Access Type

Open Access Dissertation

Date of Award

January 2017

Degree Type

Dissertation

Degree Name

Ph.D.

Department

Electrical and Computer Engineering

First Advisor

Harpreet Singh

Abstract

For the last four decades, the implementation of very large-scale integrated systems has largely based on complementary metal-oxide semiconductor (CMOS) technology. However, this technology has reached its physical limitations. Emerging nanoscale technologies such as quantum-dot cellular automata (QCA), single electron tunneling (SET), and tunneling phase logic (TPL) are major candidate for possible replacements of CMOS. These nanotechnologies use majority and/or minority logic and inverters as circuit primitives. In this dissertation, a comprehensive methodology for majority/minority logic networks synthesis is developed. This method is capable of processing any arbitrary multi-output Boolean function to nd its equivalent optimal majority logic network targeting to optimize either the number of gates or levels. The proposed method results in different primary equivalent majority expression networks. However, the most optimized network will be generated as a nal solution. The obtained results for 15 MCNC benchmark circuits show that when the number of majority gates is the rst optimization priority, there is an average reduction of 45.3% in the number of gates and 15.1% in the number of levels. They also show that when the rst priority is the number of levels, an average reduction of 23.5% in the number of levels and 43.1% in the number of gates is possible, compared to the majority AND/OR mapping method. These results are better compared to those obtained from the best existing methods.

In this dissertation, our approach is to exploit QCA technology because of its capability to implement high-density, very high-speed switching and tremendously lowpower integrated systems and is more amenable to digital circuits design. In particular, we have developed algorithms for the QCA designs of various single- and multi-operation arithmetic arrays. Even though, majority/minority logic are the basic units in promising nanotechnologies, an XOR function can be constructed in QCA as a single device. The basic cells of the proposed arrays are developed based on the fundamental logic devices in QCA and a single-layer structure of the three-input XOR function. This process leads to QCA arithmetic circuits with better results in view of dierent aspects such as cell count, area, and latency, compared to their best counterparts. The proposed arrays can be formed in a pipeline manner to perform the arithmetic operations for any number of bits which could be quite valuable while considering the future design of large-scale QCA circuits.

Share

COinS