Photonic Network Communications (2006) 11:5–14 DOI 10.1007/s11107-006-5319-x ORIGINAL ARTICLE Link-state-based algorithms for dynamic routing in all-optical networks with ring topologies Rolland Mêwanou · Samuel Pierre Received: 29 July 2004 / Revised: 13 July 2005 / Accepted: 20 July 2005 © Springer Science + Business Media, Inc. 2006 Abstract The increased usage of large bandwidth in opti- cal networks raises the problems of efficient routing to allow these networks to deliver fast data transmission with low blocking probabilities. Due to limited optical buffering in optical switches and constraints of high switching speeds, data transmitted over optical networks must be routed with- out waiting queues along a path from source to destination. Moreover, in optical networks deprived of wavelength con- verters, it is necessary for each established path to transfer data from source to destination by using only one wavelength. To solve this NP-hard problem, many algorithms have been proposed for dynamic optical routing like Fixed-Paths Least Congested (FPLC) routing or Least Loaded Path Routing (LLR). This paper proposes two heuristic algorithms based on former algorithms to improve network throughput and reduce blocking probabilities of data transmitted in all-opti- cal networks with regard to connection costs. We also intro- duce new criteria to estimate network congestion and choose better routing paths. Experimental results in ring networks show that both new algorithms achieve promising perfor- mance. Keywords Optical networks · Dynamic routing · Routing and wavelength assignment · WDM (wavelength division multiplexing) · Link-state routing R. Mêwanou Mobile Computing and Networking Research Laboratory (LARIM) e-mail: s-r.mewanou@ieee.org S. Pierre (B ) Department of Computer Engineering, École Polytechnique de Montréal, P.O. Box 6079, Station Centre-ville, Quebec, Canada H3C 3A7 e-mail: samuel.pierre@polymtl.ca Introduction Over the last few years, many applications have been hin- dered by bandwidth limits imposed by carrier networks. Nowadays, optical networks based on wavelength division multiplexing (WDM) offer high-speed rate of data transfer combined with highly reliable transmission channels. Opti- cal networks convey data on light wavelengths through opti- cal fibers. Physically, as two requests cannot be transmitted at the same time on the same wavelength, it is necessary to find the best way to allocate network resources in order to support heavy traffic. Since optical switches have constraints on high switching speeds and no waiting queues to buffer payload data, it is critical to find the best routing paths from source to destination for all requests arriving at these network nodes. In all-optical networks, so called because they are deprived of opto-electronic converters, a single wavelength is used along each routing path over its lifetime. Such paths are called lightpaths and the fact of using one wavelength per path is called the wavelength continuity constraint. Although many routing algorithms have been developed for various types of optical networks, they all share some common characteris- tics. Network topologies are known. Each routing algorithm computes a set of possible paths for all source–destination pairs before determining the best path to route requests upon their arrival. Routing algorithms which use one precomputed path for any request between two nodes are called static, whereas those which take into account the state of the links before choosing a path are known as dynamic [10]. In dynamic routing problems, for a set of requests between source–destination pairs (s –d ) arriving dynamically in the network, the best routing path among K candidate paths must be determined. This must be achieved while considering the link states, the cost constraints, and the wavelength continu- ity constraint. For example, consider the network shown in