.8%; the SCMECT transfers part of tasks to a cooperative service pool for execution, so that local load balancing is realized, and avalanche effect is effectively avoided;
b) Compared with AODV and SR, the SCMECT improves the flow of the cooperative service pool by 2.4 times and 2.5 times respectively, and improves the concurrency by 1.3 times and 1.5 times; the maximum service capacity of the system is improved by 185% and 200%;
c) The SCMECT selects trusted cooperative service alliers to participate in cooperation through a trust mechanism, so that the cooperative service efficiency can be effectively improved;
d) The collaborative service cluster constructed by SCMECT is superior to AODV and SR in response delay, packet loss rate, collaborative success rate and task receiving rate;
e) The overall service efficiency of the SCMECT system is 64.41% and 40.62% higher than that of AODV and SR.
5) SCMECT selects more effective alliance participation synergy, and improves 185% and 200% of equalization capability of an AODV and SR system and 64.41% and 40.62% of service efficiency; the trust mechanism constructed by SCMECT effectively selects the trusted cooperative service nodes to efficiently complete the cooperative service, improves the service quality, realizes load balancing and provides guarantee for large-scale trusted edge computing service.
6) The SCMECT algorithm can effectively improve the performance of selecting alliers, improve the efficiency of collaborative services, and realize efficient and reliable resource collaboration.
Summarizing:
because of the limited load and capacity of the edge devices, a close collaboration is required when processing large amounts of edge computation data. However, the edge device has the characteristics of different performances, dynamic flexibility, autonomy and the like, and selfish, rationality and other behaviors appear when the collaborative service is executed, so that the collaborative service fails. An edge computing collaboration member discovery method based on comprehensive trust evaluation is provided. The method evaluates the reliability of resources by the trust degree of the nodes, constructs a cooperative service alliance cluster by the availability, reliability, robustness, resource sharing and the like of the nodes, and measures the performance of the constructed cooperative alliance cluster by taking load balancing capability, packet loss rate, delay, task completion rate and the like as indexes. And establishing a node bad behavior punishment mechanism, and inhibiting cheating behaviors of the nodes such as selfish, rational, strategy and the like. By establishing the node cooperative service utility model, the node cooperative game is proved to have Nash equilibrium steady state. In order to verify and analyze the performance of the algorithm model, an offshore area edge calculation simulation system is realized based on the Route Views public data set design, and the system simulates cooperative service, calculation migration, task cooperation, interaction behavior, reputation evaluation and the like in edge calculation in a task-driven mode, optimizes the SCMECT model according to simulation experiment results, and enables the SCMECT model to be closer to practical application. The comparison simulation experiment result shows that SCMECT can effectively realize load balancing smoothness, inhibit hot zone effect and improve the performance of member selection and the efficiency of collaborative service.
TABLE 1 System Performance characterization Table
Claims (4)
1. An edge computing cooperative alliance member discovery method based on comprehensive trust evaluation is characterized in that n nodes m sides are assumed to exist in an edge computing system at a certain moment, formed network topology formalization is described as G (V, E), E is an adjacent matrix of a network, and if a node i is connected with a node j, E is formed ij =1, otherwise e ij =0, v= {1,2,3,..n } represents all clothesThe server node set, presume that node join its calculation power, store the resource information autonomously after the network, the main node buffers the relevant information of the resource through the automatic flushing in the information life cycle, keep in the local database and constantly update; the method specifically comprises the following steps:
1) Modeling service probability among nodes through node connectivity, analyzing stability of a network topological structure according to average service duration of the nodes, and constructing a cooperative communication probability model among the nodes; evaluating the node direct trust degree by using the characteristics of the interaction times and the interaction success rate among the nodes, recommending trust by using neighbors, filtering cheating nodes by adopting improved mean square error, and fusing the direct trust degree and the indirect trust degree to construct an aging node comprehensive trust degree quantization model;
2) Evaluating the comprehensive performance of the nodes by using node computing power, storage, bandwidth and comprehensive trust, selecting cooperative service nodes meeting requirements according to a cooperative member discovery strategy, and constructing a trusted cooperative service member set;
3) Constructing a node income utility model by using node selfish and rational behavior characteristics and using node trust and available service resources, and verifying that the discovered collaborative service set has Nash equilibrium steady state by evolution repeated game to construct a trusted collaborative service cluster;
4) Designing an improved greedy algorithm to optimize aggregation clustering, constructing a virtual cooperative service pool, cooperatively executing computing tasks, measuring the performance of the constructed cooperative alliance cluster by taking load balancing capability, packet loss rate, delay and task completion rate as indexes, and verifying and finding out the policy efficiency of the cooperative service alliance;
step 4), specifically comprising the following steps:
4-1) building a priority function g for member node j j The expression is:
wherein,for the comprehensive energy of node jThe force, rt (i, j), is the network delay from the federation master node i to the federation member node j;
4-2) constructing a load balancing function f (i) expression of the alliance master node according to the load of the alliance master node and the average connectivity probability, wherein the expression is as follows:
wherein Q is i For the load threshold of the alliance master node, xi receives the concurrent control coefficient of the task request of the node i, delta Q i Dynamically adjusting the increment for load adaptation;
4-3) constructing and evaluating an adaptive function eta (i) expression of the cluster formed by the alliance master node i according to the alliance master node load balance and the alliance master node priority, wherein the expression is as follows:
solving through an improved greedy algorithm, and optimizing the aggregation clustering;
the improved greedy algorithm comprises the following specific steps:
4-3-1) setting the number of expected clusters as k, and taking the first k nodes with the optimal comprehensive performance evaluation as initial aggregation centers;
4-3-2) for each cluster center, calculating the priority and the load increment of adjacent nodes according to a formula (15) and a formula (16), calculating f (i) according to the formula (16), and obtaining greedy under the condition that f (i) is more than 0 to be classified into the same collaborative service cluster; if no more nodes are added, selecting the node with the highest priority in the collaborative service cluster to be a new aggregation center for repeating 4-3-2), otherwise stopping adding;
4-3-3) if the system still has nodes which do not join the collaborative service cluster, selecting according to the greedy principleThe optimal node becomes a new aggregation center, and the process is repeated 4-3-2) until all nodes join in the collaborative serviceA cluster;
4-3-4) note the current solution x= { S 1 ,S 2 ,...,S k -adapting the value η (X) according to equation (18);
4-3-5) let t=0, t < t max Performing:
randomly selecting one node in the system, exiting all adjacent nodes of the node from the current collaborative service cluster, and repeating 4-3-2) to form a new collaborative service cluster; recalculating the current solution X' = { S 1 ′,S 2 ′,...,S k 'computing a system adaptation value η (X');
if η (X ') > η (X), then X' = { S is recorded 1 ′,S 2 ′,...,S k ' is the optimal solution; otherwise t++;
4-3-6) constructing an edge calculation collaborative service cluster according to the optimal solution, mapping collaborative service resources, and initializing a collaborative service pool;
4-4) forming a resource pool X= { S by completing cluster formation 1 ,S 2 ,…,S k |j∈S i J=1, 2, &..n }, to ensure the quality of collaborative service, dynamically aggregating optimal resources according to the performance of a resource pool, constructing an evaluation function eta (X) expression of the performance of the resource pool as follows:
wherein S is i A service cluster formed by a allied master node i, wherein k is the number of constructed clusters;
5) The cooperative service is completed, the quality of the cooperative service is evaluated, and the trust information of the pushing cooperative service node is updated in the local network; periodically updating the cooperative service node characteristic database.
2. The method for discovering edge computing co-federation based on comprehensive trust evaluation according to claim 1, wherein the step 1) specifically comprises the following steps:
1-1) the formula of the probability p (i) that an arbitrary node k reaches a node i is:
wherein ζ (i) is the connectivity of node i, N (i) is the set of neighboring nodes of node i, and j is the neighboring nodes of node i;
stability probability p of node i during arbitrary period t i online The formula of (2) is:
wherein T represents an observation period, T k,out 、t k,in Respectively representing the time of node down and on line in the observation period T;
1-2) constructing a connectivity probability P (i) that an edge network arbitrary node k can cooperate with to reach a alliance master node i by formulas (1), (2):
1-3) direct trust T of node j to node i i ：
Wherein T is i,j (t) Representing the direct trust degree of t-period node j to node i, s ij Representing the times of successful service of node j for node i in t period, f ij Representing the number of times node j fails to serve node i, f (ε) i ) As a penalty function, k 1 Representing the number of nodes with direct trust with the node i;
1-4) according to different recommendation situations and recommender behavior credibility, constructing recommendation credibility r of the node i i ：
Wherein r is i,j Representing recommendation degree of node j to node i, T i,m Representing the direct trust degree of the node i and the recommended node m, T m,j Representing the direct trust degree, k, of a node m and a recommended node j 2 The number of recommenders;
1-5) judging whether the recommended trust level is a malicious node for collaborative cheating or not by calculating the root mean square of the recommended trust level, if the root mean square isIf the node is smaller than the lower limit theta, the node is considered to be a malicious node, the recommendation is received with small probability p, otherwise, the recommendation is received with probability 1-p, and r of a recommendation trust degree formula (5) i Correcting, and recommending trust degree R after correction i The method comprises the following steps:
abandoning the recommendation
Wherein,representing the root mean square, and the cooperative communication probability of any node k of P (i) reaching the alliance node i;
1-6) the dynamic trust value of node i is expressed as follows:
ω l for a period of observation T l The corresponding weights:
wherein μ is a time decay coefficient, t 0 Is the initial time.
3. The method for discovering edge computing co-federation based on comprehensive trust evaluation according to claim 1, wherein the step 2) comprises the following steps:
2-1) assuming that the attributes are independent of each other, each node feature vector is denoted as X i ＝(c i ,s i ,Trust new (i) The overall performance of the node)The expression is:
wherein c i * 、s i * Trust (i) respectively represents the computing power, shared storage and lambda as weight factors after normalization processing;
2-2) comprehensive evaluation of nodesGreater than or equal to a threshold value psi, the node joins the trusted collaboration service set; when node comprehensive evaluation->Below the thresholdThe value psi does not participate in the construction of the trusted collaborative service set, but still remains in the network, and the dynamic stability of the network is maintained.
4. The method for discovering edge computing co-federation based on comprehensive trust evaluation according to claim 1, wherein the step 3) comprises the following steps:
3-1) assuming in any game that i represents the node itself, -i refers to other adjacent game nodes not containing node i, the benefit obtained at the end of one stage is B s The resources consumed at this stage are C s Thus, the node is at each time slot t n Utility function of (2)The expression is:
utility function U of the node during observation period T i The expression is:
wherein mu 1 Discount rate for node benefit at each time period;
3-2) assuming that the nodes start gaming with a cooperative strategy, in the later gaming stage, the node i imitates the behavior of the opponent i in the previous stage; setting the 1 st stage, wherein the cooperation of all nodes maintains the probability as 1, so that the cooperation is maintained all the time; stage 2, node i changes its own policy due to its attribute characteristics and stability, i.e. the collaboration probability changes toIts cooperative probability of opponent-i in stage 3 is changed to +.>Node i mimics the cooperative probability of an adversary in phase 2 in phase 3, i.e. +.>After repeated game, the cooperative probability sequence of the final generated node is as follows:
wherein the node is in time slot t n The cooperative probability rho of the self strategy is changed to be related to the attribute and the stability of the node, the higher the comprehensive evaluation is, the larger the cooperative probability after the change is, and the expression is as follows;
3-3) substituting the formula (12) into the formula (11) to obtain the final benefit expression of the node i, wherein the final benefit expression is as follows:
only U i The policy is adopted by the 0 node, so whenAnd the node takes the strategy to reach Nash equilibrium steady state, and builds a trusted collaborative service cluster.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010985519.5A CN112132202B (en) | 2020-09-18 | 2020-09-18 | Edge computing collaboration alliance discovery method based on comprehensive trust evaluation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010985519.5A CN112132202B (en) | 2020-09-18 | 2020-09-18 | Edge computing collaboration alliance discovery method based on comprehensive trust evaluation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112132202A CN112132202A (en) | 2020-12-25 |
| CN112132202B true CN112132202B (en) | 2023-11-17 |
Family
ID=73842938
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010985519.5A Active CN112132202B (en) | 2020-09-18 | 2020-09-18 | Edge computing collaboration alliance discovery method based on comprehensive trust evaluation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112132202B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112732442B (en) * | 2021-01-11 | 2023-08-25 | 重庆大学 | Distributed model for edge computing load balancing and solving method thereof |
| CN113225385B (en) * | 2021-04-16 | 2022-05-06 | 天津大学 | Micro-service request distribution method facing edge computing environment and based on game theory |
| CN113422803B (en) * | 2021-05-19 | 2022-11-25 | 武汉工程大学 | Seamless migration method for intelligent unmanned aerial vehicle inspection task based on end edge cloud cooperation |
| CN114172849B (en) * | 2021-11-26 | 2024-07-12 | 华中科技大学 | Deterministic traffic shaping method based on game theory |
| CN114401134B (en) * | 2022-01-14 | 2023-12-15 | 深圳市兴海物联科技有限公司 | Internet of things distributed trusted management method with end-side cooperation |
| CN114301987B (en) * | 2022-03-07 | 2022-05-20 | 天津市城市规划设计研究总院有限公司 | Dynamic scheduling method and system for virtualized network resources |
| CN114326822B (en) * | 2022-03-09 | 2022-05-10 | 中国人民解放军66136部队 | Unmanned aerial vehicle cluster information sharing method based on evolutionary game |
| CN116932228B (en) * | 2023-09-14 | 2023-12-12 | 湖南希赛网络科技有限公司 | Edge AI task scheduling and resource management system based on volunteer calculation |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110740194A (en) * | 2019-11-18 | 2020-01-31 | 南京航空航天大学 | Micro-service combination method based on cloud edge fusion and application |
| CN110851429A (en) * | 2019-11-21 | 2020-02-28 | 嘉兴学院 | Edge computing credible cooperative service method based on influence self-adaptive aggregation |
| CN111031102A (en) * | 2019-11-25 | 2020-04-17 | 哈尔滨工业大学 | Multi-user, multi-task mobile edge computing system cacheable task migration method |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100969158B1 (en) * | 2008-06-30 | 2010-07-08 | 경희대학교 산학협력단 | Method of trust management in wireless sensor networks |
| US9098344B2 (en) * | 2011-12-27 | 2015-08-04 | Microsoft Technology Licensing, Llc | Cloud-edge topologies |
-
2020
- 2020-09-18 CN CN202010985519.5A patent/CN112132202B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110740194A (en) * | 2019-11-18 | 2020-01-31 | 南京航空航天大学 | Micro-service combination method based on cloud edge fusion and application |
| CN110851429A (en) * | 2019-11-21 | 2020-02-28 | 嘉兴学院 | Edge computing credible cooperative service method based on influence self-adaptive aggregation |
| CN111031102A (en) * | 2019-11-25 | 2020-04-17 | 哈尔滨工业大学 | Multi-user, multi-task mobile edge computing system cacheable task migration method |
Non-Patent Citations (1)
| Title |
|---|
| 边缘计算可信协同服务策略建模;乐光学等;计算机研究与发展(第05期);186-208 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112132202A (en) | 2020-12-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112132202B (en) | Edge computing collaboration alliance discovery method based on comprehensive trust evaluation | |
| Jiang et al. | An energy-efficient networking approach in cloud services for IIoT networks | |
| CN112020103A (en) | Content cache deployment method in mobile edge cloud | |
| Ahvar et al. | FEAR: A fuzzy-based energy-aware routing protocol for wireless sensor networks | |
| CN110167054A (en) | A kind of QoS CR- LDP method towards the optimization of edge calculations node energy | |
| Sutariya et al. | A survey of ant colony based routing algorithms for manet | |
| Peng et al. | Real-time transmission optimization for edge computing in industrial cyber-physical systems | |
| CN112867092A (en) | Intelligent data routing method for mobile edge computing network | |
| Zhang et al. | Fine-grained service offloading in b5g/6g collaborative edge computing based on graph neural networks | |
| Han et al. | A self-learning repeated game framework for optimizing packet forwarding networks | |
| Zhang et al. | Graph theory-based QoE-driven cooperation stimulation for content dissemination in device-to-device communication | |
| Guo et al. | Computation offloading for machine learning in industrial environments | |
| CN113938978B (en) | Heterogeneous wireless sensor path finding method based on reinforcement learning | |
| CN114118444B (en) | Method for reducing equipment idle running time in federal learning by using heuristic algorithm | |
| Fathy et al. | Impact of mobility on performance of P2P content discovery protocols over MANET | |
| Jamali et al. | Adaptive peer to peer resource discovery in grid computing based on reinforcement learning | |
| CN113162638A (en) | SDWSN (software development station network) topology control algorithm based on non-cooperative game | |
| Lv et al. | Mobile Edge Computing Oriented Multi-Agent Cooperative Routing Algorithm: A Drl-Based Approach | |
| Wu et al. | A Routing Algorithm Based on Prioritized Replay Double DQN for Private Internet of Things | |
| CN114666793B (en) | Mobile edge network non-cooperative game forwarding method based on user credibility | |
| Wang et al. | Interlinked Chain Method for Blockchain-Based Collaborative Learning in Vehicular Networks | |
| Konorski et al. | A centralized reputation system for MANETs based on observed path performance | |
| Chaudhry et al. | Bio-inspired energy conserving adaptive power and rate control in MANET | |
| Hilal et al. | Mitigating the effect of mobility on cooperation in wireless ad hoc networks | |
| Adekiigbe et al. | Issues and challenges in clustering techniques for wireless mesh 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 |