Abstract
A demodulation scheme for fiber Bragg grating (FBG) sensors is presented. It is based on the generation of an electrical carrier by using a modulated multimode laser diode to illuminate the fiber grating. The change in Bragg wavelength is measured by tracking the phase of the carrier at the detector output in either an open- or a closed-loop scheme. A theoretical analysis of the interrogation technique in terms of linearity and dynamic range is presented. Experimental data were obtained for both strain and temperature measurements. Sensitivities of 0.7 mu epsilon/root Hz and 0.05 degrees C/root Hz were obtained over a dynamic range of approximate to 60 dB, The application of this demodulation scheme to a multiplexed sensing system is also demonstrated.

Abstract
Atomic operations are a key primitive in parallel computing systems. The standard implementation mechanism for atomic operations uses mutual exclusion locks. In an object-based programming system, the natural granularity is to give each object its own lock. Each operation can then make its execution atomic by acquiring and releasing the lock for the object that it accesses. But this fine lock granularity may have high synchronization overhead because it maximizes the number of executed acquire and release constructs. To achieve good performance it may be necessary to reduce the overhead by coarsening the granularity at which the computation locks objects. In this article we describe a static analysis technique - lock coarsening - designed to automatically increase the lock granularity in object-based programs with atomic operations. We have implemented this technique in the context of a parallelizing compiler for irregular, object-based programs and used it to improve the generated parallel code. Experiments with two automatically parallelized applications show these algorithms to be effective in reducing the lock overhead to negligible levels. The results also show, however, that an overly aggressive lock coarsening algorithm may harm the overall parallel performance by serializing sections of the parallel computation. A successful compiler must therefore negotiate a trade-off between reducing lock overhead and increasing the serialization. © 1998 Academic Press.

Abstract
A fiber optic sensing system for simultaneous measurement of displacement and temperature is presented. It is based on the use of a low finesse extrinsic Fabry-Perot cavity and a fiber Bragg grating for temperature compensation. A white light tandem interferometric technique is used to recover the signal from the low finesse cavity, providing both fringe visibility and phase measurements. These, together with the fiber grating signal, give two equations that can be used to simultaneously determine temperature and displacement. Theoretical analysis of the sensing scheme is performed. Experimental results are presented which validate theoretical predictions and also demonstrate the feasibility of this sensing system in practical applications.

Abstract

Abstract

Abstract
In this paper we describe a hybrid strategy for solving combinatorial optimisation problems, obtained by coupling a local search method to an evolutionary algorithm, and we provide an application to a particular variant of the vehicle routing problem. The local search method has been devised specifically for this class of problems. It is based on a composite neighbourhood, which is searched iteratively up to the point where no further improvements are made. The evolutionary structure is the niche search, an algorithm based on the evolution of several independent niches. Niches whose individuals' fitness is good remain, and the others tend to be replaced. The separation of the population into niches allows for a good compromise between intensive search (inside each niche) and diversification (through the separation between the niches). We also describe how we integrate specific problem knowledge into an evolutionary structure, in order to achieve a high performance optimisation algorithm. All the steps that we consider necessary are described in detail: finding an appropriate representation, determining what is a relevant neighbourhood, setting up a local search method and finally integrating; the local search into an evolutionary algorithm.