In wireless relay networks, there are several possibilities on how the helper (relay) nodes facilitate communication between a source and a destination. It has been shown recently (see http://licos.epfl.ch/index.php?p=research_projWNC) that we can develop simple relaying strategies that are (approximately) optimal. This theoretical development implies a radically new architecture for wireless networks. In this project we propose to develop practical implementation of these ideas on a software radio testbed based on the WARP platform (see http://warp.rice.edu/). We will develop implementation as well as algorithms that will need some theoretical work as well addressing issues such as network/channel knowledge and estimation, synchronization as well as computationally feasible coding mechanisms.