Learning WiFi Performance

Speaker :	Nidhi Hegde
	Alcatel Lucent
Date:	27/05/2015
Time:	2:00 pm - 3:00 pm
Location:	LINCS Meeting Room 40

Abstract

Accurate prediction of wireless network performance is important when performing link adaptation or resource allocation. However, the complexity of interference interactions at MAC and PHY layers, as well as the vast variety of possible wireless configurations make it notoriously hard to design explicit performance models.In this work, we advocate an approach of “learning by observation”, where we use machine learning techniques to learn implicit performance models, from a limited number of real-world measurements. While our model does not use information on the WiFi mechanism itself, our results show that the accuracy of performance prediction is significantly improved as compared to measurement-seeded models based on SINR. To demonstrate that learned models can be useful in practice, we build a new algorithm that uses such a model as an oracle to jointly allocate spectrum and transmit power. Our algorithm is utility-optimal, distributed, and it produces efficient allocations that significantly improve performance and fairness.Joint work with Julien Herzen (EPFL) and Henrik Lundgren (Technicolor)