Wireless Data Acquisition for Edge Learning: Data-Importance Aware Retransmission

Liu, Dongzhu; Zhu, Guangxu; Zeng, Qunsong; Huang, Kaibin

doi:10.1109/twc.2020.3024980

Cited by 31 publications

(24 citation statements)

References 33 publications

(48 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…• Machine Learning for Early ARQ/HARQ: Strategies for reducing the feedback delay using prediction mechanisms powered by machine learning techniques have been recently studied in the context of RF-based wireless communications [154], [209], [210]. From the machine learning perspective, the main task is to predict the decoding result of a given transmission using data, which is available after the first few decoder iterations.…”

Section: Challenges and Open Issuesmentioning

confidence: 99%

Link-Layer Retransmission-based Error-Control Protocols in FSO Communications: A Survey

Le¹

2022

Preprint

View full text Add to dashboard Cite

Free-space optical (FSO) communications have gained significant interest over the last few years, thanks to the capability to transport extremely high-speed data over long distances without exhausting radio frequency (RF) resources. FSO communication is widely considered in various network scenarios, such as inter-satellite/deep-space links, ground-station/vehicles, satellite/aerial links, or terrestrial links. It is expected to be a key enabling technology for the next generation of 6G wireless networks. Nevertheless, despite the great potential of FSO communications, its performance suffers from various limitations and challenges: atmospheric turbulence, clouds, weather conditions, and pointing misalignment. The error-control solutions, including physical layer (PHY) and link-layer methods, aim to mitigate the transmission errors caused by such adverse issues. While the existing surveys on error-control solutions in FSO systems primarily focussed on the PHY methods, we instead provide a review of link-layer solutions. In particular, we conduct an extensive literature survey of state-of-the-art retransmission protocols, both automatic repeat request (ARQ) and hybrid ARQ (HARQ), for various FSO communication scenarios, including point-to-point terrestrial, cooperative, multi-hop relaying, hybrid FSO/RF, satellite/aerial, and deep-space systems. Furthermore, we provide a survey of recent literature and insightful discussion on the cross-layer design frameworks related to link-layer retransmission protocols in FSO communication networks. Finally, the lessons learned, design guidelines, related open issues, and future research directions are exposed.

show abstract

Section: Challenges and Open Issuesmentioning

confidence: 99%

Link-Layer Retransmission-based Error-Control Protocols in FSO Communications: A Survey

Le¹

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…The interaction between edge computing and ARQ can be used to improve the system efficiency. For example, the authors in [253], [254] proposed a novel importance-aware ARQ protocol for data acquisition in edge learning systems. This protocol efficiently adapts the retransmission considering the data importance.…”

Section: E Edge Computing and Cloud-radio Access Network (Crans)mentioning

confidence: 99%

Hybrid Automatic Repeat Request (HARQ) in Wireless Communications Systems and Standards: A Contemporary Survey

Ahmed¹,

Al-Dweik²,

Iraqi³

et al. 2021

Preprint

View full text Add to dashboard Cite

<div>Automatic repeat request (ARQ) schemes, and in particular hybrid-ARQ (HARQ) schemes, which jointly adopt forward error correction (FEC) and ARQ, are essential to provide reliable data transmission in wireless communications systems. However, the feedback from the receiver to the transmitter and the retransmission process used in ARQ incurs significant cost in terms of power efficiency, throughput, computational power and delay. Unfortunately, such drawbacks can limit their applications to several current and emerging technologies. More specifically, the increasing number of wireless users has create spectrum scarcity, relying on small-size batteries create power constraints, deployment of real-time applications boost the demand for ultralow delay networks, and the ultra-small low-cost internet of things (IoT) devices has limited signal processing and computation capabilities. Consequently, extensive research efforts have been dedicated to overcome the limitations inherent in HARQ. This survey paper provides an extensive literature review of the state-of-the-art HARQ techniques and discusses their integration in various wireless technologies. Moreover, it provides insights on advantages and disadvantages of particular ARQ types and discusses open problems and future directions.</div>

show abstract

“…1) Inference Uncertainty: Inference uncertainty is measured using the entropy of posteriors commonly used for ML and DNN classifiers [20], [21]. Given the feature distribution in (1), the metric, denoted as H (x k ), for linear classification of the partial feature vector x k is given as…”

Section: Classification Metricsmentioning

confidence: 99%

“…as that in (20). Given the above straightforward procedure, the remainder of the paper assumes binary classification to simplify notation and exposition.…”

Section: ) Extension To Multi-class Classificationmentioning

confidence: 99%

Progressive Feature Transmission for Split Inference at the Wireless Edge

Lan¹,

Zeng²,

Popovski³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

In edge inference, an edge server provides remote-inference services to edge devices. This requires the edge devices to upload high-dimensional features of data samples over resource-constrained wireless channels, which creates a communication bottleneck. The conventional solution of feature pruning requires that the device has access to the inference model, which is unavailable in the current scenario of split inference. To address this issue, we propose the progressive feature transmission (ProgressFTX) protocol, which minimizes the overhead by progressively transmitting features until a target confidence level is reached. The optimal control policy of the protocol to accelerate inference is derived and it comprises two key operations. The first is importance-aware feature selection at the server, for which it is shown to be optimal to select the most important features, characterized by the largest discriminant gains of the corresponding feature dimensions. The second is transmission-termination control by the server for which the optimal policy is shown to exhibit a threshold structure. Specifically, the transmission is stopped when the incremental uncertainty reduction by further feature transmission is outweighed by its communication cost. The indices of the selected features and transmission decision are fed back to the device in each slot. The optimal policy is first derived for the tractable case of linear classification and then extended to the more complex case of classification using a convolutional neural network. Both Gaussian and fading channels are considered. Experimental results are obtained for both a statistical data model and a real dataset. It is seen that ProgressFTX can substantially reduce the communication latency compared to conventional feature pruning and random feature transmission. I. INTRODUCTIONRecent years have witnessed the extensive deployment of artificial intelligence (AI) technologies at the network edge to gain fast access to and processing of mobile data. This gives rise to two active research challenges: (1) edge learning [1], [2], where data are used to train largescale AI models via distributed machine learning; and (2) edge inference [1], [3], which is the

show abstract

Wireless Data Acquisition for Edge Learning: Data-Importance Aware Retransmission

Abstract: Low-rate noiseless label channel High-rate noisy data channel TDMA Communication System Model Edge Devices Edge Server Channel State Information

Cited by 31 publications

References 33 publications

Link-Layer Retransmission-based Error-Control Protocols in FSO Communications: A Survey

Link-Layer Retransmission-based Error-Control Protocols in FSO Communications: A Survey

Hybrid Automatic Repeat Request (HARQ) in Wireless Communications Systems and Standards: A Contemporary Survey

Progressive Feature Transmission for Split Inference at the Wireless Edge

Contact Info

Product

Resources

About