Filters








23,092 Hits in 5.1 sec

Overcoming the Network Bottleneck in Mobile Computing

M.R. Ebling, L.B. Mummert, D.C. Steere
1994 1994 First Workshop on Mobile Computing Systems and Applications  
In this paper we argue that prescient caching and smart scheduling are key techniques for overcoming the network bottleneck.  ...  But in mobile computing, it is the network, rather than CPU or storage, that will be the scarce resource.  ... 
doi:10.1109/wmcsa.1994.30 dblp:conf/wmcsa/EblingMS94 fatcat:e46ziknzjzfwbjqfwu2j24kzju

Data Routing Algorithm in Mobile Cloud Computing Network

Ch Srilakshmi Prasanna, M. Chenna Keshava
2015 International Journal of Engineering Research and  
This paper gives brief view of the routing algorithm used to enhance the capacity of access network in mobile cloud services based on merging of computation and networking in heterogeneous mobile cloud  ...  Mobile cloud computing networking [MCCN] is a novel approach. MCCN is an integration of cloud -based resources and mobility.  ...  CONCLUSION AND FUTURE WORK In heterogeneous mobile cloud computing network requires adaptive data routing algorithm to overcome the critical networking challenges.  ... 
doi:10.17577/ijertv4is040667 fatcat:irv244ffefgxxi4vmljwsf4rwe

Energy/Latency/Image Quality Tradeoffs in Enabling Mobile Multimedia Communication [chapter]

Clark N. Taylor, Sujit Dey, Debashis Panigrahi
2001 Software Radio  
In this paper, we propose a method to overcome the energy and bandwidth bottlenecks by adapting to the varying conditions and requirements of mobile multimedia communication.  ...  We present a methodology to enable selection of the appropriate image compression parameters to implement the energy latency image quality tradeo in mobile multimedia radios.  ...  We proposed a new radio architecture which, in addition to traditional radio components, includes a Network Aware Operating System and adaptive image video coder to adapt the source coding algorithms and  ... 
doi:10.1007/978-1-4471-0343-1_5 fatcat:srhknka5ebew7f3w22w2r52md4

Energy-Aware Cooperative Computation in Mobile Devices [article]

Ajita Singh and Yuxuan Xing and Hulya Seferoglu
2016 arXiv   pre-print
Especially, in device-to-device networks, where data rates are increasing rapidly, processing power and energy are becoming the primary bottlenecks of the network.  ...  In this paper, we develop an energy-aware cooperative computation framework for mobile devices.  ...  Instead of offloading cellular networks, our goal is to create energy-aware cooperation framework to overcome the processing power and energy bottlenecks of mobile devices.  ... 
arXiv:1602.04400v1 fatcat:y535kavhzvhmxpelxinrtmczbu

The Roadmap to 6G – AI Empowered Wireless Networks [article]

Khaled B. Letaief, Wei Chen, Yuanming Shi, Jun Zhang, Ying-Jun Angela Zhang
2019 arXiv   pre-print
In particular, 6G will go beyond mobile Internet and will be required to support ubiquitous AI services from the core to the end devices of the network.  ...  In this article, we discuss potential technologies for 6G to enable mobile AI applications, as well as AI-enabled methodologies for 6G network design and optimization.  ...  However, the limited bandwidth becomes the main bottleneck for global model aggregation from locally updated models computed at each mobile device.  ... 
arXiv:1904.11686v2 fatcat:34hysdm7pfd2jioixdx7tcqwgy

Neural Compression and Filtering for Edge-assisted Real-time Object Detection in Challenged Networks [article]

Yoshitomo Matsubara, Marco Levorato
2020 arXiv   pre-print
The edge computing paradigm places compute-capable devices - edge servers - at the network edge to assist mobile devices in executing data analysis tasks.  ...  by introducing a bottleneck layer in the early layers on the head model, and (ii) prefilter pictures that do not contain objects of interest using a convolutional neural network.  ...  To overcome this issue, we redefine here the head distillation technique to (i) introduce the bottleneck at the very early layers of the network, and (ii) refine the loss function used to distill the mimicking  ... 
arXiv:2007.15818v2 fatcat:rsofmyqgxfdapoam7gyltaja4y

Deep Learning based Approach for Bone Diagnosis Classification in Ultrasonic Computed Tomographic Images

Marwa Fradi, Mouna Afif, Mohsen Machhout
2020 International Journal of Advanced Computer Science and Applications  
At this light, two types of neural networks algorithms have been developed to automatically classify the Ultrasonic Computed Tomographic (USCT) images into three categories, such as healthy, fractured  ...  Artificial intelligence (AI) in the area of medical imaging has shown a developed technology to have automatically the true diagnosis especially in ultrasonic imaging area.  ...  In Fact, we overcome [27] with inception -v3 with 13% and with SOM by 14% .With AmeobaNet we have overcome the major related works with neural networks classification models against Alex Net, NasNet,  ... 
doi:10.14569/ijacsa.2020.0111210 fatcat:guz65s3ii5b3zib6o7yryvcnqa

A Hybrid Model for VANET Information Dissemination

Jeyalakshmi Jeyabalan, Sree Subha Soundarajan
2013 International Journal of Computer Applications  
VANET is an emerging mobile adhoc network used in order to improve vehicle and road safety, traffic efficiency, and convenienceas well as comfort to both drivers and passengers.  ...  But in VANET, in sparsely populated areas, the messages are not propagated to the neighbors properly, due to fading of messages with distance.There are several solutions to this issue.  ...  to overcome the Bottleneck.  ... 
doi:10.5120/13970-1876 fatcat:whycijazmnf2pgtyjlpz22rot4

Head Network Distillation: Splitting Distilled Deep Neural Networks for Resource-constrained Edge Computing Systems

Yoshitomo Matsubara, Davide Callegaro, Sabur Baidya, Marco Levorato, Sameer Singh
2020 IEEE Access  
overcomes the reduced data transmission time.  ...  models, here we propose to use it to train The head section of the network, executed at the mobile device-side, is shrunk -distilled -to reduce the computation complexity of that section, and a bottleneck  ... 
doi:10.1109/access.2020.3039714 fatcat:sszr7pxravfp3d2nmogulodtku

Supporting service differentiation for real-time and best-effort traffic in stateless wireless ad hoc networks (SWAN)

Gahng-Seop Ahn, A.T. Campbell, A. Veres, Li-Hsiang Sun
2002 IEEE Transactions on Mobile Computing  
mobile wireless ad-hoc networks n Simple: only use classic MAC protocol n Scalable: stateless, local control mechanism n Robust: changes in network topology, real-time traffic load, and link failures  ...  Basic Assumption Most of the wireless network capacity will be utilized by best effort traffic è Best effort traffic can be used as a "Buffer Zone" to absorb real-time traffic variation due to mobility  ...  n r (multiplicative decrease) n c (additive increase) n g (difference between the actual rate and the shaping rate) n admission control rate  ... 
doi:10.1109/tmc.2002.1081755 fatcat:i37eovnuyvfgtougk2apk6d7qu

Design and implementation of neural network computing framework on Zynq SoC embedded platform

Xingying Li, Zhenyu Yin, Fulong Xu, Feiqing Zhang, Guangyuan Xu
2021 Procedia Computer Science  
To overcome this problem, a new neural network computing framework "Zynq-Darknet" was proposed.  ...  In order to verify the performance of the framework and model, experiments were conducted on imagenet-1k dataset using different network structures.  ...  full use of the heterogeneous computing power of mobile chips and has extensive hardware support.  ... 
doi:10.1016/j.procs.2021.02.091 fatcat:2ctwvaxlsfdlbbf7av5hkgrpu4

FENCE: Fast, ExteNsible, and ConsolidatEd Framework for Intelligent Big Data Processing

Ramneek, Seung-Jun Cha, Sangheon Pack, Seung Hyub Jeon, Yeon Jeong Jeong, Jin Mee Kim, Sungin Jung
2020 IEEE Access  
We analyzed the key drivers for edge computing in industrial IoT domains, and discussed how the FENCE framework can overcome OS and software level bottlenecks in manycore systems.  ...  bottlenecks in the edge host.  ... 
doi:10.1109/access.2020.3007747 fatcat:2udiua57pfgklivxt26rn6jhlm

A Distributed Method for Bottleneck Node Detection in Wireless Sensor Network
무선 센서망의 병목 노드 탐색을 위한 분산 알고리즘

Haosong Gou, Jin-Hwan Kim, Young-Hwan Yoo
2009 The KIPS Transactions PartC  
The bottleneck nodes widely exist in WSNs and lead to decrease the lifetime of the whole network.  ...  The simulation results and analysis show that our algorithm achieves much better performance and our solutions can relax the bottleneck problem, resulting in the prolonging of the network lifetime.  ...  As shown in the figure, we can overcome the bottleneck node problem and prolong the lifetime of the whole network two times more than before.  ... 
doi:10.3745/kipstb.2009.16c.5.621 fatcat:qs4vkial5nemnkx7znxsnq3ede

ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices [article]

Xiangyu Zhang, Xinyu Zhou, Mengxiao Lin, Jian Sun
2017 arXiv   pre-print
on ImageNet classification task, under the computation budget of 40 MFLOPs.  ...  We introduce an extremely computation-efficient CNN architecture named ShuffleNet, which is designed specially for mobile devices with very limited computing power (e.g., 10-150 MFLOPs).  ...  In its residual branch, for the 3 × 3 layer, we apply a computational economical 3 × 3 depthwise convolution [3] on the bottleneck feature map.  ... 
arXiv:1707.01083v2 fatcat:wj4h22jtxbg3xnlwaveq5r536y

ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices

Xiangyu Zhang, Xinyu Zhou, Mengxiao Lin, Jian Sun
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition  
[12] on Ima-geNet classification task, under the computation budget of 40 MFLOPs.  ...  We introduce an extremely computation-efficient CNN architecture named ShuffleNet, which is designed specially for mobile devices with very limited computing power (e.g.,).  ...  In its residual branch, for the 3 × 3 layer, we apply a computational economical 3 × 3 depthwise convolution [3] on the bottleneck feature map.  ... 
doi:10.1109/cvpr.2018.00716 dblp:conf/cvpr/ZhangZLS18 fatcat:nfroohegrfbcdaahbuqd5l5cyu
« Previous Showing results 1 — 15 out of 23,092 results