Research On Resoration Protocol And Algorithm Of Intelligent Optical Network Based On Gmpls

The Intelligent Optical Network (ION) is the trend of the next generation optical networks. Moreover, the Automatic Switched Optical Network (ASON) has attracted many researchers'interests as a representation of ION. With the further developments of the transport network's technology, Survivability of network has become the important issue of designing and maintaining ION. Highly efficient and flexible mechanism for the protection and restoration is the important characteristic for next-generation network, and the faster fault recovery becomes ASON core functional requirements and key advantage. ASON provides a flexible and efficient transport plane for services to meet the services needs and operational characteristics. Meanwhile, it also creates the conditions for the upgrading of relevant network performance, such as the survivability of the network, which relies on the Dynamic recovery capacity of GMPLS-based control plane. ASON can implement rapid recovery of network's failure, and upgrade the survivability of network.Protection and restoration are the two mechanisms of the implementing network survivability. Compare with protection mechanism, restoration mechanism has better resources utilization, but it has longer recovery time. So, we can use the precomputing method of ASON survival mechanism, which computes restoration path before failure, but configures optical cross-connect (OXC) and assigns wavelength after failure. This can shorten the fault recovery time. This paper gives multi-initiation restoration protocol (MIRP), and compares with the traditional source node initiation restoration protocol (SIRP) using simulation.The simulation results show that MIRP can effectively shorten the recovery time, and upgrade the restoration probalility from failure.In the actual network, different fiber links have the relationship of risks if they traverse the same physical resources (e.g. fiber, conduit, etc). This characteristic can be defined as Shared Risk Link Groups (SRLG). This paper presents a restoration strategy, which firstly uses an effective method (ISRLG) of finding disjoint-path with SRLG, and then uses a new dynamic restoration routing algorithm with disjoint-SRLG based on GMPLS (RSGA) to fufill restoration from failure. Finally, this paper gives the simulation process on GLASS platform, and the results of performance comparision with Short Path Restoration (SPR) on aspect of resource utilization and blocking ratio.