First Advisor
Marek Perkowski
Term of Graduation
January 2026
Date of Publication
6-1-2026
Document Type
Dissertation
Language
English
Subjects
Automata Learning, FPGA, Memristor, Nondeterministic Automata, Quantum Automata, Quantum State Machines
Physical Description
1 online resource ( pages)
Abstract
As transistor scaling hits its limits, new forms of computation and innovative devices need to be considered to maintain technological progress. Quantum computing and memristive computing are two promising fields with great potential. A new approach to logic synthesis is required to realize practical hardware in these "stateful" computing paradigms. This dissertation defines some fundamental models of Quantum automata called Quantum State Machines. These models use permutative quantum logic gates to implement practical state machines. A key requirement to build such machines is the ability to synthesize logic functions, including incompletely specified functions and irreversible functions. This dissertation presents a new algorithm called Quantum Automata Synthesizer (QAS) to convert any incompletely specified or irreversible function into a fully specified reversible function, and realize it using an array of Toffoli, NOT, and CNOT gates. A practical application of this approach is to realize non-deterministic finite state machines (FSMs) using quantum gates, which was an unsolved problem. This dissertation presents a novel step-by-step approach to realize non-deterministic FSMs using quantum arrays and applies this approach to machine learning.
This dissertation also proposes a new concept of FPGAs based on a new nanotechnology device called memristor. Our proposed FPGA also includes CMOS drivers and switches, memristive memories, and routing resources. Memristors are two-terminal nanodevices that are non-volatile. Their ability to retain state makes them ideal for implementing memories. They also provide switching characteristics that make them suitable candidates for logic circuit realizations. Mapping Boolean logic functions to memristive logic gates requires a new FPGA architecture. A new concept of configurable logic blocks and programmable interconnects based on memristor crossbar arrays is required. Our proposed hybrid FPGA architecture, i.e., the "MemLogic-FPGA", is a novel concept of FPGA device technology, different from traditional FPGA devices in the market. A new CAD flow for implementing logic circuits on this FPGA architecture is therefore created. Compared to existing memristor FPGA designs, the proposed architecture scales better, and provides area and performance improvements for highly data-parallel applications with many pipelines.
Finally, we investigate the potential of quantum memristors to realize memory elements in quantum automata. In summary, this dissertation defines some fundamental architectures for computing using memristors and quantum logic and provides practical logic synthesis methods for these architectures.
Rights
In Copyright. URI: http://rightsstatements.org/vocab/InC/1.0/ This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s).
Recommended Citation
Kumar, Manjith, "Designing Automata and Systems in Memristive and Quantum Technologies" (2026). Dissertations and Theses. Paper 7102.