Performance Analysis on Machine Learning-Based Channel Estimation

Mei, Kai; Liu, Jun; Zhang, Xiaochen; Rajatheva, Nandana; Wei, Jibo

doi:10.1109/tcomm.2021.3083597

Cited by 24 publications

(12 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Such scaling issues can be handled more efficiently using AI-MDL approaches [45] [46]. To bring DL-aided intelligent and reconfigurable wireless PHY into reality, issues such as potential use cases, achievable gain and complexity trade-off, evaluation methodologies, dataset availability, and standard compatibility impact need to be addressed [10], [20]- [23].…”

Section: Literature Review: DL In Wireless Phymentioning

confidence: 99%

Deep Neural Network Augmented Wireless Channel Estimation for Preamble-based OFDM PHY on Zynq System on Chip

haq¹,

Gizzini²,

Shrey³

et al. 2023

Preprint

View full text Add to dashboard Cite

<p>Reliable and fast channel estimation is crucial for next-generation wireless networks supporting a wide range of vehicular and low-latency services. Recently, deep learning (DL) based channel estimation is being explored as an efficient alternative to conventional least-square (LS) and linear minimum mean square error (LMMSE) based channel estimation. Unlike LMMSE, DL methods do not need prior knowledge of channel statistics. Most of these approaches have not been realized on system-on-chip (SoC), and preliminary study shows that their complexity exceeds the complexity of the entire physical layer (PHY). The high latency of DL is another concern. This paper considers the design and implementation of deep neural network (DNN) augmented LS-based channel estimation (LSDNN) on Zynq multi-processor SoC (ZMPSoC). We demonstrate the gain in performance compared to the conventional LS and LMMSE channel estimation schemes. Via software-hardware co-design, word-length optimization, and reconfigurable architectures, we demonstrate the superiority of the LSDNN architecture over the LS and LMMSE for a wide range of SNR, number of pilots, preamble types, and wireless channels. Further, we evaluate the performance, power, and area (PPA) of the LS and LSDNN application-specific integrated circuit (ASIC) implementations in 45 nm technology. We demonstrate that the word-length optimization can substantially improve PPA for the proposed architecture in ASIC implementations.</p>

show abstract

Section: Literature Review: DL In Wireless Phymentioning

confidence: 99%

Deep Neural Network Augmented Wireless Channel Estimation for Preamble-based OFDM PHY on Zynq System on Chip

haq¹,

Gizzini²,

Shrey³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…A comparison between DL-based and conventional channel estimation methods has been provided in [24]. Moreover, performance analysis of ML-based channel estimation was carried out in [47], while recurrent neural network (RNN) channel estimation was studied in [48]. Finally, a short review of DL for channel estimation was provided in [49] without focusing on multicarrier systems.…”

Section: Related Workmentioning

confidence: 99%

Artificial intelligence for channel estimation in multicarrier systems for B5G/6G communications: a survey

Boas

Silva

Figueiredo

et al. 2022

J Wireless Com Network

View full text Add to dashboard Cite

Multicarrier modulation allows for deploying wideband systems resilient to multipath fading channels, impulsive noise, and intersymbol interference compared to single-carrier systems. Despite this, multicarrier signals suffer from different types of distortion, including channel noise sources and long- and short-term fading. Consequently, the receiver must estimate the channel features and compensate it for data recovery based on channel estimation techniques, such as non-blind, blind, and semi-blind approaches. These techniques are model-based and designed with accurate mathematical channel models encompassing their features. Nevertheless, complex environments challenge accurate mathematical channel estimation modeling, which might neither be accurate nor correspond to reality. This impairment decreases the system performance due to the channel estimation accuracy loss. Fortunately, (AI) algorithms can learn the relationship among different system variables using a model-driven or model-free approach. Thereby, AI algorithms are used for channel estimation by exploiting its complexity without unrealistic assumptions, following a better performance than conventional techniques under the same channel. Hence, this paper comprehensively surveys AI-based channel estimation for multicarrier systems. First, we provide essential background on conventional channel estimation techniques in the context of multicarrier systems. Second, the AI-aided channel estimation strategies are investigated using the following approaches: classical learning, neural networks, and reinforcement learning. Lastly, we discuss current challenges and point out future research directions based on recent findings.

show abstract

“…where R hh = 𝐸 hh 𝐻 is the channel autocorrelation matrix and X is a diagonal matrix containing the known transmitted signaling points [33]- [35]. The MSE of the LMMSE estimate is…”

Section: Decoder: Inferencementioning

confidence: 99%

Theoretical Analysis of Deep Neural Networks in Physical Layer Communication

Li¹,

Zhao²,

Ma³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Recently, deep neural network (DNN)-based physical layer communication techniques have attracted considerable interest. Although their potential to enhance communication systems and superb performance have been validated by simulation experiments, little attention has been paid to the theoretical analysis. Specifically, most studies in the physical layer have tended to focus on the application of DNN models to wireless communication problems but not to theoretically understand how does a DNN work in a communication system. In this paper, we aim to quantitatively analyze why DNNs can achieve comparable performance in the physical layer comparing with traditional techniques, and also drive their cost in terms of computational complexity. To achieve this goal, we first analyze the encoding performance of a DNN-based transmitter and compare it to a traditional one. And then, we theoretically analyze the performance of DNN-based estimator and compare it with traditional estimators. Third, we investigate and validate how information is flown in a DNN-based communication system under the information theoretic concepts. Our analysis develops a concise way to open the "black box" of DNNs in physical layer communication, which can be applied to support the design of DNN-based intelligent communication techniques and help to provide explainable performance assessment.

show abstract

Performance Analysis on Machine Learning-Based Channel Estimation

Cited by 24 publications

References 28 publications

Deep Neural Network Augmented Wireless Channel Estimation for Preamble-based OFDM PHY on Zynq System on Chip

Deep Neural Network Augmented Wireless Channel Estimation for Preamble-based OFDM PHY on Zynq System on Chip

Artificial intelligence for channel estimation in multicarrier systems for B5G/6G communications: a survey

Theoretical Analysis of Deep Neural Networks in Physical Layer Communication

Contact Info

Product

Resources

About