Digital Predistortion

The article proposes a Digital Predistortion (DPD) methodology that substantially
meliorates the linearity of limited range Mobile Front Haul links for the extant Long-Term
Evolution (LTE) and future (5G) networks. Specifically, the DPD is employed to Radio over
Fiber links that contrive of Vertical Cavity Surface Emitting Lasers (VCSELs) working at
850 nm. Both, Memory and Generalized Memory Polynomial models are implied to Single
Mode (SM) and Multi-Mode (MM) VCSELs respectively. The effectiveness of the proposed
DPD methodology is analyzed in terms of Normalized Mean Square Error, Normalized
Magnitude, Normalized phase and Adjacent Channel Power Ratio. The demonstration has
been carried out with a complete (Long Term Evolution) LTE frame of 10 ms having 5
MHz bandwidth with 64-QAM modulation configuration. Additionally, the effectuality of
the proposed DPD technique is evaluated for varying levels of input power and link lengths.
The experimental outcomes signify the novel capability of the implied DPD methodology
for different link lengths to achieve higher system linearization.

Nonlinearity is a ubiquitous phenomenon existing in actual circuit systems. It gives arise of substantial distortion when transmitting and processing signals with such circuits, wherein the analogue front-ends in transmitters and receivers have had primary impact on the overall performance of the wireless communication systems. Therefore, the nonlinear compensation (linearization) techniques become essential to suppress the distortion to an acceptable extent in order to ensure sufficient low bit error rate. Furthermore, the increasing demands on higher data rate and ubiquitous interoperability between various multi-coverage protocols lead to the development of heterogeneous networks. The former demand pushes the communication system to use wider bandwidth and the latter one brings up severe coexistence problems. Having fully considered the problems raised above, the work in this Ph.D. thesis carries out extensive researches on the nonlinear compensations utilizing advanced digital signal processing techniques. The motivation behind this is to push more pro-cessing tasks to the digital domain, as it can potentially cut down the bill of materials (BOM) costs paid for the off-chip devices and reduce practical implementation diffi-culties. The work here is carried out using three approaches: numerical analysis & computer simulations; experimental tests using commercial instruments; actual im-plementation with FPGA. The primary contributions for this thesis are summarized as the following three points:
1) An adaptive digital predistortion (DPD) with fast convergence rate and low complexity for multi-carrier GSM system is presented. Albeit a legacy system, the GSM, however, has a very strict requirement on the out-of-band emission, thus it pre-sents a much more difficult hurdle for DPD application. It is successfully implement-ed in an FPGA without using any other auxiliary processor. A simplified multiplier-free NLMS algorithm, especially suitable for FPGA implementation, for fast adapting the LUT is proposed. Many design methodologies and practical implementation issues are discussed in details. Experimental results have shown that the DPD performed robustly when it is involved in the multi-channel transmitter.
2) The next generation system (5G) will unquestionably use wider bandwidth to support higher throughput, which poses stringent needs for using high-speed data converters. Herein the analog-to-digital converter (ADC) tends to be the most expen-sive single device in the whole transmitter/receiver systems. Therefore, conventional DPD utilizing high-speed ADC becomes unaffordable, especially for small base sta-tions (micro, pico and femto). A digital predistortion technique utilizing spectral ex-trapolation is proposed in this thesis, wherein with band-limited feedback signal, the requirement on ADC speed can be significantly released. Experimental results have validated the feasibility of the proposed technique for coping with band-limited feed-back signal. It has been shown that adequate linearization performance can be achieved even if the acquisition bandwidth is less than the original signal bandwidth. The experimental results obtained by using LTE-Advanced signal of 160 MHz band-width are quite satisfactory, and to the authors’ knowledge, this is the first high-performance wideband DPD ever been reported.
3) To address the predicament that mobile operators do not have enough contig-uous usable bandwidth, carrier aggregation (CA) technique is developed and imported into 4G LTE-Advanced. This pushes the utilization of concurrent dual-band transmit-ter/receiver, which reduces the hardware expense by using a single front-end. Com-pensation techniques for the respective concurrent dual-band transmitter and receiver front-ends are proposed to combat the inter-band modulation distortion, and simulta-neously reduce the distortion for the both lower-side band and upper-side band sig-nals.

The radio frequency power amplifier (RF-PA) within a digital wireless transmitter is a critical component regarding both the energy consumption and the signal quality. Especially due to today's broadband multicarrier modulation methods that generate signals with high peak-to-average power ratio, it is very hard to construct RF-PAs that achieve good energy efficiency and fulfill the strict linearity requirements imposed by the standard. Because of this, the digital predistortion (DPD) of RF-PAs has become a key technique for implementing energy efficient, high data rate wireless transmitters.

This thesis investigates theoretical foundations and practical methods for the behavioral modeling and DPD of RF-PAs. The main contributions are a semi-physical model of the joint linearity-efficiency characteristics of RF-PAs, a detailed analysis of polynomial baseband models of RF-PAs focusing on the often neglected even-order terms in baseband, and a collection of practical methods for the dual-band DPD of RF-PAs.

The trade-off between the linearity and efficiency of RF-PAs is investigated based on the semi-physical RF-PA model. For this purpose, linearity and efficiency quantification methods are introduced and applied to the model. Furthermore, an overview on highly efficient RF-PA operation modes and efficiency enhancement methods is given.

A central result regarding polynomial baseband models is the first formal justification of even-order terms in baseband. This is achieved by deriving explicit passband-baseband pairs for the quasi-memoryless polynomial and the Volterra series, which show that even-order terms in baseband correspond to modified basis functionals in passband.

Another central result of the presented analysis of polynomial baseband models is the formulation and proof of the phase homogeneity requirement, which represents a necessary symmetry of all complex baseband models of time-invariant passband systems.

The practical methods for the dual-band DPD of RF-PAs include a method for dual-band crest factor reduction by clipping and error-filtering, a dual-band DPD model based on a vector-switched generalized memory polynomial, and a dual-band DPD training algorithm based on the indirect and direct learning architectures. These methods were implemented in MATLAB and validated at the 2017 DPD competition at the IEEE International Microwave Symposium, reaching the first place within the competition.

This paper proposes a spectra-folding feedback (SFFB) architecture for digital predistortion (DPD) of concurrent dual-band power amplifiers (PAs). Conventional dual-band DPD contains a dual-branch feedback path to receive the PA output signals at respective bands. The proposed architecture, however, is constructed by a single-branch receiving path, so that the hardware expense (such as down converters, filters, analog-to-digital converters, etc.) of the feedback path is significantly reduced. In this architecture, simple modifications are made to the down converter to allow the use of only one ADC. With the proposed signal processing in the baseband, the SFFB DPD can achieve comparative performance as the conventional dual-branch DPD. Experiments were performed using various configurations of LTE/LTE-Advanced (long term evolution/ long term evolution advanced) signals of up to 80 MHz contiguous bandwidth and carrier spacing of 1 GHz. It is clearly shown by the experimental results that the proposed SFFB DPD can achieve satisfactory linearization performance, with significantly reduced hardware complexity

This brief presents the measurement/modeling of a dual-input outphasing Chireix power amplifier (PA) for the analysis of nonlinear and memory effects behavior when excited with modulated signals. Moreover, we formulate a cubic-spline (CS) memory model with a closed-loop for the digital predistortion (DPD) linearization algorithm. An FPGA system has been developed to automate the outphasing angles and their respective asymmetrical incident power levels applied at the Class-F PAs of the Chireix under test. Based on experimental results, the CS model offers a DPD with a normalized mean square error (NMSE) of −21.87 dB, and −52.47 dB for a WCDMA 5-MHz and an LTE-A 10-MHz with −26.84 dB and −43.91 dB in extraction and prediction, respectively. An output power improvement of −43.5 dBc in adjacent channel power ratio (ACPR) with 9.3 dB PAPR yield 50% average efficiency enhance at 2 GHz, is achieved.

This article proposes a Digital Predistortion (DPD) technique in order to enhance the linearity of short and limited range Mobile Front Haul (MFH) links for the present Long-Term Evolution (LTE) and future (5G) cellular networks. The proposed technique is applied to 850 nm Single Mode Vertical Cavity Surface Emitting Lasers (VCSELs) and Standard Single-Mode Fibers (SSMFs) low cost and low consumption Radio-over-Fiber (RoF) link. Memory Polynomial is employed using Indirect Learning Architecture (ILA). Moreover, the resilience of the DPD approach to changes in input power levels is investigated. The improvements in the quality of the received signal due to the proposed solution are illustrated with reference to scenarios of applicative interest. The performance of proposed technique is evaluated in terms of Adjacent Channel Power Ratio (ACPR) and Normalized Mean Square Error (NMSE) for entire LTE frame of 10ms for 5 MHz band signal having 64 QAM modulation format.

— Since its introduction, compressed sampling (CS) has found use in various applications ranging from image restoration, radar and sensing to channel and system identification. Recently in the radio frequency (RF) power amplifier (PA) design and linearization communities, there have been many attempts to utilize the CS technique to enable the development of efficient wireless transmitters. This paper provides a brief review of the use of CS in PA and transmitter linearization. Mainly two approaches are discussed: the use of CS to recover amplitude-distorted signals, and the use of CS to reduce the complexity of the digital predistorters. Experimental results obtained using an envelope tracking (ET) PA prototype show the potential and value of the CS technique in developing efficient predistorters at a low computational cost.

Digital predistortion (DPD) using baseband signals is commonly used for power amplifier linearization. This paper is devoted to this subject and aims to reduce DPD complexity. In this study, we propose a structure that allows to decrease the number of DPD parameters by using multiple blocks, with each one of them dedicated to characterizing the non-linear behavior and/or memory effects. Such a structure is based on the feedback Wiener system, involving a FIR filter used as a feedback path to reproduce the PA inverse dynamics. A memory polynomial block (MP) is inserted as the final element to minimize the modeling errors. A relevant model identification method, based on an iterative algorithm, has been developed as well. The proposed architecture is used for the linearization of a commercial class-AB LDMOS RF PA by NXP Semiconductors, in wideband communication systems. Comparison of performance with the conventional generalized memory polynomial model (GMP) shows that the proposed mo...