| A Frequency synthesizer is one of the most critical components in digital and analog systems. The Fractional-N frequency synthesis has been popular since its invention in the 1980s. The Flying-adder synthesizer was invented by TI in 2000 and has been widely used in many of their products. The purpose of this thesis is to perform a thorough comparison between these two architectures and gain deeper insights into the advantages and limitations of both architectures.;The thesis starts with an in-depth study and simulation on both architectures individually. For the fractional-N synthesizer, two cases are studied: with and without sigma-delta modulator. Different structures of phase detectors and the influence of 1st and 3rd order sigma-delta modulators are analyzed. In addition, through the analyses of power spectrum and phase noise, we are able to predict the locations of the fractional spurs.;For the Flying-Adder synthesizer, two cases are also studied: with and without random dithering. We find that the number of VCO stages plays a key role in the tuning range and the noise performances. Phase noise, power spectrum analysis and the prediction of the fractional spurs are studied and simulated extensively.;With the understanding of both architectures, a comparison is performed between them in terms of system speed, noise performance, tuning range, and the locations of fractional spurs. The pros and the cons of the fractional-N and Flying-Adder architectures are illustrated in the thesis.;Finally, with an in-depth understanding of both architectures, a new approach that combines the advantages of both fractional-N and Flying-Adder techniques is proposed at the end of the thesis. Extensive simulations are performed to validate the concept. |
