Transputer Network Research Articles

Development of forceps manipulator for assisting laparoscopic surgery

Telemanipulation for use in surgery is an increasing field of robotics which combines new technical and medical developments. Some drawbacks of minimally invasive surgery can be reduced by using telemanipulation systems. This article describes the development of a hydraulic gripper with force-feedback which is designed for use in laparoscopic surgery. With respect to the manipulation of bowels during a colon resection a gripper is designed which exerts a force with a significantly better distribution on the tissue compared to conventional grippers. This is done by a flexible tube which curls itself around the tissue (like the finger of a surgeon by open surgery) and prevents damaging due to the absence of sharp (metal) edges. The force exerted on the tissue is measured by the pressure of the water in the tube and used to control the driving system of the forceps. Using a computer system, the force exerted on the tissue can be limited to a certain amount by a control algorithm. The driving system consists of an electromotor with spindle and a piston and is driven directly by the computer. Finally, a telemanipulation system is presented, which consists of the hydraulic forceps, a transputer network and a master manipulator with force actuation. With this instrument the surgeon can drive the forceps to a desired force, which is directly presented to his finger. Future developments include the positioning of the gripper inside its working area.

Performance-based control system design automation via evolutionary computing

This paper develops an evolutionary algorithm (EA) based methodology for computer-aided control system design (CACSD) automation in both the time and frequency domains under performance satisfactions. The approach is automated by efficient evolution from plant step response data, bypassing the system identification or linearization stage as required by conventional designs. Intelligently guided by the evolutionary optimization, control engineers are able to obtain a near-optimal “off-the-computer” controller by feeding the developed CACSD system with plant I/O data and customer specifications without the need of a differentiable performance index. A speedup of near-linear pipelineability is also observed for the EA parallelism implemented on a network of transputers of Parsytec SuperCluster. Validation results against linear and nonlinear physical plants are convincing, with good closed-loop performance and robustness in the presence of practical constraints and perturbations.

00/00259 Power system parallel computation by a transputer network

00/00259 Power system parallel computation by a transputer network

Parallel-Processing Diagnostic Program for a Transputer-Based Readout System

A parallel-processing program for system-diagnostic purposes was developed for a transputer-based readout system of a silicon vertex detector (SVD). The data-acquisition (DAQ) system for SVD was constructed as a part of research and development work for the BELLE [name of a detector for an asymmetric collider experiment at High Energy Accelerator Research Organization (KEK)] experiment consisting of data scanners (DS), slave memory modules (SMM) and direct memory access master modules (DMM). The DS, SMM and DMM have a transputer as a core processor. Each transputer was linked by a point-to-point connection with transputer serial interfaces (Inmos-link). The DAQ system was equipped with a transputer network system with the Inmos-links. The diagnostic program has the capability to detect malfunctioning readout components. To minimize the processing time during an experiment, the program runs in parallel on each transputer. For this purpose, the program was written using the programming language, occam. With the benefits of its distributed system architecture and without the need for any operating system, the program is compact with high modularity. In this paper, we report the system architecture of this diagnostic program.

Power system parallel computation by a transputer network

Abstract The load flow and transient stability analysis are two of the most elementary computation problems that planners and operators must face to ensure power system security. With the growth in size and complexity of interconnected power systems, the electric power system computation, especially the time domain transient stability simulation for security assessment, is becoming more time-consuming or too expensive. There is always a considerable incentive to find better computation schemes and to speed-up electric power system computation. In this paper, we present an efficient and unified parallel algorithm with the coincident parallel Transputer system for both load flow and transient stability studies. The implementation results on an actual Chinese power system are discussed and showed that, the whole developed parallel processing system is very cost-effective to speed-up the power system computation.

AN AIDED GENETIC ALGORITHM FOR MULTIPROCESSOR SCHEDULING

Genetic algorithms have been used for solving the problem of scheduling the tasks represented by a task graph onto parallel computing architectures to minimize the schedule length of the task graph. Due to the random nature of the initial population they however face the local extrema problem which could make the resulting schedules sub-optimal. To minimize this problem, an Aided Genetic Algorithm(AGA) is proposed in this paper, in which a member of the initial population of the Genetic algorithm is obtained from a heuristic pre-scheduler. It is found that the AGA achieves the required convergence in (a) lesser number of iterations, and (b) lesser number of trials in obtaining the near-optimal solution compared to the conventional genetic algorithm. The proposed AGA also takes the inter-task communication into account while scheduling. The method is then applied to the problem of optimally scheduling the Kalman filtering algorithm onto a multi-transputer network. The results are experimentally on a network of T-805 transputers.

Parallel transient dynamic non-linear analysis of reinforced concrete plates

This article describes how parallel processing may improve the computational efficiency of the transient dynamic non-linear analysis of reinforced concrete plates subjected to blast or seismic loading. A parallel scheme for the time marching procedure is presented using the explicit Newmark’s algorithm. The finite element formulation with the material modelling (of a strain-rate sensitive hardening–softening, elasto–viscoplastic model accounting for cracking and crushing) is described. The complexity of the material laws and their modelling, induced by the extreme loading conditions, requires vast computational effort per time step. It will be shown that very high computational efficiency may be obtained by decomposing the finite element mesh into a number of sub-domains for distributed analysis on multiple processors. This high efficiency is achieved using distributed memory MIMD architectures (including transputer networks), where the routing of the messages represents significant overhead in terms of inter-processor communication because of limitations resulting from the processor network topology. Through examples it will be demonstrated how this efficiency depends on the problem size (i.e. the level of refinement of the problem idealisation), the number of sub-domains and the status of the analysis with regards to the states of the material.

Design of a fuzzy walking pattern (FWP) for a shape memory alloy (SMA) biped robot

The Shape Memory Alloy (SMA) is a device which is lightweight and small in volume. The SMA can be used as the actuator of a micro-robot, but it is difficult to design a controller to handle the highly nonlinear properties of the SMA. In this paper, a Fuzzy Walking Pattern (FWP) is proposed to control a small biped robot, using an SMA as the actuator. In fact, the desired walking pattern of the small biped robot is used to construct the FWP. The proposed FWP can control the biped robot under the desired walking pattern, and handle the exceptional case when the biped robot is subject to disturbance. The proposed FWP not only solves the control problem of the SMA, but also provides a new method in controller design of the biped robot. In addition, a transputer network is designed to impelement the FWP. Experimental results demonstrate the functions of the FWP.

An active telescope secondary mirror control system

One of the main constraints for a modern astronomical telescope is the active control of the secondary mirror, mainly for the correction of the decentering coma and defocus induced by thermomechanical distortions. The mirror movements should be smooth, as precise as the optical design requires, and restricted to within predefined limits. It should be possible to perform mirror alignment and focus corrections without pausing the exposure (online control). The Galileo telescope achieves all this by using a support structure driven via six actuator bars (a hexapod system or Stewart platform) and a real-time control system based on a transputer network that allows parallel control of each actuator. Both the hexapod secondary support and the control system have been built and tested at the telescope. The results show that the errors introduced during mirror positioning lead to optical aberrations well below the diffraction figure of the telescope, and the systems work smoothly enough to allow online control.

Transputer network for data acquisition in a multichannel FADC system

Transputer network for data acquisition in a multichannel FADC system

Evaluating computation/communication balance in an object recognition task

Abstract I present performance measurements for some compute- and communications-intensive image processing tasks required by an object recognition application. The application was implemented as a task farm on a network of transputers, which still are one of the cleanest and best-balanced building blocks for parallel systems. I report on relevant properties of the basic algorithms the application is comprised of; some relevant details of the task farm software; performance measurements on a range of system sizes for tasks of very different granularity and characteristics (including broadcast of global data) with measured loads on processors and communications links; and the detrimental effects on performance of hardware variants that increase communications latency and reduce available bandwidth. The measurements show that for some of the image processing tasks, the transputer system is `on the edge' with respect to available bandwidth and latency. Some possible improvements to the communications infrastructure are discussed in light of these results. Finally, current approaches to affordable parallel computing, such as networks or clusters of workstations, are put into perspective by comparing them with the transputer system, using computation/communication balance as a figure of merit for the comparison. This shows that for current microprocessors, even their pin bandwidth is not sufficient to sustain equivalent (scaled) performance for this typical image processing application; any multiprocessor system built with current networking hardware is so unbalanced compared to the transputer system that it will be severely limited by communications bandwidth and latency.

PARCIS: a robust parallel VLSI circuit simulator

The accurate verification of VLSI circuits is essential for their successful and economic production but is an extremely time consuming process for the large circuits of today. This paper describes a robust parallel circuit simulator, PARCIS, designed for a message passing multiprocessing system. It uses a demand driven technique, based on the analysis of hierarchically partitioned circuits. The computation time is reduced by decoupling the circuit equations and distributing the computational load over many processors. On each processor, the circuit blocks, compacted in hierarchical levels, are analyzed asynchronously according to their temporal activity. Currently the PARCIS system is running on a network of transputers. To demonstrate the effectiveness of the proposed simulation program, results are presented for the simulation of typical digital circuits, showing that the execution time decreases in a constant rate as the number of processors (transputers) increases.

Rapid prototyping of distributed algorithms

Rapid prototyping is expected to be especially effective in designing, evaluating, and tuning new software whose real-time performance and quality of service are difficult to determine analytically. Distributed algorithms and protocols fall into this category of software. In this paper we present the design and implementation of a rapid prototyping system for distributed algorithms. The system provides a library of interprocess communication routines and supports dynamic reconfiguration of network topology, automatic process loading and performance monitoring of the prototyped algorithms. We discuss our experiences in developing the rapid prototyping system on a transputer network and describe examples of using the system.

Compressing cube‐connected cycles and butterfly networks

We consider the simulation of large cube-connected cycles (CCC) and large butterfly networks (BFN) on smaller ones, a problem that arises when algorithms designed for an architecture of an ideal size are to be executed on an existing architecture of a fixed size. We show that large CCCs and BFNs can be embedded into smaller networks of the same type with (a) dilation 2 and optimum load, (b) dilation 1 and optimum load in most cases, and (c) dilation 1 and nearly optimum load in all cases. Our results show that large CCCs and BFNs can be simulated very efficiently on smaller ones. Additionally, we implemented our algorithm for compressing CCCs and ran several experiments on a Transputer network, which showed that our technique also behaves very well from a practical point of view. © 1998 John Wiley & Sons, Inc. Networks 32: 47–65, 1998

Compressing cube-connected cycles and butterfly networks

We consider the simulation of large cube-connected cycles (CCC) and large butterfly networks (BFN) on smaller ones, a problem that arises when algorithms designed for an architecture of an ideal size are to be executed on an existing architecture of a fixed size. We show that large CCCs and BFNs can be embedded into smaller networks of the same type with (a) dilation 2 and optimum load, (b) dilation 1 and optimum load in most cases, and (c) dilation 1 and nearly optimum load in all cases. Our results show that large CCCs and BFNs can be simulated very efficiently on smaller ones. Additionally, we implemented our algorithm for compressing CCCs and ran several experiments on a Transputer network, which showed that our technique also behaves very well from a practical point of view.

Portable parallel FFT for MIMD multiprocessors

A portable parallelization of the Cooley–Tukey FFT algorithm for MIMD multiprocessors is presented. The implementation uses the virtual machine for multiprocessors (VMMP) and PVM portable software packages. Since VMMP provides the same set of services on all target machines, a single version of the parallel FFT code was used for shared memory (25-processor Sequent Symmetry), shared bus (MOS-running distributed UNIX) and distributed memory multiprocessor (transputer network and 64-processor IBM SP2). It is accompanied with detailed performance analysis of the implementations. The algorithm achieved high efficiencies on all target machines. The analysis indicates that most overheads are caused by the target architecture and not by VMMP or PVM inefficiencies. The portability analysis of the FFT provides several important insights. On the message passing architecture, the parallel FFT algorithm can obtain linearly increasing speedup with respect to the number of processors with only a moderate increase in the problem size. The parallel FFT can be executed by any number of processors, but generally the number of processors is much less than the length of the input data. The results indicate that the parallel FFT is portable: it achieves very good speedups on either a shared memory multiprocessor with high memory bandwidth or on a message passing multiprocessor without any change in the programs. © 1998 John Wiley & Sons, Ltd.

Vector and Parallel Computing Comparisons for Block Pre-Conditioned Conjugate Gradient

This work deals with the Block Preconditioning for the Conjugate Gradient Method on block tridiagonal systems of linear equations. The results obtained by using Transputer networks, IBM-SP2 and Cray-J90 confrrm the high perfonnance of this method with parallel and vector processing.

Transputer-based human hearing simulation

Simulations of the human hearing system can help in a number of research fields; including work with the speech and hearing impaired as well as improving the accuracy of speech recognition systems and the naturalness of speech synthesis technology. The results from psychoacoustic experiments carried out over the last few decades have enabled models of the human peripheral hearing system to be developed. Conventionally, analyses such as the Fast Fourier Transform are used to analyze speech and other sounds to establish the acoustic cues which are important for human perception. Such analyses can be shown to be inappropriate in a number of ways. Additional insights could be gained into the importance of various acoustic cues if analyses based on hearing models were used. This paper describes an implementation of a real-time spectrograph based on a contemporary model of the peripheral human hearing system, executing on a network of T9000 transputers. The differences between it and conventional spectrographs are illustrated by means of test signals and speech sounds.

Transputer based acquisition/reconstruction system as a development tool for positron emission tomography

We present an acquisition and reconstruction system for positron emission tomography (PET), to be used as a development tool for testing PET reconstruction algorithms and for detector characteristics studies. It is based on a transputer network and modular electronics: acquired data are stored in memories associated with parallel processors and offline data analysis is performed, allowing for fast comparison of performances of different reconstruction algorithms. Its simple architecture and low cost makes it an easy-to-use and convenient tool to test new detectors for PET applications. We introduce here some results obtained with two CdZnTe radiation detectors on a rotating ring, as a prototype example of a small-animal PET camera.

Parallel execution of Prolog with granularity control

This paper presents a system for parallel execution of Prolog supporting both independent conjunctive and disjunctive parallelism. The system is intended for distributed memory architecture and is composed of a set of workers with a hierarchical structure scheduler. The execution model has been designed in such a way that each worker's environment does not contain references to terms in other environments, thus reducing communication overhead. In order to guarantee the improvement of the performance by the parallelism exploitation, a granularity control has been introduced for each kind of parallelism. For conjunctive parallelism PDP applies a control based on the estimation provided by CASLOG. The features of the system allow to introduce this control without adding overhead. For disjunctive parallelism PDP controls granularity by applying a heuristic-based method, which can be adapted to other parallel Prolog systems. Different scheduling policies have also been tested. The system has been implemented on a transputer network and performance results show that it provides a high speedup for coarse grain parallel programs.