Inherent complexity of heterogeneous system design has been traditionally faced by means of abstraction and design automation. At the highest level of abstraction, the Unified Modeling Language (UML) can be extended t...
详细信息
Inherent complexity of heterogeneous system design has been traditionally faced by means of abstraction and design automation. At the highest level of abstraction, the Unified Modeling Language (UML) can be extended to model the whole system under design in an homogeneous fashion. In this paper, a design methodology based on the UML is presented. The benefits of using such a methodology are highlighted taking the design of an MPEG-4 decoder as an example.
Decision Diagrams (DDs) are the state-of-the-art data structure in VLSI CAD. They are used in many safety critical applications, like verification. Recently a new approach based on recursive checksum computation has b...
详细信息
Decision Diagrams (DDs) are the state-of-the-art data structure in VLSI CAD. They are used in many safety critical applications, like verification. Recently a new approach based on recursive checksum computation has been presented that showed how the correctness of the data structures could be verified by on-line and off-line tests. In this paper it is shown that these techniques can also be integrated in DD packages making use of dynamic reordering methods, like variable reordering and reordering based synthesis. The correctness of the data structures can be verified by (nearly) no overhead. Experimental results are presented to demonstrate the efficiency of this approach.
This paper presents a reconfiguration mechanism for an on-chip interconnection scheme called Heterogeneous IP Block Interconnection (HIBI). Required memory structures and logical signal operations for the different co...
详细信息
This paper presents a reconfiguration mechanism for an on-chip interconnection scheme called Heterogeneous IP Block Interconnection (HIBI). Required memory structures and logical signal operations for the different configurations are explained. The possible applications for this kind of reconfiguration are discussed, including ways to enhance system performance, ease of design re-use, low power designs and fault tolerance. An overview of HIBI is given as a background information for the reader. The HIBI architecture is designed to exploit VHDL synthesis but the concept could conceivably be transferred to any synthesis environment.
Software synthesis for system level design languages becomes feasible because the current technology, pricing and application trends will most likely alleviate the industrial emphasis on real-time operating systems mi...
详细信息
Software synthesis for system level design languages becomes feasible because the current technology, pricing and application trends will most likely alleviate the industrial emphasis on real-time operating systems minimisation. Automatic code generation also becomes necessary, because of increasing product complexity and decreasing design time. This paper discusses software synthesis for a realistic system level design language, to generate an executable model for implementation, simulation and verification purposes. A completely automatic mapping of both the architectural aspects and data objects is shown, including real-time garbage collection. Process execution trees (PETs) are introduced to schedule real-time, concurrent processes. This paper explains the functioning of these self-modifying data structures based on the operational semantics of POOSL (Parallel Object-Oriented Specification Language). Process execution trees are generally applicable to other process algebras as well (e.g. CCS, CSP, ACP), and follow quite naturally from the inference rules of these algebras.
This paper is concerned with a pipelined reconfigurable architecture for implementing image processing algorithms useful for robotic navigation. The computational complexity and real-time application of these algorith...
详细信息
This paper is concerned with a pipelined reconfigurable architecture for implementing image processing algorithms useful for robotic navigation. The computational complexity and real-time application of these algorithms suggest a custom hardware approach, but the large quantity of different algorithms useful for different tasks suggest a software reconfigurable approach. A logic reconfigurable based architecture combines both properties: hardware speed and software flexibility. Moreover, the large quantity of data to be processed using conventional Cartesian image processing algorithms, combined with autonomous vehicle constraints (power consumption, size and weight) makes the use of reconfigurable existent machines impossible. The log-polar vision reduces the amount of data to be processed and simplifies the computation of several interesting algorithms, allowing a reconfigurable approach which is scalable, powerful and small enough for an autonomous robot.
Software quality assurance (SQA) is an essential process to ensure high quality products. With the maturing of the agent technology, SQA can take advantage of this new development. This paper presents a framework of a...
详细信息
Software quality assurance (SQA) is an essential process to ensure high quality products. With the maturing of the agent technology, SQA can take advantage of this new development. This paper presents a framework of a multi-agent environment (AUTOQ) that can aid software quality assurance. SQA activities that can be performed by agents have been identified. A scenario of how multi-site software development can use AUTOQ is presented. It is expected that the use of AUTOQ will improve the monitoring and controlling of SQA activities and eliminates most manual effort.
In this paper we present some aspects of time optimization in hardware accelerated simulator of VLSI integrated circuits. The acceleration of simulation is achieved thanks to special data format of driver, which is us...
详细信息
In this paper we present some aspects of time optimization in hardware accelerated simulator of VLSI integrated circuits. The acceleration of simulation is achieved thanks to special data format of driver, which is used at the most time consuming operations of a simulator, e.g. finding transaction with least delay time, active signal and driver updating etc.
Labomat 3 is a reconfigurable platform for teaching and research purposes developed by our laboratory. In this paper we describe in details the hardware and software of the board as well as three application domains: ...
详细信息
Labomat 3 is a reconfigurable platform for teaching and research purposes developed by our laboratory. In this paper we describe in details the hardware and software of the board as well as three application domains: logic design, computer architecture, and codesign. The main features of the board are: a microprocessor associated with two mid-range FPGAs; a set of powerful software tools going from a real-time operating system to a JavaVM; easy to use design and simulation tools, and last but not least a network interface. The combination of these features makes Labomat 3 a unique teaching tool. Nevertheless, the board may also be used for advanced research: several boards connected together can form a powerful reconfigurable parallel system.
The body's immune system is impressively good at coping with external and internal errors, usually known as bacteria and viruses. The body is able to distinguish the haemoglobin found in blood from the insulin sec...
详细信息
The body's immune system is impressively good at coping with external and internal errors, usually known as bacteria and viruses. The body is able to distinguish the haemoglobin found in blood from the insulin secreted by the pancreas from the vitreous humor contained in the eye from everything else. It must manage to repel innumerable different kinds of invading organisms and yet not attack the body. The question to be answered is can we mimic these mechanisms in the design of our computer systems? This paper gives some details on how the body actually performs this amazing feat and gives some suggestions as to how this might inspire our design of computer systems increasing their reliability-immunotronics.
In this paper we propose two models of multiplication blocks by using VHDL. The algorithms that are used for writing the models are suitable for high speed multiplication and have regular cellular array structures. We...
详细信息
In this paper we propose two models of multiplication blocks by using VHDL. The algorithms that are used for writing the models are suitable for high speed multiplication and have regular cellular array structures. We have simplified some equations given in the references and then have written the VHDL model accordingly. Thus, a circuit synthesized by using the models proposed in this paper will have less area and gate count on the longest path than those given in the literature. We propose explicit expressions for calculating area values and the gate count on the longest path of the circuits for both of multiplication blocks for any value of n for, 2/spl les/n/spl les/54. Also we propose a method to determine the types of gates on the longest path of the circuit.
暂无评论