Abstract
Summary form only given. Shock waves have been extensively investigated in diverse areas of physics. However, their occurrence in optics is rather rare and thus far no such optical structures have been experimentally observed. In this paper we report a new family of kink-antikink shock waves that is possible in unbiased photorefactive media via the two wave mixing process. These waves move together as locked states, and have the interesting property of moving outside their initial ±? sector of propagation. The apparent direction of propagation and the spatial widths of these shock wave beams are directly related to their relative intensity. We provide a detailed analysis of these waves, and we illustrate their behavior with pertinent examples. © 2001 Optical Soc. Of America.

Abstract
Large scale distributed systems are typically evolving environments that have to deal with interoperability, scalability, mobility and QoS adaptability requirements. Generically, these systems need adaptation mechanisms to cope with short-term (cf. programmed reconfiguration) and long-term requirements (cf. evolutionary reconfiguration). We propose a reflective component-based framework with architecture style awareness for managing architecture composition and constraining adaptation. Specifically, this framework provides the necessary tools to generate and manipulate the programming model abstractions (i.e. components, connectors and respective properties and interfaces). The framework offers a principled way to deal with both introspection and adaptation of basic and composite components. It provides the developers with the ability to choose, extend and modify architecture style managers. These managers are responsible to represent and check architecture constraints both at development and deployment time, i.e. before any architectural reconfiguration may be committed. © 2001 IEEE.

Abstract
In this paper we describe two approaches developed by two research teams to address the topology identification problem in the scope of state estimation. Both approaches aim at enlarging the traditional concept of strict state estimation, assuming that the network topology is pre-determined and is fixed. In fact, we are generalizing state estimation, enlarging its domain and aiming at obtaining topology information from a state estimation run. Apart from the description of those two techniques, the paper includes a'set of tests performed over the same test system in order to illustrate the interest of the approaches and to evaluate their performances. © 2001 IEEE.

Abstract
We propose a model, called Incremental Correction. (IC) model to address the problem of reverse engineering human control skills using the Behavioural Cloning methodology. The proposed model is based on the concept of closed loop or feedback control. The controllers are induced via Machine Learning tools from traces of human expert control performance. Controllers using the IC model exhibit an increase in robustness and a reduction in encoding complexity when compared to previous models used in behavioural cloning.

Abstract

Abstract
This paper presents a multithreaded abstract machine for the TyCO process calculus. We argue that process calculi provide a powerful framework to reason about fine-grained parallel computations. They allow for the construction of formally verifiable systems on which to base high-level programming idioms, combined with efficient compilation schemes into multithreaded architectures.