Tanmay De

Traffic grooming, routing, and wavelength assignment in an optical WDM mesh networks based on clique partitioning

In wavelength routed optical networks, the number of wavelength channels is limited due to several constraints and each wavelength as well as each lightpath support traffic in the Gbps range. On the other hand, the traffic requested by an individual connection is still in the Mbps range. Therefore, to utilize the network resources (such as bandwidth and transceivers) effectively, several low-speed traffic streams have to be efficiently groomed or multiplexed into one or more high-speed lightpaths. The grooming problem of a static demand is considered as an optimization problem. In this work, we have investigated the traffic grooming problem with the objective of maximizing the network throughput for wavelength-routed mesh networks and map this problem to the clique partitioning problem. We have proposed an algorithm to handle general multi-hop static traffic grooming based on the clique partitioning concept. The efficiency of our approach has been established through extensive simulation on different sets of traffic demands with different bandwidth granularities for different network topologies and compared the approach with existing algorithms.

Distributed dynamic grooming routing and wavelength assignment in WDM optical mesh networks

The bandwidth of a wavelength channel in WDM optical networks is very high compared to the user’s requirements for various applications. Therefore, there is a scope for better utilization of channel bandwidth by traffic grooming, in which several user’s channels are multiplexed for transmission over a single channel. Several research works have been reported on traffic grooming routing and wavelength assignment (GRWA) for static and dynamic traffic pattern under centralized environment. Distributed dynamic grooming routing and wavelength assignment (DDGRWA) is a new and quite unexplored area in WDM optical mesh networks. This article introduces the concept of distributed traffic grooming in WDM mesh networks which also includes virtual topology construction, reconfiguration, routing and wavelength assignment in the distributed environment assuming incoming traffic to be dynamic in nature. We have also presented simulation results of our algorithm on dynamically generated traffic under various network topologies.

A Cost Efficient Multicast Routing and Wavelength Assignment in WDM Mesh Network

Multicast Routing and Wavelength Assignment (MRWA) is a technique implemented in WDM optical networks, where dedicated paths are established between a source and a set of destinations, unlike unicasting where a source is connected with only one destination. For a multicast session request a multicast tree is generated to establish a connection from source to all the destinations. A wavelength is assigned to each and every branches of the generated multicast tree to create a light-tree for the session. In this work, we have tried to minimize the wavelength usage to establish multicast sessions for a set of multicast session requests. Our approach is to minimize the size of the multicast tree by sharing branches, as much as possible, to connect all the destinations from the source node. A lesser usage of links minimizes the collision probability for the assignment of wavelength, say w, in each of the selected links to be assigned the wavelength. Secondly, greater sharing implies lesser splitting. As splitters are costly, minimum usage of splitters incurs lesser infrastructure cost in the network. The effectiveness of our approach has been established through extensive simulation on different set of multicast session under different network topologies and comparing with standard Minimal Spanning Tree (MST) based algorithm. The simulation shows our algorithm performs better than the MST based algorithm.

A genetic algorithm based approach for traffic grooming, routing and wavelength assignment in optical WDM mesh networks

In wavelength routed optical networks, each optical link is capable of carrying traffic using several wavelengths and each wavelength as well as lightpath support traffic in the Gbps range. On the other hand, the traffic requested by individual connection is still in the Mbps range. Therefore, to utilize the network resources (bandwidth, transceivers, etc.) effectively, several low-speed traffic streams have to be efficiently groomed or multiplexed into a high-speed lightpath. In this work, we have investigated the traffic grooming problem with the objective of maximizing the network throughput for wavelength routed mesh networks. We have proposed an algorithm to handle general multi-hop static traffic grooming based on genetic algorithm. The efficacy of our approach has been established through extensive simulation on different sets of traffic demands under different network topologies and compared with an efficient heuristics proposed in the literature.

Multicast routing and wavelength assignment in sparse splitting all optical networks

With the rapid development of high performance networking technology, multicast routing with light path has become a very important and challenging issue in the fields of wavelength routed optical networks. In this paper, we study Steiner tree heuristics in all optical multicast routing in sparse light splitting wavelength division multiplexing (WDM) networks. We present an efficient heuristic algorithm based on greedy approach to construct multicast tree for a given multicast request. The performance of our algorithm is analyzed through simulation in a wide range of network topologies and random session. The result shows that our mechanism requires least number of wavelengths per fiber and total wavelength channels for a given session with respect to existing multicast tree generation algorithms

Survivable of multicast traffic grooming against single link failures in WDM mesh networks

In Wavelength Division Multiplexing (WDM) optical networks, the failure of network resources (e.g., fiber link or node) can disrupt the transmission of information to several destination nodes on a light-tree based multicast sessions. Thus, it is essential to protect multicast sessions by reserving resources along back-up trees. So that if primary tree fails to transmit the information back-up tree will forward the message to the desired destinations. In this paper, we address the problem of survivable of multicast routing and wavelength assignment with sub-wavelength traffic demands in a WDM mesh networks. In this work, we extend the approach of segment disjoint protection methodology to groom the multicast sessions in order to protect them from single link failures. We have proposed an efficient approach for protecting multicast sessions named light-tree based shared segment protection grooming (LTSSPG) scheme and compared with existing multicast traffic grooming with segment protection (MTG-SP) approach. In case of MTG-SP, each segment of primary tree is protected by dis-joint segment in the back-up tree to share the edges or segment. Whereas in case of LTSSPG approach, the segment are shared between the primary as well as back-up trees. The main objective of this work is to minimize the cost in terms of number of wavelengths requirement and optical splitters as well as minimizing the blocking probability of network resources. The performance of various algorithms are evaluated based on extensive simulations in standard networks.

Extending light-trail into elastic optical networks for dynamic traffic grooming

Abstract A light-trail in a traditional ITU − T fixed wavelength grid optical network is a generalization of a lightpath. Multiple nodes in a light-trail are allowed to communicate each other along the light-trail, leading to sub-wavelength granularity utilized traffic grooming. Architecturally, a light-trail is analogous to a shared wavelength optical bus. Moreover, a new light-trail is always equipped by a guard band for non-interference of optical signals. Therefore, multiple light-trails must be accompanied by multiple guard bands, which indeed is an absolute wastage of optical spectrum resources. In the recent years, besides RWA (Routing and Wavelength Assignment) in WDM (Wavelength Division Multiplexing) optical networks, the Routing and Spectrum Assignment (RSA) algorithm is widely being adopted in many new and emerging elastic optical mesh networks to perform traffic grooming. In such networks, Orthogonal Frequency Division Multiplexing (OFDM) technology is a promising candidate to execute a RSA algorithm and implement an elastic lightpath as a result. The algorithm is capable of transmitting high-speed data stream using multiple low-speed spectrum overlapped sub-carriers (mini-grids). Hereafter, in this paper we have introduced a novel concept (based on our knowledge and belief) elastic light-trail (a variant of an elastic lightpath) in elastic optical networks and proposed an algorithm “Multi-hop Elastic Light-Trail – MELT” that performs an elastic light-trail expansion and/or contraction (in frequency domain), instead of creating a new elastic light-trail where applicable. Otherwise, the algorithm creates a new elastic light-trail that facilitates required data transportation. In this work, we investigate optical spectrum utilization efficiency for proposed elastic light-trail, existing elastic lightpath (Multi-hop Elastic Lightpath algorithm – MEL) and ITU − T wavelength grid fixed lightpath (Multi-hop Lightpath algorithm – ML) and perform comparison measurements.

An efficient heuristic-based algorithm for wavelength converter placement in all-optical networks

Wavelength conversion and routing are the two key techniques used for improvement of the overall blocking performance in wavelength routed all-optical networks. As wavelength converters are very expensive, the effective use of wavelength converters or more precisely placement of minimum number of wavelength converter in the appropriate nodes in the network is a challenging issue in the field of wavelength division multiplexing (WDM) optical networks. In this paper, we have studied the performance of sparse complete wavelength conversion (SCWC) network architecture. An efficient heuristic-based algorithm is proposed based on alternate routing minimum blocking probability (ARMBP) for efficient placement of wavelength converters under K-alternate routing with shortest path first-first fit (AR-SPF-FF) routing and wavelength assignment. The performance of our proposed algorithm is analyzed with the help of extensive simulations for a wide range of network topology. The results show that our proposed approach produces better performance compared to the existing wavelength converter placement algorithms.

A load balanced approach of multicast routing and wavelength assignment in WDM networks

Multicast routing and wavelength assignment MRWA deals with the problem of one-to-many communication in a wavelength division multiplexed optical fibre networks. As physical optical networks are equipped with finite number of optical channels in each fibre cable, throughput of the network is treated as a major performance issue. The aim of this work is to maximise throughput in a finite wavelength WDM network. As to establish multicast session a single message may need to split into multiple messages fanning out of a node, the node must contain costly splitters. This work also considers the cost of splitting at the split nodes while establishing a light-tree for a particular multicast session request. Here a load-balancing technique is used to optimise the optical channel requirements to establish a multicast connection. The algorithm proposed here is compared with standard all-shortest path algorithm combined with first-fit wavelength assignment for multicasting by simulation on various network topologies and different network configurations. The simulation result shows that our proposed algorithm performs much better than the standard one.

Knapsack based multicast traffic grooming for optical networks

Abstract This paper proposes a light-tree based heuristic algorithm, called 0/1 knapsack based multicast traffic grooming, in order to minimize the network cost by reducing the number of higher layer electronic and optical devices, such as transmitters, receivers, and splitters, and used wavelengths in the network. The proposed algorithm constructs light-trees or sub light-trees, which satisfy sub bandwidth demands of all multicast requests. We present a light-tree based integer linear programming (ILP) formulation to minimize the network cost. We solve the ILP problem for sample four-node and six-node networks and compare the ILP results with the proposed heuristic algorithm. We observe that the performance of the proposed algorithm is comparable to the ILP in terms of cost. When the introduced ILP is not tractable for large network, the proposed algorithm still able to find the results. Furthermore, we compare the proposed heuristic algorithm to existing heuristic algorithms for different backbone networks. Numerical results indicate that the proposed heuristic algorithm outperforms the conventional algorithms in terms of cost and resource utilization.