Enabling Large Intelligent Surfaces With Compressive Sensing and Deep Learning

By Abdelrahman Taha, Muhammad Alrabeiah, Ahmed Alkhateeb. IEEE Access, 2019

737 citations recorded by the source corpus.

Employing large intelligent surfaces (LISs) is a promising solution for improving the coverage and rate of future wireless systems. These surfaces comprise massive numbers of nearly-passive elements that interact with the incident signals, for example by reflecting them, in a smart way that improves the wireless system performance. Prior work focused on the design of the LIS reflection matrices assuming full channel knowledge. Estimating these channels at the LIS, however, is a key challenging problem. With the massive number of LIS elements, channel estimation or reflection beam training will be associated with (i) huge training overhead if all the LIS elements are passive (not connected to a baseband) or with (ii) prohibitive hardware complexity and power consumption if all the elements are connected to the baseband through a fully-digital or hybrid analog/digital architecture. This paper proposes efficient solutions for these problems by leveraging tools from compressive sensing and deep learning. First, a novel LIS architecture based on sparse channel sensors is proposed. In this architecture, all the LIS elements are passive except for a few elements that are active (connected to the baseband). We then develop two solutions that design the LIS reflection matrices with negligible training overhead. In the first approach, we leverage compressive sensing tools to construct the channels at all the LIS elements from the channels seen only at the active elements. In the second approach, we develop a deep-learning based solution where the LIS learns how to interact with the incident signal given the channels at the active elements, which represent the state of the environment and transmitter/receiver locations. We show that the achievable rates of the proposed solutions approach the upper bound, which assumes perfect channel knowledge, with negligible training overhead and with only a few active elements, making them promising for future LIS systems.

Subfields: Aerospace Engineering, Electrical and Electronic Engineering

View Paper: https://www.semanticscholar.org/paper/23569428fea624f07bf4cac88c3da263404b2c21

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

In Electrical and Electronic Engineering

Citations in the corpus

Capacity Characterization for Intelligent Reflecting Surface Aided MIMO Communication (2019, neutral): “To overcome this challenge, [7], [8] advocated”
A Survey on Channel Estimation and Practical Passive Beamforming Design for Intelligent Reflecting Surface Aided Wireless Communications (2021, methodology): “Deep learning • Deep neural network-based training [50], [68], [72] • Nonlinear signal model Various cases including narrowband/broadband and single-/multi-user cases • Convolutional neural network-based training [69]–[71], [95], [96], [105] • Federated/supervised/reinforcement learning [52]–[54], [”
Holographic MIMO Communications: Theoretical Foundations, Enabling Technologies, and Future Directions (2022, methodology): “[61] employed an LIS with the majority of its elements being passive and a few randomly distributed active elements.”
Deep Reinforcement Learning Based Intelligent Reflecting Surface Optimization for MISO Communication Systems (2020, methodology): “Dueto the recent advances of artiﬁcial intelligence, especially deep learning (DL), in wireless communication, [9] and [10] utilized DL methods to the phase shift design.”
Reconfigurable Intelligent Surfaces: Principles and Opportunities (2020, methodology): “[110] exploited the DL method for learning the RIS reflection matrices directly from the sampled channel knowledge without any knowledge of the RIS array geometry.”
Smart Radio Environments Empowered by Reconfigurable Intelligent Surfaces: How It Works, State of Research, and The Road Ahead (2020, background): “In [277], the authors propose an RIS-based architecture”
Active Reconfigurable Intelligent Surface-Aided Wireless Communications (2021, background): “Also, some active sensors can be deployed at the RIS to facilitate the channel estimation [38], [39].”
Weighted Sum-Rate Maximization for Reconfigurable Intelligent Surface Aided Wireless Networks (2019, background): “• Semi-passive RIS: In [40], a semi-passive structure is suggested by integrating active elements on the RIS which have the channel estimation capability.”
Target Sensing With Intelligent Reflecting Surface: Architecture and Performance (2022, methodology): “To tackle this issue, an alternative IRS-aided sensing method is to utilize the semi-passive IRS, where the IRS is equipped with dedicated (low-cost) sensors to receive signals for facilitating channel estimation in communications as well as target localization [10], [28]–[30].”
Deep Denoising Neural Network Assisted Compressive Channel Estimation for mmWave Intelligent Reflecting Surfaces (2020, methodology): “Besides, two channel estimation schemes respectively based on compressive sensing and deep learning (DL) were proposed in [11] and [12], whereby the angular-domain channel sparsity was utilized for reduced pilot overhead.”
Reconfigurable Intelligent Surfaces for Wireless Communications: Overview of Hardware Designs, Channel Models, and Estimation Techniques (2022, background): “In [55], the authors consider the RIS with some active elements for partial channels sensing.”
RIS-Aided Wireless Communications: Prototyping, Adaptive Beamforming, and Indoor/Outdoor Field Trials (2021, background): “In previous literature, the RIS conﬁguration is generally divided into two disjoint subproblems: 1) selecting an SNR-maximizing conﬁguration when having full Channel State Information (CSI) [10], [11]; 2) Acquiring CSI from explicit pilot signals [31]–[33].”
Reconfigurable Intelligent Surfaces for Wireless Communications: Principles, Challenges, and Opportunities (2020, methodology): “This information can be in terms of partial channel coefficients [57] or alternatively, user positions [35].”
Intelligent Reflecting Surface-Aided Wireless Communications: A Tutorial (2020, methodology): “In [102]–[104], semi-passive IRS channel estimation algorithms based on compressed sensing and deep learning techniques were proposed for the IRS-aided SISO system, where the impact of both the number of IRS sensors and channel sensing time on the achievable rate was evaluated by simulations.”