Multi-device radio frequency power amplifiers (PAs) often exhibit strongly non-linear behavior in combination with long-term memory effects, leading to an extremely challenging model development cycle. This paper presents a new, dynamic, behavioral modeling technique, based on a combination of the real-valued decomposed piecewise method and concepts from the field of machine learning. The underlying theory of the proposed modeling technique is provided, along with a detailed modeling procedure. Experimental results show that the proposed decomposed piecewise support vector regression (SVR) model leads to significant performance improvements when compared with standard SVR models for both single transistor and multi-transistor PAs. Different model thresholds are used to test the proposed model performance for both PA types. For the single-transistor PA, modeled using only one partition, an approximately 10 dB normalized mean square error (NMSE) reduction is seen when compared with the standard SVR model. For the same PA, when utilizing two partitions, the reduction improves to 14 dB. When applied to a multi-device Doherty PA, the NMSE between model and measurement data is −50 dB, representing more than 10 dB improvement compared with the standard SVR model.

A methodology is presented to predict odd-mode instability in power amplifiers under output mismatch effects, as in the case of amplifiers connected to an antenna. This kind of instability is often observed in multi-device configurations, due to their symmetry properties. Unlike the single-ended situation, there is a cancellation of odd multiples of the oscillation frequency at the circuit output, so there is no impact of the load-impedance values at the sideband frequencies. The odd-mode instability only depends on the impedance terminations at the fundamental frequency and its harmonic terms, and can only be detected within the circuit unstable loop, instead of the antenna-connection terminals. The possible unstable modes are related with the eigenvectors of an outer tier conversion matrix accounting for the symmetry properties of the circuit topology. Under sufficient low-pass filtering of the amplifier output network, the analysis parameters can be limited to the magnitude and phase of the reflection coefficient at the fundamental frequency. This analysis involves a computationally efficient graphical technique to detect potential instabilities and a bifurcation-detection method to determine the instability boundaries in the Smith chart. The two main types of instability from periodic regime are considered, respectively associated with incommensurable and subharmonic oscillations. Results have been validated through pole-zero identification and experimental measurements.

This paper presents a very flexible tool for numerically evaluating the small-signal and noise parameters of a transmission line subject to an arbitrary thermal gradient. Contrary to previous methods, the proposed approach allows straightforwardly taking into account possible variations of electrical parameters along the propagation direction, such as may be expected when temperature ranges between very different values. The main application of the proposed method is cable modeling in noise-figure measurement setups under cryogenic operation: in such circumstances, indeed, the coaxial cables (or waveguide) at the interface between the outside and the inside of the cryogenic chamber are subject to remarkable temperature excursions. As a consequence, significant de-embedding errors may arise if the cables are not correctly modeled, given the very low values of noise figure which are commonly exhibited by cryo-cooled active devices.

In this paper, the small- and large-signal modeling of InP heterojunction bipolar transistors (HBTs) in transferred substrate (TS) technology is investigated. The small-signal equivalent circuit parameters for TS-HBTs in two-terminal and three-terminal configurations are determined by employing a direct parameter extraction methodology dedicated to III–V based HBTs. It is shown that the modeling of measured S-parameters can be improved in the millimeter-wave frequency range by augmenting the small-signal model with a description of AC current crowding. The extracted elements of the small-signal model structure are employed as a starting point for the extraction of a large-signal model. The developed large-signal model for the TS-HBTs accurately predicts the DC over temperature and small-signal performance over bias as well as the large-signal performance at millimeter-wave frequencies.

In this paper, authors report on an enhanced approach for the design of monolithic microwave integrated circuit (MMIC) power combining amplifiers. Commonly used techniques for the stabilization of such circuits are empirical and too conservative. This leads very often to a non-desired degradation of the radio frequency (RF) performances that are inherent to the physical properties of such stabilization networks at the fundamental frequency of operation. The methodology proposed here is based on the use of large-signal optimization processes that combine RF and stability analyses from the early stages of the design. This approach results in an improvement of the RF performances while sufficient stability margins are preserved. The optimization procedure is explained on a Ku-band MMIC power amplifier for space-borne communications.