| 1. |
- Bengtsson, Lars, 1958-, et al.
(författare)
-
A high-performance embedded massively parallel processing system
- 1994
-
Ingår i: Proceedings of the First International Conference on Massively Parallel Computing Systems (MPCS) The Challenges of General-Purpose and Special-Purpose Computing. - Piscataway, N.J. : IEEE. - 0818663227 - 9780818663222 ; , s. 201-206
-
Konferensbidrag (refereegranskat)abstract
- A need to apply the massively parallel computing paradigm in embedded real-time systems is foreseen. Such applications put new demands on massively parallel systems, different from those of general purpose computing. For example, time determinism is more important than maximal throughput, physical distribution is often required, size, power, and I/O are important, and interactive development tools are needed. The paper describes an architecture for high-performance, embedded, massively parallel processing, featuring a large number of nodes physically distributed over a large area. A typical node has thousands of processing elements (PEs) organized in SIMD mode and is the size of the palm of a hand, Intermodule communication over a scalable optical network is described. A combination of wavelength division multiplexing (WDM) and time division multiplexing (TDM) is used. © 1994 IEEE.
|
|
| 2. |
- Bengtsson, Lars, 1958-, et al.
(författare)
-
A processor array module for distributed, massively parallel, embedded computing
- 1993
-
Ingår i: Microprocessing and Microprogramming. - Amsterdam : Elsevier. - 0165-6074. ; 38:1-5, s. 529-537
-
Tidskriftsartikel (refereegranskat)abstract
- With the increased degree of miniaturization resulting from the use of modem VLSI technology and the high communication bandwidth available through optical connections, it is now possible to build massively parallel computers based on distributed modules which can be embedded in advanced industrial products. Examples of such future possibilities are ''action-oriented systems'', in which a network of highly parallel modules perform a multitude of tasks related to perception, cognition, and action. The paper discusses questions of architecture on the level of modules and inter-module communication and gives concrete architectural solutions which meet the demands of typical, advanced industrial real-time applications. The interface between the processors arrays and the all-optical communication network is described in some detail. Implementation issues specifically related to the demand for miniaturization are discussed.
|
|
| 3. |
- Bengtsson, Lars, 1958-, et al.
(författare)
-
Brains for Autonomous Robots : Hardware and Surgery Tools
- 1994
-
Ingår i: Proceedings of PerAc '94. From Perception to Action. - Los Alamitos : IEEE. - 0818664827 - 9780818664823 ; , s. 436-439
-
Konferensbidrag (refereegranskat)abstract
- This paper presents a hardware architecture and a software tool needed for future autonomous robots. Specific attention is given to the execution of artificial neural networks and to the need for a good inspection and visualization tool when developing this kind of systems. Achievable performance using state-of-the-art technology is estimated and module miniaturization issues are discussed. © 1994 IEEE.
|
|
| 4. |
- Bengtsson, Lars, 1958-, et al.
(författare)
-
REMAP massively parallel computer platform for neural computations
- 1997
-
Ingår i: Proceedings of the Third International Conference on Microelectronics for Neural Networks (MicroNeuro’93). ; , s. 47-62
-
Konferensbidrag (refereegranskat)abstract
- The REMAP project addresses questions related to the use of massively parallel, distributed computing in embedded systems. Of specific interest is the execution of artificial neural network algorithms on multiple, cooperating processor arrays. This paper concentrates on the recently finished, and currently used, processor array prototype, REMAP-β, of SIMD (Single Instruction stream, Multiple Data streams) type. The architecture and implementation of the computer is described, both its overall structure and its constituent parts. Following this comes a discussion of its use as an architecture laboratory, which stems from the fact that it is implemented using FPGA (Field Programmable Gate Array) circuits. As an architecture laboratory the prototype can be used to implement and evaluate, e.g., various Processing Element (PE) designs. A couple of examples of PE architectures, including one with floating-point support, are given. The mapping of important neural network algorithms on processor arrays of this kind is shown, and possible tuning of the architecture to meet specific processing demands is discussed. Performance figures are given as well as implications for future VLSI implementations of the array.
|
|
| 5. |
- Boström, Ola, 1963, et al.
(författare)
-
A new neck injury criterion candidate-based on injury findings in the cervical spinal ganglia after experimental neck extension trauma
- 1996
-
Ingår i: PROCEEDINGS OF THE 1996 INTERNATIONAL IRCOBI CONFERENCE ON THE BIOMECHANICS OF IMPACT, SEPTEMBER 11-13, DUBLIN, IRELAND. ; , s. 123-136
-
Konferensbidrag (refereegranskat)abstract
- In this study a mathematical model, based on Navier Stokes equations, was developed and validated against experimental data. This model predicts the pressure changes in the spinal canal as a function of the volume change inside the canal during neck bending in the x-z (sagittal) plane. Another aim of the study was to investigate pressure phenomena and ganglion injuries at static neck extension loading and dynamic neck extension trauma with a head-restraint present. Experiments on pigs were conducted. Preliminary results indicate that ganglion injuries, as well as pressure transients inside the spinal canal, seem to correlate to the phase shift when the neck passes an s-shape (or maximal retraction) during the rearward motion of the head. That is, when the upper neck quickly changes from a flexion to an extension shape. Static loading of the neck resulted in no signs of injuries to the ganglia. A possible candidate for a neck injury criterion is presented, based on the relative acceleration between the top and the bottom of the cervical spine. A tolerance level based on the pig tests is also discussed.
|
|
| 6. |
- Boström, Ola, 1963, et al.
(författare)
-
Prediction of neck injuries in rear impacts based on accident data and simulations
- 1997
-
Ingår i: PROCEEDINGS OF THE 1997 INTERNATIONAL IRCOBI CONFERENCE ON THE BIOMECHANICS OF IMPACT. ; , s. 251-264
-
Konferensbidrag (refereegranskat)abstract
- Whiplash associated disorders, occurring in car accidents, are an increasing problem worldwide. According to real-life data from police records, the struck car's velocity change (delta V) and occupant gender are two of the most important factors related to Abbreviated Injury Scale (AIS) 1 neck injuries. In this study, a new rear-impact ranking of cars based on 4432 police reported accidents is presented. The ranking concerns the relative neck injury risk and compensates for the influences of car weight and gender. Moreover, some important factors influencing the risk of AIS 1 neck injury are proposed. These include: the stiffness, damping and yielding characteristics of the seat back, the muscle response of the occupant, and the delta V of the struck car and acceleration pulse. Using a mathematical model it is shown that the influence from these factors can be explained by a recently proposed neck injury criterion (NIC). This criterion is based on the neck motion at the passage of full neck retraction. The NIC, based on a number of volunteer tests, is analysed and validated. The consequence of injury outcome of an observed overall seat back stiffening is also discussed. In conclusion, for delta V below 20 km/h, real-life data show that the geometry of the head restraint is of minor importance. A seat back with low yielding limit or soft performance may be preferable. Moreover, the new NIC seems to be a good predictor of real-life neck injuries.
|
|
| 7. |
|
|
| 8. |
- Jonsson, Magnus, 1969-, et al.
(författare)
-
A fiber-optic interconnection concept for scaleable massively parallel computing
- 1995
-
Ingår i: Proceedings of the Second International Conference on Massively Parallel Processing Using Optical Interconnections. - Piscataway, NJ. : IEEE. - 0818671017 ; , s. 313-320
-
Konferensbidrag (refereegranskat)abstract
- One of the most important features of interconnection networks for massively parallel computer systems is scaleability. The fiber-optic network described in this paper uses both wavelength division multiplexing and a configurable ratio between optics and electronics to gain an architecture with good scaleability. The network connects distributed modules together to a huge parallel system where each node itself typically consists of parallel processing elements. The paper describes two different implementations of the star topology, one uses an electronic star and fiber optic connections, the other is purely optical with a passive optical star in the center. The medium access control of the communication concept is presented and some scaleability properties are discussed involving also a multiple-star topology.
|
|
| 9. |
- Jonsson, Magnus, 1969-, et al.
(författare)
-
Fiber-ribbon pipeline ring network for high-performance distributed computing systems
- 1997
-
Ingår i: Proceedings of the Third International Symposium on Parallel Architectures, Algorithms, and Networks, 1997. (I-SPAN '97). - : IEEE. ; , s. 138-143
-
Konferensbidrag (refereegranskat)abstract
- In this paper, we propose a high-bandwidth ring network built up with fiber-ribbon point-to-point links. The network has support for both packet switched and circuit switched traffic. Very high throughputs can be achieved in the network due to pipelining, i.e., several packets can be traveling through the network simultaneously but in different segments of the ring. The network can be built today using fiber-optic off-the-shelf components. The increasingly good price/performance ratio for fiber-ribbon links indicates a great success potential for the proposed kind of networks. We also present a massively parallel radar signal processing system with exceptionally high demands on the communication network. An aggregated throughput of tens of Gb/s is needed in this application, and this is achieved with the proposed network.
|
|
| 10. |
- Jonsson, Magnus, 1969-
(författare)
-
High performance fiber-optic interconnection networks for real-time computing systems
- 1999
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- Parallel and distributed computing systems become more and more powerful and hence place increasingly higher demands on the networks that interconnect their processors or processing nodes. Many of the applications running on such systems, especially embedded systems applications, have real-time requirements and, with increasing application demands, high-performance networks are the hearts of these systems. Fiber-optic networks are good candidates for use in such systems in the future. This thesis contributes to the relatively unexplored area of fiber-optic networks for parallel and distributed real-time computer systems and suggests and evaluates several fiber-optic networks and protocols. Two different technologies are used in the networks, WDM (Wavelength Division Multiplexing) and fiber-ribbon point-to-point links. WDM offers multiple channels, each with a capacity of several Gbit/s. A WDM star network in which protocols and services are efficiently integrated to support different kinds of real-time demands, especially hard ones, has been developed. The star-of-stars topology can be chosen to offer better network scalability. The WDM star architecture is attractive but its future success depends on components becoming more commercially mature. Fiber-ribbon links, offering instead an aggregated bandwidth of several Gbit/s, have already reached the market with a promising price/performance ratio. This has motivated the development and investigation of two new ring networks based on fiber-ribbon links. The networks take advantage of spatial bandwidth reuse, which can greatly enhance performance in applications with a significant amount of nearest downstream neighbor communication. One of the ring networks is control channel based and not only has support for real-time services like the WDM star network but also low level support for, e.g., group communication. The approach has been to develop network protocols with support for dynamic real-time services, out of time-deterministic static TDMA systems. The focus has been on functionality more than pure performance figures, mostly on real-time features but also on other types of functionality for parallel and distributed systems. Worst-case analyses, some simulations, and case studies are reported for the networks. The focus has been on embedded supercomputer applications, where each node itself can be a parallel computer, and it is shown that the networks are well suited for use in the radar signal processing systems studied. Other application examples in which these kinds of networks are valuable are distributed multimedia systems, satellite imaging and other image processing applications.
|
|
| 11. |
- Jonsson, Magnus, 1969-, et al.
(författare)
-
On inter-cluster communication in a time-deterministic WDM star network
- 1997
-
Ingår i: Proc. 2nd Workshop on Optics and Computer Science (WOCS), Geneva, Switzerland, Apr. 1, 1997. ; , s. 10-
-
Konferensbidrag (refereegranskat)abstract
- Future real-time applications requiring massively parallel computer systems also put high demands on the interconnection network. By connecting several WDM star clusters by a backbone star, forming a star-of-stars network, we get a modular high-bandwidth network. In this paper we show how to achieve time-deterministic packet switched communication in such networks, even for inter-cluster communication. An analysis of how the deterministic latency and node bandwidth vary with design parameters is presented. We also propose a general clock-synchronization scheme, improving the worst-case latency with up to 33 percentages.
|
|
| 12. |
- Jonsson, Magnus, 1969-, et al.
(författare)
-
Time-deterministic WDM star network for massively parallel computing in radar systems
- 1996
-
Ingår i: Proceedings of the Third International Conference on Massively Parallel Processing Using Optical Interconnections. - Los Alamitos, California : IEEE Computer Society Press. - 0818675918 ; , s. 85-93
-
Konferensbidrag (refereegranskat)abstract
- In massively parallel computer systems for embedded real-time applications there are normally very high bandwidth demands on the interconnection network. Other important properties are time-deterministic latency and services to guarantee that deadlines are met. In this paper we analyze how these properties vary with the design parameters for a passive optical star network, specifically when used in a massively parallel radar signal processing system. The aggregated bandwidth and computational power of the radar system are approximately 45 Gb/s and 100 GOPS, respectively. The analysis is focused on the medium access control protocol, called TD-TWDMA, for the time and wavelength multiplexed network. It is concluded that the proposed network is very well suited to this kind of signal-processing applications. We also present a new distributed slot-allocation algorithm with real-time properties.
|
|
| 13. |
- Kruusmaa, Maarja, et al.
(författare)
-
A low-risk approach to mobile robot path planning
- 1998
-
Ingår i: Lecture Notes in Computer Science. - Heidelberg : Springer Berlin/Heidelberg. - 0302-9743 .- 1611-3349. ; 1416, s. 132-141
-
Tidskriftsartikel (refereegranskat)abstract
- This paper presents a self-organizing approach for mobile robot path planning problems in dynamic environments by using case-based reasoning together with a more conventional method of grid-map based path planning. The map-based path planner is used to suggest new innovative solutions for a particular path planning problem. The case-base is used to store the paths and evaluate their traversability. While planning the route those paths are preferred which, according to former experience, are least risky. As the environment changes, the exploration as well as the evaluation of the paths will allow the system to self-organize by forming a set of low-risk paths that are safest to follow. The experiments in a simulated environment show that the robot is able to adapt in a dynamic environment and learns to use the least risky paths. © Springer-Verlag Berlin Heidelberg 1998.
|
|
| 14. |
- Nilsson, Kenneth, 1951-, et al.
(författare)
-
A modular, massively parallel computer architecture for trainable real-time control systems
- 1993
-
Ingår i: Control Engineering Practice. - Kidlington, Oxford : Pergamon Press. - 0967-0661 .- 1873-6939. ; 1:4, s. 655-661
-
Tidskriftsartikel (refereegranskat)abstract
- A new system-architectural concept for trainable real-time control systems is based on resource adequacy both in processing and communication. Cyclically executing programs in distributed nodes communicate via a shared high-speed medium. Static scheduling of programs and communication implies that the maximum possible work-load can always be handled in a time-deterministic manner. The use of Artificial Neural Networks (ANN) algorithms and trainability implies a new system development strategy based on a Continuous Development paradigm. An implementation of the Architectural concept is presented. The communication speed is measured in Gbps and the access method is TDMA. An implementation of the system-development strategy is also presented. © 1993.
|
|
| 15. |
|
|
| 16. |
- Nordström, Tomas, 1963-, et al.
(författare)
-
Using and designing massively parallel computers for artificial neural networks
- 1992
-
Ingår i: Journal of Parallel and Distributed Computing. - Orlando : Academic Press. - 0743-7315 .- 1096-0848. ; 14:3, s. 260-285
-
Tidskriftsartikel (refereegranskat)abstract
- During the past 10 years the fields of artificial neural networks (ANNs) and massively parallel computing have been evolving rapidly. The authors study the attempts to make ANN algorithms run on massively parallel computers as well as designs of new parallel systems tuned for ANN computing. Following a brief survey of the most commonly used models, the different dimensions of parallelism in ANN computing are identified, and the possibilities for mapping onto the structures of different parallel architectures are analyzed. Different classes of parallel architectures used or designed for ANN are identified. Reported implementations are reviewed and discussed. It is concluded that the regularity of ANN computations suits SIMD architectures perfectly and that broadcast or ring communication can be very efficiently utilized. Bit-serial processing is very interesting for ANN, but hardware support for multiplication should be included. Future artificial neural systems for real-time applications will require flexible processing modules that can be put together to form MIMSIMD systems
|
|
| 17. |
|
|
| 18. |
- Oskarsson, Åke, et al.
(författare)
-
trans influence on bond distances. Structures of trans-diiodobis(tetrahydrothiophene)palladium(II) and trans-diiodobis(tetrahydrothiophene)platinum(II)
- 1990
-
Ingår i: Acta Crystallographica. Section B: Structural Science, Crystal Engineering and Materials. - 2052-5206. ; 1990:46, s. 748-752
-
Tidskriftsartikel (refereegranskat)abstract
- Crystals of trans-diiodobis(tetrahydrothiophene)-palladium(II) and trans-diiodobis(tetrahydrothiophene)platinum(II) were prepared by dissolving the appropriate metal iodide in tetrahydrothiophene at 353 K and cooling to room temperature. Crystal data at room temperature (Mo K-alpha, lambda = 0.7107-angstrom) are: [PdI2(C4H8S)2], M(r) = 536.54, triclinic, P1BAR, a = 8.4920 (8), b = 8.8248 (1), c = 9.4580 (11) angstrom, alpha = 97.595 (9), beta = 91.963 (9), gamma = 99.482 (9)-degrees, V = 691.8 (1) angstrom-3, Z = 2, D(x) = 2.576 Mg m-3, mu = 5.997 mm-1, F(000) = 496, R = 0.028 for 2090 unique reflections with I > 3-sigma-(I); [PtI2(C4H8S)2], M(r) = 625.24, triclinic, P1BAR, a = 7.889 (2), b = 8.676 (1), c = 10.616 (2) angstrom, alpha = 103.75 (1), beta = 90.36 (1), gamma = 94.70 (2)-degrees, V = 703.2 (2) angstrom-3, Z = 2, D(x) = 2.953 Mg m-3, mu = 14.67 mm-1, F(000) = 560, R = 0.024 for 3988 reflections with I > 3-sigma-(I). Both compounds consist of van-der-Waals-packed mononuclear trans-[MI2(C4H8S)2] complexes. There are two centrosymmetric metal complexes in the asymmetric unit of the Pt compound but only one pseudoplanar complex in the Pd compound. The acceptor-donor distances are Pd-I 2.603 (1), 2.625 (1); Pd-S 2.316 (2), 2.329 (2); Pt-I 2.606 (1), 2.616 (1); and Pt-S 2.309 (1), 2.310 (1) angstrom. A literature survey of compounds in which ligand-I or -S (in thioethers) atoms are trans to themselves shows average M-I and M-S distances of Pd-I 2.605 (20), Pd-S 2.324 (15), Pt-I 2.609 (9) and Pt-S 2.313 (13) angstrom. The spread of the M-I and M-S distances is significantly increased if complexes are included where I or S are not trans to themselves, indicating that for some donor atoms the trans influence dominates over other factors such as the cis influence and packing forces.
|
|
| 19. |
- Svensson, Bertil, et al.
(författare)
-
Execution of neural network algorithms on an array of bit-serial processors
- 1990
-
Ingår i: Proceedings, 10th International Conference on Pattern Recognition. - Piscataway, NJ : IEEE Communications Society. - 0818620625 ; , s. 501-505
-
Konferensbidrag (refereegranskat)abstract
- Large processor arrays are candidates for performing computations of neural network models at speeds required for real time applications, e.g. in pattern recognition. The paper gives a general model of an array of bit-serial processors and demonstrates the mapping of neural net models on such an array. The approach maps a neuron on each processing element and makes communication all-to-all available by connection weight matrices. The required communication structure is very simple. The bit-serial approach allows trade-offs between speed and precision, even dynamically. Performance figures are given. A bitserial multiplier is an important part of the design. Implementation aspects are discussed and it is shown that a one-board realization of a 1024 processor system is feasible with current, commonly available, technology.
|
|
| 20. |
- Svensson, Mats, 1960, et al.
(författare)
-
Nervenzellschäden bei Schleudertraumen. Tierexperimentelle Untersuchungen
- 1998
-
Ingår i: Orthopäde. ; :27, s. 820-826
-
Tidskriftsartikel (refereegranskat)abstract
- Die mechanische Belastung der Halswirbelsäule (HWS) bei Autounfällen verursacht ofteine Reihe von Halsverletzungen, die unterdem Begriff Schleudertrauma zusammengefaût werden. Mehrere dieser Symptomekönnen möglicherweise durch Verletzungenim Bereich der Nervenwurzeln der HWS erklärt werden. Die Hypothese wurde aufgestellt,daû während der Extensions- und Flexionsbewegungendes Halses, aufgrund vonhämodynamischen Auswirkungen, ¾nderungendes inneren Volumens des Halswirbelkanalskurzzeitige Druckveränderungenim ZNS verursachen und daû dadurch Gewebeschäden durch die mechanische Belastungder Nervenwurzel entstehen. Um dieHypothese zu überprüfen, wurden anästhesierteSchweine einem experimentellenSchleudertrauma mit Extensions-, FlexionsundSeitneigungsbewegungen ausgesetzt.Die traumatische Belastung wurde unterhalbeiner Stufe gehalten, bei der eine Halswirbelfrakturstattfinden könnte. Während derBelastung wurde der Pulsdruck im Halswirbelkanalgemessen. In diesem Zusammenhangergaben sich Hinweise einer Dysfunktionder Zellmembran bei Nervenzellkörpernder Spinalganglien. Die ganglionären Verletzungenkönnen möglicherweise einige dermit Weichteilverletzungen des Halses nachAutounfällen einhergehenden Symptomeerklären. Um die Situation bei einem Auffahrunfallzu simulieren, wurde der Kopf desSchweines rückwärts vom Torso weggezogen.Dabei wurde festgestellt, daû die ganglionären Verletzungen zu einem sehr frühenZeitpunkt während der Halsbewegungenstattfinden und zwar in der Phase, in der dieBewegung von der Retraktion zur Extensionwechselt. Bei einer ähnlichen, jedoch statischenBelastung des Halses wurden keineganglionären Verletzungen bei den Schweinenfestgestellt. Dies ist ein Hinweis darauf,daû diese Verletzungen durch dynamischeFaktoren verursacht werden und bietet somitweitere Unterstützung für die Druckhypothese.Auf der Basis eines theoretischen Modellswurden Kriterien für Halsverletzungen(neck injury criterion = NIC) entwickelt. Esweist darauf hin, daû das Risiko von ganglionären Verletzungen am Punkt der maximalenHalsretraktion durch die unterschiedlichehorizontale Beschleunigung und Geschwindigkeitzwischen Kopf und oberemTorso bestimmt wird.
|
|
| 21. |
|
|
| 22. |
|
|
| 23. |
- Svensson, Mats Y., 1960, et al.
(författare)
-
Pressure Effects in the Spinal Canal During Whiplash Extension Motion - A Possible Cause of Injury to the Cervical Spinal Ganglia
- 1993
-
Ingår i: Proceedings of the International Research Committee on the Biomechanics of Impacts (IRCOBI) conference, Eindhoven, Netherlands. ; , s. 189-200
-
Konferensbidrag (refereegranskat)abstract
- impact-velocities (<20 km/h) often cause pain in the neck region as well as a number of other neurological symptoms, most of which can be related to the nerve paths that pass through the cervical intervertebral foramina. When the neck is flexed or extended in the sagittal plane the length of the cervical spinal canal alters but the cross-sectional area of the canal remains almost constant. During flexion-extension motion of the cervical spine, the size of the inner volume of the spinal canal will change. Since the tissues inside the canal can be considered incompressible, an alteration will take place of either the amount of cerebro spinal fluid or the amount of blood in the veinplexa of the epidural space, or both. This requires fluid transportation through the intervertebral foramina as well as along the spinal canal. During a whiplash extension motion, the flow velocity can be expected to rise far above physiologically normal levels and pressure gradients can thus be expected to occur. In turn, the soft tissues inside and around the cervical spine and particularly in the intervertebral foramina will sustain mechanical strain and stress. Anaesthetised pigs were exposed to a swift extension-flexion motion of the neck while the pressure inside the spinal canal and the skull was measured. Pressure pulses of magnitudes up to 150 mmHg (20 kPa) were observed during the motion. The magnitude of pressure is for each moment dependent on the position of the neck, the velocity and the acceleration of the motion. Plasma membrane dysfunction was indicated by the results from light microscopic analyses of the cervical and the three upper thoracic spinal ganglia revealing the staining of nerve cells and satellite cells by protein complexed to the Evans Blue dye.
|
|
| 24. |
- Taveniku, Mikael, et al.
(författare)
-
A multiple SIMD mesh architecture for multi-channel radar processing
- 1996
-
Ingår i: Proceedings of: ICSPAT'96, international conference on signal processing applications & technology, Boston MA, USA, October 7-10. - : Miller Freeman. ; , s. 1421-1427
-
Konferensbidrag (refereegranskat)abstract
- In modern and future radar applications there are high demands on the signal processing chain in terms of computational power and generality. At the same time, there are hard size and power consumption constraints. This paper reports on a project whose aim is to find a good scalable computer architecture that is flexible, programmable and as general-purpose as possible without too much performance loss.The proposed architecture consists of multiple SIMD computing modules, each based on a number of small mesh arrays. The modules are fully programmable and work globally as a MIMD machine and locally as SIMD machines. The data network is modular and provides both high bandwidth capacity and fast response. It has a fiber-optic stars topology, and employs time and wavelength division multiplexing, together with a medium access method specially developed for real-time systems.In this paper, we use a radar system with 64 processing channels to illustrate the algorithms and the usage of the processor modules. We show that it is possible to use a machine, consisting of small mesh processor arrays forming larger modules, with good efficiency. The building blocks show good balance between computational power and I/O bandwidth, and the SIMD approach seems good from algorithm-mapping point of view.
|
|
| 25. |
- Taveniku, Mikael, et al.
(författare)
-
The VEGA moderately parallel MIMD, moderately parallel SIMD, architecture for high performance array signal processing
- 1998
-
Ingår i: Proceedings of the first merged International Parallel Processing Symposium & Symposium on Parallel and Distributed Processing. - Los Alamitos, Calif. : IEEE Computer Press. ; , s. 226-232
-
Konferensbidrag (refereegranskat)abstract
- In array radar signal processing applications, the processing demands range from tens of GFLOPS to several TFLOPS. To address this, as well as the, size and power dissipation issues, a special purpose “array signal processing” architecture is proposed. We argue that a combined MIMD-SIMD system can give flexibility, scalability, and programmability as well as high computing density. The MIMD system level, where SIMD modules are interconnected by a fiber-optic real-time network, provides the high level flexibility while the SIMD module level provides the compute density. In this paper we evaluate different design alternatives and show how the VEGA architecture was derived. By examining the applications and the algorithms used, the SIMD mesh processor is found be sufficient. However, the smaller the meshes are the better is the flexibility and efficiency. Then, based on prototype VLSI implementations and on instruction statistics, we find that a relatively large pipelined processing element maximises the performance per area. It is thereby concluded that the small SIMD mesh processor array with powerful processing elements is the best choice. These observations are further exploited in the design of the single-chip SIMD processor array to be included in the MIMD-style overall system. The system scales from 6.4 GFLOPS to several TFLOPS peak performance.
|
|
| 26. |
- Wickström, Nicholas, et al.
(författare)
-
Estimating pressure peak position and air-fuel ratio using the ionization current and artificial neural networks
- 1997
-
Ingår i: IEEE Conference on Intelligent Transportation Systems. - Piscataway, N.J. : IEEE. - 0780342690 ; , s. 927-977
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- We propose two artificial neural network models which use the ionization current for estimation of the position of the pressure peak and the air-fuel ratio. The pressure peak position model produces estimates on a cycle-by-cycle basis for each of the cylinders. These estimates are twice as good as estimates obtained from a linear model. The air-fuel ratio model uses the universal exhaust gas oxygen sensor as reference; it produces estimates that are ten times better than estimates obtained fi om a linear model.
|
|
| 27. |
- Wickström, Nicholas, 1970-, et al.
(författare)
-
Neural Virtual Sensors — Estimation of Combustion Quality in SI Engines using the Spark Plug
- 1998
-
Ingår i: ICANN 98. - London : Springer. - 9783540762638 - 9781447115991 ; , s. 215-220
-
Konferensbidrag (refereegranskat)abstract
- We propose two virtual sensors which estimate the location of the pressure peak and the air-fuel ratio from measurements of the ionization current across the spark plug gap.The location of pressure peak virtual sensor produces estimates on a cycle-by-cycle basis for each of the cylinders. These estimates are twice as good as estimates obtained from a linear model.The air-fuel ratio virtual sensor uses the universal exhaust gas oxygen sensor as reference; it produces estimates that are ten times better than estimates obtained from a linear model.
|
|
| 28. |
- Zuccarello, Guido, et al.
(författare)
-
HIV-1 protease inhibitors based on acyclic carbohydrates
- 1998
-
Ingår i: Journal of Organic Chemistry. - : American Chemical Society (ACS). - 0022-3263 .- 1520-6904. ; 63:15, s. 4898-4906
-
Tidskriftsartikel (refereegranskat)abstract
- A series of acyclic C2-symmetric HIV protease inhibitors readily accessible from D-mannitol have been developed. Several of the compounds synthesized showed significant in vitro activity against HIV-1 protease.
|
|
| 29. |
- Åhlander, Anders, et al.
(författare)
-
A multiple SIMD approach to radar signal processing
- 1996
-
Ingår i: 1996 IEEE TENCON. - Piscataway, NJ : IEEE Press. - 0780336801 - 0780336798 ; , s. 852-857
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- Next generation radar systems, with phase-controlled array antennas, will have to process data that is many times larger than in current systems. This requires an enormous computing power. Even in a relatively small airborne radar system, with hard size and power consumption constraints, a sustained computing power of 40 GOPS (or 40 GFLOPS, if floating point calculations are used) will be needed to perform only the subset of the calculations known as the space-time adaptive processing, STAP Consequently, there is a need for new parallel computing modules, as well as new overall system architectures and application development environments. In this paper, a modular architecture with highly parallel SIMD-modules is presented as a promising solution, capable of handling STAP. A version of the architecture, equipped with bit-serial floating point PEs, is described and evaluated. Implementation technology aspects are discussed.
|
|
