Joo-Gwang Lee

A redetermination of the argon content of air for buoyancy corrections in mass standard comparisons

A redetermination of the argon mole fraction in air has been undertaken in two samples of dried natural air using mass spectrometric analysis with reference to a suite of gravimetrically prepared synthetic dry air mixtures. The resulting measurement of the argon mole fraction was 9.332 mmol mol −1 with a combined standard uncertainty of 3 µmol mol −1 . This is significantly different from the value, 9.17 mmol mol −1 , conventionally employed in the CIPM formula for the determination of the density of moist air during mass standard comparisons. Using the presently reported argon mole fraction value in the CIPM formula rather than the conventional value removes the recently identified discrepancy between the two methods of determining the density of moist air during mass standard comparisons: the CIPM formula method and the air buoyancy artefacts method. Nitrogen, oxygen and carbon dioxide mole fractions in the dry air samples were obtained simultaneously.

Characterization Method of Electric Field Probe by Using Transfer Standard in GTEM Cell

In this paper, we introduce a simple and wideband characterization method for electric field probes using a transfer standard. As a transfer standard, a thin disk-type reference probe that can operate from 50 to 1000 MHz is used and calibrated using a micro transverse electromagnetic (mu-TEM) cell. A gigahertz transverse electromagnetic (GTEM) cell is used to generate a reference electric field. It has been shown that the uncertainty of the proposed method can be increased due to the imperfection of the field condition in the cell. According to the measurement result, the proposed characterization method for the electric probes agrees within 6.3%, compared with the method that uses the TEM cell and standard antennas in a fully anechoic chamber.

Novel attenuation standards at microwave frequencies and evaluation of their uncertainty

We have developed two coaxial-type microwave attenuation standards using the power ratio transfer method. One standard operates in the frequency range of 18 GHz to 26.5 GHz, and has a dynamic range of 80 dB. Its expanded uncertainty (k = 2) is estimated to be 0.01 dB/80 dB. The other operates in the frequency range of 26.5 GHz to 40 GHz, and has a dynamic range of 60 dB and an expanded uncertainty of 0.009 dB/60 dB. We present an example for calibration of a 40 dB step attenuator using the attenuation standard in a series radio-frequency substitution scheme. The expanded uncertainty (k = 2) is 0.01 dB including the mismatch uncertainty.

Determining Noise Temperature of a Noise Source Using Calibrated Noise Sources and an RF Attenuator

A system to measure the noise temperature (NT) of noise sources is implemented in the frequency range of 18-26.5 GHz for the case where noise sources commercially available have only one nominal NT, e.g., 9500 K being equivalent to 15-dB excess noise ratio. For the Y-factor method, two noise sources, i.e., a noise source and another with a radio-frequency attenuator, serve as the standard noise sources. The noise power output of the noise sources are measured using a commercial noise figure measurement instrument. Measurement results are presented, and its uncertainty has been evaluated to be 0.23-0.25 dB (k = 2) .

Uncertainty Analysis in EVM Measurement Using a Monte Carlo Simulation

This paper proposes an error vector magnitude (EVM) measurement method for a wideband code-division multiple-access source using a real-time oscilloscope. The EVM values are extracted from the measured waveform using a signal processing that finds the appropriate carrier phase and symbol timing. Measurement uncertainty is also evaluated based on a Monte Carlo simulation, where the errors from the real-time oscilloscope and the signal processing are taken into account. The measured EVM of a source at 900 MHz is (0.2586 ± 0.0040)%, (0.2617 ± 0.0060)%, and (0.2543 ± 0.0078)% at 95% confidence level when the real-time oscilloscope has a bandwidth of 2, 4, and 20 GHz, respectively.

RF Peak Power Calibration of Modulated Signals

A new calibration method of a power sensor for measuring the peak power of digitally modulated signals is introduced. This method uses an amplitude-modulated sine-wave source and is relatively simple to the currently available method that uses a digitally modulated signal source and an oscilloscope. Basic theory and calibration procedures for measuring the correction factor for peak power are described. Using a commercial peak power sensor that is calibrated with the proposed method, the peak power of a Global System for Mobile Communications at 900 MHz modulated signal is measured and compared with that measured by an oscilloscope. Various examples of choosing the calibration parameters are also discussed. The best uncertainty in the calibration of a peak power sensor is estimated to be 0.34% (k = 2).

Uncertainty evaluation of a broadband attenuation standard

A standard step attenuator has been developed. It comprises a build-up chain of ten steps of power ratio measurement and operates in a frequency range of 10 MHz to 18 GHz. The overall system has a dynamic range of 90 dB and an expanded uncertainty of 0.006 dB (k=2). The system is used as an attenuation standard in a series radio-frequency substitution scheme. The uncertainty factors in the attenuation measurement of 40 dB at 5 GHz are fully described. For the calibration of a typical step attenuator, the expanded uncertainties are 0.011 dB for the attenuation range up to 80 dB, including the mismatch uncertainty.

A broadband attenuation standard

A standard step attenuator, composed of ten steps of a build-up chain of power measurement and operating from 10 MHz to 18 GHz, has been developed. Each step shows a nominal attenuation of about 9 dB and an expanded uncertainty of 0.0018 dB at 30 MHz. The overall system has a dynamic range over 90 dB.

Uncertainty analysis in EVM measurement using a Monte-Carlo simulation

An error vector magnitude (EVM) measurement method for W-CDMA source is proposed using a real time oscilloscope with traceability to SI unit. Also measurement uncertainty is evaluated based on a Monte-Carlo simulation, which takes into account the errors from the oscilloscope and signal processing. The measured EVM of a source at 900 MHz is 0.2535% ± 0.0057% at 95% confidence level.

Spectro-Temporal Mismatch Analysis of a Transmission Line Based on on-Wafer Optical Sampling

We present an optical sampling technique that enables exploration of mismatches of a microstrip transmission line based on reflection analyses of electromagnetic pulses. The external electro-optic sampling scheme with a minute crystal detects high-speed electrical pulses over arbitrary locations of a line with very low-intrusiveness. The temporal pulsed signals measured with an on-wafer optical probing system and the corresponding spectra are obtained to analyze the transfer characteristics of a microstrip transmission line with 20 GHz bandwidth. The spectro-temporal response was cross-checked with commercial instruments. Applications of this optical probing technique to explore mismatches at the terminal port — based on both time and frequency domain reflectometry analyses — are also presented.