RF CMOS Noise Parameters Extraction Technology And Noise Figure Automation Measurement

MOSFET (Metal-Oxide Semiconductor Field-Effect Transistor) has several advantages such as high integration density, low power and increasing cutoff frequency. With the down-scaling of MOSFET to deep-submicron dimensions and the resulting high unity-gain frequencies of hundreds of GHz make MOSFET increasingly attractive for applications in integrated high frequency analog electronics and wireless communications. However, when working at high frequencies, the effect of the noise generated within the device itself will play an increasingly important role in the overall system sensitivity characteristics, dynamic range and signal-to-noise ratio. Thus, it is crucial that we understand the noise mechanisms in sub-mircon MOSFET, an accurate extraction method for a small-signal noise equivalent circuit of MOSFET is vital for designing a circuit and optimizing device performance. Recently, several papers have reported noise measurement and noise parameters extraction which is based on equivalent-circuit parameter-extraction methods for MOSFET.This paper introduce noise theory, noise measurement and noise parameter extraction technique. Due to noise models that include physical noise mechanisms are based on small-signal models, firstly, open-short structure de-embedding technique has been used to de-embed the probe pad and extrinsic parasitic from the measured noise and S-parameters of the device, and small signal parameters are extracted by a semi-analytical approach. Although we can calculate the value of each noise parameter by using analytical method directly, considering practical measurement equipments in our lab and obtaining more accurate outcome, a new noise parameter extraction method called semi-analytical method based on 50 ohm noise measurement system without tuner is proposed, good fitting results are presented.In this paper, a noise figure automatic measurement system is designed, which include discussing theory of noise figure measurement, building hardware platform, the structure and usage method of software. The hardware platform consist of noise figure analyzer N8975A, vector network analyzer E5230A, signal generator E4438C, noise source N4002A, DC source E3631A, GPIB connecting line and PC. The software which has friendly interface, convenient operating characteristic, some based functions including calibration, measurement, data access and display and several advanced functions including measurement mode selection and parameters extraction is programmed by VC++6.0.