Ronald Romero Reyes

Reward-based connection admission control in flex-grid optical networks

In this paper we propose a novel method for online connection admission and grade of service control in flex-grid optical networks, which aims at maximizing a definable reward objective. The method, denoted as MDP-CAC, is derived from the theory of Markov Decision Processes (MDP) based on a simplified state-dependent link model. The MDP-CAC scheme admits connection requests depending on the traffic load and on the expected reward increase they bring to the system. Different objectives (e.g. maximization of either carried traffic or economic revenue) can be realized by an appropriate definition of the connection reward parameters. For a flex-grid optical link, the reward loss (i.e. the ratio of the reward of rejected to offered connections) caused by the new method is compared to that of a basic CAC scheme, which only rejects connections violating the contiguity and capacity constraints. It is observed that the MDP-CAC yields a lower reward loss. Furthermore, it is shown how the MDP-CAC method can be applied to provide GoS control (i.e. control of blocking probabilities) in multiservice flex-grid optical networks.

Techno-economic analysis of multiservice networks: Cost allocation to services: A study for flex-grid optical networks

Based on an existing approach to techno-economic analysis of telecommunication networks, a calculation framework for cost allocation to services is proposed. In particular, the relation between network planning and cost calculation is revealed. This framework is then used to investigate the allocation of infrastructure costs to transport services for flex-grid optical networks. The impact of different off-line routing and spectrum allocation algorithms on the cost per service is evaluated. For large networks it is shown that flex-grid can be a cost-efficient technology for the provisioning of elastic transport services.

Adaptive and state-dependent online resource allocation in dynamic optical networks

In this paper we tackle the problem of online resource allocation in dynamic flex-grid optical networks. To meet the stringent connection setup times envisioned for bandwidth-on-demand services in the optical layer, we propose adaptive and state-dependent algorithms for connection admission control, online routing and spectrum assignment, and as needed also grade of service control. We also describe how an existing three-way handshake protocol can be used to implement the algorithms. By modeling the resource allocation problem as a Markov decision process, the algorithms can be tuned to optimize an objective arbitrarily defined by the network operator (e.g., maximization of carried traffic or economic revenue). Simulation results show that the proposed approach significantly improves the performance of dynamic optical networks.

Online routing and spectrum assignment in flexgrid optical networks

In this paper we propose a novel method for online routing and spectrum assignment (RSA) in flexgrid optical networks based on the theory of Markov Decision Processes (MDP). This theory has already been successfully applied to optimum state-dependent routing in multiservice circuit switched networks. For flexgrid optical networks we introduce enhancements and extensions to existing MDP routing models so as to consider the spectrum continuity and contiguity constraints. We first present the mathematical formulation of the MDP-based optimization model, and then the proposed online RSA optimization algorithm is outlined. Finally, we provide performance evaluation results obtained from simulations of different network and load scenarios.

Analytical performance evaluation of connection setup latency in dynamic optical networks

With the emergence of flex-grid Wavelength Division Multiplexing (WDM) together with reconfigurable optical add/drop multiplexers and bandwidth-variable tunable transponders, a changeover from quasi-static to highly dynamic optical networks is expected. To perform resource allocation in such networks, signalling protocols that are able to rapidly set up optical connections are needed. In this paper, we present an analytical approach to evaluating the performance of an existing signalling protocol that provides fast connection setup in dynamic optical networks. The protocol is modelled as a task graph which represents the signalling latency during the connection establishment phase. By using reduction techniques, the graph is simplified so as to obtain a performance model that estimates the mean connection setup time. Simulations are used to validate the analytical performance model. The results show that the analytical model gives precise estimates of the mean connection setup time.

Reward-based online routing and spectrum assignment in flex-grid optical networks

In this paper we propose an online, reward-based routing and spectrum assignment (RSA) method for multirate flex-grid optical networks. The method, denoted as MDP-RSA, is derived from a flex-grid optical link model based on the theory of Markov decision processes. Assuming that connections arrive randomly offering a reward to the system, the objective is to take the best decision on admission, routing and spectrum assignment for each incoming connection request, such that the rate of reward from the network is maximized. This is accomplished by only admitting connections that bring a positive reward gain to the network. For selected network scenarios, the reward loss (i.e. the ratio of the reward of rejected to offered connections) caused by the MDP-RSA is compared to that of an online scheme, which admits connections on the shortest path that fulfils the contiguity and continuity constraints. It is observed that the MDP-RSA yields a lower reward loss. Furthermore, it is shown how to provide grade of service control on individual connection classes. Compared to existing online RSA schemes, the MDP-RSA is much more flexible as the connection reward parameters can be tailored to different optimization objectives (e.g. maximization of either carried traffic or economic revenue).

State-dependent connection admission control and routing and spectrum assignment in multirate flex-grid optical networks

In this paper we propose an online method for routing and spectrum assignment (RSA) in flex-grid optical networks, which aims at maximizing of a definable reward objective (e.g. carried traffic, economic revenue). The method is based on the theory of Markov Decision Processes (MDP). Connections are admitted and established depending on the traffic load (network state) and on the expected reward increase they bring to the network. In particular, for any incoming request, the routing component of the method recommends the lightpath that yields the highest reward gain in the network. The admission control function accepts the connection if the lightpath gain is positive. A policy iteration algorithm is executed to determine the optimal path and optical channel for each incoming connection request. The method is evaluated for different network scenarios. Furthermore, it is shown how it can provide differentiated grade of service (GoS) for individual connection classes.