On the Scientific Value of Large-scale Testbeds for Wireless Multi-hop Networks

Abstract

Large-scale wireless testbeds have been setup in the last years with the goal to study wireless multi-hop networks in more realistic environments. Since the setup and operation of such a testbed is expensive in terms of money, time, and labor, the crucial question rises whether this effort is justified with the scientific results the testbed generates. In this paper, we give an answer to this question based on our experience with the DES-Testbed, a large-scale wireless sensor network and wireless mesh network testbed. The DES-Testbed has been operated for almost 5 years. Our analysis comprises more than 1000 experiments that have been run on the testbed in the years 2010 and 2011. We discuss the scientific value in respect to the effort of experimentation.