Due the proliferation of smartphones, demands for wireless data has grown dramatically. The wireless spectrum, already a scarce resource, is ever-dwindling and becoming more expensive: an FCC auction in 2015 gathered $41B for 1,611 licenses around the 1700MHz band. Projections indicate an even further increase in wireless data usage: Cisco predicts a 7-fold increase in mobile data usage from 2016 to 2021. One emerging trend to mitigate the spectrum overuse is to explore whitespace, locally unused frequency bands that were previously reserved for TV and maritime communications. However, the current circuit solutions to alleviate spectrum crunch such as duplexers and circulators inherently frequency-selective, and therefore not tunable. Also, most current wireless systems use TDD (time-division duplexing), which is not the most efficient way to use the spectrum. Frequency-division duplexing is more spectrally efficient, but in a non-MIMO system the transmitter is usually at a high enough power level to saturate or sometimes even destroy the receiver. These problems are exacerbated in full duplex systems.

In this thesis, we present a comprehensive solution to the spectrum scarcity problem by attacking on two fronts: instead of a narrow-band, fixed frequency solution we propose a software defined radio that has a wide range of tunability. Furthermore, we take advantage of duplexing in frequency with only a single antenna to more efficiently use the spectrum available. We also analyze the theoretical bounds of our transceiver to inform system parameters. Furthermore, we propose a novel schematic level model to accurately simulate harmonic distortion products in switching circuits. We use this model to better predict linearity numbers in passive mixers in deep triode.