For elastic optical networks (EONs), maximizing spectrum usage, minimizing call blocking, and ensuring fair, high-quality service for various traffic types requires efficient resource allocation. This research proposes a novel data-driven approach using a neural network for an adaptive resource allocation algorithm based on the relative cost formula. Unlike traditional methods with limited mathematical models, our neural network learns complex relationships from data to estimate the relative cost of carrying new calls on the current network state. This leads to high accuracy and efficiency, both during training and real-time use.

For an efficient usage of the sensor technology, several design factors (e.g., topology and sensing coverage) should be taken into account. In this paper, we focus on the underlying topology of sensor networks in two-dimensional environments and enhance a set of recently proposed graphs. The new enhanced graphs are referred to as the Derived Circles version 2 (DCαv2) graphs. We show that DCαv2 graphs are locally constructed, connected, have the rotation-ability property, and have the Euclidean Minimum Spanning Tree (EMST) as their subgraphs. Moreover, we show that the new set of graphs has a bounded Euclidean/length and power dilation when 0.5 ≤ α ≤ 1. Furthermore, via simulations, we confirm most of these properties, and demonstrate that the DCαv2 graphs also have bounded Euclidean and power dilations when 0αv2 graphs outperform the Half Space Proximal (HSP) and the Relative Neighbourhood Graph (RNG) graphs in terms of the network dilation, Euclidean dilation, and power dilation. This, in turn, increases the speed for message delivery, reduces the energy consumption of nodes and accordingly prolongs the network lifetime.

Abstract-Recently, sensor network technology has brought the attention of many researchers due to its scalability and efficiency. However, to improve its efficiency level, several design challenging factors should be taken into account. Among these challenges, we focus on the underlying topology of sensor networks in three-dimensional environments and extend a new set of graphs referred to as the Derived Sphere (DSα) graphs from their 1-D version Il][30]. We show that DSαgraphs are locally constructed, connected, and have the rotation-ability property. Achieving the connectivity and the rotation-ability properties imply strong reliability for these graphs. Moreover, we show that the new set of graphs has a bounded Euclidean (or length) when 0αgraphs outperform 3-D Half Space Proximal (HSP) and 3-D Gabriel Graph (GG) graphs in terms of network and Euclidean dilations. This, in turn, increases the speed for message delivery. Therefore, the DSαgraphs are considered timely-efficient graphs.

Dynamic routing and wavelength assignment RWA is one of the most important issues in wavelength routed all optical networks. We introduce the learning automata technique for dynamic RWA in WDM networks without conversion under different load conditions. Learning automata are used to choose the shortest route from source to destination if more than one shortest route exist. Furthermore, learning automata are used to select which wavelength is to be used on the chosen route. We compare our wavelength assignment technique with some exhaustive wavelength assignment algorithms that scan all wavelengths on a predetermined shortest route. The use of the learning automata wavelength assignment technique reduces the call setup time by pursuing a small number of wavelengths. In addition, the technique is used to achieve fairness among different source/destination pairs. Simulation results are presented which indicate the benefits of using the learning automata technique for dynamic routing and wavelength assignment in WDM networks.

If a network supports only certain rates, we say its bandwidth is quantized. In a quantized-bandwidth network, a customer may be forced to use the next higher rate if the requested rate is not supported. The result is a higher blocking probability and a lower throughput. The throughput loss due to quantization is called the quantization loss, and its related issues are the focus of this paper. We show that the quantization loss is insensitive to capacity scaling and traffic-loading. More important, it is hardly as large as is thought. Even for a network, such as ATM, intending to support a continuous bit rate, bandwidth quantization can be a powerful tool for states reduction. States reduction is shown to be indispensable for solving many problems, such as routing and capacity planning, of a broadband network.>

The authors indicate that the number of rates supportable in an asynchronous transfer mode (ATM) broadband network has to be limited unless a network is acceptable without knowing its grade of service. The quantization of the channel bandwidth, given the traffic arrival rate as a function of requested bandwidth, is considered. Capacity sharing models in ATM networks are presented. The calculation procedure for blocking probability which will be affected by the quantization problem is described. A class of control methods that can render a product-form solution is introduced. The results indicate that there is little performance difference between a quantized channel and a continuous-rate channel as the level of quantization exceeds ten.>