An adaptive multipath algorithm to overcome the unpredictability of heterogeneous wireless networks for high-speed railway

Accessing Internet services in high-speed mobile scenario is an increasing demand for passengers and vendors. Owing to the bandwidth limitation of a single wireless network, researchers attempt to utilize the heterogeneous wireless networks along tracks to achieve multipath parallel transmission. These multipath transmission schemes usually depend on the accurate estimation of network quality to achieve high performance. However, due to the unpredictability of wireless networks in high-speed mobile scenario, current multipath transmission schemes perform poorly. In this paper, first, we make quantitative analysis for the unpredictability of wireless networks. With lots of results of real experiments, we make quantitative analysis for the estimation error of classical algorithms in different scenarios. Second, aiming at the unpredictability of wireless networks, we propose a multipath transmission algorithm named receiver adaptive incremental delay (RAID) that can aggregate bandwidth for heterogeneous networks independent of accurate network quality estimation. Final, we deploy the RAID algorithm into a real system. Abundant of real experiments and simulations prove that our proposed algorithm has a better performance than the earliest completion first algorithm and the weighted round Robin (WRR) algorithm in high-speed mobile scenario.