Title of Invention	METHOD FOR ACCESSING INTEGRATED SERVICE BY AN ACCESS NETWORK
Abstract	The present invention relates to a method for accessing integrated services by an access network, including: establishing at least one Multi-Protocol Label Switching (MPLS) Label Switch Path (LSP) in a broadband access network; MPLS encapsulating at a broadband access node a service to be transmitted, and transmitting the encapsulated service in the broadband access network via the LSP. As MPLS is a technology for providing multi-layered connection, in the embodiments of the present invention, MPLS is utilized as the means for networking of broadband access network and private line passing through the core network, so that the problem faced by the broadband access network serving as the bearer network of integrated services can be solved, the integrated services of video, voice, data and private line(including IP private line, and traditional private lines of ATM (Asynchronous Transfer Mode), TDM (Time Division Multiplexing) and FR (Frame Relay)) can be carried over the broadband access network at a lower cost.

Title of Invention

METHOD FOR ACCESSING INTEGRATED SERVICE BY AN ACCESS NETWORK

Abstract

The present invention relates to a method for accessing integrated services by an access network, including: establishing at least one Multi-Protocol Label Switching (MPLS) Label Switch Path (LSP) in a broadband access network; MPLS encapsulating at a broadband access node a service to be transmitted, and transmitting the encapsulated service in the broadband access network via the LSP. As MPLS is a technology for providing multi-layered connection, in the embodiments of the present invention, MPLS is utilized as the means for networking of broadband access network and private line passing through the core network, so that the problem faced by the broadband access network serving as the bearer network of integrated services can be solved, the integrated services of video, voice, data and private line(including IP private line, and traditional private lines of ATM (Asynchronous Transfer Mode), TDM (Time Division Multiplexing) and FR (Frame Relay)) can be carried over the broadband access network at a lower cost.

Full Text	Field of the Invention The present invention relates to the field of network communication technologies, and particularly to a method for accessing integrated services by an access network. Background of the Invention Network providers have always been pursuing the implementation methods of carrying services of voice, video, data, private line, etc., over the same network at the same time so as to increase the utilization of network and reduce the costs of investment and operation. An IP (Internet Protocol) network has the advantages of high efficiency of bandwidth utilization, low price, and high bandwidth, and is open and easy for providing flexible services. With the mass deployment of broadband access network, the IP network gradually becomes the practical platform for carrying integrated services. Providers wish to access voice, video and enterprise private line and the like, besides the originally carried data services, via the broadband access nodes. In addition to IP-based voice and enterprise private line which are newly deployed on the broadband network, some providers also wish to carry the traditional services, such as voice or private lines based on TDM (Time Division Multiplexing), ATM (Asynchronous Transfer Mode), or FR (Frame Relay), via the broadband access network. The existing IP network is mainly designed and constructed for the transmission of data services. The IP network is connectionless and can meet the requirement of the transmission of data services which need only reachability, but can not meet the requirements of reliable transmission (packet loss ratio), time characteristics(order, delay and delay variation) Dsecurity(isolation between various services, privacy of private network services) and survivability(the recovery time of backup link after failure) for transmission of voice, video and private line interconnection services. In order to meet the requirement of QoS (quality of service) required by the transmission of voice, video and private line interconnection services, it is necessary for a bearer network to provide connection-oriented service for these services, and protect these services on the basis of connection. To this end, the IP network needs to provide a connection-oriented bearer service. The MPLS (Multi-Protocol Label Switching) technology is a new technology for providing connection over the IP network. The MPLS technology has been applied to the broadband core network and can provide connection-oriented bearer service. At present, however, the MPLS technology is applied only for the broadband core network; while for an access network, different access approaches are employed to implement the access of different types of services over the access network. At present, these different types of services are generally transmitted over different independent access networks established respectively for the services of voice, video, data and private lin. The voice service is accessed by PSTN (Public Switched Telephone Network, the ordinary telephone network using TDM technology). SDH (Synchronous Digital Hierarchy) system is used for the networking of the accessing network; The data service is accessed by broadband. Direct fiber connection, Ethernet ring (using RSTP (Rapid Spanning Tree Protocol) for protection) or RPR(Resilient Packet Ring) are employed for the networking of the accessing network;; The video service is accessed by CATV(Cable Television) system. CATV system is employed for the networking of the accessing network;; The private line service is accessed by DDN (Digital Data Network, which is a private line technology based on TDM), ATM or FR network. SDH system is employed for the networking of the accessing network. It can be seen that, presently, if a variety of services are to be transmitted, investment for the corresponding network establishment are needed respectively, and the individual networks need to be maintained independently, which increases the provider's CAPEX (Capital Expenditures) and OPEX (Operating Expenditures). At present, an access network may use the MSTP (Multi-service Transport Platform) system based on SDH, to carry the data, voice and private line altogether. However, MSTP is a technology based on SDH, with low availability of bandwidth, poor flexibility of services, complex administration and maintenance, and high costs for networking and maintenance. Another way for implementing integrated service access network is based on ATM, that is, data, voice and private line can be carried together over the ATM network. However, with the enlargement of IP network, more and more applications have been transplanted to IP platform to make use of its advantages. The ATM network is not the actual development direction of network. It has low efficiency, high cost, and poor expandability, and is difficult to implement whole-network and whole-path coverage when carrying the IP services. ATM is also difficult to implement video multicast services. Furthermore, when ATM network is applied, two kinds of networks based on ATM protocol and IP protocol need to be maintained, thus complicating the administration and maintenance. So, the ATM network can not serve as the integrated service bearer platform for all services covering whole network and whole path in the aspects of bandwidth, scale and tendency, etc. Presently, the access network based on Ethernet protocol is also employed. That is, the Ethernet technology is applied between a broadband access node and a broadband access server to carry the services of data, voice, video and private line together. With the Ethernet access network, the access of pure data services can be implemented simply and efficiently at a low cost. However, there are still some issues in the bearing of integrated services. 1. survivability: the Ethernet access network of tree networking can not meet the requirements for service protection. The protection switching time of an Ethernet ring network constructed with RSTP protocol is at the level of seconds, which can not meet the switching time requirements for high quality services. 2. QoS (Quality of Service) and security: the Ethernet forwarding based on MAC(Media Access Control) address learning can not meet the security requirements for the bearing of integrated services, because its forwarding direction is unpredictable, and traffic cannot be isolated completely, and can not meet the QoS requirements for the bearing of integrated services, because bandwidth is hard to be planned. 3. A whole-network and whole-path connection across BAS (Broadband Access Server) or edge router of core network, which ensures a better quality of service for voice and private line, can not be established. 4. The traditional services of TDM, ATM, FR can not be carried, there is no standard method for carrying TDM, ATM, FR across the BAS or the edge router of core network on Ethernet. Therefore, at present, there is no unified access network which can access all the existing services, such as voice, data, video and private line, etc., at the same time with a low cost. Summary of the Invention In view of the issues described above, an embodiment of the present invention provides a method for accessing integrated services by an access network, thereby integrated services can be transmitted in broadband access network at a low cost. A method for accessing integrated services by an access network, including: establishing at least one Multi-Protocol Label Switching (MPLS) Label Switch Path (LSP) in broadband access network; at a broadband access node, MPLS encapsulating a service to be transmitted and transmitting the encapsulated service in the broadband access network via the LSP. Preferably, the process of establishing at least one LSP includes: establishing at least one LSP between a broadband access node device and a Broadband Access Server (BAS) in the broadband access network, in which the LSP terminates at the BAS or passes through the BAS; Preferably, in the process of establishing at least one LSP, the LSP may be established according to different service types and/or quality of service (QoS) provided. Preferably, in the method, the bandwidth of the access network may be planned and managed on the basis of LSP, and different QoS may be provided and guaranteed for different types of services by use of different LSPs. Preferably, the established LSP contains one layer of label, the LSP terminates at a Broadband Access Server (BAS) or an edge router of a core network, and the access network distinguishes the QoS of different types of services on the basis of the label. Preferably, the established LSP contains two layers of label including an outer layer label and an inner layer label, wherein when two layers of a two-layer LSP identified respectively by the two layers of label terminate at a Broadband Server (BAS) or an edge router of a core network simultaneously, the access network distinguishes the QoS of different types of services on the basis of the outer layer label, the broadband access node in the access network, and the BAS or the edge router of the core network identify different services and subscribers on the basis of inner layer label; or, the inner layer label does not provide extra information, and the information of different services and subscribers is provided by payload carried over MPLS. Preferably, the established LSP contains two layers of label including an outer layer label and an inner layer label, wherein the outer layer LSP identified by the outer layer label terminate at a Broadband Server (BAS) or an edge router of a core network, and the inner layer LSP identified by the inner layer label passes across the BAS or the edge router of the core network and terminates at an MPLS Provider Edge (PE) of the opposite side, the access network distinguishes the QoS of different types of services on the basis of the outer layer label; the broadband access node in the access network, the BAS or the edge router of the core network, and the MPLS PE of the opposite side identify different services and subscribers on the basis of the inner layer label; after the inner layer LSP passes through the BAS or the edge router of the core network, it is encapsulated in another outer layer LSP. Preferably, the established LSP contains two layers of label including an outer layer label and an inner layer label; wherein when two layers of the two-layer LSP identified respectively by the two layers of label pass through the BAS or the edge router of the core network simultaneously and terminates at an MPLS Provider Edge (PE) of the opposite side, the outer layer label provides a tunnel to the MPLS PE of the opposite side, the access network and the core network distinguish the QoS of different types of services on the basis of the outer layer label; broadband access node and the MPLS PE at two ends of the inner layer LSP identify different services and subscribers on the basis of the inner layer label.. Preferably, in the process of MPLS encapsulating a service to be transmitted and transmitting the encapsulated service, the encapsulation formats for MPLS encapsulation at the broadband access node include the encapsulation formats of Ethernet, Asynchronous Transfer Mode (ATM), Time Division Multiplexing (TDM), Frame Relay (FR) services carried over MPLS; before encapsulating, the method further includes: when a service is needed to be transmitted by the broadband access network, determining a corresponding LSP for the service at a broadband access node which is the source end of the service. Preferably, in the process of MPLS encapsulating a service to be transmitted and transmitting the encapsulated service, the LSP for transmission of IP multicast video over the access network is an MPLS multicast LSP. Preferably, the process of determining a corresponding LSP for the service at a broadband access node which is the source end of the service includes selecting a corresponding LSP for the service according to the type of the service and/or QoS required by the service. Preferably, the process of determining a corresponding LSP for the service at a broadband access node which is the source end of the service also includes selecting a corresponding LSP for a service according to the subscribers initiating the service. Preferably, the type of the service include: voice service, data service, video service and private line service. Preferably, the process of determining a corresponding LSP for the service at a broadband access node which is the source end of the service includes: when Ethernet is carried over MPLS, if a one-layer LSP or a two-layer LSP terminate at a Broadband Access Server (BAS) or an edge router of a core network, selecting a corresponding LSP for the service according to a service tag (S-TAG) of Ethernet packet of the service. Preferably, the process of determining a corresponding LSP for the service at a broadband access node which is the source end of the service includes: when Ethernet is carried over MPLS, if two layers of label are used and at least the inner layer LSP of the two-layer LSP terminates at an MPLS Provider Edge (PE) of the opposite side across a Broadband Access Server (BAS) or an edge router of the core network, selecting a corresponding LSP for the service according to a service tag (S-TAG) of Ethernet packet of the service. Preferably, in the process of determining a corresponding LSP for the service at a broadband access node which is the source end of the service, when services of ATM, TDM or FR are carried over MPLS, two layers of label are used, in which at least the inner layer LSP of two-layer LSP terminates at the MPLS Provider Edge (PE) of the opposite side across a Broadband Access Server (BAS) or an edge router of a core network; corresponding LSPs are selected for respective services according to Virtual Channel Connection (VCC) of ATM packet, Data Link Connection Identifier (DLCI) of FR packet, port of TDM and timestots in TDM frame of the services; the inner layer LSP terminates at the MPLS PE of the opposite side, implementing private line connection of ATM, TDM or FR. Preferably, the method also includes: detecting connectivity state of an LSP by using MPLS Operation And Maintenance (OAM); protecting an LSP which needs to be protected by using MPLS protection switching. Preferably, the process of protecting an LSP which needs to be protected by using MPLS protection switching includes: establishing in advance a standby LSP for the LSP which needs to be protected; switching a service to the standby LSP for transmission, when the LSP used by the service is detected to be failed by the function of MPLS OAM. Preferably, the method also includes: implementing network failure location, and performance monitoring of access network by use of the function of MPLS Operation And Maintenance (OAM). Preferably, the method also includes: the broadband access node recovering Ethernet frame, ATM packet, FR packet, TDM frame from MPLS messages arriving at the broadband access node, then performing subsequent processing. Preferably, the process of the broadband access node recovering Ethernet frame, ATM packet, FR packet, TDM frame from MPLS messages arriving the broadband access node and performing subsequent processing also includes: for a downstream MPLS packet which is transmitted over Ethernet private line across Broadband Access Server or edge router of core network, the broadband access node replacing the outer layer S-TAG of Ethernet after de-encapsulating the MPLS packet, so S-TAG can be allocated independently in access networks at both ends. Preferably, the method also includes: if broadband Access Server (BAS) or edge router of core network is the end point of an LSP entering at a downlink port connected to the broadband access network, identifying services and subscribers by 802.1Q, or 802.IAD, or pseudo-wire label of PWE3 in the MPLS PWE3 packet; BAS or edge router extracting Ethernet packet from the LSP after terminating the LSP; in which if BAS or edge router of core network is an intermediate node of an LSP entering at the downlink port connected to the broadband access network,, which is the case for private line across the BAS or the edge router of the core network, the BAS or the edge router should support MPLS exchange of the MPLS packets in the LSP. It can be seen that, in the embodiments, MPLS is introduced into the broadband access network so as to implement the carrying of integrated services which deploy IP protocol, such as video, voice, data and private line and the like. Also, the traditional private line of ATM, DD"N and FR can be carried over the broadband access network. Thus the bandwidth utilization and profit of broad network can be increased, and the CAPEX and OPEX can be reduced compared with the solutions deploying several independent access networks. With the method according to the embodiments of the present invention, the access network can be conveniently unified in technology with the core network of MPLS , thus reducing the cost for operating administrating and maintaining the network. In addition, with the function of MPLS OAM and protection switching, fault location and performance monitoring are implemented in the network to decrease the maintenance cost, and the protection switch time of broadband access network can be shortened to the telecommunication level of 50 ms. Since the IP core network has been MPLSized substantially, the introduction of MPLS into access network will enable the MPLS connection covering whole network and whole path in the IP network. Besides the solution to the problem of broadband access network, the MPLSized access network also gives a route map for the aggregation of data, voice, video of three networks,. Brief Description of the Drawings Fig.l is a schematic diagram of a protocol stack applied in a broadband access network device; Fig.2 is a schematic diagram of access layer MPLS networking of IP services; Fig.3 is a schematic diagram of access layer MPLS networking of integrated services; Fig.4 is a schematic diagram of entire network MPLS connection networking of IP services; Fig.5 is a schematic diagram of entire network MPLS connection networking of integrated services. Detailed Description of the Embodiments The essence of the present invention lies in the service transmission by use of the MPLS technology over a broadband access network, by which the integrated services of IPized video, voice, data and private line (including IP private line and traditional private lines of ATM, TDM and FR) and the like, can be carried over the access network at a lower cost. MPLS is a technology which can provide multi-layer connection. The problem faced by the broadband access network serving as the bearer network of integrated services can be solved, by taking the MPLS as the means of networking of the broadband access network and private line transmission across the core network. MPLS OAM and protection switching are applied in the embodiments of the present invention, so that the protection switching time of the broadband access network can meet the requirements of the telecommunication level service. In addition, because of the label-based forwarding of MPLS, the forwarding path of an MPLS packet (including the forwarding path of a packet after protection switching) can be predicted,. Thus, the isolation between service traffics can be achieved and the privacy of information can be ensured. Furthermore, MPLS makes bandwidth planning possible, thereby the QoS of different services are guaranteed. In order to implement the method according to an embodiment of the present invention, it is necessary for a broadband access node in a broadband access network to: 1. Support MPLS forwarding at the uplink port. That is, it is needed to add an uplink MPLS interface. 2. Support PWE3 (Pseudo Wire Emulation Edge-to-Edge) forwarding at the uplink port. If the carrying of the IPized MPLS integrated services needs to be supported, Ethernet PWE3 is needed. If the carrying of private line of ATM, TDM, and FR needs to be supported, MPLS PWE3 of ATM, TDM, and FR are needed correspondingly. It is necessary to be conformed with the related standards presented by the PWE3 workgroup of IETF (Internet Engineering Task Force). The broadband access node described in this context includes broadband access devices at central office (CO) and remote(RT) location under the CO, such as digital subscriber line access multiplexer (DSLAM), Ethernet access device, etc.. The protocol stack of an MPLS interface of the broadband access node is shown in Fig.l, wherein: Fig. la shows that the Ethernet frame of an IPized service is carried over MPLS via PWE3, so that integrated services, such as data, VoIP(Voice over IP), and IPTV(IP Television), can be carried between the broadband access node and the broadband access server(BAS) or the edge router of core network; or an end-to-end MPLS connection to MPLS PE(Provider Edge) of the opposite side across the BAS or the edge router of core network, such as Ethernet private line service, can be implemented. Fig. lb shows that the traditional services of ATM, TDM, and FR are carried over MPLS via PWE3, and an end-to-end private line of ATM, TDM, and FR is formed to the PE device of the opposite side across the BAS or the edge router of core network, getting rid of the independent networks of ATM, DDN, and FR. The detailed implementation procedure of the method according to the embodiment of the present invention is as follows: Step 21: multiple LSPs (Label Switching Path) in a broadband access network are established. An LSP may terminate at a BAS device, or pass through the BAS device. That is, the established LSP can be the LSP only between the DSLAM (Digital Subscriber Line Access Multiplexer) and the BAS in the broadband access network, or can be the LSP starting from the DSLAM via the BAS device across the core network. The LSP includes one layer of Label or two layers of label. 1. When the LSP uses one layer of label, the LSP may terminate at the BAS or the edge router of core network. The label is used by the access network to distinguish the quality of service (QoS) of different types of services. 2. When the LSP uses two layers of label which include an outer layer label and an inner layer label, there may be a variety of application scenarios in practice. (21). When the two layers of a two-layer LSP identified respectively by the two layers of label terminate at the BAS or the edge router of core network simultaneously, the outer layer label is used by the access network to distinguish the QoS of different types of services, and the inner layer label is used by the broadband access node in the access network and BAS/the edge router of core network to identify different services and subscribers. Optionally, the inner layer label does not provide extra information, and different services and subscribers are identified by the payload carried in the inner layer LSP; (22). When the outer layer LSP identified by the outer layer label terminates at the BAS or the edge router of core network, and the inner layer LSP identified by the inner layer label passes across the BAS or the edge router of core network and terminates at the provider edge router MPLS PE of the opposite side, the outer layer label is used by the access network to distinguish the QoS of different types of services, and the inner layer label is used by the broadband access node in the access network, the BAS or the edge router of core network, and the MPLS PE of the opposite side to identify different services and subscribers. The inner layer LSP needs to be encapsulated in another outer layer LSP when it enters the core network across the BAS or the edge router of core network (if the opposite side of communication is in this access network, the inner layer LSP returns to this access network); (23). When the two layers of the two-layer LSP identified respectively by the two layers of label pass across the BAS or the edge router of core network simultaneously and terminate at the MPLS PE of the opposite side, the outer layer label provides a tunnel to the MPLS PE of the opposite side, and is used by the access network and the core network to distinguish the QoS of different types of services, and the inner layer label is used by the broadband access nodes and the MPLS PE to identify different services and subscribers. The service transmission on the basis of various LSPs described above is handled in the following manners: When an Ethernet is carried over MPLS, if the LSP established in the manner described in 1, (21) is used, a corresponding LSP can be selected for a service according to the service tag (S-TAG) in the Ethernet packet of the service. At this time, the one-layer LSP or the two-layer LSP terminates at the BAS or the edge router of core network, the LSP which is mapped from the S-TAG can be used to distinguish the QoS of different types of services in the access network. The S-TAG is the outer layer VLAN TAG of the two layers of Virtual Local Area Network Tag (VLAN TAG) defined in IEEE 802.1 AD. The S-TAG mentioned in the embodiments of the present invention also includes the one layer VLAN TAG defined in IEEE 802.1Q; When an Ethernet is carried over MPLS, if the LSP established in the manner described in (22), (23) is employed, a corresponding LSP can be selected for a service according to the service tag (S-TAG) in the Ethernet packet of the service. At this time, the LSP which is mapped from the S-TAG can be used in the access network to distinguish the QoS of different types of services. The outer layer LSP established in the manner described in (23) can also provide a tunnel through the core network to the MPLS PE of the opposite side. After the inner layer LSP which is mapped from the S-TAG terminates at the MPLS PE of the opposite side, the MPLS PE of the opposite side can distinguish different subscribers and services on the basis of the inner layer label, thus an Ethernet private line connection can be implemented between the broadband access node and the PE of the opposite side; When services of ATM, FR or TDM are carried over MPLS, the LSP established in the manner described in (22), (23) can be used to select the corresponding LSPs for the respective services according to the virtual channel connection (VCC) of an ATM packet, the data link connection identifier (DLCI) of a FR packet, and the port of TDM and the timeslots in a TDM frame. At this time, the outer layer LSP is used by the access network to distinguish the QoS of different types of services, the outer layer LSP established in the manner described in (23) can also provide a tunnel to the MPLS PE of the opposite side through the core network. After the inner layer LSP terminates at the MPLS PE of the opposite side, the MPLS PE of the opposite side can distinguish different subscribers and services on the basis of the inner layer label, thus the private line connection of ATM, FR or TDM can be implemented between the broadband access node and the PE of the opposite side; Step 22: when the broadband access node needs to transmit a service upward, a corresponding LSP is selected at the source broadband access node of the service according to the type of the service, the quality of service (QoS) and the subscriber information of the service, and then the service is transmitted over the corresponding LSP. In step 22, the processing procedure of the broadband access node includes the process in the case that the Ethernet frame is carried over MPLS and the process in the case that the ATM/TDM/FR is carried over MPLS, which are described in detail respectively as follows: 1. when the Ethernet frame is carried over MPLS, if an MPLS packet in the broadband access network terminates at the BAS, or terminates at the access aggregation device prior to the BAS, the payload carried in the MPLS packet will identify the services and subscribers by way of 802.1Q or 802.IAD of Ethernet. After the MPLS connection is terminated, the BAS or the device terminating the MPLS packet prior to BAS extracts the payload carried in the MPLS packet directly and identifies the services and subscribers by way of 802.1Q or 802.1 AD. At this time, MPLS is only the networking means for carrying the integrated services of the access aggregation layer. Thus, the VLAN plan can be kept the same as of the old network, and the processing of the broadband access node and the BAS can be simplified. A pseudo-wire is mapped on the basis of the S-TAG for an upstream service on the broadband access node. The S-TAG and pseudo-wire are mapped to each other in the one-to-one map. That is, a corresponding LSP is selected for the upstream service according to the S-TAG of the service traffic, and each S-TAG corresponds to one LSP. After the downstream pseudo-wire sent down to the broadband access node is terminated by the broadband access node, the Ethernet frame is recovered from the MPLS packet. The subsequent processing procedure by the broadband access node is the same as the procedure without MPLS transmission. For the Ethernet private line across the BAS or the edge router of core network, since the S-TAGs are allocated locally in the access networks, the S-TAGs allocated in the access networks at the two ends of the same private line may be different. Therefore, after the LSP is terminated, it may be needed to replace the outer layer S-TAG, that is, the outer layer S-TAG may be replaced by the actual S-TAG value of the local end access network. In an embodiment of the present invention, the MPLS label can also be used as the means of access network for identifying the subscribers and services. At this time, the outer layer label of Ethernet PWE3 is used by access layer networking to distinguish the QoS of different type services. The inner layer label is used by the BAS to identify different services and subscribers. At this time, the BAS is needed to be changed correspondingly so as to identify different services and subscribers on the basis of the inner layer label. 2. when the ATM/TDM/FR is carried over MPLS if the ATM/TDM/FR is carried over MPLS, generally, one pseudo-wire starting from the broadband access node and passing across the BAS or the edge router of core network is needed to be established. That is, a private line across the BAS or the edge router of core network is emulated by carrying the traditional services of ATM, TDM and FR over MPLS via PWE3. For an ATM service, it is necessary for the broadband access node to support the ATM exchange, that is, the VPI (Virtual Path Identifier)/VCI (Virtual Circuit Identifier) in the ingress/egress ATM cell is replaced, and the VCC is used to select the pseudo-wire LSP carried in the outer layer LSP of MPLS. The VCC and the pseudo-wire are mapped to each other one to one, that is, one VCC corresponds to one LSP. For a TDM service, the timeslots in TDM frame and the port are used to select the pseudo-wire LSP. For a FR service, DLCI is used to select the pseudo-wire LSP. In step 22, in addition to the support of MPLS at the uplink port, BAS or edge router of core network also needs to: 1. support MPLS forwarding at the downlink port, and 2. support the related standards of IETF PWE3 workgroup, if the BAS or the edge router of core network is the end point of the corresponding pseudo-wire LSP of Ethernet, ATM, TDM and FR. If the BAS or the edge router of core network is the end point of LSP entering at the downlink port connected to the broadband access network, 802.1Q or 802.1 AD in a PWE3 packet can still be employed to identify the services and subscribers. At this time, after terminating the LSP, the BAS extracts the Ethernet packet from the LSP. Then the processing is the same as the method before the MPLS was introduced. Furthermore, the pseudo-wire labels of PWE3 packets can also be used as identities of the services and subscribers. When the BAS or the edge router of core network serves as the intermediate node of LSP entering at the downlink port connected to the broadband access network, for the private line across the BAS or the edge router of core network, the BAS or the edge router of core network is required to support MPLS exchange of MPLS packets entering at the downlink port. The broadband access node, the BAS or the edge router of core network according to the embodiments of the present invention are required to the support MPLS OAM and protection switching, particularly, they are required to: (1) support the OAM described in ITU-T Y.1711, so as to acquire the connectivity state of each LSP and to know in real time whether each LSP can transmit the services normally; (2) support the MPLS protection switching described in ITU-T Y.1720. Thus, when a failure is detected by the OAM function showing that the LSP is disconnected, the services carried and transmitted by the LSP are switched onto a standby LSP. A particular application in an actual network of the present invention will be described below with reference to the accompanying drawings. The access layer networking application, as well as the MPLS connection application over whole network and whole path will be described respectively hereinafter. The access layer networking application includes the access layer networking of the IPized services and the access layer networking of the integrated services, as shown respectively in Fig.2 and Fig.3. The thick solid lines in the figures represent the LSPs of MPLS. CO represents the physical location of a central office; RT represents the physical location of a remote access device below the central offices; The broadband access device (such as DSLAM, Ethernet switch, etc.), the PSTN access device, the ATM access device or the FR access device or the like, can be deployed at the COs and RTs according to the different types of the access services. RG represents the residence gateway of a broadband home subscriber. As shown in Fig.2, the IP services are transmitted via the MPLS networking, while the traditional services of ATM, TDM and FR and the like are still accessed to the central offices via other access networks and are then accessed to different core networks. In Fig.3, the IP services are transmitted via the MPLS networking, the services of ATM, TDM and FR and the like are also carried over MPLS by PWE3; thus, only the MPLS access network is reserved in the broadband access network, and the integrated services can be carried in the MPLS access network; that is, the IP services and the services of ATM, TDM and FR are all transmitted in the broadband access network via LSP. In Fig.2, as described above, for the IP services, besides that the MPLS is used for access layer networking, the MPLS label can also be used by the BAS to identify the subscribers and services. MPLS label, instead of VLAN label, can overcome the insufficiencies of space in identifying subscribers and services, and make the service management more flexible. The entire network MPLS connection can be classified into the entire network MPLS connection of IP service and the entire network MPLS connection of integrated service, as shown respectively in Fig.4 and Fig.5. Similarly, the thick solid lines in the figures represent LSPs based on MPLS protocol. As shown in Fig.4, for services based on session, VPN (Virtual private network), private line or other service passing across the tunnel of the BAS or the edge router of core network among all the IP services, an Ethernet pseudo-wire passing across the BAS or the edge router of core network may be established to the PE (Provider Edge) of the opposite side through the core network to implement the connection covering whole network and whole path. Accordingly, the QoS, security and survivability can be better ensured by taking the advantages of MPLS. As shown in Fig.5, besides that the IP services can establish the PE-to-PE whole-path connection, the statically configured private lines of TDM, ATM and FR can also migrate in whole path to the IP network via PWE3, getting rid of the DDN(Digital Data Network), ATM or FR networks. The PSTN core network is not needed for the call-based voice services until the voice service migrate to VoIP, i.e. until the TDM voice service is IPized in the NGN (Next Generation Network). At this time, the IP network can serve as the unified bearer platform over the entire network. While the present invention has been illustrated and described with reference to some preferred embodiments, the present invention is not limited to these. Those skilled in the art shall readily recognize that various variations and modifications can be made without departing from the protection scope of the present invention. These variations and modifications shall be encompassed within the protection scope of the present invention as defined by the accompanying claims. We Claim 1. A method for accessing integrated services by an access network□ characterized by comprising : establishing at least one Label Switch Path (LSP) of Multi-Protocol Label Switching (MPLS) in a broadband access network; MPLS encapsulating, at a broadband access node, a service to be transmitted, and transmitting the encapsulated service in the broadband access network via the at least one established LSP, wherein each of the at least one LSP contains two layers of label comprising an outer layer label for identifying quality of service, QoS, of different types of services and an inner layer label for identifying different services and subscribers. 2. The method as claimed in claim 1, wherein the process of establishing at least one LSP comprises: establishing at least one LSP between the broadband access node and a Broadband Access Server (BAS) in the broadband access network, in which the at least one LSP terminates at the BAS or passes through the BAS. 3. The method as claimed in claim 1, wherein in the process of establishing at least one LSP, the at least one LSP is established according to different service types and/or provided quality of service (QoS). 4. The method as claimed in claim 1, wherein the bandwidth of the access network is planned and managed on the basis of LSP, and different quality of service (QoS) is provided and guaranteed for different types of services by use of different LSPs. 5. The method as claimed in claim 1, wherein if two layers of a two-layer LSP identified respectively by the two layers of label terminate at a Broadband Server (BAS) or an edge router of a core network simultaneously, the broadband access network distinguishes the quality of service (QoS) of different types of services on the basis of the outer layer label, the broadband access node in the access network, and the BAS or and the edge router of the core network identify different services and subscribers on the basis of inner layer label; or, the inner layer label does not provide extra information, and the information of different services and subscribers is provided by pay load carried over MPLS. 6. The method as claimed in claim 1, wherein if the outer layer LSP identified by the outer layer label terminates at a Broadband Server (BAS) or an edge router of a core network, and the inner layer LSP identified by the inner layer label passes through the BAS or the edge router of the core network and terminates at an MPLS Provider Edge (PE) of the opposite side, the access network distinguishes the quality of service (QoS) of different types of services on the basis of the outer layer label; the broadband access node in the broadband access network, the BAS or the edge router of the core network, and the MPLS PE of the opposite side identify different services and subscribers on the basis of the inner layer label; the inner layer LSP is encapsulated in another outer layer LSP after it passes through the BAS or the edge router of the core network. 7. The method as claimed in claim 1, wherein if two layers of a two-layer LSP identified respectively by the two layers of label pass across the BAS or the edge router of the core network simultaneously and terminates at an MPLS Provider Edge (PE) of the opposite side, the outer layer label provides a tunnel to the MPLS PE of the opposite side, the access network and the core network distinguish the quality of service (QoS) of different types of services on the basis of the outer layer label; broadband access nodes and the MPLS PE at two ends of the inner layer LSP identify different services and subscribers on the basis of the inner layer label. 8. The method as claimed in any one of claims 1 to 7, wherein the encapsulation formats for MPLS encapsulation at the broadband access node comprises the encapsulation formats of Ethernet, Asynchronous Transfer Mode (ATM), Time Division Multiplexing (TDM), Frame Relay (FR) services carried over MPLS; and before encapsulating, the method comprises: when a service is needed to be transmitted by the broadband access network, determining a corresponding LSP for the service at a broadband access node which is the source end of the service. 9. The method as claimed in claim 8, wherein, in the process of MPLS encapsulating a service to be transmitted and transmitting the encapsulated service in the broadband access network via the LSP, the LSP for transmission of IP multicast video over the access network is an MPLS multicast LSP. 10. The method as claimed in claim 8, wherein the process of determining a corresponding LSP for the service at the broadband access node which is the source end of the service comprises: selecting a corresponding LSP for the service according to the type of the service and/or quality of service (QoS) required by the service. 11. The method as claimed in claim 10, wherein the process of determining a corresponding LSP for the service at a broadband access node which is the source end of the service comprises: selecting a corresponding LSP for a service according to the subscribers initiating the service. 12. The method as claimed in claim 10, wherein, the type of the service comprises: voice service, data service, video service and private line service. 13. The method as claimed in claim 8, wherein the process of determining a corresponding LSP for the service at a broadband access node which is the source end of the service comprises: when Ethernet is carried over MPLS, if a one-layer LSP or a two-layer LSP terminates at a Broadband Access Server (BAS) or an edge router of a core network, selecting a corresponding LSP for the service according to a service tag (S-TAG) of Ethernet packet of the service. 14. The method as claimed in claim 8, wherein the process of determining a corresponding LSP for the service at a broadband access node which is the source end of the service comprises: when Ethernet is carried over MPLS, if two layers of label are used and at least the inner layer LSP of the two-layer LSP terminates at an MPLS Provider Edge (PE) of the opposite side across a Broadband Access Server (BAS) or an edge router of core network, selecting a corresponding LSP for the service according to a service tag (S-TAG) of Ethernet packet of the service. 15. The method as claimed in claim 8, wherein in the process of determining a corresponding LSP for the service at a broadband access node which is the source end of the service, when services of ATM, TDM or FR are carried over MPLS, two layers of label are used, in which at least the inner layer LSP of two-layer LSP terminates at the MPLS Provider Edge (PE) of the opposite side across a Broadband Access Server (BAS) or an edge router of a core network; corresponding LSPs are selected for respective services according to Virtual Channel Connection (VCC) of ATM packet, Data Link Connection Identifier (DLCI) of FR packet, port of TDM and timeslots in TDM frame of the services; the inner layer LSP terminates at the MPLS PE of the opposite side, implementing private line connection of ATM, TDM or FR. 16. The method as claimed in any one of claims 1 to 7, comprising: detecting connectivity state of an LSP by using MPLS Operation And Maintenance (OAM); protecting an LSP which needs to be protected by using MPLS protection switching. 17. The method as claimed in claim 16, wherein the process of protecting an LSP which needs to be protected by using MPLS protection switching comprises: establishing in advance a standby LSP for the LSP which needs to be protected; switching a service to the standby LSP for transmission, when the LSP used by the service is detected to be failed by the function of MPLS OAM. 18. The method as claimed in any one of claims 1 to 7, comprising: implementing network failure location, and performance monitoring of the access network by use of the function of MPLS Operation And Maintenance (OAM). 19. The method as claimed in any one of claims 1 to 7, comprising: the broadband access node recovering Ethernet frame, ATM packet, FR packet, TDM frame from MPLS packets arriving at the broadband access node, then performing subsequent processing. 20. The method as claimed in claim 19, wherein the process of the broadband access node recovering Ethernet frame, ATM packet, FR packet, TDM frame from MPLS packets arriving at the broadband access node and then performing subsequent processing comprises: for a downstream MPLS packet which is transmitted over Ethernet private line across Broadband Access Server or edge router of core network, the broadband access node replacing the outer layer S-TAG of Ethernet after de-encapsulating the MPLS packet, so that S-TAG can be allocated independently in access networks at both ends. 21. The method as claimed in claim 19, comprising: if broadband Access Server (BAS) or edge router of core network is the end point of an LSP entering at a downlink port connected to the broadband access network, identifying services and subscribers by 802.1Q, or 802.1AD, or pseudo-wire label of PWE3 in the MPLS PWE3 packet; the BAS or edge router extracting Ethernet packet from the LSP after terminating the LSP; wherein if BAS or edge router of core network is an intermediate node of an LSP entering at the downlink port connected to the broadband access network,, which is the case for private line across the BAS or the edge router of the core network, the BAS or the edge router of the core network supports MPLS exchange of MPLS packets in the LSP. The present invention relates to a method for accessing integrated services by an access network, including: establishing at least one Multi-Protocol Label Switching (MPLS) Label Switch Path (LSP) in a broadband access network; MPLS encapsulating at a broadband access node a service to be transmitted, and transmitting the encapsulated service in the broadband access network via the LSP. As MPLS is a technology for providing multi-layered connection, in the embodiments of the present invention, MPLS is utilized as the means for networking of broadband access network and private line passing through the core network, so that the problem faced by the broadband access network serving as the bearer network of integrated services can be solved, the integrated services of video, voice, data and private line(including IP private line, and traditional private lines of ATM (Asynchronous Transfer Mode), TDM (Time Division Multiplexing) and FR (Frame Relay)) can be carried over the broadband access network at a lower cost.

Full Text

Field of the Invention
The present invention relates to the field of network communication technologies, and
particularly to a method for accessing integrated services by an access network.
Background of the Invention
Network providers have always been pursuing the implementation methods of
carrying services of voice, video, data, private line, etc., over the same network at the same
time so as to increase the utilization of network and reduce the costs of investment and
operation.
An IP (Internet Protocol) network has the advantages of high efficiency of bandwidth
utilization, low price, and high bandwidth, and is open and easy for providing flexible
services. With the mass deployment of broadband access network, the IP network gradually
becomes the practical platform for carrying integrated services. Providers wish to access
voice, video and enterprise private line and the like, besides the originally carried data
services, via the broadband access nodes. In addition to IP-based voice and enterprise
private line which are newly deployed on the broadband network, some providers also wish
to carry the traditional services, such as voice or private lines based on TDM (Time Division
Multiplexing), ATM (Asynchronous Transfer Mode), or FR (Frame Relay), via the
broadband access network.
The existing IP network is mainly designed and constructed for the transmission of
data services. The IP network is connectionless and can meet the requirement of the
transmission of data services which need only reachability, but can not meet the
requirements of reliable transmission (packet loss ratio), time characteristics(order, delay
and delay variation) Dsecurity(isolation between various services, privacy of private network
services) and survivability(the recovery time of backup link after failure) for transmission of
voice, video and private line interconnection services.
In order to meet the requirement of QoS (quality of service) required by the
transmission of voice, video and private line interconnection services, it is necessary for a

bearer network to provide connection-oriented service for these services, and protect these
services on the basis of connection. To this end, the IP network needs to provide a
connection-oriented bearer service.
The MPLS (Multi-Protocol Label Switching) technology is a new technology for
providing connection over the IP network. The MPLS technology has been applied to the
broadband core network and can provide connection-oriented bearer service. At present,
however, the MPLS technology is applied only for the broadband core network; while for an
access network, different access approaches are employed to implement the access of
different types of services over the access network.
At present, these different types of services are generally transmitted over different
independent access networks established respectively for the services of voice, video, data
and private lin.
The voice service is accessed by PSTN (Public Switched Telephone Network, the
ordinary telephone network using TDM technology). SDH (Synchronous Digital Hierarchy)
system is used for the networking of the accessing network;
The data service is accessed by broadband. Direct fiber connection, Ethernet ring
(using RSTP (Rapid Spanning Tree Protocol) for protection) or RPR(Resilient Packet Ring)
are employed for the networking of the accessing network;;
The video service is accessed by CATV(Cable Television) system. CATV system is
employed for the networking of the accessing network;;
The private line service is accessed by DDN (Digital Data Network, which is a private
line technology based on TDM), ATM or FR network. SDH system is employed for the
networking of the accessing network.
It can be seen that, presently, if a variety of services are to be transmitted, investment
for the corresponding network establishment are needed respectively, and the individual
networks need to be maintained independently, which increases the provider's CAPEX
(Capital Expenditures) and OPEX (Operating Expenditures).

At present, an access network may use the MSTP (Multi-service Transport Platform)
system based on SDH, to carry the data, voice and private line altogether.
However, MSTP is a technology based on SDH, with low availability of bandwidth,
poor flexibility of services, complex administration and maintenance, and high costs for
networking and maintenance.
Another way for implementing integrated service access network is based on ATM,
that is, data, voice and private line can be carried together over the ATM network.
However, with the enlargement of IP network, more and more applications have been
transplanted to IP platform to make use of its advantages. The ATM network is not the actual
development direction of network. It has low efficiency, high cost, and poor expandability,
and is difficult to implement whole-network and whole-path coverage when carrying the IP
services. ATM is also difficult to implement video multicast services. Furthermore, when
ATM network is applied, two kinds of networks based on ATM protocol and IP protocol
need to be maintained, thus complicating the administration and maintenance. So, the ATM
network can not serve as the integrated service bearer platform for all services covering
whole network and whole path in the aspects of bandwidth, scale and tendency, etc.
Presently, the access network based on Ethernet protocol is also employed. That is, the
Ethernet technology is applied between a broadband access node and a broadband access
server to carry the services of data, voice, video and private line together.
With the Ethernet access network, the access of pure data services can be implemented
simply and efficiently at a low cost. However, there are still some issues in the bearing of
integrated services.
1. survivability: the Ethernet access network of tree networking can not meet the
requirements for service protection. The protection switching time of an Ethernet ring
network constructed with RSTP protocol is at the level of seconds, which can not meet the
switching time requirements for high quality services.
2. QoS (Quality of Service) and security: the Ethernet forwarding based on

MAC(Media Access Control) address learning can not meet the security requirements for
the bearing of integrated services, because its forwarding direction is unpredictable, and
traffic cannot be isolated completely, and can not meet the QoS requirements for the bearing
of integrated services, because bandwidth is hard to be planned.
3. A whole-network and whole-path connection across BAS (Broadband Access
Server) or edge router of core network, which ensures a better quality of service for voice
and private line, can not be established.
4. The traditional services of TDM, ATM, FR can not be carried, there is no standard
method for carrying TDM, ATM, FR across the BAS or the edge router of core network on
Ethernet.
Therefore, at present, there is no unified access network which can access all the
existing services, such as voice, data, video and private line, etc., at the same time with a low
cost.
Summary of the Invention
In view of the issues described above, an embodiment of the present invention
provides a method for accessing integrated services by an access network, thereby
integrated services can be transmitted in broadband access network at a low cost.
A method for accessing integrated services by an access network, including:
establishing at least one Multi-Protocol Label Switching (MPLS) Label Switch Path
(LSP) in broadband access network;
at a broadband access node, MPLS encapsulating a service to be transmitted and
transmitting the encapsulated service in the broadband access network via the LSP.
Preferably, the process of establishing at least one LSP includes: establishing at least
one LSP between a broadband access node device and a Broadband Access Server (BAS)
in the broadband access network, in which the LSP terminates at the BAS or passes through
the BAS;

Preferably, in the process of establishing at least one LSP, the LSP may be established
according to different service types and/or quality of service (QoS) provided.
Preferably, in the method, the bandwidth of the access network may be planned and
managed on the basis of LSP, and different QoS may be provided and guaranteed for
different types of services by use of different LSPs.
Preferably, the established LSP contains one layer of label, the LSP terminates at a
Broadband Access Server (BAS) or an edge router of a core network, and the access
network distinguishes the QoS of different types of services on the basis of the label.
Preferably, the established LSP contains two layers of label including an outer layer
label and an inner layer label, wherein when two layers of a two-layer LSP identified
respectively by the two layers of label terminate at a Broadband Server (BAS) or an edge
router of a core network simultaneously, the access network distinguishes the QoS of
different types of services on the basis of the outer layer label, the broadband access node
in the access network, and the BAS or the edge router of the core network identify different
services and subscribers on the basis of inner layer label; or, the inner layer label does not
provide extra information, and the information of different services and subscribers is
provided by payload carried over MPLS.
Preferably, the established LSP contains two layers of label including an outer layer
label and an inner layer label, wherein the outer layer LSP identified by the outer layer
label terminate at a Broadband Server (BAS) or an edge router of a core network, and the
inner layer LSP identified by the inner layer label passes across the BAS or the edge router
of the core network and terminates at an MPLS Provider Edge (PE) of the opposite side,
the access network distinguishes the QoS of different types of services on the basis of the
outer layer label; the broadband access node in the access network, the BAS or the edge
router of the core network, and the MPLS PE of the opposite side identify different services
and subscribers on the basis of the inner layer label; after the inner layer LSP passes
through the BAS or the edge router of the core network, it is encapsulated in another outer
layer LSP.

Preferably, the established LSP contains two layers of label including an outer layer
label and an inner layer label; wherein when two layers of the two-layer LSP identified
respectively by the two layers of label pass through the BAS or the edge router of the core
network simultaneously and terminates at an MPLS Provider Edge (PE) of the opposite
side, the outer layer label provides a tunnel to the MPLS PE of the opposite side, the access
network and the core network distinguish the QoS of different types of services on the basis
of the outer layer label; broadband access node and the MPLS PE at two ends of the inner
layer LSP identify different services and subscribers on the basis of the inner layer label..
Preferably, in the process of MPLS encapsulating a service to be transmitted and
transmitting the encapsulated service, the encapsulation formats for MPLS encapsulation at
the broadband access node include the encapsulation formats of Ethernet, Asynchronous
Transfer Mode (ATM), Time Division Multiplexing (TDM), Frame Relay (FR) services
carried over MPLS; before encapsulating, the method further includes:
when a service is needed to be transmitted by the broadband access network,
determining a corresponding LSP for the service at a broadband access node which is the
source end of the service.
Preferably, in the process of MPLS encapsulating a service to be transmitted and
transmitting the encapsulated service, the LSP for transmission of IP multicast video over
the access network is an MPLS multicast LSP.
Preferably, the process of determining a corresponding LSP for the service at a
broadband access node which is the source end of the service includes selecting a
corresponding LSP for the service according to the type of the service and/or QoS required
by the service.
Preferably, the process of determining a corresponding LSP for the service at a
broadband access node which is the source end of the service also includes selecting a
corresponding LSP for a service according to the subscribers initiating the service.
Preferably, the type of the service include: voice service, data service, video service

and private line service.
Preferably, the process of determining a corresponding LSP for the service at a
broadband access node which is the source end of the service includes: when Ethernet is
carried over MPLS, if a one-layer LSP or a two-layer LSP terminate at a Broadband Access
Server (BAS) or an edge router of a core network, selecting a corresponding LSP for the
service according to a service tag (S-TAG) of Ethernet packet of the service.
Preferably, the process of determining a corresponding LSP for the service at a
broadband access node which is the source end of the service includes: when Ethernet is
carried over MPLS, if two layers of label are used and at least the inner layer LSP of the
two-layer LSP terminates at an MPLS Provider Edge (PE) of the opposite side across a
Broadband Access Server (BAS) or an edge router of the core network, selecting a
corresponding LSP for the service according to a service tag (S-TAG) of Ethernet packet of
the service.
Preferably, in the process of determining a corresponding LSP for the service at a
broadband access node which is the source end of the service, when services of ATM, TDM
or FR are carried over MPLS, two layers of label are used, in which at least the inner layer
LSP of two-layer LSP terminates at the MPLS Provider Edge (PE) of the opposite side
across a Broadband Access Server (BAS) or an edge router of a core network;
corresponding LSPs are selected for respective services according to Virtual Channel
Connection (VCC) of ATM packet, Data Link Connection Identifier (DLCI) of FR packet,
port of TDM and timestots in TDM frame of the services; the inner layer LSP terminates at
the MPLS PE of the opposite side, implementing private line connection of ATM, TDM or
FR.
Preferably, the method also includes:
detecting connectivity state of an LSP by using MPLS Operation And Maintenance
(OAM);
protecting an LSP which needs to be protected by using MPLS protection switching.

Preferably, the process of protecting an LSP which needs to be protected by using
MPLS protection switching includes: establishing in advance a standby LSP for the LSP
which needs to be protected; switching a service to the standby LSP for transmission, when
the LSP used by the service is detected to be failed by the function of MPLS OAM.
Preferably, the method also includes: implementing network failure location, and
performance monitoring of access network by use of the function of MPLS Operation And
Maintenance (OAM).
Preferably, the method also includes: the broadband access node recovering Ethernet
frame, ATM packet, FR packet, TDM frame from MPLS messages arriving at the
broadband access node, then performing subsequent processing.
Preferably, the process of the broadband access node recovering Ethernet frame, ATM
packet, FR packet, TDM frame from MPLS messages arriving the broadband access node
and performing subsequent processing also includes: for a downstream MPLS packet which
is transmitted over Ethernet private line across Broadband Access Server or edge router of
core network, the broadband access node replacing the outer layer S-TAG of Ethernet after
de-encapsulating the MPLS packet, so S-TAG can be allocated independently in access
networks at both ends.
Preferably, the method also includes: if broadband Access Server (BAS) or edge router
of core network is the end point of an LSP entering at a downlink port connected to the
broadband access network, identifying services and subscribers by 802.1Q, or 802.IAD, or
pseudo-wire label of PWE3 in the MPLS PWE3 packet; BAS or edge router extracting
Ethernet packet from the LSP after terminating the LSP; in which if BAS or edge router of
core network is an intermediate node of an LSP entering at the downlink port connected to
the broadband access network,, which is the case for private line across the BAS or the
edge router of the core network, the BAS or the edge router should support MPLS
exchange of the MPLS packets in the LSP.
It can be seen that, in the embodiments, MPLS is introduced into the broadband access
network so as to implement the carrying of integrated services which deploy IP protocol,

such as video, voice, data and private line and the like. Also, the traditional private line of
ATM, DD"N and FR can be carried over the broadband access network. Thus the bandwidth
utilization and profit of broad network can be increased, and the CAPEX and OPEX can be
reduced compared with the solutions deploying several independent access networks.
With the method according to the embodiments of the present invention, the access
network can be conveniently unified in technology with the core network of MPLS , thus
reducing the cost for operating administrating and maintaining the network. In addition,
with the function of MPLS OAM and protection switching, fault location and performance
monitoring are implemented in the network to decrease the maintenance cost, and the
protection switch time of broadband access network can be shortened to the
telecommunication level of 50 ms.
Since the IP core network has been MPLSized substantially, the introduction of MPLS
into access network will enable the MPLS connection covering whole network and whole
path in the IP network. Besides the solution to the problem of broadband access network, the
MPLSized access network also gives a route map for the aggregation of data, voice, video of
three networks,.
Brief Description of the Drawings
Fig.l is a schematic diagram of a protocol stack applied in a broadband access
network device;
Fig.2 is a schematic diagram of access layer MPLS networking of IP services;
Fig.3 is a schematic diagram of access layer MPLS networking of integrated services;
Fig.4 is a schematic diagram of entire network MPLS connection networking of IP
services;
Fig.5 is a schematic diagram of entire network MPLS connection networking of
integrated services.

Detailed Description of the Embodiments
The essence of the present invention lies in the service transmission by use of the
MPLS technology over a broadband access network, by which the integrated services of
IPized video, voice, data and private line (including IP private line and traditional private
lines of ATM, TDM and FR) and the like, can be carried over the access network at a lower
cost. MPLS is a technology which can provide multi-layer connection. The problem faced
by the broadband access network serving as the bearer network of integrated services can be
solved, by taking the MPLS as the means of networking of the broadband access network
and private line transmission across the core network.
MPLS OAM and protection switching are applied in the embodiments of the present
invention, so that the protection switching time of the broadband access network can meet
the requirements of the telecommunication level service.
In addition, because of the label-based forwarding of MPLS, the forwarding path of an
MPLS packet (including the forwarding path of a packet after protection switching) can be
predicted,. Thus, the isolation between service traffics can be achieved and the privacy of
information can be ensured. Furthermore, MPLS makes bandwidth planning possible,
thereby the QoS of different services are guaranteed.
In order to implement the method according to an embodiment of the present
invention, it is necessary for a broadband access node in a broadband access network to:
1. Support MPLS forwarding at the uplink port. That is, it is needed to add an uplink
MPLS interface.
2. Support PWE3 (Pseudo Wire Emulation Edge-to-Edge) forwarding at the uplink
port. If the carrying of the IPized MPLS integrated services needs to be supported, Ethernet
PWE3 is needed. If the carrying of private line of ATM, TDM, and FR needs to be supported,
MPLS PWE3 of ATM, TDM, and FR are needed correspondingly. It is necessary to be
conformed with the related standards presented by the PWE3 workgroup of IETF (Internet
Engineering Task Force).

The broadband access node described in this context includes broadband access
devices at central office (CO) and remote(RT) location under the CO, such as digital
subscriber line access multiplexer (DSLAM), Ethernet access device, etc..
The protocol stack of an MPLS interface of the broadband access node is shown in
Fig.l, wherein:
Fig. la shows that the Ethernet frame of an IPized service is carried over MPLS via
PWE3, so that integrated services, such as data, VoIP(Voice over IP), and IPTV(IP
Television), can be carried between the broadband access node and the broadband access
server(BAS) or the edge router of core network; or an end-to-end MPLS connection to
MPLS PE(Provider Edge) of the opposite side across the BAS or the edge router of core
network, such as Ethernet private line service, can be implemented.
Fig. lb shows that the traditional services of ATM, TDM, and FR are carried over
MPLS via PWE3, and an end-to-end private line of ATM, TDM, and FR is formed to the PE
device of the opposite side across the BAS or the edge router of core network, getting rid of
the independent networks of ATM, DDN, and FR.
The detailed implementation procedure of the method according to the embodiment of
the present invention is as follows:
Step 21: multiple LSPs (Label Switching Path) in a broadband access network are
established. An LSP may terminate at a BAS device, or pass through the BAS device. That
is, the established LSP can be the LSP only between the DSLAM (Digital Subscriber Line
Access Multiplexer) and the BAS in the broadband access network, or can be the LSP
starting from the DSLAM via the BAS device across the core network.
The LSP includes one layer of Label or two layers of label.
1. When the LSP uses one layer of label, the LSP may terminate at the BAS or the
edge router of core network. The label is used by the access network to distinguish the
quality of service (QoS) of different types of services.
2. When the LSP uses two layers of label which include an outer layer label and an

inner layer label, there may be a variety of application scenarios in practice.
(21). When the two layers of a two-layer LSP identified respectively by the two layers
of label terminate at the BAS or the edge router of core network simultaneously, the outer
layer label is used by the access network to distinguish the QoS of different types of services,
and the inner layer label is used by the broadband access node in the access network and
BAS/the edge router of core network to identify different services and subscribers.
Optionally, the inner layer label does not provide extra information, and different services
and subscribers are identified by the payload carried in the inner layer LSP;
(22). When the outer layer LSP identified by the outer layer label terminates at the
BAS or the edge router of core network, and the inner layer LSP identified by the inner layer
label passes across the BAS or the edge router of core network and terminates at the
provider edge router MPLS PE of the opposite side, the outer layer label is used by the
access network to distinguish the QoS of different types of services, and the inner layer label
is used by the broadband access node in the access network, the BAS or the edge router of
core network, and the MPLS PE of the opposite side to identify different services and
subscribers. The inner layer LSP needs to be encapsulated in another outer layer LSP when it
enters the core network across the BAS or the edge router of core network (if the opposite
side of communication is in this access network, the inner layer LSP returns to this access
network);
(23). When the two layers of the two-layer LSP identified respectively by the two
layers of label pass across the BAS or the edge router of core network simultaneously and
terminate at the MPLS PE of the opposite side, the outer layer label provides a tunnel to the
MPLS PE of the opposite side, and is used by the access network and the core network to
distinguish the QoS of different types of services, and the inner layer label is used by the
broadband access nodes and the MPLS PE to identify different services and subscribers.
The service transmission on the basis of various LSPs described above is handled in
the following manners:
When an Ethernet is carried over MPLS, if the LSP established in the manner

described in 1, (21) is used, a corresponding LSP can be selected for a service according to
the service tag (S-TAG) in the Ethernet packet of the service. At this time, the one-layer LSP
or the two-layer LSP terminates at the BAS or the edge router of core network, the LSP
which is mapped from the S-TAG can be used to distinguish the QoS of different types of
services in the access network. The S-TAG is the outer layer VLAN TAG of the two layers
of Virtual Local Area Network Tag (VLAN TAG) defined in IEEE 802.1 AD. The S-TAG
mentioned in the embodiments of the present invention also includes the one layer VLAN
TAG defined in IEEE 802.1Q;
When an Ethernet is carried over MPLS, if the LSP established in the manner
described in (22), (23) is employed, a corresponding LSP can be selected for a service
according to the service tag (S-TAG) in the Ethernet packet of the service. At this time, the
LSP which is mapped from the S-TAG can be used in the access network to distinguish the
QoS of different types of services. The outer layer LSP established in the manner described
in (23) can also provide a tunnel through the core network to the MPLS PE of the opposite
side. After the inner layer LSP which is mapped from the S-TAG terminates at the MPLS PE
of the opposite side, the MPLS PE of the opposite side can distinguish different subscribers
and services on the basis of the inner layer label, thus an Ethernet private line connection
can be implemented between the broadband access node and the PE of the opposite side;
When services of ATM, FR or TDM are carried over MPLS, the LSP established in the
manner described in (22), (23) can be used to select the corresponding LSPs for the
respective services according to the virtual channel connection (VCC) of an ATM packet,
the data link connection identifier (DLCI) of a FR packet, and the port of TDM and the
timeslots in a TDM frame. At this time, the outer layer LSP is used by the access network to
distinguish the QoS of different types of services, the outer layer LSP established in the
manner described in (23) can also provide a tunnel to the MPLS PE of the opposite side
through the core network. After the inner layer LSP terminates at the MPLS PE of the
opposite side, the MPLS PE of the opposite side can distinguish different subscribers and
services on the basis of the inner layer label, thus the private line connection of ATM, FR or
TDM can be implemented between the broadband access node and the PE of the opposite

side;
Step 22: when the broadband access node needs to transmit a service upward, a
corresponding LSP is selected at the source broadband access node of the service according
to the type of the service, the quality of service (QoS) and the subscriber information of the
service, and then the service is transmitted over the corresponding LSP.
In step 22, the processing procedure of the broadband access node includes the process
in the case that the Ethernet frame is carried over MPLS and the process in the case that the
ATM/TDM/FR is carried over MPLS, which are described in detail respectively as follows:
1. when the Ethernet frame is carried over MPLS,
if an MPLS packet in the broadband access network terminates at the BAS, or terminates
at the access aggregation device prior to the BAS, the payload carried in the MPLS
packet will identify the services and subscribers by way of 802.1Q or 802.IAD of
Ethernet. After the MPLS connection is terminated, the BAS or the device terminating
the MPLS packet prior to BAS extracts the payload carried in the MPLS packet directly
and identifies the services and subscribers by way of 802.1Q or 802.1 AD. At this time,
MPLS is only the networking means for carrying the integrated services of the access
aggregation layer. Thus, the VLAN plan can be kept the same as of the old network, and
the processing of the broadband access node and the BAS can be simplified.
A pseudo-wire is mapped on the basis of the S-TAG for an upstream service on the
broadband access node. The S-TAG and pseudo-wire are mapped to each other in the
one-to-one map. That is, a corresponding LSP is selected for the upstream service according
to the S-TAG of the service traffic, and each S-TAG corresponds to one LSP. After the
downstream pseudo-wire sent down to the broadband access node is terminated by the
broadband access node, the Ethernet frame is recovered from the MPLS packet. The
subsequent processing procedure by the broadband access node is the same as the procedure
without MPLS transmission.
For the Ethernet private line across the BAS or the edge router of core network, since

the S-TAGs are allocated locally in the access networks, the S-TAGs allocated in the access
networks at the two ends of the same private line may be different. Therefore, after the LSP
is terminated, it may be needed to replace the outer layer S-TAG, that is, the outer layer
S-TAG may be replaced by the actual S-TAG value of the local end access network.
In an embodiment of the present invention, the MPLS label can also be used as the
means of access network for identifying the subscribers and services. At this time, the outer
layer label of Ethernet PWE3 is used by access layer networking to distinguish the QoS of
different type services. The inner layer label is used by the BAS to identify different services
and subscribers. At this time, the BAS is needed to be changed correspondingly so as to
identify different services and subscribers on the basis of the inner layer label.
2. when the ATM/TDM/FR is carried over MPLS
if the ATM/TDM/FR is carried over MPLS, generally, one pseudo-wire starting from
the broadband access node and passing across the BAS or the edge router of core network
is needed to be established. That is, a private line across the BAS or the edge router of core
network is emulated by carrying the traditional services of ATM, TDM and FR over MPLS
via PWE3. For an ATM service, it is necessary for the broadband access node to support the
ATM exchange, that is, the VPI (Virtual Path Identifier)/VCI (Virtual Circuit Identifier) in
the ingress/egress ATM cell is replaced, and the VCC is used to select the pseudo-wire LSP
carried in the outer layer LSP of MPLS. The VCC and the pseudo-wire are mapped to each
other one to one, that is, one VCC corresponds to one LSP. For a TDM service, the
timeslots in TDM frame and the port are used to select the pseudo-wire LSP. For a FR
service, DLCI is used to select the pseudo-wire LSP.
In step 22, in addition to the support of MPLS at the uplink port, BAS or edge router
of core network also needs to:
1. support MPLS forwarding at the downlink port, and
2. support the related standards of IETF PWE3 workgroup, if the BAS or the edge
router of core network is the end point of the corresponding pseudo-wire LSP of Ethernet,

ATM, TDM and FR.
If the BAS or the edge router of core network is the end point of LSP entering at the
downlink port connected to the broadband access network, 802.1Q or 802.1 AD in a PWE3
packet can still be employed to identify the services and subscribers. At this time, after
terminating the LSP, the BAS extracts the Ethernet packet from the LSP. Then the
processing is the same as the method before the MPLS was introduced. Furthermore, the
pseudo-wire labels of PWE3 packets can also be used as identities of the services and
subscribers.
When the BAS or the edge router of core network serves as the intermediate node of
LSP entering at the downlink port connected to the broadband access network, for the
private line across the BAS or the edge router of core network, the BAS or the edge router
of core network is required to support MPLS exchange of MPLS packets entering at the
downlink port.
The broadband access node, the BAS or the edge router of core network according to
the embodiments of the present invention are required to the support MPLS OAM and
protection switching, particularly, they are required to:
(1) support the OAM described in ITU-T Y.1711, so as to acquire the connectivity
state of each LSP and to know in real time whether each LSP can transmit the services
normally;
(2) support the MPLS protection switching described in ITU-T Y.1720. Thus, when a
failure is detected by the OAM function showing that the LSP is disconnected, the services
carried and transmitted by the LSP are switched onto a standby LSP.
A particular application in an actual network of the present invention will be
described below with reference to the accompanying drawings. The access layer
networking application, as well as the MPLS connection application over whole network
and whole path will be described respectively hereinafter.
The access layer networking application includes the access layer networking of the

IPized services and the access layer networking of the integrated services, as shown
respectively in Fig.2 and Fig.3. The thick solid lines in the figures represent the LSPs of
MPLS. CO represents the physical location of a central office; RT represents the physical
location of a remote access device below the central offices; The broadband access device
(such as DSLAM, Ethernet switch, etc.), the PSTN access device, the ATM access device
or the FR access device or the like, can be deployed at the COs and RTs according to the
different types of the access services. RG represents the residence gateway of a broadband
home subscriber.
As shown in Fig.2, the IP services are transmitted via the MPLS networking, while
the traditional services of ATM, TDM and FR and the like are still accessed to the central
offices via other access networks and are then accessed to different core networks. In Fig.3,
the IP services are transmitted via the MPLS networking, the services of ATM, TDM and
FR and the like are also carried over MPLS by PWE3; thus, only the MPLS access network
is reserved in the broadband access network, and the integrated services can be carried in
the MPLS access network; that is, the IP services and the services of ATM, TDM and FR
are all transmitted in the broadband access network via LSP.
In Fig.2, as described above, for the IP services, besides that the MPLS is used for
access layer networking, the MPLS label can also be used by the BAS to identify the
subscribers and services. MPLS label, instead of VLAN label, can overcome the
insufficiencies of space in identifying subscribers and services, and make the service
management more flexible.
The entire network MPLS connection can be classified into the entire network MPLS
connection of IP service and the entire network MPLS connection of integrated service, as
shown respectively in Fig.4 and Fig.5. Similarly, the thick solid lines in the figures
represent LSPs based on MPLS protocol.
As shown in Fig.4, for services based on session, VPN (Virtual private network),
private line or other service passing across the tunnel of the BAS or the edge router of core
network among all the IP services, an Ethernet pseudo-wire passing across the BAS or the

edge router of core network may be established to the PE (Provider Edge) of the opposite
side through the core network to implement the connection covering whole network and
whole path. Accordingly, the QoS, security and survivability can be better ensured by taking
the advantages of MPLS.
As shown in Fig.5, besides that the IP services can establish the PE-to-PE whole-path
connection, the statically configured private lines of TDM, ATM and FR can also migrate
in whole path to the IP network via PWE3, getting rid of the DDN(Digital Data Network),
ATM or FR networks. The PSTN core network is not needed for the call-based voice
services until the voice service migrate to VoIP, i.e. until the TDM voice service is IPized in
the NGN (Next Generation Network). At this time, the IP network can serve as the unified
bearer platform over the entire network.
While the present invention has been illustrated and described with reference to some
preferred embodiments, the present invention is not limited to these. Those skilled in the art
shall readily recognize that various variations and modifications can be made without
departing from the protection scope of the present invention. These variations and
modifications shall be encompassed within the protection scope of the present invention as
defined by the accompanying claims.

We Claim
1. A method for accessing integrated services by an access network□ characterized
by comprising :
establishing at least one Label Switch Path (LSP) of Multi-Protocol Label Switching
(MPLS) in a broadband access network;
MPLS encapsulating, at a broadband access node, a service to be transmitted, and
transmitting the encapsulated service in the broadband access network via the at least
one established LSP,
wherein each of the at least one LSP contains two layers of label comprising an outer
layer label for identifying quality of service, QoS, of different types of services and an
inner layer label for identifying different services and subscribers.
2. The method as claimed in claim 1, wherein the process of establishing at least
one LSP comprises:
establishing at least one LSP between the broadband access node and a Broadband
Access Server (BAS) in the broadband access network, in which the at least one LSP
terminates at the BAS or passes through the BAS.
3. The method as claimed in claim 1, wherein in the process of establishing at least
one LSP, the at least one LSP is established according to different service types and/or
provided quality of service (QoS).
4. The method as claimed in claim 1, wherein the bandwidth of the access network
is planned and managed on the basis of LSP, and different quality of service (QoS) is
provided and guaranteed for different types of services by use of different LSPs.
5. The method as claimed in claim 1, wherein
if two layers of a two-layer LSP identified respectively by the two layers of label
terminate at a Broadband Server (BAS) or an edge router of a core network
simultaneously, the broadband access network distinguishes the quality of service
(QoS) of different types of services on the
basis of the outer layer label, the broadband access node in the access network, and the

BAS or and the edge router of the core network identify different services and
subscribers on the basis of inner layer label; or, the inner layer label does not provide
extra information, and the information of different services and subscribers is provided
by pay load carried over MPLS.
6. The method as claimed in claim 1, wherein if the outer layer LSP identified by the
outer layer label terminates at a Broadband Server (BAS) or an edge router of a core
network, and the inner layer LSP identified by the inner layer label passes through the
BAS or the edge router of the core network and terminates at an MPLS Provider Edge
(PE) of the opposite side, the access network distinguishes the quality of service (QoS)
of different types of services on the basis of the outer layer label; the broadband access
node in the broadband access network, the BAS or the edge router of the core network,
and the MPLS PE of the opposite side identify different services and subscribers on the
basis of the inner layer label; the inner layer LSP is encapsulated in another outer layer
LSP after it passes through the BAS or the edge router of the core network.
7. The method as claimed in claim 1, wherein if two layers of a two-layer LSP
identified respectively by the two layers of label pass across the BAS or the edge router
of the core network simultaneously and terminates at an MPLS Provider Edge (PE) of
the opposite side, the outer layer label provides a tunnel to the MPLS PE of the opposite
side, the access network and the core network distinguish the quality of service (QoS) of
different types of services on the basis of the outer layer label; broadband access nodes
and the MPLS PE at two ends of the inner layer LSP identify different services and
subscribers on the basis of the inner layer label.
8. The method as claimed in any one of claims 1 to 7, wherein the encapsulation
formats for MPLS encapsulation at the broadband access node comprises the
encapsulation formats of Ethernet, Asynchronous Transfer Mode (ATM), Time Division
Multiplexing (TDM), Frame Relay (FR) services carried over MPLS; and
before encapsulating, the method comprises: when a service is needed to be
transmitted by the broadband access network, determining a corresponding LSP for the
service at a broadband access node which is the source end of the service.
9. The method as claimed in claim 8, wherein, in the process of MPLS

encapsulating a service to be transmitted and transmitting the encapsulated service in
the broadband access network via the LSP, the LSP for transmission of IP multicast
video over the access network is an MPLS multicast LSP.
10. The method as claimed in claim 8, wherein the process of determining a
corresponding LSP for the service at the broadband access node which is the source
end of the service comprises:
selecting a corresponding LSP for the service according to the type of the service
and/or quality of service (QoS) required by the service.
11. The method as claimed in claim 10, wherein the process of determining a
corresponding LSP for the service at a broadband access node which is the source end
of the service comprises:
selecting a corresponding LSP for a service according to the subscribers initiating the
service.
12. The method as claimed in claim 10, wherein, the type of the service comprises:
voice service, data service, video service and private line service.
13. The method as claimed in claim 8, wherein the process of determining a
corresponding LSP for the service at a broadband access node which is the source end
of the service comprises:
when Ethernet is carried over MPLS, if a one-layer LSP or a two-layer LSP
terminates at a Broadband Access Server (BAS) or an edge router of a core network,
selecting a corresponding LSP for the service according to a service tag (S-TAG) of
Ethernet packet of the service.
14. The method as claimed in claim 8, wherein the process of determining a
corresponding LSP for the service at a broadband access node which is the source end
of the service comprises:
when Ethernet is carried over MPLS, if two layers of label are used and at least the
inner layer LSP of the two-layer LSP terminates at an MPLS Provider Edge (PE) of the
opposite side across a Broadband Access Server (BAS) or an edge router of core
network, selecting a corresponding LSP for the service according to a service tag

(S-TAG) of Ethernet packet of the service.
15. The method as claimed in claim 8, wherein in the process of determining a
corresponding LSP for the service at a broadband access node which is the source end
of the service,
when services of ATM, TDM or FR are carried over MPLS, two layers of label are
used, in which at least the inner layer LSP of two-layer LSP terminates at the MPLS
Provider Edge (PE) of the opposite side across a Broadband Access Server (BAS) or an
edge router of a core network;
corresponding LSPs are selected for respective services according to Virtual
Channel Connection (VCC) of ATM packet, Data Link Connection Identifier (DLCI) of FR
packet, port of TDM and timeslots in TDM frame of the services;
the inner layer LSP terminates at the MPLS PE of the opposite side, implementing
private line connection of ATM, TDM or FR.
16. The method as claimed in any one of claims 1 to 7, comprising:
detecting connectivity state of an LSP by using MPLS Operation And Maintenance
(OAM);
protecting an LSP which needs to be protected by using MPLS protection switching.
17. The method as claimed in claim 16, wherein the process of protecting an LSP
which needs to be protected by using MPLS protection switching comprises:
establishing in advance a standby LSP for the LSP which needs to be protected;
switching a service to the standby LSP for transmission, when the LSP used by the
service is detected to be failed by the function of MPLS OAM.
18. The method as claimed in any one of claims 1 to 7, comprising:
implementing network failure location, and performance monitoring of the access
network by use of the function of MPLS Operation And Maintenance (OAM).
19. The method as claimed in any one of claims 1 to 7, comprising:
the broadband access node recovering Ethernet frame, ATM packet, FR packet,
TDM frame from MPLS packets arriving at the broadband access node, then performing

subsequent processing.
20. The method as claimed in claim 19, wherein the process of the broadband
access node recovering Ethernet frame, ATM packet, FR packet, TDM frame from MPLS
packets arriving at the broadband access node and then performing subsequent
processing comprises:
for a downstream MPLS packet which is transmitted over Ethernet private line across
Broadband Access Server or edge router of core network, the broadband access node
replacing the outer layer S-TAG of Ethernet after de-encapsulating the MPLS packet, so
that S-TAG can be allocated independently in access networks at both ends.
21. The method as claimed in claim 19, comprising:
if broadband Access Server (BAS) or edge router of core network is the end point of
an LSP entering at a downlink port connected to the broadband access network,
identifying services and subscribers by 802.1Q, or 802.1AD, or pseudo-wire label of
PWE3 in the MPLS PWE3 packet;
the BAS or edge router extracting Ethernet packet from the LSP after terminating the
LSP;
wherein if BAS or edge router of core network is an intermediate node of an LSP
entering at the downlink port connected to the broadband access network,, which is the
case for private line across the BAS or the edge router of the core network, the BAS or
the edge router of the core network supports MPLS exchange of MPLS packets in the
LSP.

The present invention relates to a method for accessing integrated services by an access
network, including: establishing at least one Multi-Protocol Label Switching (MPLS) Label
Switch Path (LSP) in a broadband access network; MPLS encapsulating at a broadband access
node a service to be transmitted, and transmitting the encapsulated service in the broadband
access network via the LSP. As MPLS is a technology for providing multi-layered connection,
in the embodiments of the present invention, MPLS is utilized as the means for networking of
broadband access network and private line passing through the core network, so that the
problem faced by the broadband access network serving as the bearer network of integrated
services can be solved, the integrated services of video, voice, data and private line(including IP
private line, and traditional private lines of ATM (Asynchronous Transfer Mode), TDM (Time
Division Multiplexing) and FR (Frame Relay)) can be carried over the broadband access
network at a lower cost.

METHOD FOR ACCESSING INTEGRATED SERVICE BY AN ACCESS NETWORK

Documents:

Inventors:

PCT Conventions: