Reinforcement Learning for Power-Efficient Grant Prediction in LTE

Beitrag bei einer Tagung
(Konferenzbeitrag)


Details zur Publikation

Autor(en): Brand P, Falk J, Ah Sue J, Brendel J, Hasholzner R, Teich J
Herausgeber: ACM
Jahr der Veröffentlichung: 2018
Tagungsband: Proc. of the 21st International Workshop on Software and Compilers for Embedded Systems
Seitenbereich: 1-8
Sprache: Englisch


Abstract


Reducing the energy consumption of mobile phones is a major concern in the design of LTE 5G.

So far, dynamic power management techniques act reactive rather than proactive, which leads to the inability to exploit a significant amount of opportunities to power down components.

We propose a technique that is capable of exploiting said opportunities by prediction.

For this, we employ reinforcement learning, capable of being trained online without prior system knowledge.

Through simulated LTE traffic, our experiments show significant savings in energy compared to standard LTE power management.

 



FAU-Autoren / FAU-Herausgeber

Ah Sue, Jonathan
Lehrstuhl für Informatik 12 (Hardware-Software-Co-Design)
Brand, Peter
Lehrstuhl für Informatik 12 (Hardware-Software-Co-Design)
Falk, Joachim
Lehrstuhl für Informatik 12 (Hardware-Software-Co-Design)
Teich, Jürgen Prof. Dr.-Ing.
Lehrstuhl für Informatik 12 (Hardware-Software-Co-Design)


Zitierweisen

APA:
Brand, P., Falk, J., Ah Sue, J., Brendel, J., Hasholzner, R., & Teich, J. (2018). Reinforcement Learning for Power-Efficient Grant Prediction in LTE. In ACM (Eds.), Proc. of the 21st International Workshop on Software and Compilers for Embedded Systems (pp. 1-8). Sankt Goar, Deutschland, DE.

MLA:
Brand, Peter, et al. "Reinforcement Learning for Power-Efficient Grant Prediction in LTE." Proceedings of the 21st International Workshop on Software and Compilers for Embedded Systems (SCOPES'18), Sankt Goar, Deutschland Ed. ACM, 2018. 1-8.

BibTeX: 

Zuletzt aktualisiert 2018-19-04 um 04:36