A comparative analysis of the complexity/accuracy tradeoff in the mitigation of RF MIMO transmitter impairments

Abstract

Abstract-This paper presents a comparative analysis of the complexity accuracy tradeoff in state-of-the-art RF MIMO transmitter mitigation models. The complexity and accuracy of the candidate models depends on the basis functions considered in these models. Therefore, a brief description of the mitigation models is presented accompanied by derivations of the model complexities in terms of the number of FLOPs. Consequently, the complexity accuracy tradeoff in the candidate models is evaluated for a 2 × 2 RF MIMO transmitter. Furthermore, the model complexities are analyzed for increasing nonlinear orders and number of antennas.I. INTRODUCTION RF MIMO transmitters are the contemporary technology in wireless communication for achieving high date rate capacities. However, an energy efficient operation of RF MIMO transmitters results in hardware impairments that are more severe than SISO transmitter impairments. This severity is caused by additional distortion effects such as crosstalk between the MIMO transmission paths [1].Several mitigation models for SISO transmitter impairments have been proposed in literature [2]. Given an infinite order and complexity, these models can achieve a good error performance. However, at a given finite complexity, only part of the accuracy can be achieved [2]. Hence, it is necessary to analyze the complexity accuracy tradeoff in these models to get a complete picture of their effectiveness.The complexity of SISO RF transmitter mitigation models is often determined based on the number of parameters which can be determined form the nonlinear order and memory. However, this representation may not always be an appropriate measure for complexity since mitigation models for RF transmitters are implemented on digital signal processing (DSP) hardware [3]. The computational resources of such digital platforms are predominantly allocated for performing additions, subtractions and multiplications for which the number of floating point of operations (FLOPs) is an accurate measure and not the number of parameters. Hence, a complexity accuracy tradeoff analysis for SISO transmitter mitigation models in terms of the number of FLOPs has been presented in [3].Transitioning to RF MIMO transmitters, several mitigation models have also been proposed based on the MIMO Volterra series [4]. For example, the MIMO parallel Hammerstein (PH) model [5], the MIMO memory polynomial (MP) model [6] and the generalized (GMP) and extended generalized MIMO memory polynomial models (EGMP) [7]. These models incorporate higher complexity basis functions to account for the additional

Predistortion-Based Linearization for 5G and Beyond Millimeter-Wave Transceiver Systems: A Comprehensive Survey

“…Neural Network-Based Model Waveform Processing at Transmission and Observation Path [67], [81], [89]- [103] [110]- [120], [122], [123] [67], [75], [82]- [88] DAC and ADC nonidealities [29], [130], [133]- [141] RF Sub-Components Effects [130]- [132], [142], [145] [130], [131] Amplitude-Compensated Phase Shifter [78], [79], [148]- [155], [157], [158] Phase-Compensated and Low-Noise Variable Gain Amplifier [44], [149], [159]- [169] SVM-Based Model better output performance. Since the invention of wireless communication, the linearity of transmitting signals hold paramount importance [73], [74].…”

“…However, in this case, the VoS is considered as a complete polynomial model to account for the distortion information of IB and OOB spectrum, measured as normalized mean square error (NMSE) and ACLR/ACPR, respectively. But it comes at the cost of very high computational inefficiency, which increases severely at the MIMO level due to the density of an overdetermined coefficient matrix (a) for each input (X) and output (y) polynomial function [87], [88], where output signal Y out (n) of equation ( 1) can be expressed in a vector (matrix) form as:…”