CN111885113B - Self-adaptive selection and resource allocation method for anchor nodes in social network - Google Patents
Self-adaptive selection and resource allocation method for anchor nodes in social network Download PDFInfo
- Publication number
- CN111885113B CN111885113B CN202010590757.6A CN202010590757A CN111885113B CN 111885113 B CN111885113 B CN 111885113B CN 202010590757 A CN202010590757 A CN 202010590757A CN 111885113 B CN111885113 B CN 111885113B
- Authority
- CN
- China
- Prior art keywords
- network
- node
- nodes
- anchor
- core
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- 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
- H04L67/104—Peer-to-peer [P2P] networks
- H04L67/1074—Peer-to-peer [P2P] networks for supporting data block transmission mechanisms
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/104—Peer-to-peer [P2P] networks
- H04L67/1074—Peer-to-peer [P2P] networks for supporting data block transmission mechanisms
- H04L67/1078—Resource delivery mechanisms
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W72/00—Local resource management
- H04W72/50—Allocation or scheduling criteria for wireless resources
- H04W72/52—Allocation or scheduling criteria for wireless resources based on load
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W72/00—Local resource management
- H04W72/50—Allocation or scheduling criteria for wireless resources
- H04W72/535—Allocation or scheduling criteria for wireless resources based on resource usage policies
Abstract
The invention discloses a self-adaptive selection and resource allocation method of an anchor node in a social network, which is used for an edge base station end for processing data of a mobile social network; the central cloud end is used for performing resource allocation data on the social network data; an uplink for transmitting social network data to a central cloud; a downlink for transmitting resource allocation data to the edge base station; according to the invention, dynamic allocation of multimedia resources and computing resources is realized through end edge cloud cooperation, repeated downloading is reduced, and the efficiency of network flow unloading is effectively improved.
Description
Technical Field
The invention relates to self-adaptive selection of an anchor node and a method for distributing computing resources in a group mobile point-to-point (P2P) social network, which are used for maintaining the stability of a network group, improving the reliability of resource transmission in the network and realizing the efficient unloading of central network traffic. The method is mainly based on a mobile edge computing technology and a data mining technology, and belongs to the field of edge computing and data mining.
Background
With the rapid development of social networks, more and more people tend to share and download multimedia resources through mobile devices. This will result in an explosive increase in central network traffic load as repeated downloads of popular resources will consume a large amount of communication and computing resources. Mobile Edge Computing (MEC) is considered an effective solution to this problem, placing computational and multimedia resources on edge nodes of the network, which we call cellular Base Stations (BS). The mobile device can obtain the required resources through the edge base station without accessing the central cloud. In addition, mobile peer-to-peer (P2P) communication technologies are also commonly used in multimedia sharing, including bluetooth technology, WIFI wireless hotspot technology, device-to-device (D2D) communication technologies, and the like. Based on the above problems and technologies, we combine the mobile P2P technology with MEC, and effectively reduce repeated downloading by expanding the sharing of popular resources in the mobile social network, thereby improving the traffic offload efficiency and improving the computing performance of the mobile network. We call a base station and its under-coverage mobile P2P users together form a mobile P2P network.
While there have been some studies to improve the computational performance of mobile social networks by moving the computational task to the edge by combining MEC and mobile P2P technologies, current work has ignored studies on the stability of social networks. The stability of the network plays a critical role in the reliability of resource sharing and the efficiency of network traffic offload in the network, and if people in a social network run away, the resource sharing is interrupted, so that the resource loss and waste are caused, and the efficiency of traffic offload and the computing performance of the network are reduced. While the stability of a network is closely related to the user's engagement and social relationships.
User engagement is used to simulate the personal behavior of users in a social network, where each user may choose to remain engaged or exit the network, which behavior may be influenced by the behavior of their neighbors. The k-core model is a basic model based on node degree constraint and is widely used for measuring user participation in the network. If we convert a network into an undirected graph, k-core is a maximally connected subgraph of the graph, where the degree of each point is greater than or equal to k. Based on the k-core model, if a user has fewer than k friends, the user leaves the network, and the user's leaving may cause a series of users associated with him to follow the leaving. This phenomenon is called a collapse of the social network, and we call the user following the departure a "follower". To prevent network collapse, Bhawalk et al propose the problem of anchoring the k-core. The core of the problem is to find b users which play a crucial role in the overall participation of the network, which are called anchor nodes, and ensure that they do not leave the network through a reward mechanism, so as to keep the participation of the users in the network to the maximum extent, that is, when the network crashes and stops, the number of users in the network can reach the maximum, thereby effectively maintaining the stability of the network, and the new k-core obtained after the anchor point is fixed is called anchor k-core (anchored k-core). In addition, Bhawalk et al also propose an algorithm named on Layer anchor k-core (OLAK) to efficiently find the best anchor point in the network, i.e., the key node with the most followers leaving the network.
Based on the above theoretical work, considering that the association and participation between offline P2P mobile users are also affected by time and space, we take the Global Positioning System (GPS) spatio-temporal sequence information of users into account and represent the closeness of the connection between users using the GPS similarity of users calculated by a Dynamic Time Warping (DTW) method. We introduce a (k, r) -core model, where r is the GPS similarity threshold between users, in which each node not only satisfies the degree constraint k, but also the spatio-temporal similarity between each pair of nodes satisfies the threshold r. In addition, the number of base stations in reality is large, which means that there are a large number of mobile P2P networks, and the existing methods for anchoring k-core problem are all based on a single social network. Therefore, we propose an adaptive anchor (k, r) -core problem based on the massive mobile P2P network, which dynamically picks different numbers of anchor points for each network under limited resources, i.e. the total number of anchor points does not exceed a given value, to maximize the total number of users participating in all networks. Then, the multimedia resources are distributed to the anchor points, so that the resources can be guaranteed to be shared by other owners through the anchor points, repeated downloading of the resources by a mobile user to a central cloud is reduced, and the problem of flow load is effectively solved.
Disclosure of Invention
In order to solve the problem of anchoring (k, r) -core of a large-scale network, the invention provides a Self-Adaptive algorithm named as Self-Adaptive one Layer Anchored (k, r) -core (SA-OLAK) algorithm to realize the dynamic selection of an anchor point. The invention discloses an end edge cloud cooperative network architecture, and dynamic allocation of resources is realized based on a proposed algorithm. The edge base station end collects and calculates the structure information of the mobile P2P network covered by the edge base station end and sends the structure information to the cloud end, the cloud end executes an SA-OLAK algorithm to formulate a resource allocation strategy, resources are sent to the edge base station, and the base station allocates the resources to the anchor points to achieve the unloading of network flow.
The invention aims to ensure that the user participation in all networks is maximized, namely the final total number of users is maximized, so as to maintain the stability of the network and the reliability of resource transmission by dynamically selecting a corresponding number of key users as anchor nodes for each network based on a large-scale mobile P2P social network under the limit of limited resources and user space-time similarity. And dynamic allocation of multimedia resources and computing resources is realized through end edge cloud cooperation, repeated downloading is reduced, and the efficiency of network flow unloading is effectively improved.
Aiming at the existing anchoring k-core problem, the invention provides an anchoring (k, r) -core problem based on a large-scale network under the consideration of space-time factors, ensures that the maximum number of users in all networks always participate by dynamically selecting a specific number of anchor points for each network, maintains the overall stability of the network, and provides an SA-OLAK algorithm to effectively solve the problem. Meanwhile, dynamic allocation of network resources is realized by combining the proposed end edge cloud cooperation model. The method comprises the following steps: a self-adaptive selection and resource allocation method of an anchor node in a social network is characterized in that,
-an edge base station side for mobile social network data processing;
the central cloud end performs resource allocation by using an algorithm according to the social network data;
-an uplink for transmitting social network data to the central cloud;
-a downlink for transmitting resource allocation data to the edge base station;
wherein: the central cloud performs a resource allocation process through an algorithm according to the social network data;
the cloud receives data information of the base station end, and sets a k value and an r value as a node degree limit and a node similarity threshold, and a B value is used as the total number of anchor points;
according to the information of each social network, finding out a node set L (G) which is possibly selected as an anchor node in the network, and using the node set L (G) as a candidate anchor point set, namely a node in (k-1) -shell and a neighbor node thereof outside k-core in the network:
L(G)＝Sk1(G)∪{NB(Sk1(G),G)\Ck(G)}
wherein S isk-1(G) Represents (k-1) -shell, i.e. the set of nodes in (k-1) -core but not in k-core, NB represents the set of neighboring nodes, Ck(G) (ii) a The candidate anchor point and the k-core jointly form a core node of the network;
according to the candidate anchor point sets L (G) found out, the candidate anchor point sets are converted into a hierarchical structure called Onion Layers so as to improve the efficiency of selecting anchor points;
the method comprises the steps of obtaining an index of the number of anchor points to be selected in a network by calculating two indexes of the network, namely a candidate anchor point occupation ratio and a stable node occupation ratio; representing stable nodes in the candidate anchor points by the number P (G) of the nodes with the degree greater than k in the candidate anchor points:
P(G)＝{v|deg(v,L(G)∪Ck(G))≥k&v∈L(g)}
where deg represents the degree of the node, and the ratio of P (G) total to candidate anchor point total is used to represent the stable node occupation ratio gamma1Namely:
calculating the occupation ratio of the candidate anchors, and expressing the occupation ratio gamma of the candidate anchors by the ratio of the total number of the candidate anchors to the total number of the core nodes2Namely:
calculating the index of the estimated value of the number of the anchors to be selected
As a final indicator of the number of selected anchor points, namely:
wherein beta represents the weight occupied by the two indexes, and is obtained by further normalizing:
according to anchor point quantity index
And dividing the optimal anchor points with corresponding quantity for each network, distributing the multimedia resources with corresponding quantity for the network, and sending the resources to the base station corresponding to the network and distributing the network resources according to the optimal anchor points.
Advantageous effects
The method is based on a mobile P2P network covered by a large-scale base station in reality, and introduces a (k, r) -core model to simulate the participation behavior of users in the network by considering the structural relationship among users and the geographical position space-time information. Aiming at a large-scale P2P network, an anchoring (k, r) -core problem is provided to maintain the overall stability of a network group and improve the reliability of resource transmission, an SA-OLAK algorithm and an end edge cloud cooperative framework are provided to solve the problem, the self-adaptive distribution of network resources is realized, and the load of cloud traffic is effectively reduced. Experimental results show that the overall number of the participants in the network can be greatly increased by dynamically selecting the corresponding number of anchor points for each network under limited resources, namely the stability of the network is improved, and the performance of the method is higher than that of a non-adaptive method, a random selection algorithm of the anchor points and a greedy algorithm based on the degrees of the nodes.
Drawings
FIG. 1 is a schematic diagram of a mobile P2P network;
FIG. 2 is a schematic diagram of k-core and anchored k-core;
FIG. 3 is a schematic diagram of an on Layer;
FIG. 4 is a schematic diagram of an end edge cloud collaboration framework;
FIG. 5 is a schematic flow chart of edge cloud cooperation and SA-OLAK algorithm for adaptively allocating anchor points and resources to a network;
fig. 6 is a graph comparing experimental results with several other algorithms at different anchor budgets.
Best mode for carrying out the invention
In order to make the technical scheme and the purpose of the invention clearer, the invention is further described with reference to the accompanying drawings and specific implementation examples. It should be understood that the detailed description and specific examples, while indicating the invention, are intended for purposes of illustration only and are not intended to limit the scope of the invention.
In order to effectively solve the problem of network flow unloading, starting from participation of users in a mobile P2P network under the coverage of a base station and maintenance of network stability, the problem is effectively solved by considering the connection and structural relationship among the users in a social network and geographical position space-time information, proposing an anchoring (k, r) -core problem and proposing an end edge cloud cooperative frame and an SA-OLAK algorithm, and the overall stability of a large-scale network can be effectively maintained under limited resources through verification, so that the total number of the users participating in the network is maximized.
Fig. 4 is a schematic flow chart of a method for adaptively allocating a corresponding number of anchor nodes to a large-scale network based on an edge cloud cooperative framework according to the present invention, which includes the following steps:
(1) each base station acquires data records transmitted between users in the P2P network under its coverage, and other required user dimension information, including GPS records, GPS record time information, and the like. The method and the device can be used for preprocessing the data and extracting the required effective data, and the extracted effective data information comprises a sender ID, a receiver ID, a timestamp and a user GPS. Processing all GPS information of each user into a GPS time sequence with the style of T according to the time stamp sequencei＝(Ti(1),Ti(2),....Ti(n)), which is called a GPS track, where i is the user number and T is a GPS record.
(2) Each base station constructs a social network graph according to the collected transmission relations among the users, represents the users as nodes in the network graph, represents the transmission relations among the users as undirected edges in the network graph, and assigns the space-time similarity values among the users as weight attributes of the edges. The mobile social network graph is an undirected weighted graph, which represents the relationship between users in the social network and users, and can be represented as G (V, E, A), where V is a set of nodes representing users, E is a set of edges representing transmission relationships between users, and A is a set of weights of all edges representing the degree of spatio-temporal similarity between users. For example, Vi and Vj represent two users, Eij represents that one or more transfer relationships exist between users i and j, and the weight Aij on the edge is assigned as a spatiotemporal similarity between the two users.
(3) And (3) calculating the GPS similarity between points on the basis of the GPS track of each user calculated in the step (1) by utilizing a DTW algorithm on the base station to express the space-time similarity between the users. Specifically, for two points Vi and Vj, the GPS locus is TiAnd TjFirstly, an m x n matrix (m and n are the lengths of two tracks) is constructed, the elements in the matrix are the distance between two GPS records in a GPS sequence, and then a minimum regular path is found based on the matrix, so that the distance between the two GPS tracks is minimum. I.e. a path which is monotonously continuous and does not exceed the boundary, and is represented by W, wherein the k-th element of W is defined as Wk＝(i,j)kDefining a mapping between two sequences, W ═ W1,w2,...,wk,...wKWhere K represents the path length. We then get the minimum regular path length and normalize it as the GPS spatio-temporal similarity between two users, i.e.:
(4) each base station transmits the collected and calculated P2P network structure data information to the cloud end through an uplink, a k value is set as a node degree limit at the cloud end according to the collected data, an r value is set as a GPS similarity threshold value between nodes, and a B value is set as a budget of anchors, namely the total number of the anchors allowed to be selected in all networks.
(5) And executing the SA-OLAK algorithm at the cloud end. First, according to the information of the P2P network, k-core, (k-1) -core and (k-1) -shell of each network are calculated. And deleting the nodes with the degree smaller than k through an iterative algorithm in each iteration until the number of the nodes in the network is not reduced any more, so that the k-core of the network is obtained. The same is true for the (k-1) -core, and the (k-1) -shell represents the set of nodes in the (k-1) -core but not in the k-core, which is obtained by deleting the corresponding nodes in the (k-1) -core.
(6) Finding a candidate anchor point set L (G) in the network, namely a node set in (k-1) -shell in the network and neighbor nodes of the nodes outside k-core, and dividing the nodes into an on Layers hierarchy, as shown in FIG. 3. The on Layer can be obtained through an on Peeling algorithm, that is, through an iteration method, each iteration finds out a node which does not satisfy the degree limit k in the candidate anchor point and an edge of which the similarity of the GPS is less than a threshold r, and the node is put on a corresponding level and deleted in the original set until all points in the original set are points in k-core.
(7) And quantifying an estimation index of the number of anchor points selected by each network. Two indexes in the network, namely a candidate anchor occupation ratio and a stable node occupation ratio, which have influence on the possible number of anchors are determined. The ratio of the number P (G) of nodes with degrees greater than k in the candidate anchor points to the total number of the candidate anchor points is used for representing the stable node occupation ratio gamma1The ratio of the total number of candidate anchors to the total number of core nodes is used to represent the anchor occupation ratio gamma2Combining the two indexes as the final estimation value
As a final indicator of the number of anchor points selected. The three index parameters are calculated by the formulas (1) to (4).
(8) Anchor point number estimated value calculated according to each network
Dividing the precalculated value B according to the proportion, determining the quantity of anchor points which are finally selected by each network, and selecting the optimal anchor points with corresponding quantity layer by layer in the candidate anchor point set on Layers by using a greedy algorithm, namely nodes which have the most followers when leaving the network,so as to maximize the total number of nodes participating in the network all the time by anchoring a certain number of nodes, and further maximize the total number of nodes in the network group. Completion of SA-OLAK algorithm
(9) The cloud distributes multimedia network resources corresponding to the number of the selected anchor points to each network according to the distribution strategy obtained through the SA-OLAK algorithm, the resources are sent to the base station corresponding to the network through a downlink, the base station distributes the resources to the anchor points covering the P2P network, namely, the network flow is unloaded to the edge base station, and then the resources are reliably shared to the whole network through the anchor points, so that the repeated resource acquisition of a mobile user to the cloud is effectively reduced.
(10) Finally, in order to verify the effectiveness of the SA-OLAK algorithm, the algorithm is compared with a non-adaptive method and other three adaptive algorithms based on a transmission record data set of real offline mobile P2P software called as flash transfer (Xender), the number of followers (followers), namely the total number of users reserved through anchor points, is used for representing the degree of superiority and inferiority of the algorithm, and the algorithm performance is compared by selecting different total numbers of anchor points. Wherein the non-adaptive method, i.e. the original OLAK algorithm, selects the same number of anchor points for each network instead of dynamic selection. Adaptive algorithm we use the Random algorithm (Random), the Degree-based algorithm (Degree) and the optimal algorithm (optimal). The random algorithm dynamically and randomly selects different numbers of nodes as anchor points in each network according to the internal structure, the degree-based algorithm dynamically selects the node with the highest degree as an anchor point in each network, the optimal algorithm sequentially traverses each point assumed as an anchor point, and then selects the node with the largest corresponding anchor (k, r) -core as the most true anchor point. We compare the SA-OLAK algorithm (self-adaptive in the figure) with the non-adaptive OLAK (figure 6-1) in the case of k being 3 and r being 0.45, and find that our algorithm can effectively retain more users. We compare SA-OLAK with three adaptive methods (fig. 6-2), setting k to 3 (upper panel in fig. 6-2) and 4 (lower panel in fig. 6-2) and already r to 0.45, and found that our algorithm still works well, although the optimal algorithm gets more wells, it consumes much time. Secondly, we look at the number change of the followers by changing the threshold r of the GPS similarity (fig. 6-3), and the larger the threshold is, the more the limitation on the similarity is, the less the corresponding result is obtained. Finally we compare the running times of several algorithms, and it can be seen that the SA-OLAK algorithm consumes little time. The SA-OLAK method is an optimal method for solving the problem of self-adaptive anchoring (k, r) -core and improving the network stability and the file propagation reliability by integrating the experimental results and considering the performance and efficiency of the algorithm.
Claims (2)
1. A self-adaptive selection and resource allocation method of an anchor node in a social network is characterized in that,
-an edge base station side for mobile social network data processing;
the central cloud end performs resource allocation by using an algorithm according to the social network data;
-an uplink for transmitting social network data to the central cloud;
-a downlink for transmitting resource allocation data to the edge base station;
wherein: the central cloud performs a resource allocation process through an algorithm according to the social network data;
the cloud receives data information of the base station side, and sets a k value and an r value as a node degree limit and a node similarity threshold, and a B value as the total number of anchor points;
according to the information of each social network, finding out a node set L (G) which is possibly selected as an anchor node in the network, and using the node set L (G) as a candidate anchor point set, namely a node in (k-1) -shell and a neighbor node thereof outside k-core in the network:
L(G)＝Sk-1(G)∪{NB(Sk-1(G)，G)\Ck(G)}
wherein, Ck(G) Represents k-core, Sk-1(G) Represents (k-1) -shell, i.e., the set of nodes in (k-1) -core but not in k-core, and NB represents the set of neighboring nodes; the candidate anchor point and the nodes in the k-core jointly form a core node set of the network;
according to the candidate anchor point sets L (G) found out, the candidate anchor point sets are converted into a hierarchical structure called Onion Layers so as to improve the efficiency of selecting anchor points;
the method comprises the steps of obtaining an index of the number of anchor points to be selected in a network by calculating two indexes of the network, namely a candidate anchor point occupation ratio and a stable node occupation ratio; representing stable nodes in the candidate anchor points by the number P (G) of the nodes with the degree greater than k in the candidate anchor points:
P(G)＝{v|deg(v，L(G)∪Ck(G))≥k&v∈L(G)}
wherein v represents a certain node in the network, deg represents degree of the node v, and the ratio of P (G) sum to candidate anchor point sum represents stable node occupation ratio y1Namely:
calculating a candidate anchor occupation ratio, and expressing the anchor occupation ratio y by the ratio of the total number of the candidate anchors to the total number of the core nodes2Namely:
calculating the index of the estimated value of the number of the anchors to be selected
As a final indicator of the number of selected anchor points, namely:
wherein beta represents the weight occupied by the two indexes, and is obtained by further normalizing:
according to anchor point quantity index
And dividing the optimal anchor points with corresponding quantity for each network, distributing the multimedia resources with corresponding quantity for the network, and sending the resources to the base station corresponding to the network and distributing the network resources according to the optimal anchor points.
2. The method of claim 1, wherein the anchor node is selected and allocated adaptively according to the candidate anchor set l (g) and converted into an online Layers hierarchy:
screening out some nodes through successive iteration by an iteration method, and putting the nodes into corresponding levels, wherein the level number is increased by 1 once in each iteration;
and screening out nodes which do not meet the degree limit k and edges of which the GPS similarity is less than a threshold r from the candidate anchor point set L (G) in each iteration, putting the screened nodes which do not meet the degree limit k and isolated nodes after the edges which do not meet the threshold r are deleted into corresponding levels, and deleting the nodes in the set L (G) until all points in the original set are points in k-core.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010590757.6A CN111885113B (en) | 2020-06-24 | 2020-06-24 | Self-adaptive selection and resource allocation method for anchor nodes in social network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010590757.6A CN111885113B (en) | 2020-06-24 | 2020-06-24 | Self-adaptive selection and resource allocation method for anchor nodes in social network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111885113A CN111885113A (en) | 2020-11-03 |
| CN111885113B true CN111885113B (en) | 2021-09-28 |
Family
ID=73156985
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010590757.6A Active CN111885113B (en) | 2020-06-24 | 2020-06-24 | Self-adaptive selection and resource allocation method for anchor nodes in social network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111885113B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021045661A1 (en) * | 2019-09-04 | 2021-03-11 | Telefonaktiebolaget Lm Ericsson (Publ) | Edge cloud anchoring |
| CN114726742A (en) * | 2020-12-22 | 2022-07-08 | 华东师范大学 | K-Core maximization method based on anchoring edge and application thereof |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10588044B2 (en) * | 2010-10-05 | 2020-03-10 | Cisco Technology, Inc. | System and method for offloading data in a communication system |
| US10616294B2 (en) * | 2015-05-14 | 2020-04-07 | Web Spark Ltd. | System and method for streaming content from multiple servers |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9292884B2 (en) * | 2013-07-10 | 2016-03-22 | Facebook, Inc. | Network-aware product rollout in online social networks |
| CN108009710A (en) * | 2017-11-19 | 2018-05-08 | 国家计算机网络与信息安全管理中心 | Node test importance appraisal procedure based on similarity and TrustRank algorithms |
| CN109871619B (en) * | 2019-02-22 | 2022-12-27 | 中南大学 | Static charging pile deployment method based on grid division |
| CN110312277B (en) * | 2019-04-08 | 2022-01-28 | 天津大学 | Mobile network edge cooperative cache model construction method based on machine learning |
| CN110717903A (en) * | 2019-09-30 | 2020-01-21 | 天津大学 | Method for detecting crop diseases by using computer vision technology |
| CN110855649A (en) * | 2019-11-05 | 2020-02-28 | 西安交通大学 | Method and device for detecting abnormal process in server |
-
2020
- 2020-06-24 CN CN202010590757.6A patent/CN111885113B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10588044B2 (en) * | 2010-10-05 | 2020-03-10 | Cisco Technology, Inc. | System and method for offloading data in a communication system |
| US10616294B2 (en) * | 2015-05-14 | 2020-04-07 | Web Spark Ltd. | System and method for streaming content from multiple servers |
Non-Patent Citations (1)
| Title |
|---|
| 基于网络结构的社交网络稳定性研究;刘陈亮;《中国优秀硕士学位论文全文数据库信息科技辑》;20150331;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111885113A (en) | 2020-11-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Liu et al. | Deep reinforcement learning for offloading and resource allocation in vehicle edge computing and networks | |
| CN109947545B (en) | Task unloading and migration decision method based on user mobility | |
| CN110247793B (en) | Application program deployment method in mobile edge cloud | |
| Barbera et al. | VIP delegation: Enabling VIPs to offload data in wireless social mobile networks | |
| CN102918887B (en) | For the method and apparatus that dynamic channel and transfer rate select | |
| CN110502328B (en) | Offshore edge computing trusted cooperative task migration method | |
| CN111885113B (en) | Self-adaptive selection and resource allocation method for anchor nodes in social network | |
| Bellavista et al. | A social-driven edge computing architecture for mobile crowd sensing management | |
| CN112020103A (en) | Content cache deployment method in mobile edge cloud | |
| Liu et al. | Wireless distributed learning: A new hybrid split and federated learning approach | |
| Hu et al. | Caching as a service: Small-cell caching mechanism design for service providers | |
| Li et al. | Distributed task offloading strategy to low load base stations in mobile edge computing environment | |
| CN114390057B (en) | Multi-interface self-adaptive data unloading method based on reinforcement learning under MEC environment | |
| CN112437156B (en) | Distributed cooperative caching method based on MEC-D2D | |
| CN115802389B (en) | Federal learning method trained by digital twin auxiliary model | |
| Huang et al. | Multi-agent reinforcement learning for cost-aware collaborative task execution in energy-harvesting D2D networks | |
| CN111405646B (en) | Base station dormancy method based on Sarsa learning in heterogeneous cellular network | |
| CN113641504A (en) | Information interaction method for improving multi-agent reinforcement learning edge calculation effect | |
| Jiang et al. | Dynamic user pairing and power allocation for NOMA with deep reinforcement learning | |
| Li et al. | Learning-based hierarchical edge caching for cloud-aided heterogeneous networks | |
| Tan et al. | Femtocaching in wireless video networks: Distributed framework based on exact potential game | |
| Yuan et al. | Caching hit ratio maximization in mobile edge computing with node cooperation | |
| Seyfollahi et al. | Enhancing mobile crowdsensing in Fog-based Internet of Things utilizing Harris hawks optimization | |
| Luo et al. | Joint game theory and greedy optimization scheme of computation offloading for UAV-aided network | |
| CN111885551B (en) | Selection and allocation mechanism of high-influence users in multi-mobile social network based on edge cloud collaborative mode |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20220424 Address after: 200120 rooms 902A and 902B, block B, No. 112, liangxiu Road, pilot Free Trade Zone, Pudong New Area, Shanghai Patentee after: PPLABS NETWORK TECHNOLOGY (SHANGHAI) Co.,Ltd. Address before: 300072 Tianjin City, Nankai District Wei Jin Road No. 92 Patentee before: Tianjin University |
|
| CP01 | Change in the name or title of a patent holder | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 200120 rooms 902A and 902B, block B, No. 112, liangxiu Road, pilot Free Trade Zone, Pudong New Area, Shanghai Patentee after: Pioneer Cloud Computing (Shanghai) Co.,Ltd. Address before: 200120 rooms 902A and 902B, block B, No. 112, liangxiu Road, pilot Free Trade Zone, Pudong New Area, Shanghai Patentee before: PPLABS NETWORK TECHNOLOGY (SHANGHAI) Co.,Ltd. |