VLSI implementations of low-power leading-one detector circuits

被引：14

作者：

Abed, Khalid H. ^{[1
]}

Siferd, Raymond E. ^{[2
]}

机构：

[1] Jackson State Univ, Dept Comp Engn, Jackson, MS 39217 USA

[2] Wright State Univ, Dept Elect Engn, Dayton, OH 45435 USA

来源：

PROCEEDINGS OF THE IEEE SOUTHEASTCON 2006 | 2006年

关键词：

D O I：

10.1109/second.2006.1629364

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

This paper presents two approaches to design leading-one detector (LOD) circuits, which locate the leading-one position in a binary word. The first approach provides fast leading-one detectors (LODs), and the second emphasizes the design of novel hardware-efficient and low-power LODs. Each approach is used to obtain 0.6 mu m CMOS VLSI implementations of 16-, 32-, and 64-bit LOD circuits. Simulations of the 16-, 32-, and 64-bit fast LODs run at 310, 265, and 215 MHz, respectively. The 16-, 32-,and 64-bit low-power LODs consume 14.85, 44.70, and 61.67 milliwatts, respectively while operating at VDD equal to 5 volts and their maximum speed.

引用

页码：279 / 284

页数：6

共 13 条

[1]

BRUGUERA J, 1999, IEEE T COMPUT, P298

[2] Leading-one Prediction scheme for latency improvement in single datapath floating-point adders [J].

Bruguera, JD ;

Lang, T .

INTERNATIONAL CONFERENCE ON COMPUTER DESIGN: VLSI IN COMPUTERS AND PROCESSORS, PROCEEDINGS, 1998, :298-305

[3] COMPUTATION OF BASE 2 LOGARITHM OF BINARY NUMBERS [J].

COMBET, M ;

VANZONNE.H ;

VERBEEK, L .

IEEE TRANSACTIONS ON ELECTRONIC COMPUTERS, 1965, EC14 (06) :863-&

[4]

FRANGAKIS GP, 1983, P IEE SEP, P169

[5]

HOEFFLINGER B, 1991, ELECTRON LETT, P1132

[6]

Mitchell J. N., 1962, IRE T ELECTRON COMPU, VEC-11, P512, DOI [DOI 10.1109/TEC.1962.5219391, 10.1109/TEC.1962.5219391]

[7]

OKLOBDZIJA V, 1993, IEEE T VLSI SYST, P124

[8]

OKLOBDZIJA V, 1992, P 26 IEEE AS C SIGN, P391

[9]

OKLOBDZIJA V, 1997, IEEE J SOLID-ST CIRC, P292

[10]

SanGregory S. L., 1999, IEEE MIDW S CIRC SYS

← 1 2 →