Une architecture massivement parallèle intégrée

This paper presents a new massively parallel MIMD architecture, halfway between the Connection Machine and hypercubes based on 32-bit processors. It is built from specific 8-bit processors arranged in a 2-D grid and communicating by message transfers. We discuss the communication problems, the instruction set of the basic processing unit, the programmation of the whole array and the use of a high level data-flow language.     

Onr-partition designs in hamming spaces

The concept of the combinatorial matrix of an unrestricted code and the notion of anr-partition design admitted by a code are introduced and discussed in detail. The theory includes a characterization of completely regular codes, and a combinatorial interpretation of the fact that the distinct rows of the distance distribution matrix of a code are linearly independent. In general, it is possible to compute the distance distribution matrix of any code admitting a given partition design by solving a well-defined system of linear equations; this is an efficient technique provided the number of classes in the partition is relatively small.     

Multihour engineering and traffic routing optimization in hierarchical telephone networks

Classical hierarchical routing in telephone networks is extended to a wider class called out-of-chain routing in such a way that some useful properties of hierarchical routing are retained. This new routing pattern offers more potential paths than the fixed hierarchical one and can be introduced as a dynamic routing where the fixed alternate sequences change at some predetermined instants during the day. The effect of this new routing pattern on the network performances is examined. The main topic of this paper is to present heuristic methods used to optimise such routings in large networks. We show on artificial networks that the throughput of a given network can be significantly improved by suitable routing choices. We demonstrate that the integration of routing changes within a multihour dimensioning process is possible but the lack of realistic data does not permit at this time to quantify the value of routing optimization on real networks.     

Inhibition et mémoire de travail : effet d'une privation aigu? de sommeil sur une tache de génération aléatoire.

The literature contains inconsistent data on the effects of acute sleep deprivation on the superior cognitive functions. The primary purpose of this study is to determine the effectiveness of inhibition, one of the functions of the working memory executive centre (EC), over an extended, 36-hour waking period. Inhibition is a cognitive mechanism whereby individuals ignore non-relevant information recorded in their working memory. We also tested the effects of a 36- hour period of acute sleep deprivation on simple reaction time. Twelve young, healthy volunteers (M=21.5 years, ?=2.3) performed a random generation task involving letters and a simple reaction time psychomotor test over four sessions held at 10-hour intervals. Each participant was assigned a "constant routine." Participants were kept awake in a prone position within a room whose environment was held strictly constant (light, noise, temperature, meals, etc.). This control procedure provided assurance that any variation in participant performance was solely caused by sleep deprivation. The random generation task, nearly two minutes in length, consisted in verbally producing a sequence of 100 letters in a random fashion (i.e. by inhibiting, for example, alphabetical order) and by keeping to a set rhythm... (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Cast bulk ZrTiAlCuNi amorphous alloys

Cylindrical bulk amorphous samples with diameters up to 10 mm have been prepared by casting ZrTiAlCuNi alloys in a copper mould. In order to rank glass-forming ability as a function of alloy composition, alloys were also cast into wedge-shaped moulds; to a first approximation, the thickness of the amorphous region obtained can be taken as an indication of glass-forming ability. The compositions which lead to the production of bulk glasses all have reduced glass transition temperatures in excess of 0.65 and the extremely high glass-forming ability of these compositions is discussed. We suggest that both the Al and Ti contents are determining factors for the production of bulk amorphous samples and these are believed to reduce the driving force for, and hence the rate of, crystallisation. These amorphous alloys have been found to display high thermal stability and can be annealed for several minutes in the supercooled liquid region. They are ductile at room temperature and have a high value of yield stress.     

A clustering method to distribute a database on a grid

Clusters and grids of workstations provide available resources for data mining processes. To exploit these resources, new distributed algorithms are necessary, particularly concerning the way to distribute data and to use this partition. We present a clustering algorithm dubbed Progressive Clustering that provides an “intelligent” distribution of data on grids. The usefulness of this algorithm is shown for several distributed datamining tasks.     

Theoretical analysis of a fractal surface radiation

We compare, in this paper, the far field radiated by two fractal distributions of current. The first one is obtained by a fractal cut following the Cantor arrangement, the second one keeps the length of each previous current element, but modifies the distance between the elements which is now taken as a constant. We establish the analytical formulations of the far field at any step n of the Cantor set. Then, we extract the array factor and we analyse and compare the following properties : convergence of the radiation pattern, wide of the main lobe, side lobes level and directivity.     

Nonlinear black-box modeling in system identification: a unified overview

A nonlinear black-box structure for a dynamical system is a model structure that is prepared to describe virtually any nonlinear dynamics. There has been considerable recent interest in this area, with structures based on neural networks, radial basis networks, wavelet networks and hinging hyperplanes, as well as wavelet-transform-based methods and models based on fuzzy sets and fuzzy rules. This paper describes all these approaches in a common framework, from a user's perspective. It focuses on what are the common features in the different approaches, the choices that have to be made and what considerations are relevant for a successful system-identification application of these techniques. It is pointed out that the nonlinear structures can be seen as a concatenation of a mapping form observed data to a regression vector and a nonlinear mapping from the regressor space to the output space. These mappings are discussed separately. The latter mapping is usually formed as a basis function expansion. The basis functions are typically formed from one simple scalar function, which is modified in terms of scale and location. The expansion from the scalar argument to the regressor space is achieved by a radial- or a ridge-type approach. Basic techniques for estimating the parameters in the structures are criterion minimization, as well as two-step procedures, where first the relevant basis functions are determined, using data, and then a linear least-squares step to determine the coordinates of the function approximation. A particular problem is to deal with the large number of potentially necessary parameters. This is handled by making the number of ‘used’ parameters considerably less than the number of ‘offered’ parameters, by regularization, shrinking, pruning or regressor selection.     

Human chromosome 5 sequence primer amplifies Alu polymorphisms on chromosomes 2 and 17

Members of the Alu family of repetitive elements occur frequently in the human genome and are often polymorphic. Techniques involving Alu element mediated polymerase chain reactions (Alu PCR) allow the isolation of region-specific human DNA fragments from mixed DNA sources. Such fragments are a source of region-specific Alu elements useful for the detection of Alu-related polymorphisms. A clone from human chromosome 5, corresponding to locus D5F40S1, was isolated using Alu PCR differential hybridization. Alu elements within this clone were investigated for the presence of potentially polymorphic 3' polyA tails. Primers were devised to amplify the 3' polyA tail of an Alu element present within the clone. One primer, D5F40S1-T, was specific to the DNA flanking the 3' end of the Alu element, and the other primer was homologous to sequences within the element. When these primers were used in PCR reactions, products from chromosomes 2 and 17 (loci D2F40S2 and D17F40S3) were amplified in addition to the expected product from chromosome 5. The most likely explanation for this nonspecific amplification is that the D5F40S1-T primer is located within a low-copy repetitive element that is 3' of the Alu element. This phenomenon presents a potential problem for the identification of region-specific Alu polymorphisms.