C-HEAP: A Heterogeneous Multi-Processor Architecture Template and Scalable and Flexible Protocol for the Design of Embedded Signal Processing Systems
Authors: Nieuwland A.1; Kang J.2; Gangwal O.P.3; Sethuraman R.4; Busá N.5; Goossens K.6; Peset Llopis R.7; Lippens P.8
Source: Design Automation for Embedded Systems, Volume 7, Number 3, October 2002 , pp. 205-232(38)
Publisher: Springer
- Free Content
- New Content
- Subscribed Content
- Free Trial Content
Abstract:
The key issue in the design of Systems-on-a-Chip (SoC) is to trade-off efficiency against flexibility, and time to market versus cost. Current deep submicron processing technologies enable integration of multiple software programmable processors (e.g., CPUs, DSPs) and dedicated hardware components into a single cost-efficient IC. Our top-down design methodology with various abstraction levels helps designing these ICs in a reasonable amount of time. This methodology starts with a high-level executable specification, and converges towards a silicon implementation. A major task in the design process is to ensure that all components (hardware and software) communicate with each other correctly. In this article, we tackle this problem in the context of the signal processing domain in two ways: we propose a modular, flexible, and scalable heterogeneous multi-processor architecture template based on distributed shared memory, and we present an efficient and transparent protocol for communication and (re)configuration. The protocol implementations have been incorporated in libraries, which allows quick traversal of the various abstraction levels, so enabling incremental design. The design decisions to be taken at each abstraction level are evaluated by means of (co-)simulation. Prototyping is used too, to verify the system's functional correctness. The effectiveness of our approach is illustrated by a design case of a multi-standard video and image codec.
Keywords: Communication; design methodology; distributed shared memory; embedded systems; heterogeneous multi-processor architectures; Kahn process networks; signal processing applications; synchronization protocol
Language: English
Document Type: Research article
Affiliations: 1: Philips Research Laboratories, Prof. Holstlaan, 5656 AA Eindhoven, The Netherlands E-mail: andre.nieuwland@philips.com 2: Philips Research Laboratories, Prof. Holstlaan, 5656 AA Eindhoven, The Netherlands E-mail: jeffrey.kang@philips.com 3: Philips Research Laboratories, Prof. Holstlaan, 5656 AA Eindhoven, The Netherlands E-mail: o.p.gangwal@philips.com 4: Philips Research Laboratories, Prof. Holstlaan, 5656 AA Eindhoven, The Netherlands E-mail: ramanathan.sethuraman@philips.com 5: Philips Research Laboratories, Prof. Holstlaan, 5656 AA Eindhoven, The Netherlands E-mail: natalino.busa@philips.com 6: Philips Research Laboratories, Prof. Holstlaan, 5656 AA Eindhoven, The Netherlands E-mail: kees.goossens@philips.com 7: Philips Research Laboratories, Prof. Holstlaan, 5656 AA Eindhoven, The Netherlands E-mail: rafael.peset.llopis@philips.com 8: Philips Research Laboratories, Prof. Holstlaan, 5656 AA Eindhoven, The Netherlands. He currently works with Magma Design Automation. E-mail: lippens@magma-da.com
- Free Content
- New Content
- Subscribed Content
- Free Trial Content
Click here for Page Help