Xingbin Zeng

Accurate Time-Delay Estimation and Alignment for RF Power Amplifier/Transmitter Characterization

This paper presents a cross-covariance-based time- delay estimation method for accurate determination of the time- delay between the input and the output of a RF power amplifier/transmitter. In order to overcome the time resolution limitation caused by the relative low sampling rate of the analog- to-digital converter in the baseband signal acquisition system, the Lagrange interpolation is utilized to improve the time resolution. A two-step method for time-delay estimation/alignment is proposed to increase the dynamic range for time-delay estimation/alignment under the certain computation resource consumption condition. The experimental validation results have been provided to illustrate the effectiveness of the proposed approach for accurate time-delay estimation and alignment.

Generalized Hammerstein-Based Dynamic Nonlinear Behavior Models for Wideband RF Transmitters

This paper presents a generalized Hammerstein-based nonlinear model to simulate the dynamic behaviors of a wideband RF transmitter suitable for a 3G communication system. Such a model can be extracted easily with the singular value decomposition (SVD) or the recursive least squares (RLS) algorithms after converting the series input data sequence into a parallel data sequence. The experimental validation results reveal that the generalized Hammerstein-based nonlinear model can accurately predict the dynamic nonlinear characteristics of the wideband RF transmitter.

Memory Effect Modeling of Wideband Wireless Transmitters Using Neural Networks

In this paper a three-layer real-valued time-delayed neural network (RVTDNN) is employed to simulate the memory effects of a wideband wireless transmitter. The RVTDNN is trained at first using a Matlab program, and then it is implemented in Agilent Advanced Design System software. Different training algorithms have been applied to the neural network to extract its weights and biases, and it is found that the Levenberg-Marquardt (LM) algorithm exhibits the best performance. A look-up-table based memoryless predistorter is cascaded to the RVTDNN model to validate the capability of the RVTDNN model in simulating the memory effects of the transmitter. The validation results demonstrate that the identified RVTDNN model can accurately mimic the memory effects of a wideband wireless transmitter prototype, which is based on a 60-watt push-pull GaAs FET power amplifier, under a two-carrier 3GPP-FDD excitation signal.

Accurate Permittivity Measurement Using the Cavity Perturbation Technique at ISM 5.8GHz Radio Band

In this paper, a new calibration method with two-filling factors for the permittivity measurement is proposed. With this calibration method, it is not necessary to measure the volume of the small sample, which not only eliminates the measurement error of the sample volume, but also simplifies the measurement procedure. To improve the measurement accuracy, the Cursor method, the resonance curve area (RCA) method and the Lorentzian fit method are utilized to process the measured resonant curve data so as to obtain accurate resonant frequency and quality factor. Finally, the obtained permittivity measurement results of a liquid sample validate the accuracy of this permittivity measurement approach.

Accurate modeling of wideband RF Doherty power amplifiers using dynamic nonlinear models

In this paper, a 100-watte LDMOS Doherty power amplifier is designed to test the modeling capability of the widely used memory polynomial model and the augmented Wiener model for the dynamic nonlinear behaviors of a wideband PA. The augmented Wiener model and its identification algorithm are discussed in details. The experimental validation results demonstrate that the augmented Wiener model can simulate the dynamic nonlinearity of this Doherty PA more accurately than the memory polynomial model.

Spectral Methods for Accurate Identification and Quantification of Memory Effects of Wideband RF Power Amplifiers

In this paper, a spectral method based on the memoryless predistortion technique is presented to identify accurately the memory effects of wideband RF power amplifiers stimulated by practical modulated signals. According to this new definition of the memory effects, a novel metric to measure quantitatively the memory effects in wideband power amplifiers is proposed. With the proposed memory effect metrics, the intensity of the memory effects for different power amplifiers, under a variety of the excitation modulated signals, can be quantitatively evaluated.

Hammerstein-Like Predistortion Techniques for Wideband Wireless Power Amplifier Linearization

In this paper, a novel Hammerstein-like predistortion technique is proposed to linearize the wideband wireless power amplifiers. This Hammerstein-like predistortion technique is based on a Hammerstein-like block configuration, including a memoryless strong nonlinear block followed by a dynamic weak nonlinear block. Here the memoryless strong nonlinear block is utilized to account for the static memoryless nonlinear property, and the weak nonlinear block is employed to eliminate the memory effects. The validation results demonstrate the superiority of the proposed Hammerstein-like predistortion technique in suppressing the memory effects in comparison to a traditional one.

Linearization of wideband RF Doherty power amplifiers with complex dynamic nonlinearities

In this paper, a traditional look-up-table (LUT) predistorter and a memory polynomial predistorter are applied to a LDMOS Doherty power amplifier to pre-compensate its nonlinearities so as to make its linearity be able to satisfy the requirements of a wideband wireless communication system. The experimental validation results reveal that the widely used memory polynomial cannot effectively suppress the out-of-band spectrum regrowth caused by the complex dynamic nonlinearity of this Doherty power amplifier. In fact, the traditional memoryless LUT predistorter provides unexpected better linearization performance than the memory polynomial predistorter.

Characterization and Modeling of Wideband Wireless Transceivers

In this paper, we proposed a practical approach to accurately characterize a 3G wideband wireless transceiver, which will exhibit strong memory effects. The memory effects can be easily identified by accurate evaluation and alignment of the time delay between the input and output of the transceiver. Using the extracted transceiver characteristic data, a memory polynomial model is constructed to mimic the dynamic properties of the transceiver. The experimental validation results, which are obtained with a transceiver prototype based on an L-band 60- Watt GaAs FET push-pull amplifier, a vector signal generator, a spectrum analyzer and a vector signal analyzer, demonstrate the effectiveness of the transceiver characterization approach and the accuracy of the dynamic nonlinear models discussed.

Memory Effect Pre-compensation for Wideband RF Power Amplifiers Using FIR-Based Weak Nonlinear Filters

This paper proposes an effective memory effect pre-compensation technique, which is based on a weak nonlinear finite impulse response (FIR) filter. This weak nonlinear FIR filter comprises two FIRs in parallel with different inputs. The performance evaluation of this pre-compensation technique is carried out on an L-band push-pull GaAs FET 45-dBm peak-envelope-power amplifier using 3GPP-FDD WCDMA signals. The validation results in spectrum domain demonstrate noticeable superiority of the proposed weak nonlinear FIR filter in suppressing the memory effects in comparison to a traditional FIR filter.