Uploading high-dimensional features from edge devices to an edge server over wireless channels creates a communication bottleneck for edge inference. To tackle the challenge, we propose the progressive feature transmission (ProgressFTX) protocol, which minimizes the overhead by progressively transmitting features until a target confidence level is reached. The control of the protocol to accelerate inference is designed with two key operations. The first, importance-aware feature selection, guides the server to select the most discriminative feature dimensions. The second is transmission-termination control such that the feature transmission is stopped when the incremental uncertainty reduction by further 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 sub-optimal policy is obtained for classification using a convolutional neural network. Experimental results on a real-world dataset shows that ProgressFTX can substantially reduce the communication latency compared to conventional feature pruning and random feature transmission.

Progressive Transmission of High-Dimensional Data Features for Inference at the Network Edge / Lan, Q.; Zeng, Q.; Popovski, P.; Gunduz, D.; Huang, K.. - 2022-:(2022), pp. 1-5. ((Intervento presentato al convegno 23rd IEEE International Workshop on Signal Processing Advances in Wireless Communication, SPAWC 2022 tenutosi a fin nel 2022 [10.1109/SPAWC51304.2022.9834000].

Progressive Transmission of High-Dimensional Data Features for Inference at the Network Edge

Lan Q.;Gunduz D.;
2022

Abstract

Uploading high-dimensional features from edge devices to an edge server over wireless channels creates a communication bottleneck for edge inference. To tackle the challenge, we propose the progressive feature transmission (ProgressFTX) protocol, which minimizes the overhead by progressively transmitting features until a target confidence level is reached. The control of the protocol to accelerate inference is designed with two key operations. The first, importance-aware feature selection, guides the server to select the most discriminative feature dimensions. The second is transmission-termination control such that the feature transmission is stopped when the incremental uncertainty reduction by further 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 sub-optimal policy is obtained for classification using a convolutional neural network. Experimental results on a real-world dataset shows that ProgressFTX can substantially reduce the communication latency compared to conventional feature pruning and random feature transmission.
23rd IEEE International Workshop on Signal Processing Advances in Wireless Communication, SPAWC 2022
fin
2022
2022-
1
5
Lan, Q.; Zeng, Q.; Popovski, P.; Gunduz, D.; Huang, K.
Progressive Transmission of High-Dimensional Data Features for Inference at the Network Edge / Lan, Q.; Zeng, Q.; Popovski, P.; Gunduz, D.; Huang, K.. - 2022-:(2022), pp. 1-5. ((Intervento presentato al convegno 23rd IEEE International Workshop on Signal Processing Advances in Wireless Communication, SPAWC 2022 tenutosi a fin nel 2022 [10.1109/SPAWC51304.2022.9834000].
File in questo prodotto:
File Dimensione Formato  
Progressive_Transmission_of_High-Dimensional_Data_Features_for_Inference_at_the_Network_Edge.pdf

non disponibili

Tipologia: Versione dell'editore (versione pubblicata)
Dimensione 2.14 MB
Formato Adobe PDF
2.14 MB Adobe PDF   Visualizza/Apri   Richiedi una copia
Pubblicazioni consigliate

Caricamento pubblicazioni consigliate

Licenza Creative Commons
I metadati presenti in IRIS UNIMORE sono rilasciati con licenza Creative Commons CC0 1.0 Universal, mentre i file delle pubblicazioni sono rilasciati con licenza Attribuzione 4.0 Internazionale (CC BY 4.0), salvo diversa indicazione.
In caso di violazione di copyright, contattare Supporto Iris

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/11380/1286019
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? ND
social impact