KR101032477B1 - Method for configuring Peer to Peer Network using Network distances and Delaunay Triangulation - Google Patents

Abstract

The present invention relates to a method for constructing an end-to-end network using network distance based and dillonis triangulation. The present invention first sets the coordinates of nodes based on the network distance for each node having coordinates in the two-dimensional virtual Euclidean space. When the coordinate setting is completed, the node selects peripheral nodes to which the node is to be connected by performing delaunay triangulation based on the set coordinate. When neighboring nodes are selected, the networks are connected to each other to configure data transmission in real time. If a node is added while the network is configured, a temporary node is selected to allow data transmission until the node is placed in an appropriate position and connected to the network structure, and at the same time, the coordinate calculation for the added node and When the connection with other nodes is completed by performing dillon triangulation, the connection with the temporary node is released. In addition, even if a node is separated while the network is configured, the network is reconfigured by performing dillonitrigonation based on the coordinates of the nodes connected to the detached node. According to the present invention as described above, the data transmission delay between nodes is reduced, and data can be transmitted without duplication or loss even when adding and removing nodes, thereby providing a stable end-to-end network.

 Network distance based, dilling triangulation, end-to-end network

Description

Method for configuring Peer to Peer Network using Network distances and Delaunay Triangulation}

The present invention relates to an end-to-end network structure, and more particularly, to a method of constructing a network by network distance based and Delaunay triangulation.

The end-to-end multimedia streaming service forms a source-to-peer network with a peer-to-peer network, receives the packet data of the source node from the neighboring node, and the neighboring nodes are nodes that exist in their surroundings. This is how you send it. This allows packet data to be delivered to all nodes participating in the network.

An example of such a service is an IPTV service that provides a user with a broadcast program, video, audio, text, etc. on an IP-based network. The IPTV service must satisfy a quality of service (QoS), a quality of experience (QoE), security, bidirectionality, reliability, and the like of a prescribed quality or more. For this reason, the IPTV service can maintain normal communication quality such as packet loss rate, packet delay time, or jitter, so that normal service can be provided.

In addition, the IPTV service may be used in various transmission mechanisms according to the size of the viewer or the service method thereof, and is often designed in an end-to-end network configuration for effective use of such a transmission mechanism.

The end-to-end network can be designed in various ways. For example, a method of transmitting packet data by directly connecting each node participating in a network, forming a network by forming each node in a tree structure, a method of transmitting a packet data, setting virtual coordinates for each node, and A method of constructing a network and transmitting packet data by performing Delroni triangulation based on coordinates has been proposed.

However, the above-described method causes a lot of problems.

First, if the nodes are directly connected, the connections between the nodes may be easy, but since the distances between the nodes are not calculated efficiently, the transmission delay of the packet data will inevitably occur.

In addition, when providing a network structured in a tree structure, if a node is detached or fails during service, the network must be reconfigured. However, it takes a lot of time to reconfigure the actual network, during which the service is interrupted, there is a problem that duplicate or loss of packet data occurs.

In addition, the conventional end-to-end network uses virtual coordinates or temporary coordinates when calculating the coordinates of the nodes, so that the coordinate distances between the nodes participating in the system and the actual distances for transmitting and receiving actual data may be designed differently. In other words, if the coordinates of the source node and the neighboring nodes are not optimized, the dillonitrigonation is applied, and when the nodes are connected to each other, an error occurs in the designed distance and the actual distance, which increases the transmission delay time of the packet data and There is a possibility of loss. After all, this causes a problem that the service such as image quality is degraded when using the IPTV service.

Accordingly, an object of the present invention is to solve the above problems and to optimize the distance between the source node constituting the network and its neighboring nodes.

Another object of the present invention is to ensure that the network configuration is maintained in a stable and efficient state even when a problem occurs in any node constituting the network.

According to a feature of the present invention for achieving the above object, a plurality of nodes setting their coordinates based on the network distance; Selecting a neighboring node to which the plurality of nodes connect to each other by performing delunay triangulation based on the set coordinates; And configuring the network such that the plurality of nodes are connected to the selected neighboring nodes to transmit data in real time between the plurality of nodes and the neighboring nodes; It includes.

If a node is newly added after the network is configured, selecting a node from the plurality of nodes to transmit data to the added node; Connecting the added node and an arbitrary node through a temporary path to transmit data; During the step of selecting the arbitrary node, the coordinates of the added node are calculated based on the network distance, and the neighboring nodes to which the added node is to be connected are selected by performing delodynization based on the network distance. step; And reconfiguring the network by disconnecting the temporary path when data transmission is performed through the peripheral node when the peripheral nodes are selected.

The node coordinate based on the network distance is such that the distance in Euclidean space is close to the actual network distance.

When a node leaves the network in the configured state, each node that is connected to the detached node collects coordinates of a node adjacent to and a non-adjacent node; Each node that has been connected to the detached node may perform a delodyne triangulation based on the collected coordinates to select a node to be connected with itself among the adjacent nodes and non-adjacent nodes and reconfigure the network. It is made to include.

In the present invention, each node participating in the network sets its own coordinates based on the network distance, performs dilloni triangulation based on the set coordinates, and selects other neighboring nodes to be connected to itself, thereby configuring the network. . Thus, the message transmission delay time between nodes is reduced in constructing a network using arbitrary arbitrary coordinates.

If a node is added while the network is configured, the node can continue to receive data through a random configuration path with the temporarily selected node, during which the node can join its network after calculating its coordinates. have. In addition, even if a node leaves, the neighboring nodes are reconfiguring the network by dilling triangulation.

As a result, an incomplete or unstable network design enables an efficient and stable end-to-end network that can be transmitted without duplication or loss of data even when nodes are added or removed.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT A preferred embodiment of a method for constructing an end-to-end network using network distance-based and dillonitrigonation according to the present invention will be described in detail with reference to the accompanying drawings.

FIG. 1 is a flowchart illustrating a method of constructing an end-to-end network using network distance-based and dillonis triangulation in a state where a predetermined number of nodes exist according to an exemplary embodiment of the present invention.

Referring to FIG. 1, first, each node present in the 2D Euclidean space measures the distance between nodes to determine their coordinates when a program for network configuration is executed (S101). The distance measurement is performed by using a coordinate calculation method based on the network distance with respect to several nodes in which one node is located. In an embodiment, the coordinate calculation method does not use a specific coordinate calculation method. Examples of coordinate calculation methods include 'Golbal network positioning (GNP)' and 'Vivaldi'.

As such, when the distance measurement between the nodes is completed, each node calculates and sets its coordinates (S102). When the coordinates of the nodes are set, the node is selected as the neighboring nodes that are actually located close to the node. This is because network distance-based coordinates are used instead of arbitrary coordinates. As such, using coordinates based on network distances, the time between message transmission and reception and the actual node can be configured similarly. This can reduce message latency between nodes.

When the coordinates of the nodes are set, each node selects a node to connect from among neighboring nodes (s103). It searches through its neighbors by performing dillonitrigonation to determine its relationship with the neighbors.

When the network configuration between the nodes is completed by the dilling triangulation (s104), packet data transmission is possible according to the packet transmission scheme of the DT protocol (s105).

That is, when the network is configured by determining the neighboring nodes by setting the coordinates based on the network distance and dilling triangulation as described above, any one node is selected as the neighboring node, and thus the packet to the neighboring node is selected. The transmission delay of data is reduced, and the stability of the network structure is improved.

2 shows an example of the network structure constructed by the method of FIG.

That is, FIG. 2 is a network configuration in which each node receives its own coordinate value by measuring the distance between nodes based on the network distance, and selects a neighboring node by performing a dillon triangulation method based on the coordinate value. . In addition, the configuration of FIG. 2 is for explaining an embodiment of the case of adding or detaching a node described below.

3 is a flowchart illustrating a method of configuring a network when a node is added in a state where an end-to-end network is configured according to an exemplary embodiment of the present invention, and FIG. 4 is a diagram illustrating a network of the process of FIG. Cite the network configuration of 2.

First, it is assumed that the node 'p' is added in a state where the end-to-end network is configured as shown in FIG. 4A (s110).

In the end-to-end network configuration, when the node 'p' is added as described above (s111), a random node is randomly selected (herein referred to as node 'b') (s112). Then, the node 'p' establishes a connection with the node 'b' which is the arbitrary node through the temporary path, and the node 'p' receives the packet data from the node 'b' via the temporary path ( s113). At this time, the temporary path is maintained until the node 'p' is coordinated based on the network distance and is connected to the correct position in the network structure.

On the other hand, the reason for establishing a connection between the added node 'p' and any node (b) is to prevent the loss of data transmitted from other nodes before the node 'p' fully participates in the network. . That is, as described above, the node participating in the end-to-end network sets its own coordinates, and based on this, the node is connected to the neighboring nodes by dillonitrigonization, and the node 'p' also calculates its own coordinates. The process is essential. In this case, however, since the coordinate calculation takes a predetermined time by itself, a situation may occur in which packet data is not transmitted from another node during the coordinate calculation.

In addition, the random node selection process is applied to a random walk (random walk) method. In the random walk scheme, assuming that node 'p' has been added as described above in the end-to-end network configuration, the node 'p' requests that a random node be selected. If the node receiving the request message repeatedly transmits the request message to its neighboring nodes, the node finally receiving the request message becomes an arbitrary node. However, it is possible to select any node by any other way without necessarily selecting any node in the above manner.

On the other hand, while the node 'p' receives data from the node 'd' via the temporary path, at the same time, the end-to-end network starts calculating the coordinates of the node 'p' (s114). The coordinate calculation uses an approximate solution such as a 'linear least square' method with neighboring nodes.

When the coordinates of the node 'p' are calculated, the neighboring nodes, that is, the node 'a', the node 'b', and the node 'c' are selected by performing dillon triangulation based on the coordinates (s115).

As such, when the node 'p' is connected to its neighboring nodes selected by dillonitriangulation and receives packet data in the state in which the temporary path is connected (s116), the connection of the temporary path is released (s116). s117). The network configuration in which the temporary path is released is shown in FIG. 4B. If the node 'p' is not connected to the nodes 'a', the nodes 'b', and the nodes 'c' which are peripheral nodes thereof, the node 'p' continues to receive data through the temporary path.

5 and 6 are flowcharts illustrating a method of reconfiguring a network when a node is detached in a state where an end-to-end network is configured according to a preferred embodiment of the present invention, and FIG. A network diagram showing the process when a node leaves, which refers to the network diagram of FIG. 2.

First, there are two ways in which a node can leave the end-to-end network. The first is when any node of the network constituting the normal departure, the second is when any node of the node constituting the network abnormally. This will be described separately.

Normal exit of node mode

In this case, refer to FIGS. 5 and 7.

Assume that the node leaving in FIG. 7A is 'd'. In the state where the network is configured (s120), the normal leaving mode is when the node 'd' leaves (s121), its neighbor nodes node 'b', node 'e', node 'f', and node 'g' , All the coordinates of the node 'c' are provided to the node 'c' (s122).

Then, the node 'b', node 'e', node 'f', node 'g', and node 'c' may know coordinates of other nodes. For example, node 'b' is directly connected to node 'c' and node 'e' and can receive coordinates directly therefrom, and coordinates of node 'f' and node 'g' are transmitted from node 'd'. To receive. Of course, the coordinates of the node 'c' and the node 'e' may also be transmitted from the node 'd'.

In this way, since the node 'd' leaves the neighboring nodes after providing its own information, the node 'b', node 'e', node 'f', node 'g', and node 'c' Know each other's coordinates. Therefore, the node 'b', node 'e', node 'f', node 'g', and node 'c' do not need to calculate coordinates based on network distance.

Then, the node 'b', node 'e', node 'f', node 'g', and node 'c' perform dillonitrigonation so that each node selects a neighboring node to which it will connect ( s123), network reconfiguration is completed (s124). This state is shown in FIG. 7B. In FIG. 7B, it is assumed that dillonia triangulation is performed so that the nodes 'b', node 'e', node 'f', node 'g', and node 'c' become optimal triangles.

Unhealthy exit of node mode

In this case, see FIGS. 6 and 7.

This indicates that any node has failed or the power is suddenly turned off. Therefore, it is not possible to provide the coordinate information to the neighboring nodes when the node is detached, as in the normal detachment mode of the node.

As in the above example, it is assumed that a node abnormally detached in FIG. 7A is 'd'. When the network is configured (s130), when the node 'd' is separated (s131), its neighboring nodes 'b', node 'e', node 'f', node 'g', and node 'c' Does not provide the coordinates of its own nodes.

Thus, node 'b' knows only the coordinate values of node 'c' and node 'e', node 'e' knows only the coordinate values of node 'b' and node 'f', and node 'f' , 'g' and 'c' also know only the coordinates of two adjacent nodes. In other words, since the node 'b' is disconnected from the node 'g' and the node 'f' due to the departure of the node 'd', the coordinate values of the node 'g' and the node 'f' are not known.

In this case, one of the neighboring nodes transmits its coordinate value in a predetermined direction, and a node located in the direction transmits its coordinate value together with the transmitted coordinate value in the same direction. Through this process, the first node that transmits its coordinates can know the coordinates of all nodes. For example, node 'b' sends its coordinates to node 'e', node 'e' sends its coordinates and its coordinates to node 'f', and node 'f' Sends node 'b', 'c' and its coordinates to node 'g', and node 'g' sends node 'b', 'e', 'f' and its coordinates to node 'c' Node 'c' transmits node 'b', 'e', 'f', 'g' and its coordinates to node 'b'. Then, the node 'b' can know the coordinate values of the node 'e', node 'f', node 'g', and node 'c'.

The node 'b' provided with the coordinate values of all the nodes notifies the node 'e', the node 'f', the node 'g', and the node 'c' that are other nodes (s132). Then, the adjacent nodes 'b', node 'e', node 'f', node 'g', and node 'c' which are adjacent to the separated node 'd' may know their respective coordinate values.

Thereafter, dillonia triangulation is performed based on the coordinate values of each node, and each node newly selects a neighboring node to be connected to itself by the dillion triangulation (S133). Thereafter, each node connects with the selected node to complete network reconfiguration (s134). The state of the network reconfigured embodiment is shown in FIG. 7B.

The rights of the present invention are not limited to the embodiments described above, but are defined by the claims, and those skilled in the art can make various modifications and adaptations within the scope of the claims. It is self-evident.

1 is a flowchart illustrating a method of constructing an end-to-end network using network distance-based and dillonis triangulation in a state where a predetermined number of nodes exist according to an exemplary embodiment of the present invention.

2 is a diagram illustrating the network configuration of FIG. 1.

3 is a flowchart illustrating a method of configuring a network when a node is added in the state where the end-to-end network of FIG. 2 is configured.

4 is a diagram illustrating a network configuration of the process shown in FIG. 3.

5 and 6 are flowcharts illustrating a method of reconfiguring a network when a node is separated while the end-to-end network of FIG. 2 is configured.

FIG. 7 is a diagram illustrating a network configuration when a node leaves in FIGS. 5 and 6.

Claims (4)

A plurality of nodes setting their coordinates based on the network distance; Selecting a neighboring node to which the plurality of nodes connect to each other by performing delunay triangulation based on the set coordinates; And, Configuring a network such that the plurality of nodes are connected to the selected neighboring nodes so that data is transmitted in real time between the plurality of nodes and the neighboring nodes; Including, If a node is newly added after the network is configured, Selecting an arbitrary node from the plurality of nodes for transmitting data to the added node; Connecting the added node and an arbitrary node through a temporary path to transmit data; During the step of selecting the arbitrary node, the coordinates of the added node are calculated based on the network distance, and the neighboring nodes to which the added node is to be connected are selected by performing delodynization based on the network distance. step; And Reconfiguring a network by disconnecting the temporary path when data transmission is performed through the peripheral nodes when the peripheral nodes are selected; End-to-end network configuration method using a network distance-based and dillonis triangulation comprising a. delete The method of claim 1, The node coordinates based on the network distance is such that the distance in the Euclidean space is closer to the actual network distance. The method of claim 1, When a node leaves the network in the configured state, each node that has been connected to the detached node collects coordinates of a node adjacent to and a non-adjacent node; And, Selecting nodes to be connected to the node and reconfiguring a network by performing dillonitrigonation based on the collected coordinates for each node that has been connected to the detached node; End-to-end network configuration method using a network distance-based and dillonis triangulation characterized in that it further comprises.

Images