Argy Krikelis

In this paper we propose the use of Timed Coloured Petri Nets for the Performance Evaluation of Hardware/Software systems for DSP applications. Complex systems on chip, composed by hardware and software parts, are often required to meet strict timing constraints, both in terms of throughput and latency. However, the verification of the suitability...

In this paper, we present an efficient method for supporting full interactive functions of MPEG-2 video streaming. The method is based on storing multiple differently encoded versions of the video stream at the server. A normal version is used for normal playback, while several other versions are used for fast/slow/jump forward (rewind)/rewind at v...

In this paper, we present an efficient method for supporting full interactive functions of MPEG-2 video stream. This method is based on storing multiple different encoded versions of the same movie at the server. A normal version is used for normal playback, while several interactive versions are used for Fast/Slow/Jump Play. Interactive versions a...

Over the past few years, technology drivers for processor designs have changed significantly. Media data delivery and processing -- such as telecommunications, networking, video processing, speech recognition and 3D graphics -- is increasing in importance and will soon dominate the processing cycles consumed in computer-based systems. This paper de...

Over the past few years, technology drivers for microprocessors have changed significantly. Media data delivery and processing--such as telecommunications, networking, video processing, speech recognition and 3D graphics--is increasing in importance and will soon dominate the processing cycles consumed in computer-based systems. This paper presents...

Media data delivery and processing such as telecommunications, networking, video processing, speech recognition and 3D graphics is increasing in importance and will soon dominate the processing cycles consumed in computer-based systems. This paper describes a processor called Linedancer, that provides high media performance with low energy consumpt...

Over the past few years, technology drivers for microprocessors have changed significantly. Media data delivery and processing--such as telecommunications, networking, video processing, speech recognition and 3D graphics--is increasing in importance and will soon dominate the processing cycles consumed in computer-based systems. This paper presents...

The success of future generations of wireless technology will depend on their capability to provide reliable access to multimedia networks. Methods to diminish unpredictable channel errors in high-resolution applications, such as video telephony, are likely to dominate future research and development activities. Existing approaches based on intra-f...

The simultaneous evolution of mobile communications and multimedia applications points toward a future where both technologies play a significant role in shaping mobile multimedia communication. However, the destiny of mobile communications and, to a great extent, of sophisticated mobile multimedia, heavily depends on the success of the emerging th...

In multimedia and hypermedia systems, the importance of mobile links to a stationary infrastructure and of mobile data terminals is increasing. Current advances in hardware technology for hand-held terminals (for example, display size and resolution, processing capabilities, and power consumption) let multimedia systems be implemented in mobile env...

The ultimate goal of mobile multimedia systems is to assist users all the time and everywhere by providing the right information at the right place in the right manner. With such assistance, users will not have to waste effort in interacting and interfacing with the mobile device, but will be able to dedicate their attention to the actual environme...

We show that the latest version of massively parallel processing associative string processing architecture (System-V) is applicable for fast Monte Carlo simulation if an effective on-processor random number generator is implemented. Our lagged Fibonacci generator can produce 10 ⁸ random numbers on a processor string of 12 K PE-s. The ti...

Massively parallel architectures require decentralized and scalable system policies. The paper presents and evaluates a set of local dynamic load balancing strategies inspired by diffusion and characterized by different scopes of locality: the goal is ...

The greatest challenges for the Infoverse and a global information visualization environment concern the inherent limitations of mobile data appliances. The biggest limitations are low communication bandwidth, limited computing resources (in terms of storage capacity, computing power, and data entry and presentation capabilities) for complex inform...

In large-scale multimedia applications, such as video on demand (VoD), interactive television (ITV), and Web browsers, the sheer number of users and the data size of the systems exacerbate the difficulty of supporting continuous-time presentation. For example, a single video stream requires between 1.5 and 6 Mbps of bandwidth. At 6 Mbps, the typica...

Although the development and use of Internet multimedia applications are increasing, the ability to manipulate and process multimedia information on the Internet is missing. The Internet's rich connectivity makes it an ideal blueprint for a high-performance computing system: millions and millions of heterogeneous computing nodes that can support op...

The author discusses the management of multimedia information with classical indexing and semiological indexing. When dealing with very large amounts of data, as is the case with multimedia data streams, the only way to extract useful information is to process as much of the data in parallel as possible. The author discusses the role of parallel pr...

Not Available

Not Available

Parallel and distributed high-performance computing will be generally available both implicitly and explicitly for digital multimedia applications. This is not a conjecture; it's a reflection of current activity in engineering systems for visual- and data-intensive multimedia applications. As a result, parallel high-performance computing must matur...

Multimedia-information processing, in terms of creation and generation, interpretation, enhancement, rendering, and encoding and decoding, has existed for a long time. However, its recent transition to digital processing has increased the demand for it to prodigious levels and placed performance and storage constraints on computational engines. For...

The performance capabilities of virtual shared-memory systems still lead to the impression that the use of VSM is purely academic, providing conceptual simplicity at the expense of performance. The heaviness of the VSM implementation itself, especially ...

Not Available

Until now the scientific and engineering fields were the frontiers and beneficiariers of the advances in parallel computing, However, already there is a shifting in the motivation for the development of such systems, with multimedia being the epicentre of the attention. The reason is not one of fashion, but the result of stimulation from the unique...

A modular massively parallel computing (Modular-MPC) philosophy to image-related processing is discussed in this paper The approach is based on application-specific configurations of generic fine-grain Single Instruction stream operating on Multiple Data (SIMD) streams massively parallel computing modules to achieve high performance and maximal fle...

Massively parallel processing architectures have matured primarily through image processing and computer vision application. The similarity of processing requirements between these areas and video processing suggest that they should be very appropriate for video processing applications. This research describes the use of an associative massively pa...

Massively parallel processing architectures have matured primarily through image processing and computer vision application. The similarity of processing requirements between these areas and video processing suggest that they should be very appropriate for video processing applications. This research describes the use of an associative massively pa...

Numerous objects and natural phenomena in the spatial and temporal universe and in computational models are three-dimensional (3D) volumes of data. Volumetric visualisation is concerned with the representation, manipulation, interpretation rendering of such volumetric data in order to delve into the voluminous and complex structures and their dynam...

Real-time near Earth objects (NEO) detection is a problem requiring large processing capabilities due to the large CCD arrays and computation intensive algorithms to ensure system efficiency. The paper summarises the results and recommendations from the study for the use of a modular-MPC (massively parallel computer) as a cost-effective solution fo...

Associative memory concerns the concept that one idea may trigger the recall of a different but related idea. Traditional computers, however, rely upon a memory design that stores and retrieves data by its address rather than its content. In such a search, every accessed data word must travel individually between the processing unit and the memory....

The ASP System Testbed for Research and Applications (ASTRA) is an associative processor-based parallel processing system developed for image processing applications. The machine implements a multilayer heterogeneous architecture with an extendable linear array of simple associative processing elements at its base and more powerful processors at hi...

The suitability of the ASP computer for image processing tasks has been evaluated by implementing the Abingdon Cross benchmark. The results obtained indicate that the ASP is among the fastest and most cost effective image processors available.

The performance of the ASP computer on vision tasks has been evaluated by applying the Abingdon Cross benchmark using a number of different algorithms. In this paper, these algorithms are compared and contrasted on the basis of their performance

The ASP (Associative String Processor) is a massively parallel fault-tolerant fully associative processor designed primarily for the implementation of very compact, easily extensible, modular low-MIMD/high-SIMD parallel processing systems capable of supporting real-world applications of continuous data input and tightly integrated numeric and symbo...

A parallel algorithm for detecting moving targets by imaging sensors and estimating their trajectories is proposed. The algorithm uses the temporal continuity of the smooth trajectories of moving targets and successfully detects and simultaneously tracks all the target trajectories by mapping them from the image sequence onto a single target frame....

The author presents the ASP (associative string processor) a massively parallel, programmable, fault-tolerant architecture, which can efficiently support low-MIMD/high-SIMD and other parallel computation paradigms by providing comprehensive cover of both numeric and symbolic processing, supporting content-addressing, and a dynamically reconfigurabl...

The rebirth of activity in the area of neural computation is stimulated by the increasing frequency with which traditional computational paradigms appear to inefficiently handle fuzzy problems of large dimensionality (e.g. pattern recognition, associative information retrieval, etc.) and the technological advances. Indeed, with the huge strides in...

The use of the ASP (associative string processor), a massively parallel processing architecture, to implement representative tasks of integrated computer vision applications is discussed. An overview of the ASP architecture is provided. Two-dimensional image convolution and graph matching are then examined. The performance and versatility of the AS...

A description is given of four systolic array designs for the computation of the DFT (discrete Fourier transform). The designs are simple, modular, expandable and characterized by high performance. The designs are representative cases of larger system families which can be structured by altering the ways that the input data and weight streams are b...

The paper describes a modular, highly versatile and cost-effective massively parallel processing architecture, the ASP (associative string processor), its application to integrated computer vision and reports its performance to a number of independent computer vision and image understanding tasks

An implementation of a fully connected artificial neural network using the multi-layered perceptron model is described. The neural network is implemented on the ASP (Associative String Processor). The ASP is a fine-grain massively parallel SIMD architecture, emerging from research at Brunel University and being developed by Aspex Microsystems Ltd.,...

A description is given of research in the use of parallel computer architectures for continuous word speech recognition. In particular, the implementation of a continuous speech recognition algorithm on the associative string processor (ASP) is reported. The general principle of connected-word recognition, based on the Viterbi recognition algorithm...

The associative string processor (ASP) is a homogeneous, reconfigurable, programmable massively parallel processor which offers step-function advantages in cost performance and application flexibility due to its unique architecture and its use of state-of-the-art microelectronics. The authors briefly describe the ASP architecture and its implementa...