Machine Learning for Digital Front‐End

Gilabert, Pere L.; López‐Bueno, David; Pham, Thi Quynh Anh; Montoro, Gabriel

doi:10.1002/9781119562306.ch17

Cited by 4 publications

(6 citation statements)

References 115 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The technology development of AI for wireless communication and RF systems has become an important challenge in the academic, and in industrial communities. Additionally, AI addresses major challenges in communication networks by optimizing parameters in terms of specifications, such as latency and flexibility of efficiency 147 . Figure 23 depicts branches of AI include machine learning and deep learning with explanation of the related subsets, where it is used to characterize communication channels in the digital domain.…”

Section: Optimization Methods In Communicationsmentioning

confidence: 99%

Overview of evolutionary algorithms and neural networks for modern mobile communication

Kouhalvandi

Shayea

Özoğuz

et al. 2022

Trans Emerging Tel Tech

View full text Add to dashboard Cite

The sixth generation (6G) of mobile networks must support the huge growth of mobile connections and provides various intelligent services in future mobile networks. For that, designing high-performance network architecture and communication systems (CSs) as well as finding coherent solutions for the determined problems is critical demand that needs to be achieved in 6G networks.Optimal solutions are mainly sought by using modifiable, smart, and perceptive algorithms aimed at optimizing more specific tasks. Therefore, advanced optimization methods are highly required to accommodate the requirements of CSs efficiently. That will be in various parts of the future networks such as advanced mobility management, multi communication links, efficient power consumption, ultra-lower latency, ultra-security, high speed, and reliable connectivity.Accordingly, highly accurate and smart functions must be modeled to optimize the required communication parameters in the network. This study provides a comprehensive overview of optimization methods that may need further investigations and developments to be applied in 6G networks at various parts of networks. A detailed theoretical description for each method is presented and discussed to elucidate future research directions for optimizing specific characteristics with multi-objective optimizations. Moreover, this article illustrates the conceptual and structural viewpoints of reported optimization methods. Also, the capability of various optimization methods that can offer industrial solutions in 6G is discussed. The potential applications of each method are also analyzed. Finally, this article presented the research issues and future directions of optimization technology and research gaps that need to be addressed before the standardization of 6G networks. INTRODUCTIONRecently, innovative solutions of communication systems (CSs) have been developed impressively, as data traffic is increasing exponentially by more than 50% per year. 1 Several factors will contribute for increasing data demand massively, for example, as predicted by CISCO, the video will make up 82% of the all IP traffic by 2022. 2,3 This massive increase in data traffic will cause several new issues that will facing the implementation of future mobile networks. That will include a dramatic increase for wider spectrum need, 4 the necessity for big system capacity, 5 further mobility issues, 6,7 and higher energy consumption, 8 in which cannot be handled efficiently by the current cellular networks.

show abstract

Section: Optimization Methods In Communicationsmentioning

confidence: 99%

Overview of evolutionary algorithms and neural networks for modern mobile communication

Kouhalvandi

Shayea

Özoğuz

et al. 2022

Trans Emerging Tel Tech

View full text Add to dashboard Cite

show abstract

“…architecture (left) and illustration of the spectra of the complex baseband signal and the RF signal at a given PA output with typical RF impairments (right). Source: [5].…”

Section: A State-of-the-artmentioning

confidence: 99%

“…We define how to apply the original OLS technique to reduce the polynomial model basis functions and ANN dataset features in MIMO DPD, knowing that it is able to outperform OMP in DPD basis selection and perform equally well than its later variants [34]. In polynomial-based DPD, PCA is commonly used in the identification subsystem to reduce the number of basis and avoid an ill-conditioned estimation while reducing the complexity of the LS calculation [5]. We propose using PCA to reduce the number of features in large ANN MIMO DPD datasets.…”

Section: B Contribution and Noveltymentioning

confidence: 99%

Training Data Selection and Dimensionality Reduction for Polynomial and Artificial Neural Network MIMO Adaptive Digital Predistortion

López‐Bueno

Montoro

Gilabert

2022

IEEE Trans. Microwave Theory Techn.

Self Cite

View full text Add to dashboard Cite

In 5G and beyond radios, the increased bandwidth, the fast-changing waveform scenarios, and the operation of large array multiple-input multiple-output (MIMO) transmitter architectures have challenged both the polynomial and the artificial neural network (ANN) MIMO adaptive digital predistortion (DPD) schemes. This paper proposes training data selection methods and dimensionality reduction techniques that can be combined to enable relevant reductions of the DPD training time and the implementation complexity for MIMO transmitter architectures. In this work, the combination of an efficient uncorrelated equation selection (UES) mechanism together with orthogonal least squares (OLS) is proposed to reduce the training data length and the number of basis functions at every behavioral modeling matrix in the polynomial MIMO DPD scheme. For ANN MIMO DPD architectures, applying UES and principal component analysis (PCA) is proposed to reduce the input dataset length and features, respectively. The UES-OLS and the UES-PCA techniques are experimentally validated for a 2x2 MIMO test setup with strong power amplifier input and output crosstalk.

show abstract

“…The proposed NN shown in Figure 3 uses a feedforward fully connected (FC) structure. Based on the interconnection pattern or architecture, we can distinguish between feedforward networks (FNNs) and recurrent (or feedback) networks (RNNs) [27]. The feedforward network is considered since it is the most used NN and according to the universal approximation theorem, it can approximate any non-linear function with any desired error [28].…”

Section: Nn Trainingmentioning

confidence: 99%

“…The feedforward network is considered since it is the most used NN and according to the universal approximation theorem, it can approximate any non-linear function with any desired error [28]. An FC structure in a densely populated NN may increase requirements for hardware resources, but in many applications, the weight of some interconnections can be set to zero without loss of accuracy, which results in sparsely connected layers [27]. The sparse structure is out of the scope of this work.…”

Section: Nn Trainingmentioning

confidence: 99%

Highly non‐linear and wide‐band mmWave active array OTA linearisation using neural network

Jalili

Zhang

Hintsala³

et al. 2021

IET Microwaves Antenna & Prop

View full text Add to dashboard Cite

This paper proposes a neural network (NN)-based over-the-air (OTA) linearisation technique for a highly non-linear and wide-band mmWave active phased array (APA) transmitter and compares it with the conventional memory polynomial model (MPM)based technique. The proposed NN effectively learns the distinctive non-linear distortions, which may not easily fit to existing MPM solutions, and can, therefore, successfully cope with the challenges introduced by the high non-linearity and wide bandwidth. The proposed technique has been evaluated using a state-of-the-art 4 � 4 APA operating in highly non-linear regions at 28 GHz with a 100-MHz-wide 3GPP base-station signal as input. Experimental results show the pre-distortion signal generated by the NN exhibits the peak-to-average power ratio (PAPR) much lower than the one generated by MPM and consequently superior linearisation performance in terms of adjacent channel leakage ratio (ACLR) and error vector magnitude (EVM) for high non-linearity cases. Using the proposed NN-based linearisation technique, an improvement of 5-dB ACLR and 7% points in EVM are achieved, which demonstrates the promising potential of this technique for emerging broadband communication systems such as 5G/6G and low Earth orbit (LEO) satellite networks. K E Y W O R D S 5G mobile communication, active antenna arrays, learning (artificial intelligence), linearisation techniques, millimetre wave amplifiers, millimetre wave antenna arrays, satellite communication 1 | INTRODUCTION Recent wireless communication systems operating at mmWave are using active phased array (APA) transmitters together with multi-input multi-output (MIMO) systems to improve the system capacity and data rates of the wireless networks. The traditional linearisation techniques like digital pre-distortion (DPD) that are mainly based on the memory polynomial model (MPM) are facing new challenges [1] such as the following.� Highly integrated front-ends and a large number of PA's do not allow placement of feedback circuits for each branch so a single-input single-output (SISO) over-the-air (OTA) model is needed.� mmWave frequencies and wide-band linearisation must be handled. � The increasing desire for high power efficiency requires linearisation of power-efficient but highly non-linear APAs.Several solutions using a SISO model and modified DPD algorithms for linearisation have been proposed to combat the above challenges [2][3][4][5][6]. A SISO modelling where the entire transmitter is considered as a two-port system has been presented in [7-10] using an observation receiver in far-field together with using an MPM-based DPD technique for linearisation of the antenna array in the presence of crosstalk. In [11], the potential mismatches between PAs have been This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

show abstract

Machine Learning for Digital Front‐End

Cited by 4 publications

References 115 publications

Overview of evolutionary algorithms and neural networks for modern mobile communication

Overview of evolutionary algorithms and neural networks for modern mobile communication

Training Data Selection and Dimensionality Reduction for Polynomial and Artificial Neural Network MIMO Adaptive Digital Predistortion

Highly non‐linear and wide‐band mmWave active array OTA linearisation using neural network

Contact Info

Product

Resources

About