公开(公告)号：KR100572693B1

公开(公告)日：2006-04-19

申请号：KR1020030098350

申请日：2003-12-27

Applicant: 한국전자통신연구원

Inventor： 김재영 ,  안병준 ,  김영선 ,  김학서 ,  정민영

IPC: H04L12/741

Abstract: 1. 청구범위에 기재된 발명이 속한 기술분야
본 발명은 인터넷 프로토콜 패킷 룩업 방법에 관한 것임.
2. 발명이 해결하려고 하는 기술적 과제
본 발명은, IPv6 패킷을 처리하는 노드의 128비트 길이의 로컬 인터페이스 프리픽스 엔트리 정보만을 저장하는 작은 크기의 CAM과 64비트 이하의 네트워크 프리픽스 엔트리 정보를 저장하는 다단계로 구성되는 포워딩 테이블들을 이용하여 IPv6 패킷을 이용하여 통신을 하는 모든 네트워크 상의 노드에서 인입되는 IPv6 패킷을 최소 시간 내에 해당 목적지로 전송하기 위해 필요한 IPv6 인터넷 프로토콜 패킷 룩업 방법을 제공하고자 함.
3. 발명의 해결방법의 요지
본 발명은, 인터넷 프로토콜 패킷 룩업 방법에 있어서, 패킷 수신기로부터 패킷을 수신하여 헤더 정보를 추출하고, 추출된 헤더 정보에서 목적지 주소를 추출하여 CAM(Content Addressable Memory)에서 룩업을 수행하는 CAM 룩업단계; 상기 CAM 룩업단계에서 룩업 매칭이 발생하였는지를 확인하는 CAM 룩업 확인단계; 상기 CAM 룩업 확인단계에서, 매칭이 된 경우는 중앙 제어기로 패킷을 전달하고, 매칭이 되지 않은 경우는 상기 CAM에 저장된 인덱스를 이용하여 멀티비트-트라이 테이블 위치를 파악하여 해당 멀티비트-트라이 테이블을 이용하여 룩업을 수행하는 멀티비트-트라이 테이블 룩업단계; 상기 멀티비트-트라이 테이블 룩업단계에서 매칭이 발생하는지를 확인하는 멀티비트-트라이 테이블 룩업 확인단계; 및 상기 멀티비트-트라이 테이블 룩업 확인단계에서, 매칭이 발생하면 외부 네트워크로 전송해야할 패킷이므로 패킷 송신기로 패킷을 전달하고, 매칭이 발생하지 않으면 패킷을 폐기하거나 디폴트 루트로 패킷을 전송하는 패킷 처리단계를 포함함.
4. 발명의 중요한 용도
본 발명은 네트워크 시스템 등에 이용됨.
IPv6, 인터넷 프로토콜, 프리픽스, 룩업, 포워딩 정보, 로컬 인터페이스 정보, 인덱스, CAM

Abstract translation: 1.权利要求书中描述的发明所属的技术领域

公开(公告)号：KR100431208B1

公开(公告)日：2004-05-12

申请号：KR1020020028866

申请日：2002-05-24

Applicant: 한국전자통신연구원 ,  주식회사 케이티

Inventor： 김영희 ,  정민영 ,  전병천 ,  김대영

IPC: H04L12/28

Abstract: PURPOSE: An ER-LSP(Explicit Routed-Label Switched Path) management method for admitting the fault of an MPLS(Multi Protocol Label Switching) system and an apparatus therefor are provided to prevent that unnecessary control messages are generated by the resetting of all ER-LSPs by managing LDP(Label Distribution Protocol) session information and state information according to the ER-LSPs and resetting only a corresponding ER-LSP by stored information when an LDP session is connected after a system and link fault is removed. CONSTITUTION: An ER-LSP database(40) stores an ER-LSP ID, an ER-LSP state, link information, and a link state with respect to each ER-LSP. An ER-LSP setup/release management unit(20) manages the setup and release of the ER-LSP. A state informing block performs a line recovery and LDP session setup informing function and a link fault and LDP session release informing function. If the fault of the LDP session and link is sensed by the state informing block, an ER-LSP fault management unit(60) performs the fault management function of the ER-LSP, releases assigned resources, and releases internal switch connection setup. If the state of the ER-LSP is changed by the ER-LSP setup/release management unit(20) and the ER-LSP fault management unit(60), an ER-LSP state management unit(30) changes ER-LSP state information and LDP session information of the ER-LSP database(40). When the LDP session setup is sensed by the state informing block, an ER-LSP resetting task block(90) requests resetting with respect to the ER-LSP of the setup state and the ER-LSP of a down-state.

Abstract translation: 目的：提供一种用于接纳MPLS（多协议标签交换）系统故障的ER-LSP（显式路由标签交换路径）管理方法及其装置，以防止通过重置所有ER而产生不必要的控制消息 -LSP通过根据ER-LSP管理LDP（标签分发协议）会话信息和状态信息，并在移除系统和链路故障后连接LDP会话时通过存储的信息仅重置对应的ER-LSP。 组成：ER-LSP数据库（40）存储关于每个ER-LSP的ER-LSP ID，ER-LSP状态，链路信息和链路状态。 ER-LSP建立/释放管理单元（20）管理ER-LSP的建立和释放。 状态通知块执行线路恢复和LDP会话建立通知功能以及链路故障和LDP会话释放通知功能。 如果LDP会话和链路的故障由状态通知块检测到，则ER-LSP故障管理单元（60）执行ER-LSP的故障管理功能，释放分配的资源，并且释放内部交换机连接建立。 如果ER-LSP建立/释放管理单元（20）和ER-LSP故障管理单元（60）改变ER-LSP的状态，则ER-LSP状态管理单元（30）改变ER-LSP状态 信息和ER-LSP数据库（40）的LDP会话信息。 当状态通知块检测到LDP会话建立时，ER-LSP重置任务块（90）请求关于建立状态的ER-LSP和关闭状态的ER-LSP的重置。

公开(公告)号：KR100411248B1

公开(公告)日：2003-12-18

申请号：KR1020010084606

申请日：2001-12-26

Applicant: 한국전자통신연구원 ,  주식회사 케이티

Inventor： 정민영 ,  윤현식 ,  양선희

IPC: H04L12/28

Abstract: PURPOSE: A device and a method for adaptively controlling cell transfer in an ATM(Asynchronous Transfer Mode)-based MPLS(Multi-Protocol Label Switching) system are provided to periodically calculate an ATM cell transfer rate by using IP packet characteristics information varying according to given CR-LSP(Constraint-Route Label Switched Path) band information and time, and to reflect the calculated information on channel bands, thereby securing CR-LSP service quality and efficiently using system bands. CONSTITUTION: A database(40) manages data such as band characteristics according to CR-LSPs. A packet size statistics collector(30) collects IP(Internet Protocol) packet statistical information according to CR-LSPs from a delivery engine(80) to record the collected information in the database(40). A cell transfer rate calculator(50) calculates the current maximum cell transfer rate and an average transfer rate based on the information stored in the database(40), and stores the calculated rates in the database(40). A band converter(60) calculates change rates of the maximum and average cell transfer rates by using the information stored in the database(40), if update of the maximum and average cell transfer rates is notified from the cell transfer rate calculator(50), and updates prior maximum and average cell transfer rates stored in the database(40) according to the calculated change rates, then requests change of bands allocated to the CR-LSPs to a connection controller(70).

Abstract translation: 目的：提供一种用于在基于ATM（异步传输模式）的MPLS（多协议标签交换）系统中自适应地控制信元传输的设备和方法，以通过使用根据以下各项变化的IP分组特征信息来周期性地计算ATM信元传输速率 给出CR-LSP（约束路由标签交换路径）的带宽信息和时间，并反映计算出的信道带宽信息，从而保证CR-LSP服务质量并有效利用系统带宽。 构成：数据库（40）根据CR-LSP管理诸如频带特征的数据。 分组大小统计收集器（30）根据来自递送引擎（80）的CR-LSP收集IP（互联网协议）分组统计信息，以将收集的信息记录在数据库（40）中。 单元传送速度计算器（50）基于存储在数据库（40）中的信息计算当前的最大单元传送速率和平均传送速率，并将计算出的速率存储在数据库（40）中。 如果从小区转移率计算器（50）通知最大和平均小区转移率的更新，则频带转换器（60）通过使用存储在数据库（40）中的信息来计算最大和平均小区转移率的变化率 ，并且根据计算的改变速率更新存储在数据库（40）中的先前的最大和平均小区传送速率，然后请求将分配给CR-LSP的频带改变到连接控制器（70）。

公开(公告)号：KR1020020032937A

公开(公告)日：2002-05-04

申请号：KR1020000063714

申请日：2000-10-28

Applicant: 한국전자통신연구원 ,  주식회사 케이티

Inventor： 이재섭 ,  정민영 ,  김미회

IPC: H04L12/24

CPC classification number: H04L45/28 ,  H04L41/0668 ,  H04L45/50

Abstract: PURPOSE: A method for setting a connection and restoring a fault of the connection for protecting and restoring the connection in an MPLS(Multi Protocol Label Switching) network is provided to efficiently make use of network resources by setting up a protection connection using former connection fault information in case that an LSP(Label Switching Path) connection fault is generated in an MPLS network, and by rapidly restoring the connection fault using an MPLS FIS(Fault Indication Signal). CONSTITUTION: Information capable of triggering a fault restoration system is received(801). It is analyzed whether the information is an LDP connection setting request(802). In case that a new LSP connection setting is requested, a connection setting response is waited(803). In case that the connection setting response is received, it is checked whether there is link information which experienced a PD fault(804). In case that there's the information, a protection connection is set up(805). In case that a PSL receives fault notifying information(806), it is checked whether the protection connection has already set up(807). In case that the protection connection has already set up, a traffic is switched(809). In case that the protection connection has not already set up, the protection connection is set up(808). After that the switching is performed(809). In case that the PSL receives fault restoration notifying information(810), the protection connection is maintained(811).

Abstract translation: 目的：提供一种在MPLS（多协议标签交换）网络中设置连接和恢复连接故障以保护和恢复连接的方法，以通过使用以前的连接故障设置保护连接来有效利用网络资源 在MPLS网络中生成LSP（标签交换路径）连接故障的情况下，以及使用MPLS FIS（故障指示信号）快速恢复连接故障的信息。 构成：接收到能够触发故障恢复系统的信息（801）。 分析信息是否是LDP连接设置请求（802）。 在请求新的LSP连接设置的情况下，等待连接设置响应（803）。 在接收到连接设置响应的情况下，检查是否存在经历PD故障的链路信息（804）。 如果有信息，则建立保护连接（805）。 在PSL接收到故障通知信息（806）的情况下，检查保护连接是否已经建立（807）。 在保护连接已经建立的情况下，流量被切换（809）。 在保护连接尚未建立的情况下，保护连接建立（808）。 之后进行切换（809）。 在PSL接收故障恢复通知信息（810）的情况下，保护连接（811）。

公开(公告)号：KR100875911B1

公开(公告)日：2008-12-26

申请号：KR1020060086283

申请日：2006-09-07

Applicant: 한국전자통신연구원

Inventor： 양미정 ,  김태일 ,  정민영 ,  정명희 ,  최승혁 ,  박재형

IPC: H04L12/28 ,  H04L12/56 ,  H04L12/02 ,  H04B7/26

Abstract: A method and an apparatus for updating link state information are provided to remarkably reduce the overheads due to link state update and the fluctuation of a routing protocol by reducing the number of times link state information is updated. A node, in a network where a plurality of nodes are connected through links, transmits link state information, which indicates the state of its own connected link, to a neighbor node, and constructs a whole network link state table containing link state information of the whole network, based on the transmitted link state information(S110). If a QoS routing request is made, the node configures a QoS routing table, based on the whole network link state table, and establishes a QoS path. The node calculates a value consumed for the resources of each link located on the established QoS path and acquires the present value and average value of the link state information(S120). In case the difference between the present value and average value of the link state information is larger than a threshold, the node transmits the present value to the neighbor node(S130).

Abstract translation: 提供了一种更新链路状态信息的方法和装置，通过减少链路状态信息更新的次数，可以显着降低链路状态更新和路由协议波动带来的开销。 在多个节点通过链路连接的网络中的节点向相邻节点发送表示其自身的连接链路的状态的链路状态信息，并且构建包含该链路状态信息的整个网络链路状态表 基于所发送的链路状态信息的整个网络（S110）。 如果进行QoS路由请求，则节点基于整个网络链路状态表来配置QoS路由表，并建立QoS路径。 节点计算对于位于建立的QoS路径上的每个链路的资源消耗的值，并且获取链路状态信息的当前值和平均值（S120）。 在链路状态信息的当前值和平均值之间的差值大于阈值的情况下，节点将当前值发送到邻居节点（S130）。

公开(公告)号：KR1020070059924A

公开(公告)日：2007-06-12

申请号：KR1020060086283

申请日：2006-09-07

Applicant: 한국전자통신연구원

Inventor： 양미정 ,  김태일 ,  정민영 ,  정명희 ,  최승혁 ,  박재형

IPC: H04L12/28 ,  H04L12/56 ,  H04L12/02 ,  H04B7/26

CPC classification number: H04W40/24 ,  H04W24/10

Abstract: A method and an apparatus for updating link state information are provided to remarkably reduce the overheads due to link state update and the fluctuation of a routing protocol by reducing the number of times link state information is updated. A node, in a network where a plurality of nodes are connected through links, transmits link state information, which indicates the state of its own connected link, to a neighbor node, and constructs a whole network link state table containing link state information of the whole network, based on the transmitted link state information(S110). If a QoS routing request is made, the node configures a QoS routing table, based on the whole network link state table, and establishes a QoS path. The node calculates a value consumed for the resources of each link located on the established QoS path and acquires the present value and average value of the link state information(S120). In case the difference between the present value and average value of the link state information is larger than a threshold, the node transmits the present value to the neighbor node(S130).

Abstract translation: 提供一种用于更新链路状态信息的方法和装置，通过减少链路状态信息被更新的次数，显着地减少了链路状态更新引起的开销和路由协议的波动。 通过链路连接多个节点的网络中的节点将指示其自身连接链路的状态的链路状态信息发送到邻居节点，构成包含链路状态信息的整个网络链路状态表， 整个网络，基于发送的链路状态信息（S110）。 如果进行QoS路由请求，则根据整个网络链路状态表，节点配置QoS路由表，并建立QoS路径。 该节点计算针对建立的QoS路径上的每个链路的资源消耗的值，并获取链路状态信息的当前值和平均值（S120）。 在链路状态信息的当前值和平均值之间的差大于阈值的情况下，节点将当前值发送到邻居节点（S130）。

公开(公告)号：KR100431207B1

公开(公告)日：2004-05-12

申请号：KR1020020026442

申请日：2002-05-14

Applicant: 한국전자통신연구원 ,  주식회사 케이티

Inventor： 양선희 ,  윤호선 ,  윤현식 ,  정민영 ,  김병식

IPC: H04L12/28

Abstract: PURPOSE: A method for supplying an extranet IP-VPN(Virtual Private Network) service in an MPLS(Multi Protocol Label Switching)-based network is provided to simply support the extranet service, and to assign more than one RD(Route Descriptor) for the same interface, thereby simultaneously supporting an intranet service and the extranet service. CONSTITUTION: When path information is added(231), a CE(Customer Edge)1(121) transmits the path information to a PE(Provider Edge)1(111)(232). The PE1(111) retrieves an RD, and decides whether a VPN_intra1 is connected with a VPN_extra2(233), then distributes path update information to BGP peers(234). The path update message is reflected on a VPN_extra2 table of a PE3(113)(235). The PE3(113) transmits the path information to a CE3(123)(236), and modifies route information(237). The received information is reflected on a VPN_extra2 table of a PE4(114)(239). The PE4(114) transmits the path information to a CE4(124)(240).

Abstract translation: 目的：提供一种在基于MPLS（多协议标签交换）的网络中提供外部网络IP-VPN（虚拟专用网络）服务的方法，以简单地支持外部网络服务，并为多个RD（路由描述符） 同一接口，从而同时支持内联网服务和外联网服务。 构成：当添加路径信息（231）时，CE（客户边缘）1（121）将路径信息发送给PE（提供商边缘）1（111）（232）。 PE1（111）检索RD，并确定VPN_intra1是否与VPN_extra2（233）连接，然后将路径更新信息分发给BGP对等体（234）。 路径更新消息反映在PE3的VPN_extra2表（113）（235）上。 PE3（113）将路径信息发送到CE3（123）（236），并修改路线信息（237）。 接收到的信息反映在PE4（114）（239）的VPN_extra2表中。 PE4（114）将路径信息发送给CE4（124）（240）。

公开(公告)号：KR100353850B1

公开(公告)日：2002-09-28

申请号：KR1020000066638

申请日：2000-11-10

Applicant: 한국전자통신연구원 ,  주식회사 케이티

Inventor： 김숙연 ,  정민영 ,  양선희 ,  이유경

IPC: H04L12/28

Abstract: 본 발명은 멀티 프로토콜 레이블 교환(MPLS) 망에서 라우터간의 트래픽 흐름들에 대한 서비스품질(QoS)을 보장하는 경로 보호 방법 및 그 기록매체에 관한 것으로서, MPLS 망 등에서 임의의 두 레이블 에지 라우터(LER : Label Edge Router)간의 트래픽 흐름의 집합에 대한 QoS 조건이 주어져 있을 때, 사전 보호 경로 설정을 통해 망의 고장시에 신속히 제한적 레이블 스위치 경로(CR-LSP)를 복구하여 그 QoS 조건을 보장함으로써, 망의 자원 활용도와 서비스 가용성을 최대화하기 위한 경로 보호 방법 및 그 기록매체를 제공하기 위하여, 한 쌍의 LER 사이의 트래픽 흐름의 집합에 대한 QoS 조건을 만족하는 동작/보호 경로들로 경로 집합을 설정하는 단계; 설정된 경로 집합에 속하는 모든 경로들을 동작 경로로 활용하면서 동시에 서로의 보호 경로로도 활용할 수 있도록, 트래픽 흐름 집합에 속하는 각 트래픽 흐름들이 요구하는 자원을 경로 집합에 속하는 모든 경로들에 할당하여 예약한 후에, 고장인 경로를 식별하는 단계; 및 정상인 상태에서 고장인 상태로 바뀐 경로들에 할당되었던 예약 자원을 정상인 경로로 옮겨 할당함으로써 트래픽 흐름 집합의 QoS 조건을 만족시켜, 트래픽 흐름 집합을 QoS 조건이 만족되도록 전송하는 단계를 포함함.

公开(公告)号：KR1020020053519A

公开(公告)日：2002-07-05

申请号：KR1020000083177

申请日：2000-12-27

Applicant: 한국전자통신연구원 ,  주식회사 케이티

Inventor： 윤호선 ,  양선희 ,  정민영 ,  김숙연

IPC: H04L12/20

CPC classification number: H04L45/507 ,  H04L12/4641 ,  H04L63/0272

Abstract: PURPOSE: A VPN(Virtual Private Network) service system in an MPLS(Multi Protocol Label Switching) network and a method thereof are provided so that an edge router can efficiently provide VPN service on the basis of a communication network like an MPLS network. CONSTITUTION: An operator command processing block(11) receives and processes an operator's command. A VPN_CFB(12) is a VPN service function block that provides VPN service actually. A BGP4_FB(13) indicates a BGP4+ function block to transfer generic routing information and VPN routing information. An SIG_FB(14), a function block to set up an LSP(Label Switched Path), executes a CR-LDP protocol or an RSVP-TE protocol. An IBM_FB(15) creates a resource management and forwarding entry and transfers the created entry to a VFE_FB(16). The VFE_FB(16) indicates a forwarding engine. A VPN_comp_t(17) is a table to manage information related to VPN groups and sites. A VPN_rib_t(18) is a table to manage VPN routing information. A VPN_path_t(19) is a table to manage paths by the site. An LSP_table(20) stores LSP associated information. A VPN_ft(21) indicates a VPN forwarding table.

Abstract translation: 目的：提供MPLS（Multi Protocol Label Switching，多协议标签交换）网络中的VPN（Virtual Private Network，虚拟专用网）业务系统及其方法，使得边缘路由器可以在诸如MPLS网络的通信网络的基础上高效地提供VPN业务。 构成：操作员指令处理块（11）接收并处理操作者的命令。 VPN_CFB（12）是实际提供VPN业务的VPN业务功能块。 BGP4_FB（13）表示传输通用路由信息和VPN路由信息的BGP4 +功能块。 SIG_FB（14）是建立LSP（标签交换路径）的功能块，执行CR-LDP协议或RSVP-TE协议。 IBM_FB（15）创建资源管理和转发条目，并将创建的条目传输到VFE_FB（16）。 VFE_FB（16）表示转发引擎。 VPN_comp_t（17）是管理与VPN组和站点有关的信息的表。 VPN_rib_t（18）是一个管理VPN路由信息的表格。 VPN_path_t（19）是用于管理站点路径的表。 LSP_table（20）存储LSP相关信息。 VPN_ft（21）表示VPN转发表。

公开(公告)号：KR100776789B1

公开(公告)日：2007-11-19

申请号：KR1020060112892

申请日：2006-11-15

Applicant: 한국전자통신연구원

Inventor： 양미정 ,  김태일 ,  정해원 ,  정명희 ,  최승혁 ,  정민영 ,  박재형

IPC: H04B7/26 ,  G06Q30/02 ,  G06F15/16 ,  H04L12/24

CPC classification number: H04W40/12 ,  H04L41/00 ,  H04W40/24

Abstract: A link information management method of OSPF(Open Shortest Path First)-TE(Traffic Engineering) and a device are provided to generate relatively small link state advertising messages through available bandwidth change information of accumulated links and changed available bandwidth information of links, thereby preventing an increase in traffic within a network. TE link information set by an operator is inputted(201). A neighbor with a routing peer of neighboring nodes is formed by using link management information initialized on the basis of the TE link information(202,203). The TE link information is exchanged with the neighboring nodes to configure a TE-LSDB(Link State DB) for providing TE paths(204). If a TE link resource information change event occurs, the link management information is updated(205,206). It is decided whether to advertise the TE link resource information change based on statistical information of available link bandwidths, and the routing peer of the neighboring nodes is notified with the change of the TE link resource information(207,208). The TE-LSDB is reconfigured according to the changed TE link resource information(209).

Abstract translation: 提供了OSPF（开放最短路径优先）-TE（流量工程）和设备的链路信息管理方法，用于通过可用的链路的带宽变化信息生成相对较小的链路状态广告消息，并改变链路的可用带宽信息，从而防止 网络中流量的增加。 输入由操作者设定的TE链接信息（201）。 通过使用基于TE链路信息（202,203）初始化的链路管理信息来形成具有相邻节点的路由对等体的邻居。 TE链路信息与相邻节点交换，配置TE-LSDB（Link State DB），用于提供TE路径（204）。 如果发生TE链路资源信息改变事件，链路管理信息被更新（205,206）。 决定是否根据可用链路带宽的统计信息通告TE链路资源信息变化，通过TE链路资源信息的改变来通知相邻节点的路由对等体（207,208）。 TE-LSDB根据改变的TE链路资源信息重新配置（209）。