High–Throughput Enhancements and Rate Adaptation Strategies for IEEE 802.11 Based Real–Time Communications

Speaker:  Federico Tramarin - CNR Padova
  Monday, June 13, 2016 at 3:00 PM Rinfresco 14.45, inizio seminario 15.00. DATA NUOVA.
In the last few years, IEEE 802.11–based wireless systems have proved their effectiveness in
a wide range of real–time communication scenarios, such as factory automation, distributed em-
bedded and robotic systems, localization/positioning and distributed motion control applications.
Nonetheless, this operational framework did not satisfactorily address significant features provided
by the standard and its most recent amendments. We can think, for instance, at the introduction of
the important IEEE 802.11n High Throughput (HT) amendment, commonly implemented in almost
any modern commercial device, but which has been substantially ignored in the industrial realm.
Analogously, the Multirate Support feature is widely exploited by general purpose Wireless LANs
through the definition of Rate Adaptation (RA) strategies, revealing instead unsuitable for the case of
real–time communications.
In this talk I will present a thorough investigation of the most important enhancements that the
IEEE 802.11n/HT amendment introduced both at the physical (PHY) and medium access control
(MAC) layers. I will focus mainly on the case of industrial communication systems, and on some
relevant performance indicators, such as timeliness and reliability.
The talk will provide a theoretical background of the HT amendment, and describe a set of ex-
perimental measurements carried out on real devices with the aim of providing an exhaustive perfor-
mance assessment. The final outcome of this assessment is the definition of a suitable parameters’
set for the tuning of HT enhancements, to ensure the optimization of network performance in this
specific and demanding field of application.
On these basis, this talk will also provide a deep insight on the design of a new RA strategy, fully
exploiting the HT enhancements, which is definitively better tailored for the considered scenario. The
proposed Rate Selection for Industrial Networks (RSIN) algorithm is designed with the goal of mini-
mizing the transmission error probability, while taking into account the deadline imposed to frames’
delivery. Finally, I will discuss the practical implementation of RSIN on commercial devices that has
been carried out for the in–field assessment of the algorithm. The outcomes clearly highlighted that
RSIN outperforms all the most popular and adopted RA strategies, and is able to provide satisfactory
performance in a soft real–time industrial scenario.

Ca' Vignal 3 - Piramide, Floor 0, Hall Verde

Contact person
Graziano Pravadelli

Publication date
May 8, 2016