Advanced Noise-Shaping Data Conversion Techniques

The rapidly grown digital devices market places a huge demand for Data Conversion nowadays. For high-resolution data converter, Noise-Shaping (NS) techniques are essential. This dissertation first goes through the fundamentals of data conversion and noise-shaping techniques and then introduces three advanced noise-shaping techniques that further boost data converters’ performance.

The first technique boosts up the order of NS noticeably. A system-level approach is introduced to achieve high-order NS with enhanced robustness, and a NS-SAR ADC is designed and measured as an example, which is the first 4th-order NS-SAR published.

The second technique effectively increases the bandwidth of NS converters using a time-interleaved framework. A Time-Interleaved NS-SAR ADC is designed and measured as an example to illustrate this technique. It achieves the highest bandwidth among NS-SAR up to date.

The last technique is aimed to solve the inherent problems in continuous-time DSM. A CT-DT hybrid NS is introduced, which adopts the advantages of both the CT and DT NS. A prototype ADC with such technique is designed and measured as the last example, showing the benefit of tuning-free and high tolerance to non-idealities.

Chair: Professor Michael P. Flynn