Showing papers on "Pipeline (computing) published in 1973"

Pipeline signal processor

Abstract: A signal processor including a programmable arithmetic controller and a pipeline arithmetic unit controlled by such controller is disclosed. The arithmetic unit includes a plurality of serially coupled processing levels. The arithmetic controller includes a corresponding plurality of serially coupled control levels, each one of such control levels being coupled to a corresponding one of the processing levels. Each one of the processing levels passes digital data applied thereto in accordance with a control instruction applied to such processing level by the arithmetic controller. As data passes through the various processing levels, the control instruction associated with such data passes through the corresponding control level so that such control instruction ''''follows'''' such data as both data and control instruction pass through the processor. In this way the processor is adapted to start a new process concurrently as such processor completes a prior process.

Apparatus for image processing

Abstract: A special purpose pipeline digital computer is disclosed for processing a pair of related, digitally encoded, images to produce a difference image showing any dissimilarities between the first image and the second image. The computer is comprised of a number of special purpose pipeline processors linked to a supervisory general purpose processor. First, a initial image warp transformation is computed by a spatial transformation pipeline processor using a plurality of operator selected, feature related, match points on the pair of images, and then, image correlation is performed by a dot product processor working with a square root and divide processor to identify the exact matching location of a second group matching points selected in a geometrical pattern, on the pair of images. The final image warp transformation to achieve image registration occurs in the spatial transformation processor, using a localized polylateral technique having the geometrically selected match points as the vertrices of the polylaterals. Finally, photoequalization is performed and the difference image is generated from the pair of registered images by a photoequalization processor.

A parallel variable metric optimization algorithm

Abstract: An algorithm, designed to exploit the parallel computing or vector streaming (pipeline) capabilities of computers is presented. When p is the degree of parallelism, then one cycle of the parallel variable metric algorithm is defined as follows: first, the function and its gradient are computed in parallel at p different values of the independent variable; then the metric is modified by p rank-one corrections; and finally, a single univariant minimization is carried out in the Newton-like direction. Several properties of this algorithm are established. The convergence of the iterates to the solution is proved for a quadratic functional on a real separable Hilbert space. For a finite-dimensional space the convergence is in one cycle when p equals the dimension of the space. Results of numerical experiments indicate that the new algorithm will exploit parallel or pipeline computing capabilities to effect faster convergence than serial techniques.

Method and apparatus for pipeline inspection

Abstract: A pipeline inspection system including a camera having an annular illumination source surrounding the lens so that the illumination is concentrated on the pipeline sidewalls within the viewing angle of the camera. By use of near-infrared radiation having a wavelength on the order of 900 nanometers, improved operation can be obtained with GaAs emitters and a vidicon having a silicon target. The camera and its illumination source is adapted to be towed through the pipeline, and an instantaneous picture produced on an above-ground video display device.

Pipeline leak detector

Abstract: A leak detection and location system for a pipeline carrying a thermally unstable fluid is disclosed which includes an electric circuit along the pipeline with temperature sensitive elements spaced along the pipeline, each set to produce a unique signal in the circuit when a temperature characteristic of, or influenced by, leaking fluid is sensed and means to transmit that unique signal to an alarm. There is also disclosed a trough or canopy containing elements of the leak detection and location system, said trough to be buried with the pipeline in a position to collect leaking fluid.

Pipeline signalling systems and techniques

Abstract: An AC coil is located externally of a pipeline with the pipeline passing other than through the core of the coil and with at least a part of the imaginary surface created by projecting the coil parallel to its longitudinal axis intersecting the pipeline. Alternating current is supplied to the coil. The resultant magnetic flux induces eddy currents to flow in the pipeline. A flux detector located within the pipeline detects the magnetic flux generated by the eddy currents. The flux detector may be carried by a pipeline pig and the detected signal used to locate the pig in the pipeline or control some function of the pig.

Method for controlled pressurization of a pipeline during construction thereof

Abstract: A method for maintaining a positive internal pressure in a pipeline while the pipeline is being constructed which comprises initially positioning a pair of releasable sealing devices in spaced relationship within the pipeline and in a releasable pressure tight engagement with the inner periphery thereof, introducing a pressure fluid into the interior of the pipeline between the devices and behind the first of said devices, depressurizing the interior of the pipeline between the devices; moving the second of said devices within the pipeline to a position in the proximity of the open end of the pipeline wherein construction is taking place and engaging the second device in a pressure tight position with the inner periphery of the pipeline; admitting a pressure fluid into the pipeline between the two devices for equalizing the pressure on the opposite sides of the first device; moving the first device in a direction toward the second device to a position immediately behind the second device and re-engaging the first device in a pressure tight arrangement with the inner periphery of the pipeline; and repeating the process as required during the construction of the pipeline.

Pipeline array for square-root extraction

Abstract: Pipelining an arithmetic process is a well known technique for improving the computation speed of the arithmetic algorithm. In the letter is proposed a pipeline version of the array for the extraction of square roots of binary numbers. It is shown that a significant speed improvement (on a throughout basis) can result by this modification of the conventional logic arrays.

Construction of efficient tree networks: The pipeline problem

Abstract: It is observed that sample cost functions for the pipeline problem [1] are discretely convex, i.e., their graphs are discrete subsets of graphs of convex functions. As a result, it is shown that the pipeline problem may be attacked by using either a minimum cost flow approach, or a combination of dynamic programming and sorting. Problems with concave cost functions are shown to be relatively easy. Even for problems which are neither convex nor concave, it is shown that in certain instances, only a subset of the data for each cost function is relevant. Error bounds are presented when approximations to the original cost functions are used. Results obtained in [4] for a related problem are summarized.

A Review of the Capabilities and Limitations of Parallel and Pipeline Computers

Abstract: Publisher Summary This chapter examines the capabilities and limitations of Parallel and Pipeline Computers, with reference to the University of Illinois' parallel-organized ILLIAC IV and the Control Data Corporation's pipeline organized STAR. The parallel computer, such as the ILLIAC IV, has for its conceptual basis the notion that two conventional computing machines can work at twice the rate of one machine. The major deficiency in this approach is that two machines also cost twice as much as one machine. To overcome this fundamental drawback, the parallel design economizes by having many identical copies of the conventional computer's arithmetic unit driven with only one control unit. A quite different approach underlies the design of the pipeline processor, such as the STAR, which is being developed by the Control Data Corporation. The logical control processes that must take place for the number stream in the pipeline to flow smoothly are sufficiently complex that the foreseeable pipeline machines will permit only one type of arithmetic operation per stream and only one stream at a time.

An algorithmic approach to scheduling a multi-product pipeline system

Abstract: The pipeline system with which this paper is concerned transports a act of various oil products produced by a number of gas plants located along the main pipeline. Each gas plant has a storage tank for each product it produces. The problem is to determine a feasible sequence in which the various products are pumped within each schedule period such that the interface loss is minimized. This paper proposes an algorithmic approach in which suitable dispatching rules are developed. The functions of this algorithm are to : (1) determine the sequence and batch size of products flowing in the pipeline ; (2) update the storage tanks and pipeline status after each pumping ; and (3) identify the type, identify the number, and compute the volume of interfaces created during the schedule period. A comparative evaluation of both existing and proposed procedures are conducted for seven consecutive production periods (600 hours). The criterion considered is the ratio of interface volume created to the total volume of pr...