Magnetically coupled current probe embedded on-chip and PCB for analysis of simultaneous switching current

Abstract

Recently, the rise of smart portable electronic devices presents new challenges in obtaining not only low power but also high performance. But in this low power and high performance mobile application, a large num-ber of logic block and I/O blocks operate in higher speed and demand more power so the operating voltage should be lowered continuously to keep the lower power requirement. As a consequence, a large amount of in-stantaneous switching current which consumed within a short period of time can produce a considerable switch-ing noise over the entire power distribution network (PDN). In addition, decreased power supply voltage makes this simultaneous switching noise (SSN) effects even worse because the noise margin gets lower as the power supply voltage scales with the technology. Therefore accurate prediction of the simultaneous switching noise became one of the most important issues in design of high speed digital system. Since the product of the switch-ing current and the input impedance of the PDN in frequency domain will produce PDN noise spectrum. So the time domain voltage fluctuation of the PDN can easily be predicted by applying inverse Fourier transformation. So it is clear that an accurate prediction of power supply switching noise requires an accurate PDN impedance profile over the frequency range of interest as well as an accurate switching current profile from entire switching circuit or logic block. Several methods of modeling and analysis techniques have been proposed to obtain the PDN Impedance profile such as transmission line method, cavity resonator method and transmission matrix method. And some measurement-based methods such as spectrum analyzer and oscilloscope for extracting switching current profile have also been presented because the SPICE-like circuit simulation is extremely difficult for extraction especially when the large circuit and logic block operate at random application with multiple frequencies In this dissertation, a ...