This paper proposes a design method for designing a wideband filtering power amplifier (PA) based on terminated coupled line structure (TCLS). This method generates four transmission zeros and three transmission poles by loading stepped impedance resonator and short-circuit stubs on the four ports of TCLS, which greatly optimizes the in-band return loss and stopband suppression of the output matching network. In addition, the effective suppression of the second harmonic also increases the efficiency of the PA. In order to verify the effectiveness and superiority of this design method, a wideband bandpass filtering PA is designed and manufactured in 2.9–3.7 GHz. Measurement results show that the saturated output power is from 40.1 dBm to 41.2 dBm, the drain efficiency is greater than 60.3%, and the gain is 10–11.2 dB.

A novel, broadband, nonlinear behavioral model, based on support vector regression (SVR) is presented in this paper. The proposed model, distinct from existing SVR-based models, incorporates frequency information into its formalism, allowing the model to perform accurate prediction across a wide frequency band. The basic theory of the proposed model, along with model implementation and the model extraction procedure for radio frequency transistor devices is provided. The model is verified through comparisons with the simulation of an equivalent circuit model, as well as experimental measurements of a 10 W Gallium Nitride (GaN) transistor. It is seen that the efficiency prediction throughout the Smith chart, for varying fundamental and second harmonic loads, across a wideband frequency range, show excellent fidelity to the measured results. Device dc self-biasing is also modelled to allow prediction of power amplifier (PA) efficiency, which is shown to be highly accurate when compared with corresponding measured data. Finally, a class-J PA is constructed and measured across the frequency with a large-signal input tone. The resulting measured and modelled values of key PA performance figures are shown to be in excellent agreement, indicating the model is suitable for broadband PA design.

Continuous mode class-J radio-frequency (RF) power amplifier is a promising technique that extends the operating bandwidth of the conventional class-B power amplifier. However, the maximum theoretical efficiency is limited to that of the class-B power amplifier. In this paper, an enhanced mode of operation for the class-J power amplifier is proposed by incorporating a third harmonic voltage component to produce an optimum waveform for maximizing the fundamental voltage component and thereby to increase the drain efficiency and introduce a new design space. A detailed derivation for the necessary relations of output power, drain efficiency, and the required harmonic load impedances is provided, showing a significant improvement in theoretical maximum efficiency from 78.5 to 89.8%. In order to confirm the developed analytic approach, a 10 W prototype amplifier model was designed and fabricated to operate within the global system for mobile communications (GSM) frequency band 850–950 MHz using a commercial GaN power high electron mobility transistor (HEMT). The experimental results have indicated that the drain efficiency of the circuit varies from 68 to 80% within the desired band.

This paper reports on a record-low-phase noise D-band signal source with 5 dBm output power, and 1.3 GHz tuning range. The source is based on the unconventional combination of a fundamental frequency 23 GHz oscillator in 150 nm AlGaN/GaN HEMT technology followed by a 130 nm SiGe BiCMOS MMIC including a sixtupler and an amplifier. The amplifier operates in compression mode as power-limiting amplifier, to equalize the source output power so that it is nearly independent of the oscillator's gate and drain bias voltages used for tuning the frequency of the source. The choice of using a GaN HEMT oscillator is motivated by the need for a low oscillator noise floor, which recently has been demonstrated as a bottle-neck for data rates in wideband millimeter-wave communication systems. The phase noise performance of this signal source is −128 dBc/Hz at 10 MHz-offset. To the best of the authors’ knowledge, this result is the lowest reported phase noise of D-band signal source.

Guaranteeing charge conservation of empirically extracted Gallium Nitride (GaN) High-Electron-Mobility Transistor (HEMT) models is necessary to avoid simulation issues and artifacts in the prediction. However, dispersive effects, such as thermal and charge-trapping phenomena, may compromise the model extraction flow resulting in poor model accuracy. Although GaN HEMT models should be extracted, in principle, from an isodynamic dataset, this work deals with the systematic identification of an approximate, yet most suitable, charge-conservative empirical model from standard multi-bias S-parameters, i.e., from non-isodynamic data. Results show that the obtained model maintains a reasonable accuracy in predicting both small- and large-signal behavior, while providing the benefits of charge conservation.

A nonlinear circuit model (NCM) with physical parameters is proposed for direct simulation of the RF characteristics of GaN high-electron-mobility transistors (GaN HEMTs) on the basis of device structure. The physical equations are used for the construction of the model in order to connect strongly the model parameters with the device/physical parameters. Hyperbolic tangent functions are used as the model equations to ensure good model convergence and rapid simulation (short simulation time). The usefulness of these equations is confirmed by technology computer aided design (TCAD) simulation. The number of model parameters for the nonlinear components (Ids, Cgs, Cgd) is reduced to 17 by using common physical parameters for modeling the drain current and capacitance. The accuracy of this model is verified by applying to GaN HEMTs. The modeled I–V and capacitance characteristics agree well with the measurement data over a wide voltage range. Furthermore, this model can be used for the accurate evaluation of S-parameters and large-signal RF characteristics.

Radio Frequency Micro-Electro-Mechanical System (RF-MEMS) represents a feasible solution to obtain very low power dissipation and insertion loss, very high isolation and linearity switch with respect to “solid state” technologies. In this paper, we demonstrate the full integration of RF-MEMS switches in the GaN-HEMT (Gallium Nitride/High Electron Mobility Transistor) fabrication line to develop RF-MEMS devices and LNA-MMIC (Low Noise Amplifier/Monolithic Microwave Integrated Circuit) prototype simultaneously in the same GaN wafer. In particular, two different coplanar wave (CPW) LNAs and a series of discrete RF-MEMS in ohmic-series and capacitive-shunt configuration have been fabricated. RF-MEMS performances reveal an insertion loss and isolation better than 1 and 15 dB, respectively, in the frequency range 20–50 GHz in the case of pure capacitive shunt switches and in the frequency range 5–35 GHz for the ohmic-series switches. Moreover, the GaN HEMT device shows an Fmax of about 38 GHz and a power density of 6.5 W/mm, while for the best LNA-MMIC we have obtained gain better than 12 dB at 6–10 GHz with a noise figure of circa 4 dB, demonstrating the integration achievability.

This paper presents the design and implementation of an inverse class-F power amplifier (PA) using a high power gallium nitride high electron mobility transistor (GaN HEMT). For a 3.5 GHz continuous wave signal, the measurement results show state-of-the-art power-added efficiency (PAE) of 78%, a drain efficiency of 82%, a gain of 12 dB, and an output power of 12 W. Moreover, over a 300 MHz bandwidth, the PAE and output power are maintained at 60% and 10 W, respectively. Linearized modulated measurements using 20 MHz bandwidth long-term evolution (LTE) signal with 11.5 dB peak-to-average ratio show that −42 dBc adjacent channel power ratio (ACLR) is achieved, with an average PAE of 30%, −47 dBc ACLR with an average PAE of 40% are obtained when using a WCDMA signal with 6.6 dB peak-to-average ratio (PAR).