An Eecient Vlsi Architecture for Columnsort

نویسندگان

  • Rong Lin
  • Stephan Olariu
چکیده

| This paper presents novel VLSI architecture in support of an eecient implementation of Leighton's well-known Columnsort. The designs take advantage of re-conngurable bus architectures enhanced with simple shift switches. Our main contribution is to show that (1) Column-sort can be partitioned into two components of hardware scheme involving sorting arrays of small size and a hardware or software scheme involving simple data movement tasks, (2) The dynamically reconngurable mesh architecture can be exploited to obtain a highly eecient small hardware sorter. The proposed architecture features high regularity of circuitry, simplicity of control structure, and adaptability. Both theoretical analyses and simulation tests have shown that the proposed VLSI architecture for sorting are superior to the existing designs in the context of sorting small and moderate size arrays.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

An Eecient Vlsi-architecture for Real-time Motion Estimation

An eecient VLSI-architecture that enables hierarchical motion estimation at real time for large tracking areas in applications such as HDTV is presented here. This is capable of treating (16*16)-blocks, with a maximum displacement of +/-14 pixels, at 72 MHz pixel rate. The architecture is based on the the hierarchical three-step search algorithm with a data-ow design which allows sequential inp...

متن کامل

A High-Speed VLSI chip for Parallel Image Decorrelation

We present here the architecture and design of a special purpose CMOS VLSI chip for high-speed parallel image decorrelation/inverse decorrelation scheme which is to be used for eecient real-time losslesss image compression/ decompression. The chip is designed for one processing element of the parallel architecture that performs the image decorrelation/inverse decorrelation. The architecture is ...

متن کامل

Under Revision for Ieee Transaction on Vlsi Systems Eecient Vlsi for Lempel-ziv Data Compression in Wireless Data Communication Networks

We present a parallel algorithm, architecture, and implementation for eecient Lempel-Ziv-based data compression. The parallel algorithm exhibits a scalable, pa-rameterized, and regular structure and is well suited for VLSI array implementation. Based on our parallel algorithm and systematic design methodologies 12], two semi-systolic array architectures have been developed which are low power a...

متن کامل

Modular approach for an ASIC integration of electrical drive controls

VLSI circuits design allows today to consider new modes of implementation for electrical controls. However, design techniques require an adaptation effort that few designers, too accustomed to the software approach, provide. The authors of this article propose to develop a methodology to guide the electrical designers towards optimal performances of control algorithms implementation. Thus, they...

متن کامل

A Scalable Bit - Sequential SIMD Array for Nearest - NeighborClassi cation using the City - Block

We present a fully scalable SIMD array architecture for a most eecient implementation of pattern classiication by nearest-neighbor algorithms using the city-block metric. The elementary accumulator cell is highly optimized for a sequential accumulation of absolute integer diierences, so that several hundreds of them can be easily integrated on a single chip. A two-dimensional M N array structur...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 1999