Research On The Access Theory And Experiment Of Carrier Grade Ethernet Optical Networks

The so-called carrier Ethernet is that keeping the original Ethernet frame structure unchanged, realize the telecommunications network management through the expansion of the Ethernet frame header and the layer two Signaling command. One key issue of the mature Ethernet industrial chain link is the comprehensive introducing of the standardization of Ethernet QoS, security, OAM mechanisms and extension technology. As one of the key access network technologies, Ethernet passive optical network has gradually been accepted by most operators at home and abroad and put into practice, whick as a carrier Ethernet in an integral part of the whole of its line telecommunications level network, the expansion and transformation of imperative operations must be taken.The research in this dissertation was supported by the National High Technology Research and Development Program of China (863 Program) project "Gigabits Ethernet Based on Passive Optical Networks", launched a system-depth theoretical and experimental study around the carrier-class passive optical network technology. The main contributions of this paper are as the followings:(1) Poring the Metro Ethernet Forum (MEF) of the theory of carrier Ethernet to Ethernet passive optical networks, concluding the Ethernet Passive Optical Network's technical requirements from aspects such as architecture, network management, security, QoS, multi-service, et.al.(2) Combined self-experience in the design of the prototype, summarizes the system design, particularly in the embedded subsystems. In order to enhance the system stability line and carrier-class equipment reliability, Introducing the suitable Embedded Systems platform architecture for Ethernet Passive Optical Network, and distributed systems division and function of OLT. Study also describes the FTTB-type ONU design to achieve a complete EPON prototype system, which has already been tested completely by a number of telecom operators. Result shows that the prototype system is in line with network requirements, also be able to meet current operating requirements of carrier-class.(3) According to Carrier Ethernet operations, maintenance, management, end to end requirements, Proposing a set of EPON-based extension of OAM management process, which gives the extended OAM state machine design and extend OAM the management of configuration including OLT and ONU. The design can provide a more complete variety of management functions at the operational level to improve that the existing Ethernet OAM functionality is limited to the limitations of link-level management, and have a better scalability for the standard definition of the future with the MEF to provide cross-docking facilities and a reliable interface.(4) According to the needs of Carrier Access Network carry full-service, based on EPON existing MPCP protocol framework, presents a scheduling policy named "User based Fair Service Envelope" (UFSE) with the consideration of the business and user priority. The strategy improve the fair character of existing EPON dynamic bandwidth algorithm, which concern both the different users and the differ-services. The algorithm is broaderly adaptive, can be used in the current common single-user and multi-user scenarios, and to achieve the scalability of level scheduling. Simulation result is given to show the effectiveness of the algorithm.(5) According to the expanded requirements of carrier Ethernet technology and the Ethernet services classification of MEF. Based on TR101 marking strategy in DSL systems, we demonstrated three identification architetures in the EPON system for multi-service and multi-users and meticulous management, These architetures can achieve accurate identification and management for users, business and equipment.(6) According to Carrier Ethernet for business operations security requirements and the characteristics of multicast video services. The controllable multicast technology is introduced to improve the security of the existing access network, whick can identify the multicast user certification and program validity, and also proposed the dynamic configuration and the static configuration for different user density. The delay of these two characteristics were analyzed through the experimental.