KR101588549B1 - Tcp based network routers supporting network coding and tcp based data streaming system using network coding - Google Patents

Abstract

According to embodiments of the present invention, a data streaming system performing network coding comprises: at least one transmission node to generate source data; an encoding router; and a decoding router. The encoding router drives an encoding application which performs network encoding in an application class to create a TCP session between the at least one transmission node and the encoding router, receives the source data from the transmission node through the TCP session, creates a TCP session between the decoding router and the encoding router, generates network-coded data according to a network coding coefficient matrix and the source data, and transmits the network-coded data and elements of the network coding coefficient matrix to the decoding router through the TCP session. The decoding router drives a decoding application which performs network encoding in an application class to reconstruct the source data by decoding or approximate decoding based on the network-coded data and the elements of the network coding coefficient matrix received from the encoding router through the TCP session.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a TCP-based network routing node capable of implementing network coding, and a TCP-based data streaming system implementing network coding.

The present invention relates to network coding, and more particularly, to network coding implemented on a routing based network.

Conventional TCP (Transmission Control Protocol) -based network schemes are designed to allow a receiving node to simply store data received from a transmitting node in a point-to-point manner, If the network configuration changes in the middle of the network, the existing routing path must be disconnected or a new routing path must be configured and retransmission is required. As a result, the speed is degraded or service quality is affected, There is little overhead to maintain.

On the other hand, network coding repeats the operation of neighboring nodes in the network receiving and combining packets and outputting them as one combined packet so that data can flow to a destination without a predetermined routing path, Network coding can use bandwidth, power resources, and computation resources more efficiently, and is robust against network fluctuations. As the source and path vary, that is, the number of nodes having the same information and the number of nodes requesting, the performance of such network coding can be maximized.

As a result, network coding has become increasingly used as a methodology for constructing an efficient distributed transmission algorithm in a dynamic network with a variety of paths and sources. It is known that the approximate decoding technique, which is one of the advantages of network coding, is particularly useful in case of source data having a high correlation between data, such as multimedia content or measurement data contents occurring in a sensor network.

However, the information transmitted through the network coding should not be transmitted as it is in the original form, but should be network-coded to modulate a plurality of received packets by a proper operation to form a new packet. In order to realize this, in the existing research, there is a method of newly designing the communication hardware so that the network nodes participating in the network coding can perform these operations, or a method of modifying the transport layer to network-code the received TCP segments at the transport layer Has been proposed.

However, since most of the existing large-scale networks are routing-based networks, it is impossible to implement network coding on a large scale unless modifying the lower layers of the infrastructure to be suitable for network coding. . Although the efficiency of network coding on the laboratory scale has been technically proven, it is difficult for service providers to prioritize investment in replacing the communications hardware of the entire infrastructure or simultaneously changing the operating system and drivers of the nodes to support network coding The point is the problem.

In addition, the network coding scheme can not restore the original file unless a single packet is lost or delayed, except that approximate decoding is possible because there is a high correlation between the source data Or the service quality can not be decoded within a time limit.

Korean Patent Publication No. 10-2013-0016593 (Feb.

[2] "Network coding meets TCP" Jay Kumar Sundararajan, et al. IEEE INFOCOM 2009 pp.280-288

A problem to be solved by the present invention is to provide a TCP-based network router capable of implementing network coding and a TCP-based data streaming system implementing network coding.

The present invention provides a TCP-based network router capable of implementing network coding without changing the hardware of an existing routing-based infrastructure, and a TCP-based data streaming system implementing network coding.

The present invention provides a TCP-based network router capable of implementing network coding without modifying existing routing-based lower layers and a TCP-based data streaming system implementing network coding.

The solution to the problem of the present invention is not limited to those mentioned above, and other solutions not mentioned can be clearly understood by those skilled in the art from the following description.

A data streaming system for implementing network coding according to an aspect of the present invention includes:

At least one transmitting node for generating source data;

An encoding application that performs network coding is run at an application layer to create a TCP session with at least one sending node, receive source data via a TCP session from at least one sending node, An encoding router for generating a session, generating network coded data according to a network coding coefficient matrix and source data, and transmitting elements of the network coded data and a network coding coefficient matrix to a decoding router through a TCP session; And

A decoding application that performs network coding is driven at the application layer and decodes or approximates decodes based on the elements of the network coefficient data and the network coded data received via the TCP session from the encoding router to recover the source data .

According to one embodiment, the elements of the network coded data and the network coding coefficient matrix may be loaded into the payload of the TCP segment.

According to one embodiment, the network coded data is loaded into the payload of the TCP segment, and the elements of the network coding coefficient matrix may be loaded into the TCP header of the TCP segment.

According to one embodiment, the decoding router may be operable to transmit the recovered source data to at least one receiving node.

According to one embodiment, a TCP session may be created between the encoding router and the transmitting node as the receiving node requests transmission of the source data to the transmitting node.

According to one embodiment, the encoding router may be operable to receive meta information of source data from a transmitting node.

A router of a data streaming system implementing network coding according to another aspect of the present invention comprises:

An encoding application that performs network coding is run at the application layer such that the encoding application generates a TCP session with at least one transmitting node that generates source data and sends the source data through a TCP session from at least one transmitting node Generates network coded data according to the network coding coefficient matrix and the source data, and operates to load and output elements of the network coded data and the network coding coefficient matrix to the TCP segment.

A router of a data streaming system implementing network coding according to another aspect of the present invention comprises:

A decoding application that performs network coding is driven at the application layer so that the decoding application decodes or approximates decoding based on the elements of the network coded data and the network coefficient matrix received and received in the TCP segment can do.

A network coded streaming data transmission method using a data streaming system according to another aspect of the present invention includes:

Encoding an application that performs network coding is driven at an application layer; generating a TCP session with at least one sending node generating source data;

The encoding router receiving source data from at least one sending node via a TCP session;

The encoding router comprising the steps of: generating network coded data according to a network coding coefficient matrix and source data;

The encoding router creating a TCP session between a decoding application that performs network coding and a decoding router that is running in an application layer;

Transmitting, by the encoding router, the elements of the network coded data and the network coding coefficient matrix to the decoding router via a TCP session; And

And decoding or approximate decoding based on elements of the network coded data and the network coefficient matrix received via the TCP session by the decoding router to recover the source data.

According to one embodiment, the elements of the network coded data and the network coding coefficient matrix may be loaded into the payload of the TCP segment.

According to one embodiment, the network coded data is loaded into the payload of the TCP segment, and the elements of the network coding coefficient matrix may be loaded into the TCP header of the TCP segment.

According to one embodiment, a network coded streaming data transmission method through the data streaming system comprises:

The decoding router may further comprise transmitting the restored source data to at least one receiving node.

According to one embodiment, a TCP session may be created between the encoding router and the transmitting node as the receiving node requests transmission of the source data to the transmitting node.

According to one embodiment, a network coded streaming data transmission method through the data streaming system comprises:

The encoding router may further comprise receiving meta information of the source data from the transmitting node.

According to the TCP-based network router capable of implementing the network coding of the present invention and the TCP-based data streaming system implementing the network coding, the network coding can be implemented without changing the hardware of the existing routing-based infrastructure.

According to the TCP-based network router capable of implementing the network coding of the present invention and the TCP-based data streaming system implementing the network coding, the network coding can be implemented without modifying the existing lower layers based on the routing.

The effects of the present invention are not limited to those mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the following description.

1 is a conceptual representation of transmission of network coded data occurring in a TCP based data streaming system in which the present invention can implement network coding.
2 is a conceptual diagram illustrating a TCP-based data streaming system capable of implementing network coding according to an embodiment of the present invention.
3 is a conceptual diagram illustrating a hierarchical structure of an encoding router capable of implementing network coding according to an embodiment of the present invention.
4 is a diagram illustrating a structure of a TCP segment generated by an encoding router in a TCP-based data streaming system capable of implementing network coding according to an embodiment of the present invention.
5 is a conceptual diagram illustrating a hierarchical structure of a decoding router capable of implementing network coding according to an embodiment of the present invention.
FIG. 8 is a conceptual diagram illustrating procedures in which a streaming server transmits streaming data to a streaming client in a network coding scheme in a TCP-based data streaming system capable of implementing network coding according to an embodiment of the present invention.
9 is a flowchart illustrating a method of transmitting network coded streaming data through a TCP-based data streaming system according to an embodiment of the present invention.

For the embodiments of the invention disclosed herein, specific structural and functional descriptions are set forth for the purpose of describing an embodiment of the invention only, and it is to be understood that the embodiments of the invention may be practiced in various forms, The present invention should not be construed as limited to the embodiments described in Figs.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. The same reference numerals are used for the same constituent elements in the drawings and redundant explanations for the same constituent elements are omitted.

The following description is largely divided into three sections, Sections 1 and 2, which illustrate the basic principles of approximate decoding along with an introduction to the transmission of network coded data to which the present invention may be applied. Section 3 describes a TCP based data streaming system capable of implementing network coding according to embodiments of the present invention. Note that some of the notations for vector or variable appearing in Section 3 may differ from those shown in Section 1 and Section 2.

The term "encoding" below refers to network coding unless explicitly described in context, for example, as a video encoding. The term "decoding" or "decoding " likewise implies the restoration of network coded information unless explicitly expressed in context, for example, as motion picture decoding.

1. Random linear network coding

1 is a conceptual representation of transmission of network coded data occurring in a TCP based data streaming system in which the present invention can implement network coding. And the source data is transmitted to the decoder (decoder) through the network coding node. The data transmitted in this system has a correlation with each other. The correlated data may have external correlation, such as data measured over the sensor network, or may be highly correlated data, such as images within a video sequence.

Research on the transmission of such correlation data has been widely performed in distributed systems. Network coding techniques have been studied as a technique for transmitting correlation data over a network.

In FIG. 1, an overlay ad hoc network is composed of source nodes, intermediate nodes, and receiving nodes, and the source data is transmitted from the source node to the receiving node through intermediate nodes responsible for network coding operations do. The intermediate node may transmit the received data as it is, but may combine and output the data received from the other source nodes.

Random Linear Network Coding (RLNC) techniques used in the present invention among various network coding schemes are described in Ho, T., M'edard, M., Shi, J., Effros, M. and Karger, DR, "On randomized network coding," in Proc. Allerton Annual Conf. Commun., Control, and Comput., Oct. 2003.

내의 L 개의 소스 심볼들은

라고 표시되며, 각각 이산(discrete)이면서 상관되어(correlated) 있고, 1≤i≤L일 때에

이다.

는

의 알파벳 집합(alphabet set)이고 |

|는

의 크기를 의미한다. 네트워크 코딩 연산은 갈로와 필드(GF: Galois Field) 내에서 수행되기 때문에, 각각의

은 GF 내의 원소로 고려되어야 한다.

가 실수 필드(

, field of real numbers)에 있는 수인지 아니면 GF에 있는 수인지 여부를 명확하게 특정하기 위해서, 수학식 1과 같은 식별 함수(identity function)을 정의할 수 있다.Source set

Lt; RTI ID = 0.0 > L &

And are respectively discrete and correlated, and when 1? I? L

to be.

The

Of the alphabet set (alphabet set) and |

| Is

. Since the network coding operation is performed in a Galois Field (GF), each

Should be considered as elements in the GF.

Is a real field (

, field of real numbers, or the number in GF, it is possible to define an identity function as shown in equation (1).

가

을 의미하는 GF 내의 원소임을 의미한다.

인 경우에,

는 L 개의 요소를 가지는 N 번째 소스 데이터 집합이다.

는

의 알파벳 세트를 의미한다. 또한 본 명세서에서 행렬 또는 벡터의 위 첨자로 쓰이는 T는 행렬 또는 벡터의 전치(transpose)를 의미하며, 실수값 T 또는 아래 첨자로 쓰이는 T와 구별됨을 유의한다.this is

end

Which means that it is an element in GF.

in case of,

Is the Nth source dataset with L elements.

The

≪ / RTI > Also note that T, used as a superscript of a matrix or vector herein, is a transpose of a matrix or vector, and is distinguished from a real value T or a T used as a subscript.

기호는 생략된다.

와

연산자들은 각각 GF 내에서 정의되는 합산 연산과 승산 연산을 의미한다.In the formula describing with the operation symbols in the GF, it implies that all operations occur in the GF,

The symbol is omitted.

Wow

The operators refer to summation and multiplication operations defined in GF, respectively.

The intermediate node k using the RLNC generates and transmits a datagram according to the following equation (2).

와 코딩 계수

의 GF에서의 선형 결합(linear combination in GF)을 의미한다. i는 1≤i≤L인 임의의 정수이고, T는 하나의 데이터그램 내에 결합된 데이터의 개수이다.

와

는 항상 GF 내의 원소로 간주될 수 있다. 코딩 계수들은 크기 2^M을 가지고 GF로부터 균일하고 랜덤하게 선정되는데, 이를

라고 한다. 이는 GF 크기가 M에 의해 결정되며,

임을 의미한다.Equation (2)

And a coding coefficient

Means a linear combination in the GF. i is any integer with 1? i? L, and T is the number of data combined in one datagram.

Wow

Can always be regarded as an element in the GF. The coding coefficients are chosen uniformly and randomly from GF with size 2 ^M ,

. This is because the GF size is determined by M,

.

에서 합산 연산은 XOR 연산으로 수행될 수 있다. GF의 크기는 소스 심볼에 관하여 수행될 수 있는 코딩 연산들의 집합을 결정한다. 따라서 입력 알파벳의 크기가

라고 가정할 수 있다. 만약

인 경우에는 입력 집합은 예를 들어 소스 결합(source binding) 또는 양자화 등을 이용하여 축소되고 그럼으로써 입력 집합은 GF 크기, 즉 2^M을 초과하지 않게 된다.In the present invention, in particular, a GF having a characteristic value of 2

The summation operation can be performed by an XOR operation. The size of GF determines the set of coding operations that can be performed with respect to the source symbol. Therefore, the size of the input alphabet is

. if

The input set is reduced using, for example, source binding or quantization, so that the input set does not exceed the GF size, i.e. 2 ^M.

이 수신되어 연산에 쓰일 수 있다면, 선형 시스템

는 다음 수학식 3과 같이 표현될 수 있다.The datagrams generated at each node are forwarded to neighboring nodes and finally to the receiving nodes. If at the decoder of the receiving node, there are K inversely linear or linearly independent symbols or datagrams,

Lt; / RTI > can be received and used in an operation,

Can be expressed by the following equation (3).

연산자는 유한 필드(finite field)에서 행렬들 사이의 승산 연산을 의미한다. here

The operator denotes a multiplication operation between matrices in a finite field.

는 코딩 계수 행렬(coding coefficient matrix)이라고 하는데, 컬럼 벡터들

로 구성되며, 수신된 심볼들의 유사도 모델에 기초하여 생성된다. 코딩 계수 행렬

를 유사도 모델에 기초하여 생성하는 방법은 네트워크 코딩의 복호화 기법에 숙련된 자들에게 주지된 방법이므로 설명을 생략한다.K × T matrix

Is called a coding coefficient matrix, and column vectors < RTI ID = 0.0 >

And is generated based on the similarity model of the received symbols. Coding coefficient matrix

Based on the degree-of-similarity model is a method well known to those skilled in the network coding decoding technique, and thus a description thereof will be omitted.

가 풀-랭크(full-rank), 즉 K=T인 경우에, 수학식 3의 선형 시스템의 해(solution)인

는, 코딩 계수 행렬

의 역행렬

가 고유하게 결정되고 또한 이에 상응하여

이기 때문에, 다음 수학식 4와 같이 고유하게 결정될 수 있다.If such a coding coefficient matrix

Is a full-rank, that is, K = T, the solution of the linear system of equation (3)

, A coding coefficient matrix

Inverse of

Is uniquely determined and correspondingly

, It can be uniquely determined as shown in the following equation (4).

The inverse matrix of the coding coefficient matrix can be easily obtained by a known method such as performing a Gaussian elimination on GF.

을 결정하기에 충분하지 않은 수의 심볼들만 수신된 경우에, 다시 말해, K<T인 경우에, 코딩 계수 행렬

가 풀-랭크가 아니기 때문에 수학식 3의 선형 시스템의 해

는 유일하게 결정되지 못할 뿐 아니라, 무한한 개수로 얻어질 수 있다. The problem is, if a unique inverse matrix

In the case of K < T, the coding coefficient matrix &lt; RTI ID = 0.0 &gt;

Is not a full-rank, the solution of the linear system of equation (3)

Not only can not be determined uniquely, but can be obtained in an infinite number.

The approximate decoding method is introduced to solve this problem, and the approximately determined source symbols are reconstructed with insufficiently received data.

2. General Approximate Decoding

이 특이 행렬(singular matrix)인 경우에는, 코딩 계수 행렬

의 역행렬

은 얻어질 수 없고 수학식 3의 선형 시스템에서 해가 유일하지 않을 수 있다.The decoder can recover the source data upon receipt of the symbol set y. If there are external factors such as packet delay or loss during the transmission procedure, or a randomly selected coding coefficient matrix

In the case of this singular matrix, the coding coefficient matrix

Inverse of

Can not be obtained and the solution may not be unique in the linear system of equation (3).

To solve this problem, the approximate decoding method can make the coding coefficient matrix full-rank using additional constraints.

와

를 구축하는 데에 이용될 수 있다. 추가 제한 조건들

및

의 모든 요소들(elements) 역시 GF 내의 값이다.The additional constraints that the correlation of the input data adds in the decryption procedure

Wow

Can be used for constructing. Additional Restrictions

And

All the elements of GF are also values in GF.

및

을 추가함으로써 선형 시스템은 유일한 해를 가질 수 있게 되고 유일한 해

는 다음 수학식 5와 같이 표현될 수 있다.Additional Restrictions

And

By adding the linear system can be the only solution and the only solution

Can be expressed by the following equation (5).

및

내의 계수들은 소스의 모델들에 기초하여 결정될 수 있다.

가 결정되면, 원본 데이터의 근사 심볼

은 수학식 1의 식별 함수들에 의해

과 같이 얻어질 수 있다. Additional Restrictions

And

The coefficients in the matrix can be determined based on the models of the source.

Is determined, an approximate symbol of the original data

Is determined by the identification functions of Equation (1)

Can be obtained.

의 생성에 관하여, 좀더 상세한 내용은 H. Park, N. Thomos, and P. Frossard, Approximate decoding approaches for network coded correlated data, Signal Processing (Elsevier), vol. 93, no. 1, pp. 109~213, Jan. 2013 논문을 참조할 수 있다.Additional constraints

H. Park, N. Thomos, and P. Frossard, Approximate decoding approaches for network coded correlated data, Signal Processing (Elsevier), vol. 93, no. 1, pp. 109 to 213, Jan. 2013 thesis can be referred to.

는 유사도 모델에 기초하여 생성될 수 있는데, 예를 들어 가장 유사한 데이터

및

의 위치에 상응하는 자리에 값 "1" 및 "1"을 가지는 두 개의 요소들을 제외하고 각 행이 0으로 구성되도록 생성된다. 또한

은 T-K 길이를 가지는 벡터로서 결정되는데, 예를 들어 벡터

내의 모든 요소들이 0일 수 있다(

).(TK) x T matrix

May be generated based on the similarity model, for example, the most similar data

And

Quot; 1 "and" 1 "in the place corresponding to the position of " 1 &quot; Also

Is determined as a vector having a TK length, for example, vector

All elements in the array can be zero (

).

은 다음의 수학식 6과 같이 얻어질 수 있다.Thus, an approximate symbol of the original data

Can be obtained by the following Equation (6).

This allows the decoder to approximate the original symbols even though some symbols are insufficient for complete decoding.

를 가역적으로(invertible) 만든다는 점이다.
The key idea of this approximate decryption algorithm is to add a remainder equation based on the source correlation,

To make it invertible.

3. A TCP-based data streaming system capable of implementing network coding according to embodiments of the present invention

3.1 First embodiment

Referring to FIG. 2, a TCP-based data streaming system capable of implementing network coding according to embodiments of the present invention is exemplified as follows.

2, a TCP-based data streaming system 20 capable of implementing network coding comprises at least one transmitting node 21a, 21b, 21c, an encoding router 22, a decoding router 23 and one or more receiving nodes (24a, 24b, 24c).

Basically, since the TCP-based data streaming system 20 is implemented on the TCP-based network, the decoding routers 23 and the receiving nodes 24a, 24b, and 24c, between the transmitting nodes 21a, 21b, and 21c and the encoding router 22, And TCP sessions are generated between the terminals 24a and 24c. A predetermined TCP-based routing path is also formed between the encoding router 22 and the decoding router 23, and a TCP session can be semi-persistently maintained depending on the situation.

The encoding router 22 and the decoding router 23 may be, for example, a router of an Internet service provider (ISP), a gateway of a local network composed of the transmitting nodes 21a, 21b and 21c, Or may be the gateway of another local network that is wired to nodes 24a, 24b, and 24c.

Although the TCP sessions between the transmitting nodes 21a, 21b and 21c and the encoding router 22 may be constantly generated, each time the transmitting nodes 21a, 21b and 21c transmit data, Lt; / RTI &gt; Likewise, although the TCP sessions between the decoding router 23 and the receiving nodes 24a, 24b and 24c may also be constantly generated, the decoding routers 23 may be connected to the receiving nodes 21a, 21b and 21c When the data is to be transmitted to the mobile station.

And it transmits to the transmitting nodes (21a, 21b, 21c) are each encoded router 22, the conventional respective source data _{(x 1, x 2, x} 3) in the manner in the form of TCP segments sending TCP segments. In some cases, may be transmitted on one transmission node (21a) has a plurality of data sources over a long period of time _{(x 1, x 2, x} 3) encoding a router in sequence (22).

Received from the receiving nodes (24a, 24b, 24c) is in a conventional manner for receiving TCP segments, the restored source data (x _1, x _2, x _3), each decoding a router 23 in the form of TCP segments can do.

That is, with respect to the transmitting nodes 21a, 21b, and 21c and the receiving nodes 24a, 24b, and 24c, there is substantially no hardware portion that needs to be modified or the network hierarchy to be changed to implement network coding. However, it is also possible that the encoding router 22 knows in advance that the data transmitted by the transmitting nodes 21a, 21b, 21c is data to be network coded, or that the source nodes decoded from the network coded data are received by the receiving nodes 24a, 24b Such as setting the appropriate information to be written in the TCP header, for example, a predetermined port number in advance, at the time of creation of the TCP session, so that the decoding router 23 can know in advance Processing may be required.

As described above, since the network coding is particularly suitable when the correlation of the source data is high, it is preferable that the transmitting nodes 21a, 21b, and 21c transmit data having a common attribute with high correlation. For example, the transmitting nodes 21a, 21b, and 21c may transmit data corresponding to different parts of a moving picture file to the encoding router 22, respectively. In this case, the transmitting nodes 21a, 21b, and 21c transmit the meta information about the file to be distributed and transmitted, that is, the file name, the size, the number of divided data, . The meta information is also transmitted to the decoding router 23.

On the other hand, it is preferable that the segment lengths for transmitting data by each of the transmission nodes 21a, 21b, and 21c are the same, but they may be different depending on cases. However, when the length of the source data to be network-encoded is different from each other, it is difficult for the network coding such as Equation (3) to be properly performed. Therefore, A predetermined padding string may be added to the data to make the source data of the same length.

로써 네트워크 코딩하여 인코딩된 데이터들(y₁, y₂, y₃)을 생성한다.The encoding router 22 converts the received source data (x ₁ , x ₂ , x ₃ ) into a network coding coefficient matrix

To generate encoded data (y ₁ , y ₂ , y ₃ ).

At this time, the hierarchical structure of the encoding router 22 can be described as shown in FIG. 3 is a conceptual diagram illustrating a hierarchical structure of an encoding router capable of implementing network coding according to an embodiment of the present invention.

3, the network hierarchy of the encoding router 22 may be illustrated as a simplified PHY layer 221, a MAC layer 222, a network layer 223, a transport layer 224 and an encoding application 225 .

For example, the PHY layer 221 and MAC layer 222 of the encoding router 22 may be Ethernet-based, the network layer 223 may be IP-based, and the transport layer 224 may be TCP-based.

The encoding application 225 running in the application layer operates together with the application (not shown) of the transmitting nodes 21a, 21b and 21c to generate a TCP session in advance via the transport layer 224 and set a predetermined port number do.

Then, the transmitting nodes 21a, 21b, and 21c set TCP headers appropriately according to a predetermined port number set at the time of creation of the TCP session, and add the source data (x ₁ , x ₂ , x ₃ ) to generate an IP packet, and transmits the generated IP packet to the encoding router 22 through the Ethernet network.

The PHY layer 221 and the MAC layer 222 of the encoding router 22 receive the Ethernet frame. The MAC layer 222 provides the IP packet to the network layer 223 if the received Ethernet frame includes, for example, an IP packet. The network layer 223 provides a TCP segment to the transport layer 224 if the datagram in the IP packet is a TCP segment.

The operations up to this point are those operations usually occurring in a router. A conventional TCP-based network coding implementation technique such as the prior art document [2] has added a network coding layer for network coding TCP segments between the network layer 223 and the transport layer 224. [ The network coding layer of the preceding document [2] network-encodes the received TCP segments and delivers the network-coded new TCP segments to the network layer.

Such a conventional TCP-based network coding implementation scheme speeds up the network segmentation of the received TCP segments and immediately transfers them to the IP layer, which can be matched with the existing TCP / IP scheme. However, since most of the TCP segments are network-coded, TCP segments that are not correlated with each other are network-coded. If packet loss or transmission delay occurs, the receiving side can not recover data on time. This may not be good. In addition, since the network coding layer is added between the network layer and the transport layer, the network layer and the transport layer must be reconfigured to recognize the network coding layer and exchange messages.

On the other hand, the TCP-based network routers 22 and 23 capable of implementing the network coding of the present invention perform network coding at an application layer higher than the transport layer 224. [

Accordingly, the transport layer 224 identifies, for example, a particular port number in the TCP header in the TCP segment and provides the corresponding TCP segment to the encoding application 225.

를 설정하며, 소스 데이터들(x₁, x₂, x₃)을 네트워크 코딩 계수 행렬

로써 네트워크 코딩하고, 네트워크 코딩된 데이터들(y₁, y₂, y₃)을 생성한다.The encoding application 225 collects the meta information and the source data (x ₁ , x ₂ , x ₃ ) transmitted for network coding from the transmission nodes 21a, 21b, and 21c where TCP sessions are respectively generated, Based on the meta information, an appropriate size network coding coefficient matrix

And sets the source data (x ₁ , x ₂ , x ₃ ) as a network coding coefficient matrix

And generates network coded data (y ₁ , y ₂ , y ₃ ).

의 역행렬

을 이용하여 수학식 4과 같이 이상적으로 복호되거나 또는 수학식 5 및 수학식 6와 같이 근사 복호될 수 있다. 이를 위해, 인코딩 어플리케이션(225)은 네트워크 코딩된 데이터(y₁, y₂, y₃)와 네트워크 코딩 행렬

의 계수들, 내지 근사 복호를 위한 추가 제한 조건

가 TCP 세그먼트에 포함되도록 할 필요가 있다.On the other hand, the network coded data includes a network coding coefficient matrix

Inverse of

(4) or may be approximated as shown in Equations (5) and (6). To this end, the encoding application 225 may generate the network coded data (y ₁ , y ₂ , y ₃ )

The additional constraints for approximate decoding,

Needs to be included in the TCP segment.

의 계수들, 내지 근사 복호를 위한 추가 제한 조건

을 TCP 페이로드에 탑재하도록 설정한 메시지를 전송 계층(224)에 전달할 수 있다. In accordance with an embodiment, the encoding application 225 may generate the network coded data (y ₁ , y ₂ , y ₃ )

The additional constraints for approximate decoding,

To the transport layer 224. The transport layer 224 may transmit the message to the transport layer 224. [

의 계수들 내지 근사 복호를 위한 추가 제한 조건

을 TCP 페이로드에 탑재한 TCP 세그먼트를 생성하고, 생성된 TCP 세그먼트를 네트워크 계층(223)으로 전달할 수 있다.In this case, the transport layer 224 includes network coded data (y ₁ , y ₂ , y ₃ ) and a network coding matrix

Additional constraints for the coefficients to approximate decoding

To the network layer 223, and transmits the generated TCP segment to the network layer 223. The TCP segment is transmitted to the network layer 223 via the TCP segment.

의 계수들 내지 근사 복호를 위한 추가 제한 조건

을 TCP 헤더에 탑재하도록 설정한 메시지를 전송 계층(224)에 전달할 수 있다. In another embodiment, the encoding application 225 loads the network coded data (y ₁ , y ₂ , y ₃ ) into a TCP payload,

Additional constraints for the coefficients to approximate decoding

To the transport layer 224. In this case,

의 계수들 내지 근사 복호를 위한 추가 제한 조건

을 TCP 헤더, 예를 들어 TCP 옵션 필드에 기록하여 TCP 세그먼트를 생성하고, 생성된 TCP 세그먼트를 네트워크 계층(223)으로 전달할 수 있다.In this case, the transport layer 224 mounts the network coded data (y ₁ , y ₂ , y ₃ ) in the TCP payload,

Additional constraints for the coefficients to approximate decoding

For example, in the TCP option field to generate a TCP segment, and transmit the generated TCP segment to the network layer 223. [

The operation of this transport layer 224 is to use the basic operation of the transport layer as it is and there is absolutely no part that has to be modified for the implementation of the present invention.

To illustrate the structure of a TCP segment, refer to FIG. 4 is a diagram illustrating a structure of a TCP segment generated by a network encoding router in a TCP-based data streaming system capable of implementing network coding according to an embodiment of the present invention.

In Fig. 4, the TCP segments carry network coded data (y ₁ , y ₂ , y ₃ ), respectively, in a TCP payload, and each of the three equal number of network coding matrix coefficients c ₁ to c ₉ ) in the TCP header.

2, the encoding router 22 transmits TCP segments carrying the network coded data (y ₁ , y ₂ , y ₃ ) and the network coding matrix coefficients to the decoding router 23 through a pre-generated TCP session send. The TCP segments sent by the encoding router 22 are carried in IP packets and can be transmitted to the decoding router 23 without being different from ordinary IP packets.

Depending on the embodiment, the network coded data output by the encoding router 22 may be routed through another encoding router, instead of being forwarded directly to the decoding router 23. In this case, another encoding router may again network encode the received network coded data.

It is also possible that the transmitting nodes 22a, 21b and 21c transmit the source data to the encoding router 22 by loading the source datagram in a User Datagram Protocol (UDP) segment, It is also possible to load the elements of the data and the network coding coefficient matrix into the UDP segment and transmit them to the decoding router 23.

Decoding router 23 has a network hierarchy as illustrated in FIG. 5, as well as encoding router 22. 5 is a conceptual diagram illustrating a hierarchical structure of a decoding router capable of implementing network coding according to an embodiment of the present invention.

5, the network hierarchy of the decoding router 23 may also be briefly illustrated by the PHY layer 231, the MAC layer 232, the network layer 233, the transport layer 234 and the decoding application 235 .

For example, the PHY layer 231 and the MAC layer 232 of the decoding router 23 may be Ethernet-based, the network layer 233 may be IP-based, and the transport layer 234 may be TCP-based.

The decoding application 235 running in the application layer operates together with the application (not shown) of the receiving nodes 24a, 24b and 24c to generate a TCP session in advance via the transport layer 234 and set a predetermined port number do.

The PHY layer 231 and the MAC layer 232 of the decoding router 23 receive an Ethernet frame. The MAC layer 232 provides the IP packet to the network layer 233 if the received Ethernet frame includes, for example, an IP packet. The network layer 233 provides a TCP segment to the transport layer 234 if the datagram in the IP packet is a TCP segment.

The transport layer 234 identifies, for example, a particular port number in the TCP header in the TCP segment and provides the corresponding TCP segment to the decoding application 235.

의 계수들 내지 근사 복호를 위한 추가 제한 조건

을 추출하고, 별도로 네트워크 코딩되지 않은 메타 정보를 획득할 수 있다.The decoding application 235 receives the network coded data (y ₁ , y ₂ , y ₃ ) and the network coding coefficient matrix

Additional constraints for the coefficients to approximate decoding

And obtain meta information that is not separately network coded.

의 계수들 내지 근사 복호를 위한 추가 제한 조건

을 추출할 수 있다.According to an embodiment, the decoding application 235 receives network coded data (y ₁ , y ₂ , y ₃ ) from the TCP payload and a network coding coefficient matrix

Additional constraints for the coefficients to approximate decoding

Can be extracted.

의 계수들 내지 근사 복호를 위한 추가 제한 조건

을 추출할 수 있다.According to an embodiment, the decoding application 235 extracts the network coded data (y ₁ , y ₂ , y ₃ ) from the TCP payload and extracts the network coding coefficient matrix

Additional constraints for the coefficients to approximate decoding

Can be extracted.

의 역행렬

과 네트워크 코딩된 데이터들(y₁, y₂, y₃)로부터 수학식 4과 같이 이상적으로 복호하거나 또는 수학식 5 및 수학식 6와 같이 근사 복호함으로써 소스 데이터들(x₁, x₂, x₃)을 복원할 수 있다.The decoding application 235 then uses the network coding coefficient matrix &lt; RTI ID = 0.0 &gt;

Inverse of

With the network coded data _{(y 1, y 2, y} 3) ideal decoding or the source data by approximating the decoding as shown in Equation 5 and Equation _{6 (x 1, x 2,} x a as shown in Equation 4, from ₃ ) can be restored.

The decoding application 235 sends the restored source data (x ₁ , x ₂ , x ₃ ) to at least one receiving node 24a, 24b, 24c via a TCP session, do.

In this way, it is possible to implement network coding by simply installing an encoding application and a decoding application for performing network coding at the application layer on each of two nodes serving as routers, without modifying or modifying the existing TCP-based infrastructure have.

FIG. 8 is a conceptual diagram illustrating procedures for a streaming server to transmit streaming data to a streaming client using a network coding scheme in a TCP-based data streaming system capable of implementing network coding according to an embodiment of the present invention.

6, when a user who wants a video streaming service accesses a video streaming website through a predetermined streaming client 64 and clicks a specific video link, the streaming client 64 and the decoding router 63, A TCP session is created between the encoding router 62 and the streaming server 61, respectively.

The streaming server 61 may internally have a plurality of video encoding nodes 61a, 61b, and 61c. The plurality of video encoding nodes 61a, 61b, and 61c may correspond to the transmission nodes 21a, 21b, and 21c illustrated in FIG.

Referring to FIG. 7, the streaming server 61 transmits the requested video data to the encoding router 62 in the form of, for example, a payload of a TCP segment.

로 네트워크 코딩하고, 인코딩 라우터(62)는 네트워크 코딩된 데이터들을 TCP 세그먼트에 탑재하여 디코딩 라우터(63)로 전송한다.In the encoding router 62, the encoding application 621 converts the source data extracted from the received TCP segments into a predetermined network coding coefficient matrix &lt; RTI ID = 0.0 &gt;

And the encoding router 62 loads the network coded data into the TCP segment and transmits it to the decoding router 63. [

의 역행렬

로 디코딩하여 소스 데이터들을 복원하고, 디코딩 라우터(63)는 복원된 소스 데이터들을 TCP 세그먼트에 탑재하여 스트리밍 클라이언트(64)로 전송한다.In the decoding router 63, the decoding application 631 transmits the network coded data extracted from the received TCP segments to a predetermined network coding coefficient matrix

Inverse of

And restores the source data, and the decoding router 63 loads the restored source data into the TCP segment and transmits the restored source data to the streaming client 64.

Referring to FIG. 8, when the transmission of the video streaming data is completed, the streaming client 64 ends the TCP session with the decoding router 63. So that the session between the encoding router 62 and the streaming server 61 can also be terminated. The session between the decoding router 63 and the encoding router 62 may be terminated and may be maintained so as to be used for transmission of other video streaming data depending on the situation.

9 is a flowchart illustrating a method of transmitting network coded streaming data through a TCP-based data streaming system according to an embodiment of the present invention.

Referring to FIG. 9, in the network-coded streaming data transmission method using the TCP-based data streaming system according to the embodiments of the present invention, in step S91, an encoding application for performing network coding is connected to an encoding router 22 may start with a TCP session between at least one sending node 21a.

According to the embodiment, when UDP is used instead of TCP, a session may not be separately generated.

When a TCP session is created, in step S92, the encoding router 22 can receive meta information and source data of a file to be network coded from at least one transmitting node 21a via a TCP session.

In step S93, the encoding router 22 generates network coded data according to the network coding coefficient matrix and the source data.

In step S94, the encoding router 22 creates a TCP session between the decoding application that performs network coding and the decoding router 23 that is driven in the application layer.

In step S95, the encoding router 22 transmits the elements of the network coded data and the network coding coefficient matrix to the decoding router 23 via the TCP session.

According to an embodiment, the elements of the network coded data and the network coding coefficient matrix may be loaded into the payload of the TCP segment.

According to an embodiment, network coded data may be loaded into a payload of a TCP segment, and elements of a network coding coefficient matrix may be loaded into a TCP header of a TCP segment.

In step S96, the decoding router 23 decodes or approximates and decodes the received data based on the elements of the received network coded data and the network coding coefficient matrix to recover the source data.

In the optional step S97, the decoding router 23 transmits the restored source data to at least one receiving nodes 24a.

3.2 Application examples

First, the network coding data streaming system of the present invention can be applied to an end user oriented service. In such a case, the encoding router 22 may be installed in a transmission node of a video service provider such as, for example, Youtube.com, and the decoding router 23 may be integrated with a web client to be installed in a personal computer or a mobile terminal .

Next, the network coding data streaming system of the present invention can be used to reduce traffic of a backbone network. The encoding router 22 may be installed, for example, at a location where the backbone network provider receives streaming data from a large video service provider, and the decoding router 23 is located at a location where the backbone network provider is connected to the local networks Can be installed.

Meanwhile, the network coding data streaming system of the present invention can be used to reduce traffic inside the cloud system or inside the data center. About three-quarters of the traffic coming from the cloud system or data center is known to be generated by operations such as storing, sorting, creating, searching, authenticating, and backing up data in the cloud system or data center. Therefore, if the encoding router 22 is installed in a location where data is stored and retrieved in a cloud system or a data center, and the decoding router 23 is installed in a location where data is output to the outside, the traffic generated in the cloud system or the data center And power consumption can be greatly reduced.

Furthermore, the network coding data streaming system of the present invention can be used in a digital broadcasting network. Since digital broadcasting simultaneously broadcasts a small number of image information, it is inherently suitable for network coding.

It is to be understood that both the foregoing general description and the following detailed description of the present invention are exemplary and explanatory and are intended to provide further explanation of the invention as claimed. It will be understood that variations and specific embodiments which may occur to those skilled in the art are included within the scope of the present invention.

Further, the apparatus according to the present invention can be implemented as a computer-readable code on a computer-readable recording medium. A computer-readable recording medium includes all kinds of recording apparatuses in which data that can be read by a computer system is stored. Examples of the recording medium include ROM, RAM, optical disk, magnetic tape, floppy disk, hard disk, nonvolatile memory and the like. The computer-readable recording medium may also be distributed over a networked computer system so that computer readable code can be stored and executed in a distributed manner.

20 TCP based data streaming system
21a, 21b, 21c transmitting node
22 encoding routers
23 decoding router
24a, 24b, 24c receiving node
61 streaming server
62 encoding routers
63 decoding router
64 streaming clients

Claims (14)

A routing path between the encoding router and at least one transmitting node that holds the source data when there is a transmission request for source data from the receiving node via the decoding router, Receiving source data from the at least one transmitting node via the routing path and providing the source data to the encoding application when the received data is source data transmitted at the transmitting node, Wherein the encoding application generates network coded data based on the network coding coefficient matrix and the source data, and transmits the network coded data and elements of the network coding coefficient matrix to the decoding router Encodes router; And
The method comprising: driving a decoding application that performs network coding at an application layer; transmitting the transmission request received from the receiving node to the encoding router; and wherein the received data includes network coded data and a network coefficient matrix Providing elements of the received network coded data and the network coefficient matrix to the decoding application if the elements of the network coded data and the network coefficient matrix are decoded by the decoding application based on the elements of the network coded data and the network coefficient matrix Or a decoding router for approximate decoding to recover source data and transmit the restored source data to at least one receiving node. The data streaming system of claim 1, wherein elements of the network coded data and the network coding coefficient matrix that the encoding router transmits to the decoding router are loaded in a payload of a TCP segment or a UDP segment. The network coding method according to claim 1, wherein the network coded data transmitted by the encoding router to the decoding router is loaded in a payload of a TCP segment or a UDP segment, Are mounted on the header of the TCP segment or the UDP segment. delete delete The data streaming system of claim 1, wherein the encoding router is operative to receive meta information of the source data from the transmitting node. As a router in a data streaming system,
The router
The method comprising: driving an encoding application that performs network coding at an application layer; providing the received source data to the encoding application if the data received at the router is source data received from at least one transmitting node; And to transmit to the other router a segment comprising elements of the network coding coefficients and the network coded data generated in the network coding coefficient matrix,
The encoding application
Generating a routing path between the at least one transmitting node holding the source data and the router when there is a transmission request for the source data from the receiving node via the another router, And generating the network coded data based on the network coding coefficient matrix and the source data when the source data is provided through the routing path, and generating the network coded data based on the network coded coefficient matrix and the elements of the network coding coefficient matrix, And to generate a segment. As a router in a data streaming system,
The router
The method comprising: driving a decoding application that performs network coding at an application layer, and when the data received at the router is elements of network-coded data and a network coefficient matrix transmitted from another router, the received network- To provide elements of the network coefficient matrix to the decoding application and to transmit the recovered source data to the at least one receiving node in the decoding application,
The decoding application
If the transmission request for the source data is received from the at least one receiving node, transferring the transmission request to the transmitting node via the other router connected to the transmitting node holding the source data, And if the elements of the network coefficient matrix are provided, recovering the source data by decoding or approximating decoding based on the elements of the network coded data and the network coefficient matrix. An encoding router for driving an encoding application that performs network coding at an application layer includes at least one transmission for holding the source data by the encoding application when there is a transmission request for source data from a receiving node via a decoding router, Creating a routing path between the node and the encoding router;
The encoding router receiving the source data over the routing path from the at least one transmitting node;
Wherein the encoding router provides the source data to the encoding application when the received data is source data transmitted from the transmitting node, thereby causing the encoding application to generate the network coding coefficient matrix and the network coded data based on the source data Gt;
The encoding router transmitting elements of the network coded data and a network coding coefficient matrix to a decoding router that drives a decoding application that performs network coding at an application layer;
Wherein the decoding router provides the received network coded data and elements of the network coefficient matrix to the decoding application when the received data is elements of network coded data and a network coefficient matrix transmitted from the encoding router And decoding or approximate decoding based on the network coded data and the elements of the network coefficient matrix by the decoding application to recover source data; And
Wherein the decoding router is configured to transmit the recovered source data to the at least one receiving node. &Lt; Desc / Clms Page number 19 &gt; The system of claim 9, wherein the network coded data and elements of the network coding coefficient matrix transmitted by the encoding router to the decoding router are loaded into a payload of a TCP segment or a UDP segment. A network coded streaming data transmission method. The network coding method according to claim 9, wherein the network coded data transmitted by the encoding router to the decoding router is loaded in a payload of a TCP segment or a UDP segment, Are loaded into the header of the TCP segment or the UDP segment. delete delete The method of claim 9,
Wherein the encoding router further comprises receiving meta information of the source data from the transmitting node. &Lt; RTI ID = 0.0 &gt; 31. &lt; / RTI &gt;

Images