Single electron tunneling based arithmetic computation

In this dissertation we investigate the implementation of computer arithmetic operations with Single Electron Tunneling (SET) technology based circuits. In our research we focus on the effective utilization of the SET technologys specific characteristic, i.e., the ability to control the transport of individual electrons. We pursue two main research directions: Single Electron Encoded Logic (SEEL) and Electron Counting (EC) arithmetic. All the design we propose are verified by means of simulation and evaluated in terms of area, delay and energy consumption. The first line of research, i.e., the SEEL paradigm, is based on encoding Boolean variables as a net charge of zero or one electron thus charge transport in a SEEL gate due to switching activity is limited to one electron. Within the SEEL paradigm we first propose a generic Threshold Logic Gate (TLG) from which we derive a Boolean logic gate family. Subsequently, we investigate the behavior of the SEEL TLG as a network component, argue that buffering is a method to reduce crosstalk problems and propose four SEEL buffers. A buffer choice is argued, based on TLG buffering suitability, and subsequently utilized to implement buffered Boolean gates. We next explore SEEL implementations of memory elements and propose static latches, flip-flops and dynamic memory elements. Given the above components, i.e., SEEL Boolean gates, threshold gates and memory elements, we propose building blocks for serial adders, ripple carry adders, carry-lookahead adders and tree multipliers and evaluate the above adders and multipliers. The second line of research, i.e., the EC paradigm, is based on encoding integer values as charge and performing arithmetic operations on charge encoded operands. Within the EC paradigm we first suggest a set of EC building blocks and propose generic schemes for EC based addition and multiplication. We then assume that limited amounts of charge can be transported in an EC block and propose designs for high radix EC addition and multiplication that can manage such limitations