Marc Vanden Bossche

Large-Signal Network Analysis Including the Baseband

In recent years, wireless systems have rapidly evolved due to the increasing demands of different services, functionalities, and applications. In order to provide high-speed connectivity while using the available spectrum in an efficient manner, wideband digitally modulated signals are increasingly used to transmit information.

Pulsed multi-tone measurements for time domain load pull characterizations of power transistors

Real-time active load-pull (RTALP) characterization of power transistors with a large signal network analyzer (LSNA) has the benefit of greatly accelerating the design of power amplifiers. However RTALP has been so far limited to devices operating under CW excitations and constant biasing and was therefore mostly applicable to transistors with small memory effects. In TDMA communication or radar systems, amplifiers are excited by pulsed or modulated RF signals and operate under pulsed biasing. For such applications the measurement of modulated/pulsed RF signals under pulsed biasing/RF conditions is required for a more realistic understanding of the behavior of transistors with memory effects. This paper demonstrates that under appropriate conditions it is possible to effectively combine these two excitation modes while performing large-signal measurement by sub-sampling down-conversion with a LSNA. In this demonstration, the acquisition of CW two-tone and three-tone excitations is performed using multiple-recording with pulse duty rate of 1.5% and 0.15%. The pulse-recorded measurements for the three-tone excitation are found to be consistent with CW measurements (100% duty rate). This preliminary work indicates that either intermodulation or real-time active load-pull measurements under pulsed operation can be readily realized for RF samplers with sufficiently short settling time while operating under pulsed RF modulation.

S-functions extracted from narrow-band modulated large-signal network analyzer measurements

In this paper we report for the first time on the extraction of a S-functions behavioral model from modulated large-signal network analyzer measurements. The model parameters are directly obtained from the measurements without any additional optimization procedure. A good match between the obtained values and these derived using the standard method is shown for a packaged pHEMT device. We demonstrate that the benefits of a reduced number of measurements and a simplified model extraction are counterbalanced by an increased sensitivity of the model to the phase noise characteristic of the RF sources used during the measurements.

A method to select correct stimuli levels for S-functions behavioral model extraction

S-functions and other frequency-domain behavioral models based on describing functions rely on the linearization of the response of a nonlinear RF or microwave component around a set of large-signal operating points. To assure the validity of the linearization principle, and thus the validity of the model, one has to select appropriate stimuli levels during the experiments which are performed to extract the model parameters. In this work we propose a formal procedure to identify the correct small-signal power levels using a system identification approach, i.e. taking the uncertainty of the measured data into account. The method is based on the analysis of the cost function of an errors-in-variables (EIV) estimator obtained for simple models.

S-functions behavioral model order reduction based on narrowband modulated large-signal network analyzer measurements

In this paper we report for the first time on order reduction applied to S-functions behavioral models. The most dominant model parameters are selected based on the relative uncertainty of their estimated values evaluated against a threshold value. The selection procedure is performed on the same measurement data that is used to extract the model and obtained using a large-signal network analyzer. High level of model order reduction, achieved without any substantial loss of the prediction accuracy, is demonstrated on S-functions extracted for a packaged pHEMT device.

A phase calibration methodology for accurate modulated measurements

To address the challenge of the phase calibration of modulated RF and microwave signals or the response of a linear or nonlinear system under modulated excitation, a next-generation comb generator is characterized. The latter allows kHz-level modulation tones in addition to the GHz-level RF and microwave tones, while still exhibiting an acceptable power level for each tone. This comb generator includes a PRBS generator, two dividers and a logical “and” operation to trigger the internal pulse generating unit. The pulse generator is characterized using a calibrated sampling oscilloscope. Depending upon the mode of operation this generator could be operated using a plentitude of possible values for every setting (input frequency, divider value, PRBS length). Each setting results in a different amplitude and phase spectrum. In practice it is impossible to characterize this pulse generator and provide a spectral phase dataset for every possible setting of each mode. To address this problem it is investigated if it is feasible to mathematically obtain the complete amplitude and phase spectrum of each possible setting by just using the characterized information of a single pulse. The amplitude and phase characteristic obtained by the measured and predicted spectrum are analyzed and compared. In this paper this technique will be applied to a comb generator operating under divider or PRBS mode.

Simultaneous measurement of high and low frequency response of non-linear microwave circuits

In this work, measurements of the high frequency as well as the low frequency response of a non-linear microwave circuit are reported. The developed set-up is based on an extension of the LSNA and it enables the simultaneous measurement of baseband response and RF behaviour. Thanks to this capability a direct correlation between the variation of baseband impedance and asymmetry of distortion components around the fundamental carrier is possible. Experimental results of two-tone measurements, carried out on a hybrid GaAs microwave circuit, are shown.

Negative input resistance and real-time active load-pull measurements of a 2.5GHz oscillator using a LSNA

A large-signal measurement-based methodology to design oscillators using the Kurokawa theory is presented in this paper. Measurements of the negative input resistance and device line of a 2.5 GHz HEMT oscillator versus frequency and power, and its optimization using real-time active load-pull (RTALP) for the 2nd and 3rd harmonics are performed with a large signal network analyzer (LSNA). As a result, the maximum output power of the oscillator is increased from 31.0 mW to 38.8 mW. Finally self-oscillation is verified using a load tuner to yield an output power and frequency of oscillation in reasonable agreement with the Kurokawa analysis.

A compact measurement set-up for envelope-tracking RF PAs with calibrated sensing of baseband V/I at the supply terminal

This paper presents a compact set-up to perform automated measurements of RF PAs under envelope-tracking (ET) operation. In addition to RF signal generation and acquisition, it allows calibrated measurements of the baseband voltage and current at the supply terminal with up to 40 MHz bandwidth, enabling instantaneous efficiency measurements. The presented set-up is used to extract two different shaping tables, and to test ET operation with a 5 MHz WCDMA and a 20 MHz LTE signals at 838 MHz. The used devices are an InGaP/Silicon RF power amplifier (PA) from Skyworks and a high voltage extra fast complementary bipolar (XFCB) power modulator (PM) from Analog Devices.

Live demonstration: Mixed-signal network analysis characterization and modeling platform

This demonstrator presents a novel framework for the characterization and modeling of mixed-signal systems. It will be shown that a joint analog and digital analysis is relevant for future radio-frequency (RF) components and devices. A characterization procedure suitable for mixed-signal systems is presented based on a National Instruments (NI) PXI architecture by analyzing the system in a scattering wave perspective at each port. This framework allows an RF engineer to integrate innovative mixed-signal models on available RF/microwave simulators, such as the NI AWR Microwave Office. An example of various analog-to-digital-converters (ADCs) will be carried out by measuring its linear parameters, and then using it inside a simulator tool.