Muhammad Nabeel, Bastian Bloessl and Falko Dressler, "Selective Signal Sample Forwarding for Receive Diversity in Energy-Constrained Sensor Networks," Proceedings of IEEE International Conference on Communications (ICC 2017), Paris, France, May 2017.

### Abstract

Receive diversity increases the reliability and robustness of ultra-low power Wireless Sensor Networks (WSNs) by using spatially separated antennas without modifying the physical layer. We consider a distributed ground network in the wild to track bats in their natural habitat. The bats are equipped with a sensor node of only 2 g that limits the energy budget available for communication. In this work, we exploit the distributed nature of the ground network to employ diversity combining, i.e., we use the ground nodes as a geographically distributed multi-antenna array. However, this causes several research challenges given the limited bandwidth between nodes and the need for accurate synchronization to combine signal copies constructively at a central node. Sending all signal samples from all ground nodes to the central node is prohibitive due to the required data rates in the network. As a novel concept, we propose a system that only forwards selected signal samples belonging to a packet with high probability. We study the performance in simulations as well as in a testbed using a Software Defined Radio (SDR) prototype implementation. Our results clearly indicate a substantial performance gain while keeping the data rate in the ground network in a feasible range.

### Quick access

Original Version (at publishers web site)
Authors' Version (PDF on this web site)
BibTeX

### BibTeX reference

@inproceedings{nabeel2017selective,     address = {Paris, France},     author = {Nabeel, Muhammad and Bloessl, Bastian and Dressler, Falko},     booktitle = {IEEE International Conference on Communications (ICC 2017)},     doi = {10.1109/ICC.2017.7996320},     month = {May},     publisher = {IEEE},     title = {{Selective Signal Sample Forwarding for Receive Diversity in Energy-Constrained Sensor Networks}},     year = {2017},    }