CLIFT: A Cross-Layer InFormation Tool for latency analysis based on real satellite physical traces

Abstract

New mobile technology generations succeed in achieving high good put, which results in diverse applications profiles exploiting various resource providers (Wifi, 4G, SG, ... ). Badly set parameters on one of the network component may severely impact on the transmission delay and reduce the quality of experience. The cross layer impact should be investigated on to assess the origin of latency. To run cross-layer (from physical layer to application layers) simulations, two approaches are possible:(1) use physical layer models that may not be exhaustive enough to drive consistent analysis or (2) use real physical traces. Driving realistic measurements by using real physical (MACIPHY) traces inside network simulations is a complex task.We propose to cope with this problem by introducing Cross Layer InFormation Tool (CLIFT), that translates real physical events from a given trace in order to be used inside a network simulator such as ns-2. Our proposal enables to accurately perform analysis of the impact of link layer reliability schemes (obtained by the use of real physical traces) on transport layer performance and on the latency. Such approach enables a better understanding of the interactions between the layers. The main objective of CLIFT is to let us study the protocols introduced at each layer of the OSI model and study their interaction. We detail the internal mechanisms and the benefits of this software with a running example on 4G satellite communications scenarios.

Less-than-Best-Effort Capacity Sharing over High BDP Networks with LEDBAT

“…To drive this experiment, we use a ns-2 extension called Cross-Layer InFormation Tool (CLIFT) [14] allowing to play real physical layer traces inside ns-2. The 4G satellite link trace used was provided by CNES.…”