In this letter, the second source harmonic in continuous-mode class GF (CCGF) is optimized to flatten the power amplifier’s (PA) frequency response over a wideband. A new design space is explored by considering the effects of controlling the input nonlinearity of the gate–source capacitance ( $C_{\mathrm {gs}}$ ) on the drain current waveforms under continuous-mode drain voltage waveforms. A wideband CCGF PA is designed and fabricated using a commercial 10-W gallium nitride (GaN) device and low-temperature co-fired ceramic (LTCC) technology. Results of the measurement show a flat frequency response from 3.3 to 4.3 GHz with variations less than ±0.4 dB for 40 dBm output power, and 17 dB large signal gain at 3-dB compression point. A drain efficiency of 66 ± 2% is achieved over the entire bandwidth.
In this letter, two basis function multiplexing-based behavioral modeling methods for digital predistortion (DPD) of RF power amplifiers (PAs) are proposed to reduce the running complexity of DPD. The proposed full basis-propagating selection (FBPS) model and reduced-complexity FBPS (RC-FBPS) model give two reasonable ways to multiplex even-order basis functions, extending the basis-propagating selection (BAPS) model which only uses basis function delay and odd-order basis functions. The experimental results confirm that both the proposed FBPS and RC-FBPS models can achieve a good tradeoff between running complexity and performance.
This letter presents a low-resolution multiband transmitter architecture using direct digital synthesis (DDS) with dithering. Modern modulated signals constructed by a low-resolution digital-to-analog converter (DAC) suffer from quantization errors and distortions. The proposed technique reduces the quantization problems with the use of a proper oversampling ratio (OSR) and the dithering technique. This technique has lower cost and lower complexity compared to conventional transmitters while maintaining linearity performance. A 50-MHz 5G NR 256-quadratic-amplitude modulation (QAM) signal is used for concept validation at bands n1, n40, and n50. The proposed work achieved the adjacent channel leakage ratio (ACLR) and error vector magnitude (EVM) requirements of under −30 dBc and 3.5%, respectively. A single 3-bit 10-GS/s radio frequency (RF) DAC is used where the performance is similar to a conventional transmitter using a pair of 5-bit baseband in-phase and quadrature (IQ) DACs.
A wideband common-gate (CG) cascode low-noise amplifier (LNA) is demonstrated with a 28-nm fully depleted silicon on insulator (FDSOI) CMOS process. Wide bandwidth is achieved with a CG structure in the first stage and tightly coupled ( $k=0.86$ ) transmission line transformer (TLT) matching networks. A gm-boosting technique with cross-coupled capacitors improves the gain and noise figure (NF) of the CG amplifier. A high-input 1-dB gain compression point (IP1 dB) can be achieved by introducing an inter-stage inductor at the second-stage cascode amplifier. The LNA shows 19.3 dB gain, 3.1 dB NF, 19.8 GHz (63%) of 3 dB bandwidth, and −15.7 dBm IP1 dB at 28 GHz.
This letter proposes a modeling approach to handle nonuniform transient electromagnetic field coupling to transmission lines (TLs). In this approach, the matrix pencil method is applied to express the complex-valued external electric field at the horizontal direction along the TL by the first order of the exponential function in the frequency domain. The fitting procedure of the MPM only contains several matrix computation steps and does not need the initial value or any searching process, which leads to a significantly higher fitting efficiency. Moreover, the fitted results are formed with the first order of the exponential function, which ensures that the analytical solution of the TL equation can be formulated and thus has a high calculation efficiency. The proposed method was validated experimentally using a large scale of radiated wave simulator.
