Models of traffic demand are fundamental inputs to the design and engineering of data networks. This talk will address this requirement in the context of large-scale wireless infrastructures using real-measurement data. Our modeling effort focuses on capturing the demand variation in both the spatial and temporal domain in a way that scales well with the size of the wireless network. The network traffic dynamics are studied over two different monitoring periods at various levels of spatial aggregation, from individual buildings to the whole network.
Based on these models, we generate synthetic traffic and compare it with the measured (real-life) data. The comparison clearly illustrates the trade-off between model scalability and reusability and accuracy in capturing local-scale traffic dynamics. Our main contribution is a novel approach for traffic demand modeling in large wireless networks that features high flexibility in the exploitation of the spatial and temporal resolution available in data traces.
Finally, this talk will present a performance analysis of a wireless hotspot AP with respect to throughput, goodput, delay and jitter per flow. We compare the performance of these benchmarks under various traffic inputs:
(a) the measured data, from a real-life network (baseline case),
(b) synthetic traces based on various models.
The performance of these benchmarks when the measured data is used is very close to the one when synthetic traces based on our models are employed, deviating substantially from the one when synthetic traces based on popular (frequently-used) traffic models are used.