Generating more Realistic Packet Loss Patterns for Wireless links using Neural Networks

Simulations of wireless network connections are essential forthe development of new technologies because they are farmore scalable than real-world experiments and reproducible.Modeling packet loss realistically provides a highly abstractyet powerful tool for the simulation of wirelesses links. Typi-cally, simple statistical models or replaying of recorded tracesare used for the simulation. For a proper parametrization ofsimple statistical models, recorded traces are required, too.Both approaches have drawbacks: replaying traces is limitedto the length of the traces, a repetition may lead to unwantedeffects in the simulation. The statistical models solve this, butthe resulting packet loss patterns significantly differ from realones. In this paper, we propose using a neural network in-stead. It takes the same kind of input, i.e., a real-world trace,but it can generate longer traces with more realistic loss pat-terns. We share pre-trained neural networks for multiple linksin office and industry scenarios with the community for usein future research.