Implementation of Art1 and Art2 Artificial Neural Networks on Ring and Mesh Architectures

The Artificial Neural Networks (ANNs) are being used to solve a variety of problems in pattern recognition, robotic control, VLSI CAD and other areas. In most of these applications, a speedy response from the ANNs is imperative. However, ANNs comprise a large number of artificial neurons, and a massive interconnection network among them. Hence, implementation of these ANNs involves execution of computer-intensive operations. The usage of multiprocessor systems therefore becomes necessary. In this article, we have presented the implementation of ART1 and ART2 ANNs on ring and mesh architectures. The overall system design and implementation aspects are presented. The performance of the algorithm on ring, 2-dimensional mesh and n-dimensional mesh topologies is presented. The parallel algorithm presented for implementation of ART1 is not specific to any particular architecture. The parallel algorithm for ARTE is more suitable for a ring architecture.

A Survey Of Software Dependability

This paper presents on overview of the issues in precisely defining, specifying and evaluating the dependability of software, particularly in the context of computer controlled process systems. Dependability is intended to be a generic term embodying various quality factors and is useful for both software and hardware. While the developments in quality assurance and reliability theories have proceeded mostly in independent directions for hardware and software systems, we present here the case for developing a unified framework of dependability—a facet of operational effectiveness of modern technological systems, and develop a hierarchical systems model helpful in clarifying this view. In the second half of the paper, we survey the models and methods available for measuring and improving software reliability. The nature of software “bugs”, the failure history of the software system in the various phases of its lifecycle, the reliability growth in the development phase, estimation of the number of errors remaining in the operational phase, and the complexity of the debugging process have all been considered to varying degrees of detail. We also discuss the notion of software fault-tolerance, methods of achieving the same, and the status of other measures of software dependability such as maintainability, availability and safety.

Software simulation of the EEG

The literature contains many examples of digital procedures for the analytical treatment of electroencephalograms, but there is as yet no standard by which those techniques may be judged or compared. This paper proposes one method of generating an EEG, based on a computer program for Zetterberg's simulation. It is assumed that the statistical properties of an EEG may be represented by stationary processes having rational transfer functions and achieved by a system of software fillers and random number generators.The model represents neither the neurological mechanism response for generating the EEG, nor any particular type of EEG record; transient phenomena such as spikes, sharp waves and alpha bursts also are excluded. The basis of the program is a valid ‘partial’ statistical description of the EEG; that description is then used to produce a digital representation of a signal which if plotted sequentially, might or might not by chance resemble an EEG, that is unimportant. What is important is that the statistical properties of the series remain those of a real EEG; it is in this sense that the output is a simulation of the EEG. There is considerable flexibility in the form of the output, i.e. its alpha, beta and delta content, which may be selected by the user, the same selected parameters always producing the same statistical output. The filtered outputs from the random number sequences may be scaled to provide realistic power distributions in the accepted EEG frequency bands and then summed to create a digital output signal, the ‘stationary EEG’. It is suggested that the simulator might act as a test input to digital analytical techniques for the EEG, a simulator which would enable at least a substantial part of those techniques to be compared and assessed in an objective manner. The equations necessary to implement the model are given. The program has been run on a DEC1090 computer but is suitable for any microcomputer having more than 32 kBytes of memory; the execution time required to generate a 25 s simulated EEG is in the region of 15 s.

Mesh-free approximations via the error reproducing kernel method and applications to nonlinear systems developing shocks

Error estimates for the error reproducing kernel method (ERKM) are provided. The ERKM is a mesh-free functional approximation scheme [A. Shaw, D. Roy, A NURBS-based error reproducing kernel method with applications in solid mechanics, Computational Mechanics (2006), to appear (available online)], wherein a targeted function and its derivatives are first approximated via non-uniform rational B-splines (NURBS) basis function. Errors in the NURBS approximation are then reproduced via a family of non-NURBS basis functions, constructed using a polynomial reproduction condition, and added to the NURBS approximation of the function obtained in the first step. In addition to the derivation of error estimates, convergence studies are undertaken for a couple of test boundary value problems with known exact solutions. The ERKM is next applied to a one-dimensional Burgers equation where, time evolution leads to a breakdown of the continuous solution and the appearance of a shock. Many available mesh-free schemes appear to be unable to capture this shock without numerical instability. However, given that any desired order of continuity is achievable through NURBS approximations, the ERKM can even accurately approximate functions with discontinuous derivatives. Moreover, due to the variation diminishing property of NURBS, it has advantages in representing sharp changes in gradients. This paper is focused on demonstrating this ability of ERKM via some numerical examples. Comparisons of some of the results with those via the standard form of the reproducing kernel particle method (RKPM) demonstrate the relative numerical advantages and accuracy of the ERKM.

Multipath Dissemination in Regular Mesh Topologies

Mesh topologies are important for large-scale peer-to-peer systems that use low-power transceivers. The Quality of Service (QoS) in such systems is known to decrease as the scale increases. We present a scalable approach for dissemination that exploits all the shortest paths between a pair of nodes and improves the QoS. Despite th presence of multiple shortest paths in a system, we show that these paths cannot be exploited by spreading the messages over the paths in a simple round-robin manner; nodes along one of these paths will always handle more messages than the nodes along the other paths. We characterize the set of shortest paths between a pair of nodes in regular mesh topologies and derive rules, using this characterization, to effectively spread the messages over all the available paths. These rules ensure that all the nodes that are at the same distance from the source handle roughly the same number of messages. By modeling the multihop propagation in the mesh topology as a multistage queuing network, we present simulation results from a variety of scenarios that include link failures and propagation irregularities to reflect real-world characteristics. Our method achieves improved QoS in all these scenarios.

Finite volume TVD formulation of lattice Boltzmann simulation on unstructured mesh

A numerical scheme is presented for accurate simulation of fluid flow using the lattice Boltzmann equation (LBE) on unstructured mesh. A finite volume approach is adopted to discretize the LBE on a cell-centered, arbitrary shaped, triangular tessellation. The formulation includes a formal, second order discretization using a Total Variation Diminishing (TVD) scheme for the terms representing advection of the distribution function in physical space, due to microscopic particle motion. The advantage of the LBE approach is exploited by implementing the scheme in a new computer code to run on a parallel computing system. Performance of the new formulation is systematically investigated by simulating four benchmark flows of increasing complexity, namely (1) flow in a plane channel, (2) unsteady Couette flow, (3) flow caused by a moving lid over a 2D square cavity and (4) flow over a circular cylinder. For each of these flows, the present scheme is validated with the results from Navier-Stokes computations as well as lattice Boltzmann simulations on regular mesh. It is shown that the scheme is robust and accurate for the different test problems studied.

NUVIEW:software for display and interactive manipulation of nucleic acid models

The NUVIEW software package allows skeletal models of any double helical nucleic acid molecule to be displayed out a graphics monitor and to apply various rotations, translations and scaling transformations interactively, through the keyboard. The skeletal model is generated by connecting any pair of representative points, one from each of the bases in the basepair. In addition to the above mentioned manipulations, the base residues can be identified by using a locator and the distance between any pair of residues can be obtained. A sequence based color coded display allows easy identification of sequence repeats, such as runs of Adenines. The real time interactive manipulation of such skeletal models for large DNA/RNA double helices, can be used to trace the path of the nucleic acid chain in three dimensions and hence get a better idea of its topology, location of linear or curved regions, distances between far off regions in the sequence etc. A physical picture of these features will assist in understanding the relationship between base sequence, structure and biological function in nucleic acids.

NUPARM and NUCGEN: software for analysis and generation of sequence dependent nucleic acid structures

Software packages NUPARM and NUCGEN, are described, which can be used to understand sequence directed structural variations in nucleic acids, by analysis and generation of non-uniform structures. A set of local inter basepair parameters (viz. tilt, roll, twist, shift, slide and rise) have been defined, which use geometry and coordinates of two successive basepairs only and can be used to generate polymeric structures with varying geometries for each of the 16 possible dinucleotide steps. Intra basepair parameters, propeller, buckle, opening and the C6...C8 distance can also be varied, if required, while the sugar phosphate backbone atoms are fixed in some standard conformation ill each of the nucleotides. NUPARM can be used to analyse both DNA and RNA structures, with single as well as double stranded helices. The NUCGEN software generates double helical models with the backbone fixed in B-form DNA, but with appropriate modifications in the input data, it can also generate A-form DNA ar rd RNA duplex structures.

Mesh Processing for Computerized Facial Anthropometry

Understanding of the shape and size of different features of the human body from scanned data is necessary for automated design and evaluation of product ergonomics. In this paper, a computational framework is presented for automatic detection and recognition of important facial feature regions, from scanned head and shoulder polyhedral models. A noise tolerant methodology is proposed using discrete curvature computations, band-pass filtering, and morphological operations for isolation of the primary feature regions of the face, namely, the eyes, nose, and mouth. Spatial disposition of the critical points of these isolated feature regions is analyzed for the recognition of these critical points as the standard landmarks associated with the primary facial features. A number of clinically identified landmarks lie on the facial midline. An efficient algorithm for detection and processing of the midline, using a point sampling technique, is also presented. The results obtained using data of more than 20 subjects are verified through visualization and physical measurements. A color based and triangle skewness based schemes for isolation of geometrically nonprominent features and ear region are also presented. [DOI: 10.1115/1.3330420]

Providing QoS to Real and Data Applications in WiMAX Mesh Networks

We consider the problem of centralized routing and scheduling for IEEE 802.16 mesh networks so as to provide Quality of Service (QoS) to individual real and interactive data applications. We first obtain an optimal and fair routing and scheduling policy for aggregate demands for different source- destination pairs. We then present scheduling algorithms which provide per flow QoS guarantees while utilizing the network resources efficiently. Our algorithms are also scalable: they do not require per flow processing and queueing and the computational requirements are modest. We have verified our algorithms via extensive simulations.

Reconfigurable Viterbi decoder on mesh connected multiprocessor architecture

In modern wireline and wireless communication systems, Viterbi decoder is one of the most compute intensive and essential elements. Each standard requires a different configuration of Viterbi decoder. Hence there is a need to design a flexible reconfigurable Viterbi decoder to support different configurations on a single platform. In this paper we present a reconfigurable Viterbi decoder which can be reconfigured for standards such as WCDMA, CDMA2000, IEEE 802.11, DAB, DVB, and GSM. Different parameters like code rate, constraint length, polynomials and truncation length can be configured to map any of the above mentioned standards. Our design provides higher throughput and scalable power consumption in various configuration of the reconfigurable Viterbi decoder. The power and throughput can also be optimized for different standards.

Software Pipelined Execution of Stream Programs on GPUs

The StreamIt programming model has been proposed to exploit parallelism in streaming applications on general purpose multi-core architectures. This model allows programmers to specify the structure of a program as a set of filters that act upon data, and a set of communication channels between them. The StreamIt graphs describe task, data and pipeline parallelism which can be exploited on modern Graphics Processing Units (GPUs), as they support abundant parallelism in hardware. In this paper, we describe the challenges in mapping StreamIt to GPUs and propose an efficient technique to software pipeline the execution of stream programs on GPUs. We formulate this problem - both scheduling and assignment of filters to processors - as an efficient Integer Linear Program (ILP), which is then solved using ILP solvers. We also describe a novel buffer layout technique for GPUs which facilitates exploiting the high memory bandwidth available in GPUs. The proposed scheduling utilizes both the scalar units in GPU, to exploit data parallelism, and multiprocessors, to exploit task and pipelin parallelism. Further it takes into consideration the synchronization and bandwidth limitations of GPUs, and yields speedups between 1.87X and 36.83X over a single threaded CPU.

High-Throughput Flexible Constraint Length Viterbi Decoders on De Bruijn, Shuffle-Exchange and Butterfly Connected Architectures

Flexible constraint length channel decoders are required for software defined radios. This paper presents a novel scalable scheme for realizing flexible constraint length Viterbi decoders on a de Bruijn interconnection network. Architectures for flexible decoders using the flattened butterfly and shuffle-exchange networks are also described. It is shown that these networks provide favourable substrates for realizing flexible convolutional decoders. Synthesis results for the three networks are provided and a comparison is performed. An architecture based on a 2D-mesh, which is a topology having a nominally lesser silicon area requirement, is also considered as a fourth point for comparison. It is found that of all the networks considered, the de Bruijn network offers the best tradeoff in terms of area versus throughput.

Data mining approaches to software fault diagnosis

Automatic identification of software faults has enormous practical significance. This requires characterizing program execution behavior and the use of appropriate data mining techniques on the chosen representation. In this paper, we use the sequence of system calls to characterize program execution. The data mining tasks addressed are learning to map system call streams to fault labels and automatic identification of fault causes. Spectrum kernels and SVM are used for the former while latent semantic analysis is used for the latter The techniques are demonstrated for the intrusion dataset containing system call traces. The results show that kernel techniques are as accurate as the best available results but are faster by orders of magnitude. We also show that latent semantic indexing is capable of revealing fault-specific features.

Parallelization of an unstructured data based cell centre finite volume code, HIFUN-3D

This work describes the parallelization of High Resolution flow solver on unstructured meshes, HIFUN-3D, an unstructured data based finite volume solver for 3-D Euler equations. For mesh partitioning, we use METIS, a software based on multilevel graph partitioning. The unstructured graph used for partitioning is associated with weights both on its vertices and edges. The data residing on every processor is split into four layers. Such a novel procedure of handling data helps in maintaining the effectiveness of the serial code. The communication of data across the processors is achieved by explicit message passing using the standard blocking mode feature of Message Passing Interface (MPI). The parallel code is tested on PACE++128 available in CFD Center