FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4381849893> ?p ?o ?g. }

• W4381849893 endingPage "102129" @default.
• W4381849893 startingPage "102129" @default.
• W4381849893 abstract "Nowadays, unmanned aerial vehicle (UAV) swarm supported by mobile edge computing applications is attracting more and more attention, such as smart agriculture, smart transportation, smart security surveillance, and smart environmental monitoring. Yet, small UAV devices cannot deploy advanced best-performing DNN models with insufficient computing and storage resources. Aerial computing centers serving as edges can collect tasks from UAVs, reduce UAV load, and provide high accuracy inference. Limited radio resources in offloading and computing centers load imbalance could lead significant latency in task offloading inference. Considering image processing in UAV application, we propose a hybrid inference framework combining early exit and task offloading based on distributed neural network. The bottleneck-designed DNN provide smaller intermediate data size compared to the original one. By deploying shallow neural networks at the drone terminals and advanced DNNs at the air computing servers, latency-constrained tasks can be exited after local computation, offloaded to the edge after local computation, or directly offloaded to the edge. Compared to binary or partial offloading, early-exit based framework provides fast inference as the entire workflow stops when a confident result is obtained. Combining task offloading into the inference process offers a more flexible option to release the UAV computing load. The proposed approach efficient utilize the wireless transmission capacity and balance the computation load to maximize the system inference accuracy within the latency constraints. Simulations of practical task inference flow demonstrate our approach outperforms under different channel conditions, task generation rates and local computing capacities." @default.
• W4381849893 created "2023-06-25" @default.
• W4381849893 creator A5024547118 @default.
• W4381849893 creator A5026608908 @default.
• W4381849893 creator A5076986736 @default.
• W4381849893 creator A5083856519 @default.
• W4381849893 creator A5087736924 @default.
• W4381849893 date "2023-10-01" @default.
• W4381849893 modified "2023-09-27" @default.
• W4381849893 title "A hybrid fast inference approach with distributed neural networks for edge computing enabled UAV swarm" @default.
• W4381849893 cites W1998305460 @default.
• W4381849893 cites W2796588635 @default.
• W4381849893 cites W2808052525 @default.
• W4381849893 cites W2809251854 @default.
• W4381849893 cites W2951054802 @default.
• W4381849893 cites W2962726494 @default.
• W4381849893 cites W2963122961 @default.
• W4381849893 cites W2963163009 @default.
• W4381849893 cites W2963918968 @default.
• W4381849893 cites W2964052915 @default.
• W4381849893 cites W2980150456 @default.
• W4381849893 cites W2981629859 @default.
• W4381849893 cites W3005478472 @default.
• W4381849893 cites W3022354451 @default.
• W4381849893 cites W3034971973 @default.
• W4381849893 cites W3043914740 @default.
• W4381849893 cites W3095012225 @default.
• W4381849893 cites W3101867439 @default.
• W4381849893 cites W3132370151 @default.
• W4381849893 cites W3134255974 @default.
• W4381849893 cites W3135020793 @default.
• W4381849893 cites W3160555381 @default.
• W4381849893 cites W3164182356 @default.
• W4381849893 cites W4211150976 @default.
• W4381849893 cites W4213260037 @default.
• W4381849893 cites W4226206247 @default.
• W4381849893 cites W4236099117 @default.
• W4381849893 cites W4293218370 @default.
• W4381849893 cites W4312229791 @default.
• W4381849893 cites W4321021465 @default.
• W4381849893 doi "https://doi.org/10.1016/j.phycom.2023.102129" @default.
• W4381849893 hasPublicationYear "2023" @default.
• W4381849893 type Work @default.
• W4381849893 citedByCount "0" @default.
• W4381849893 crossrefType "journal-article" @default.
• W4381849893 hasAuthorship W4381849893A5024547118 @default.
• W4381849893 hasAuthorship W4381849893A5026608908 @default.
• W4381849893 hasAuthorship W4381849893A5076986736 @default.
• W4381849893 hasAuthorship W4381849893A5083856519 @default.
• W4381849893 hasAuthorship W4381849893A5087736924 @default.
• W4381849893 hasConcept C111919701 @default.
• W4381849893 hasConcept C120314980 @default.
• W4381849893 hasConcept C138236772 @default.
• W4381849893 hasConcept C149635348 @default.
• W4381849893 hasConcept C154945302 @default.
• W4381849893 hasConcept C162307627 @default.
• W4381849893 hasConcept C162324750 @default.
• W4381849893 hasConcept C177212765 @default.
• W4381849893 hasConcept C187736073 @default.
• W4381849893 hasConcept C2776061582 @default.
• W4381849893 hasConcept C2776214188 @default.
• W4381849893 hasConcept C2778456923 @default.
• W4381849893 hasConcept C2780451532 @default.
• W4381849893 hasConcept C2780513914 @default.
• W4381849893 hasConcept C2781041963 @default.
• W4381849893 hasConcept C31258907 @default.
• W4381849893 hasConcept C41008148 @default.
• W4381849893 hasConcept C76155785 @default.
• W4381849893 hasConcept C77088390 @default.
• W4381849893 hasConcept C79403827 @default.
• W4381849893 hasConcept C79974875 @default.
• W4381849893 hasConcept C82876162 @default.
• W4381849893 hasConcept C93996380 @default.
• W4381849893 hasConceptScore W4381849893C111919701 @default.
• W4381849893 hasConceptScore W4381849893C120314980 @default.
• W4381849893 hasConceptScore W4381849893C138236772 @default.
• W4381849893 hasConceptScore W4381849893C149635348 @default.
• W4381849893 hasConceptScore W4381849893C154945302 @default.
• W4381849893 hasConceptScore W4381849893C162307627 @default.
• W4381849893 hasConceptScore W4381849893C162324750 @default.
• W4381849893 hasConceptScore W4381849893C177212765 @default.
• W4381849893 hasConceptScore W4381849893C187736073 @default.
• W4381849893 hasConceptScore W4381849893C2776061582 @default.
• W4381849893 hasConceptScore W4381849893C2776214188 @default.
• W4381849893 hasConceptScore W4381849893C2778456923 @default.
• W4381849893 hasConceptScore W4381849893C2780451532 @default.
• W4381849893 hasConceptScore W4381849893C2780513914 @default.
• W4381849893 hasConceptScore W4381849893C2781041963 @default.
• W4381849893 hasConceptScore W4381849893C31258907 @default.
• W4381849893 hasConceptScore W4381849893C41008148 @default.
• W4381849893 hasConceptScore W4381849893C76155785 @default.
• W4381849893 hasConceptScore W4381849893C77088390 @default.
• W4381849893 hasConceptScore W4381849893C79403827 @default.
• W4381849893 hasConceptScore W4381849893C79974875 @default.
• W4381849893 hasConceptScore W4381849893C82876162 @default.
• W4381849893 hasConceptScore W4381849893C93996380 @default.
• W4381849893 hasFunder F4320327101 @default.
• W4381849893 hasFunder F4320335777 @default.

• next