An Interpretable Neural Network for Configuring Programmable Wireless Environments

Liaskos, Christos; Tsioliaridou, Ageliki; Nie, Shuai; Pitsillides, Andreas; Ioannidis, Sotiris; Akyildiz, Ian F.

doi:10.1109/spawc.2019.8815428

Cited by 69 publications

(59 citation statements)

References 18 publications

(28 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [266], the authors introduce a neural-network-based approach for configuring the behavior of tiles in RIS-based environments. The wireless propagation is modeled as a custom, interpretable, back-propagating neural network, in which the RIS elements act as nodes and their cross-interactions as links.…”

Section: T Machine Learning Based Designmentioning

confidence: 99%

Smart Radio Environments Empowered by Reconfigurable Intelligent Surfaces: How It Works, State of Research, and The Road Ahead

Renzo

Zappone

Debbah

et al. 2020

IEEE J. Select. Areas Commun.

2,226

918

View full text Add to dashboard Cite

What is a reconfigurable intelligent surface? What is a smart radio environment? What is a metasurface? How do metasurfaces work and how to model them? How to reconcile the mathematical theories of communication and electromagnetism? What are the most suitable uses and applications of reconfigurable intelligent surfaces in wireless networks? What are the most promising smart radio environments for wireless applications? What is the current state of research? What are the most important and challenging research issues to tackle? These are a few of the many questions that we investigate in this short opus, which has the threefold objective of introducing the emerging research field of smart radio environments empowered by reconfigurable intelligent surfaces, putting forth the need of reconciling and reuniting C. E. Shannon's mathematical theory of communication with G. Green's and J. C. Maxwell's mathematical theories of electromagnetism, and reporting pragmatic guidelines and recipes for employing appropriate physics-based models of metasurfaces in wireless communications.

show abstract

Section: T Machine Learning Based Designmentioning

confidence: 99%

Smart Radio Environments Empowered by Reconfigurable Intelligent Surfaces: How It Works, State of Research, and The Road Ahead

Renzo

Zappone

Debbah

et al. 2020

IEEE J. Select. Areas Commun.

2,226

918

View full text Add to dashboard Cite

show abstract

“…The specifics depend on the NN used, but this model is considered inherently interpretable. NN has been used for programmable wireless environments (PWEs) [110]. TS approximation [111] is a fuzzy network approximation of other NNs.…”

Section: B Interpretability Via Mathematical Structurementioning

confidence: 99%

A Survey on Explainable Artificial Intelligence (XAI): Toward Medical XAI

Tjoa

Guan

2021

IEEE Trans. Neural Netw. Learning Syst.

1,195

547

View full text Add to dashboard Cite

Recently, artificial intelligence and machine learning in general have demonstrated remarkable performances in many tasks, from image processing to natural language processing, especially with the advent of deep learning (DL). Along with research progress, they have encroached upon many different fields and disciplines. Some of them require high level of accountability and thus transparency, for example, the medical sector. Explanations for machine decisions and predictions are thus needed to justify their reliability. This requires greater interpretability, which often means we need to understand the mechanism underlying the algorithms. Unfortunately, the blackbox nature of the DL is still unresolved, and many machine decisions are still poorly understood. We provide a review on interpretabilities suggested by different research works and categorize them. The different categories show different dimensions in interpretability research, from approaches that provide "obviously" interpretable information to the studies of complex patterns. By applying the same categorization to interpretability in medical research, it is hoped that: 1) clinicians and practitioners can subsequently approach these methods with caution; 2) insight into interpretability will be born with more considerations for medical practices; and 3) initiatives to push forward data-based, mathematically grounded, and technically grounded medical education are encouraged.

show abstract

“…It should be emphasized that the availability of perfect CSI is an idealistic assumption. Nevertheless, the algorithms proposed under this assumption are still useful as a reference point for studying the theoretical performance gain brought by the IRS, as well as providing training labels for the machine learning based joint beamforming designs, e.g., [30] and [31]. How to obtain CSI at IRS is a difficult task.…”

Section: A Channel Modelmentioning

confidence: 99%

Weighted Sum-Rate Maximization for Intelligent Reflecting Surface Enhanced Wireless Networks

Guo

Liang

Chen

et al. 2019

2019 IEEE Global Communications Conference (GLOBECOM)

256

226

View full text Add to dashboard Cite

Intelligent reflecting surface (IRS) is a promising solution to build a programmable wireless environment for future communication systems. In practice, an IRS consists of massive low-cost elements, which can steer the incident signal in fully customizable ways by passive beamforming. In this paper, we consider an IRS-aided multiuser multiple-input single-output (MISO) downlink communication system. In particular, the weighted sum-rate of all users is maximized by joint optimizing the active beamforming at the base-station (BS) and the passive beamforming at the IRS. In addition, we consider a practical IRS assumption, in which the passive elements can only shift the incident signal to discrete phase levels. This non-convex problem is firstly decoupled via Lagrangian dual transform, and then the active and passive beamforming can be optimized alternatingly. The active beamforming at BS is optimized based on the fractional programming method. Then, three efficient algorithms with closed-form expressions are proposed for the passive beamforming at IRS. Simulation results have verified the effectiveness of the proposed algorithms as compared to different benchmark schemes.Index Terms-Intelligent reflecting surface (IRS), large intelligent surface (LIS), passive radio, beamforming, multiple-inputmultiple-output (MIMO), fractional programming.

show abstract

An Interpretable Neural Network for Configuring Programmable Wireless Environments

Cited by 69 publications

References 18 publications

Smart Radio Environments Empowered by Reconfigurable Intelligent Surfaces: How It Works, State of Research, and The Road Ahead

Smart Radio Environments Empowered by Reconfigurable Intelligent Surfaces: How It Works, State of Research, and The Road Ahead

A Survey on Explainable Artificial Intelligence (XAI): Toward Medical XAI

Weighted Sum-Rate Maximization for Intelligent Reflecting Surface Enhanced Wireless Networks

Contact Info

Product

Resources

About