Scipediacontent: Scipediacontent moved page Draft Content 384218806 to Zhang Patras 2018a

2021-02-01T22:15:43Z

Scipediacontent moved page Draft Content 384218806 to Zhang Patras 2018a

Scipediacontent: Created page with " == Abstract == Forecasting with high accuracy the volume of data traffic that mobile users will consume is becoming increasingly important for precision traffic engineering,..."

2021-02-01T22:15:38Z

Created page with " == Abstract == Forecasting with high accuracy the volume of data traffic that mobile users will consume is becoming increasingly important for precision traffic engineering,..."

New page

== Abstract ==

Forecasting with high accuracy the volume of data traffic that mobile users will consume is becoming increasingly important for precision traffic engineering, demand-aware network resource allocation, as well as public transportation. Measurements collection in dense urban deployments is however complex and expensive, and the post-processing required to make predictions is highly non-trivial, given the intricate spatio-temporal variability of mobile traffic due to user mobility. To overcome these challenges, in this paper we harness the exceptional feature extraction abilities of deep learning and propose a Spatio-Temporal neural Network (STN) architecture purposely designed for precise network-wide mobile traffic forecasting. We present a mechanism that fine tunes the STN and enables its operation with only limited ground truth observations. We then introduce a Double STN technique (D-STN), which uniquely combines the STN predictions with historical statistics, thereby making faithful long-term mobile traffic projections. Experiments we conduct with real-world mobile traffic data sets, collected over 60 days in both urban and rural areas, demonstrate that the proposed (D-)STN schemes perform up to 10-hour long predictions with remarkable accuracy, irrespective of the time of day when they are triggered. Specifically, our solutions achieve up to 61% smaller prediction errors as compared to widely used forecasting approaches, while operating with up to 600 times shorter measurement intervals.

== Original document ==

The different versions of the original document can be found in:

* [http://arxiv.org/abs/1712.08083 http://arxiv.org/abs/1712.08083]

* [http://arxiv.org/pdf/1712.08083 http://arxiv.org/pdf/1712.08083]

* [http://dx.doi.org/10.1145/3209582.3209606 http://dx.doi.org/10.1145/3209582.3209606]

* [http://dl.acm.org/ft_gateway.cfm?id=3209606&ftid=1981653&dwn=1 http://dl.acm.org/ft_gateway.cfm?id=3209606&ftid=1981653&dwn=1],
: [http://dx.doi.org/10.1145/3209582.3209606 http://dx.doi.org/10.1145/3209582.3209606] under the license http://www.acm.org/publications/policies/copyright_policy#Background

* [https://dblp.uni-trier.de/db/journals/corr/corr1712.html#abs-1712-08083 https://dblp.uni-trier.de/db/journals/corr/corr1712.html#abs-1712-08083],
: [https://arxiv.org/abs/1712.08083 https://arxiv.org/abs/1712.08083],
: [https://dl.acm.org/citation.cfm?id=3209606 https://dl.acm.org/citation.cfm?id=3209606],
: [https://doi.acm.org/10.1145/3209582.3209606 https://doi.acm.org/10.1145/3209582.3209606],
: [https://www.research.ed.ac.uk/portal/files/56312541/traffic_forecasting.pdf https://www.research.ed.ac.uk/portal/files/56312541/traffic_forecasting.pdf],
: [https://doi.org/10.1145/3209582.3209606 https://doi.org/10.1145/3209582.3209606],
: [https://www.research.ed.ac.uk/portal/en/publications/longterm-mobile-traffic-forecasting-using-deep-spatiotemporal-neural-networks(824abcdc-5fb8-4d61-9f71-e77681eb6ae0).html https://www.research.ed.ac.uk/portal/en/publications/longterm-mobile-traffic-forecasting-using-deep-spatiotemporal-neural-networks(824abcdc-5fb8-4d61-9f71-e77681eb6ae0).html],
: [http://www.arxiv-vanity.com/papers/1712.08083 http://www.arxiv-vanity.com/papers/1712.08083],
: [https://academic.microsoft.com/#/detail/2963035276 https://academic.microsoft.com/#/detail/2963035276]

* [https://www.research.ed.ac.uk/portal/en/publications/longterm-mobile-traffic-forecasting-using-deep-spatiotemporal-neural-networks(824abcdc-5fb8-4d61-9f71-e77681eb6ae0).html https://www.research.ed.ac.uk/portal/en/publications/longterm-mobile-traffic-forecasting-using-deep-spatiotemporal-neural-networks(824abcdc-5fb8-4d61-9f71-e77681eb6ae0).html],
: [https://doi.org/10.1145/3209582.3209606 https://doi.org/10.1145/3209582.3209606],
: [http://hdl.handle.net/20.500.11820/824abcdc-5fb8-4d61-9f71-e77681eb6ae0 http://hdl.handle.net/20.500.11820/824abcdc-5fb8-4d61-9f71-e77681eb6ae0],
: [https://www.pure.ed.ac.uk/ws/files/56312541/traffic_forecasting.pdf https://www.pure.ed.ac.uk/ws/files/56312541/traffic_forecasting.pdf],
: [https://www.sigmobile.org/mobihoc/2018 https://www.sigmobile.org/mobihoc/2018],
: [https://dl.acm.org/citation.cfm?id=3209606 https://dl.acm.org/citation.cfm?id=3209606]

← Older revision	Revision as of 22:15, 1 February 2021
(No difference)

Zhang Patras 2018a - Revision history

Scipediacontent: Scipediacontent moved page Draft Content 384218806 to Zhang Patras 2018a

Scipediacontent: Created page with " == Abstract == Forecasting with high accuracy the volume of data traffic that mobile users will consume is becoming increasingly important for precision traffic engineering,..."