Characterization of Electrical Memory Effects in Power Amplifier using Multi-tone Measurement

A wideband Gallium Nitride (GaN) HEMT power amplifier (PA) achieving 7.6W output power over 1.1 GHz bandwidth at f o = 2.75 GHz is presented. A systematic design and synthesis of wideband low-pass matching networks realizing optimal fundamental impedance is applied. These techniques have produced amplifiers with increased output power, efficiency and very large fractional bandwidth. Additionally, the PA can achieve a maximum 61% power added efficiency. This work seeks to maximize the usable bandwidth of the GaN HEMT PA to the maximum possible operating frequency.

In this letter, a novel model is proposed for modeling the nonlinearity and memory effects of power amplifiers. The classical Volterra model is modified through a function of the sum of nonlinearity order with sum of memory length. The parameters of this model can be extracted in digital domain since the model is analyzed based on the envelope signals. The model we proposed enables a substantial reduction in the number of coefficients involved, and with excellent accuracy.

Power amplifiers (PAs) are inherently nonlinear devices and are used in all transmitters of a communications equipment. To improve the performance of telecommunication channels, digital baseband predistortion is employed to compensate for PA nonlinearities. But most existing architectures base on the memoryless PA model. For wideband signal transmission, PA memory effects can no longer be ignored and memoryless predistortion has poor performance. In this paper, a predistortion scheme based on the indirect learning architecture is proposed and analyzed with application to the wideband wireless transmitter. Then the performance of overall transmitter system is verified by simulation results. The measurement shows that, when inputting a single carrier WCDMA signal with 5 MHz bandwidth, good AM-AM and AM-PM performance and roughly 20 dB improvement of out-of-distortion at an offset frequency of 5 MHz were achieved, after adding the proposed predistortion method.

Power amplifier (PA) nonlinearity plays a crucial role in determining the performance of radio frequency (RF) communication systems. Specifically, PA nonlinearity causes the amplitude-phase modulation (APM) constellation points to get distorted, which results in performance degradation. In this paper, we study the effect of PA nonlinearity on the bit error performance of spatial modulation MIMO (SM-MIMO) systems. When the PA distortion parameters are perfectly known at the receiver, the optimal detection rule is the one which minimizes the maximum-likelihood (ML) cost over the distorted constellation. For the case when the receiver has no knowledge of PA distortion parameters, we propose a receiver compensation technique which involves estimating the points of the distorted APM constellation based on training and performing detection using the estimated constellation. Simulation results show that, for SM-MIMO with 16-QAM, the proposed scheme achieves almost the same bit error performance as that achieved using the perfect knowledge of the PA parameters. Also, the proposed scheme is found to perform within 2–4 dB of the performance achieved using an ideal (linear) PA even at low values of input backoff.

Radio frequency power amplifier (PA) is an important part of the transmitter system, which can drive numerous output devices. However, the nonlinear characteristics of PA will cause serious harmonic interference, which leads to electromagnetic interference (EMI) problems. In this article, the nonlinear characteristics and the memory effect of PA are analyzed. The strong nonlinearity region and the weak nonlinearity region are divided according to the strength of the nonlinearity. For the strong nonlinearity, an encoder–decoder-based (E-D-based) artificial neural network model is proposed to predict the harmonic interference of PA. To promote the prediction of high-order harmonics when the input signal is small, a multilayer perceptron model is used for the weak nonlinearity region. The models can effectively predict the first five harmonics of PA, in which the mean absolute error of the fundamental wave is about 0.1 dB, the one of the second-order and the third-order harmonics is about 0.5 dB. Since transfer learning (TL) can simplify the training of the model based on the similarity of different tasks, TL based on model transfer is used to predict the harmonic interference of other PAs according to the existing models. The amount of data required for the modeling of PA can be greatly reduced and the accuracy of prediction can be guaranteed by applying TL. Ultimately, the proposed method can predict the harmonic interference rapidly and accurately according to the known excitation signal so that corresponding measures can be taken to avoid the influence of radiated spurious emission on the use of the sensitive receiving devices in the same electromagnetic environment.

Joint waveform and beamforming design is demonstrated as an effective approach to improve the wireless energy transfer (WET) performance. Meanwhile, the emerging technique of reconfigurable holographic surface (RHS) without the half-wavelength limitation is able to achieve a higher spatial gain. Hence, the RHS-assisted WET system is designed to enhance WET performance by optimizing the waveform and beamforming. However, the ideal linear power amplifier is always conceived, while the nonlinearity of the power amplifier is ignored. In this paper, a more practical nonlinear power amplifier (NPA)-based RHS-assisted WET system is proposed, where the energy harvesting (EH) model by considering the channel aging effect is derived to adapt to the time-varying channel. In order to improve the WET performance, a predistortion technique is adopted to tackle the nonlinear input-output characteristic of NPAs, while the waveform and beamforming are jointly optimized. Simulation results demonstrate that the RHS outperforms the phased array benchmark having the same size on the WET performance, when the NPAs work in the unsaturated region. Furthermore, the impact of the NPAs on the WET performance by considering channel aging effect is demonstrated, while the waveform design is invalid when the NPAs work near the saturation region.

Gallium Nitride (GaN) HEMTs are the focus of intense research and development due to their potential for the realization of power amplifiers (PAs) with high gain and record levels of power delivery. Much of the work in GaN HEMT development has been concentrated on performance improvement, and the demonstration of exceedingly higher power densities (>10W/mm). To demonstrate this performance, the device is typically subjected to bias levels greater than 40V, which result in large electric field stresses in the vicinity of the gate. Additionally, junction temperatures greater than 250/spl deg/C are predicted for the corresponding density of dissipated power. There is a lack of a comprehensive understanding of the effect of these large electrical and thermal stresses on the aging and degradation of GaN devices. There has only been a limited number of reports on the stability of GaN devices over time under RF dirve. In the present paper, we investigate the stability of GaN/AlGaN HEMTs under RF and DC stress. We provide a first look at the behavior of the device output power, quiescent currents, and leakage currents as a function of time under variable RF and DC stress conditions.

<para xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> This paper presents a quantitative metric for memory effects in power amplifiers (PAs) and applies it to various active device technologies and wireless system contexts. The proposed metric is mathematically founded in the dynamic two-tone distortion response, has a clear physical meaning in the important field of PA linearization and can be easily evaluated from either harmonic balance simulations or measurement data gathered in a microwave laboratory. In addition, a memoryless PA linearizer, optimum for reducing the integrated intermodulation distortion (IMD) power in the operation bandwidth for a two-tone excitation, is derived, providing a rigorous figure-of-merit of PA linearizability under static IMD compensation. The application of this figure-of-merit is then illustrated for three different PA prototypes based on Si LDMOS, InGaP/GaAs heterojunction bipolar transistors, and GaN high electron-mobility transistors, designed for 900-MHz (global system for mobile communications), 2.1-GHz (wideband code division multiple access), and 3.5-GHz (WiMax) wireless systems, respectively. </para>

This paper presents a hardware implementation of a digital predistorter (DPD) for linearizing RF power amplifiers (PAs) for wideband applications. The proposed predistortion linearizer is based on a nonlinear auto-regressive moving average (NARMA) structure, which can be derived from the NARMA PA behavioral model and then mapped into a set of scalable lookup tables (LUTs). The linearizer takes advantage of its recursive nature to relax the LUT count needed to compensate memory effects in PAs. Experimental support is provided by the implementation of the proposed NARMA DPD in a field-programmable gate-array device to linearize a 170-W peak power PA, validating the recursive DPD NARMA structure for W-CDMA signals and flexible transmission bandwidth scenarios. To the best of the authors' knowledge, it is the first time that a recursive structure is experimentally validated for DPD purposes. In addition to the results on PA efficiency and linearity, this paper addresses many practical implementation issues related to the use of FPGA in DPD applications, giving an original insight on actual prototyping scenarios. Finally, this study discusses the possibility of further enhancing the overall efficiency by degrading the PA operation mode, provided that DPD may be unavoidable due to the impact of memory effects.

A low-power, pulse-operation-based nonlinear distortion statistics module is proposed, comprising a nonlinear distortion error conversion unit and a nonlinear distortion error extraction unit. This module enables high-accuracy characterization of Ka-band power amplifier (PA) nonlinearity, providing critical guidance for advanced self-adaptive linearization. A novel nonlinear distortion statistical characteristics theory is introduced, leveraging the cumulative probability distribution (CPD) of PA input and output envelope voltages. Based on this theory, a search method for maximizing the accuracy of nonlinearity characterization is developed, mainly focusing on the dynamic range. The proposed statistics module is implemented in a 65-nm CMOS process. The nonlinear distortion error conversion unit first characterizes nonlinear distortion through a dual-path peak detector and comparators. Subsequently, the nonlinear distortion error extraction unit extracts the nonlinear distortion error signal using only logic gates and an error integrator, which performs an <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">xor</small> operation followed by a summation operation to derive the error. This design significantly reduces hardware complexity while maintaining high accuracy. Prototype verification is conducted through two scenarios. Characterizing accuracy, mainly focusing on the dynamic range, is validated by cascading the statistics module with an analog linearizer. A CMOS PA integrated with an analog linearizer is utilized as an example to demonstrate the potential for self-adaptive PA linearization. The characterizing accuracy measurement results demonstrate that the proposed module achieves a normalized dynamic range better than 0.57. In the self-adaptive PA linearization measurement, the statistic module’s iterative guidance for linearization achieves the adjacent channel power ratio (ACPR) of −35 dBc and an improvement of 2.2% in error vector magnitude (EVM). With a low power consumption of only 30 mW, the proposed statistics module achieves high accuracy in nonlinear characterization and provides a self-adaptive linearization solution for millimeter wave (mmWave) PAs, demonstrating the significant potential for advanced self-adaptive linearization applications.

Network densification is essential to enhance the areal capacity of future wireless networks, for which the user-centric network is a promising solution. In the perspective of network scalability, user association and cooperative beamforming (BF) among the base stations (BSs) should be designed elaborately to manage strong interference from nearby BSs. However, practical systems with hardware impairments including nonlinearity of power amplifiers may disturb these techniques and make the network difficult to be scalable. The nonlinearity of power amplifiers affects not only radio frequency (RF)-level signal but also performance according to the number of user equipments (UEs), resulting in necessity of re-designing network-level operations including BS-UE association. This paper proposes a deep learning-based cooperative BF framework for distributed networks with nonlinear power amplifiers. To optimize the cooperative BF for complicated nonlinear systems, we adopt unsupervised learning approach with constraints. To reduce the communication overhead of the network, we propose a novel neural network structure, from which the BSs can perform the cooperative BF in distributed manner. In particular, the information exchange between the central unit and the local BSs is reduced by designing neural network so that the local channel state information are utilized only in the local BSs while the central unit utilizes only covariance matrices. We show that the proposed scheme achieves a higher effective sum rate compared to the baseline schemes by adjusting the user association, BF, and power allocation to control interference and nonlinear distortion with reduced communication overhead.

Next-generation wireless systems require increased spectral and energy efficiency in terminals. This can only be achieved if the nonlinear characteristics of the radio front-end, primarily the power amplifier (PA), are known and considered while designing algorithms or optimizing radio links. PA datasheets contain limited information, and wideband PA measurements typically require advanced equipment and software. This paper proposes a measurement and signal processing framework that enables the acquisition of nonlinear PA characteristics using generic complex sample transmitters and receivers, e.g., software-defined radio (SDR). This paper also proposes a multistage signal calibration and synchronization procedure that allows obtaining nonlinear PA characteristics that are non-distorted by other phenomena. The correctness and usefulness of the proposed framework are demonstrated by measuring three PAs, each under varying supply voltage and carrier frequency. At the same time, the high variability of the obtained nonlinear characteristic justifies the need for PA measurements to obtain awareness of nonlinear characteristics.

The application of digital predistortion in base band signal is an extended method of amplifier linearization to reduce the adjacent channel interference (ACI) in those systems in which a varying envelope modulation scheme like OFDM and MQAM is used. Digital baseband predistortion is a highly cost-effective way to linearize power amplifiers (PAs), but most existing architectures assume that the PA has a memoryless nonlinearity. For wider bandwidth applications such as wideband code-division multiple access (WCDMA) or wideband orthogonal frequency-division multiplexing (W-OFDM), PA memory effects can no longer be ignored. In this paper we proposed a new technique for adaptation of digital predistorter that considers memory effects in power amplifiers. This method is a combination of two techniques, memory polynomial predistortion and slope-dependent method. This new technique is validated by using a 1.9 GHz 60W LDMOS power amplifier and various signals such as 2-carrier and 3-carrier CDMA.

The single event effect (SEE) and the total ionizing dose (TID) effect of a commercial enhancement mode gallium nitride (GaN) high electron nobility transistor (HEMT) with p-type gate structure and cascode structure are studied by using the radiation of heavy ions and &lt;sup&gt;60&lt;/sup&gt;Co gamma in this paper. The safe operating areas ofSEE, the sensitive parameters degradation of TID effect and the SEE and TID characteristics of GaN HEMT device are respectively presented. The experimental results show that the SEE and TID effect have less influence on the p-type gate GaN device. The linear energy transfer (LET) threshold of the single event Burnout effect (SEB) is higher than 37 MeV·cm&lt;sup&gt;2&lt;/sup&gt;/mg and the failure threshold of TID effect is above 1M rad (Si) for p-type gate GaN device. However, the GaN HEMT device with cascode structure is much more sensitive to SEE and TID effect than p-type gate GaN device. Under heavy ions at LET of 22 MeV·cm&lt;sup&gt;2&lt;/sup&gt;/mg and a cumulative dose of 200 krad (Si), the SEB phenomenon and parameters-degradation of cascode-type GaN HEMT are respectively observed. Besides, the circuit structure of the cascode-type GaN HEMT device is analyzed by using metallographic microscope imaging and focused ions beam technology. It reveals the possible reason why it is sensitive to SEB and TID for cascode-type GaN HEMT. These results show that the extra carriers caused by heavy ion radiation can tunnel the Schottky barrier formed by gate metal and AlGaN layer, leading to a large source-drain current in GaN HEMT device. Meanwhile, it is shown that the metal oxide semiconductor field-effect transistor in cascode circuit for TP90H180PS GaN HEMT may be the main reason why the cascode-type GaN HEMT is sensitive to TID.

In a previous study, we used linear frequency response analysis to show that naturalistic stimulation of spider primary mechanosensory neurons produced different response dynamics than the commonly used Gaussian random noise. We isolated this difference to the production of action potentials from receptor potential and suggested that the different distribution of frequency components in the naturalistic signal increased the nonlinearity of action potential encoding. Here, we tested the relative contributions of first- and second-order processes to the action potential signal by measuring linear and quadratic coherence functions. Naturalistic stimulation shifted the linear coherence toward lower frequencies, while quadratic coherence was always higher than linear coherence and increased with naturalistic stimulation. In an initial attempt to separate the order of time-dependent and nonlinear processes, we fitted quadratic frequency response functions by two block-structured models consisting of a power-law filter and a static second-order nonlinearity in alternate cascade orders. The same cascade models were then fitted to the original time domain data by conventional numerical analysis algorithms, using a polynomial function as the static nonlinearity. Quadratic models with a linear filter followed by a static nonlinearity were favored over the reverse order, but with weak significance. Polynomial nonlinear functions indicated that rectification is a major nonlinearity. A complete quantitative description of sensory encoding in these primary mechanoreceptors remains elusive but clearly requires quadratic and higher nonlinear operations on the input signal to explain the sensitivity of dynamic behavior to different input signal patterns.