In this work we propose a physical layer design, based on Pulse Position Modulated (PPM) signals, for a decentralized wireless sensor network implementing an iterative consensus algorithm. The proposed scheme does not require any MAC protocol to avoid or resolve collisions, and is also suitable for a half-duplex implementation. The considered network model assumes only local coupling among the nodes, thus allowing for low transmit power even in large scale networks. Furthermore, we show how to remove the effect of propagation delays, multipath, and non perfect synchronization among the nodes, without requiring any channel parameter estimate. As an example of application, we consider a simple parameter estimation problem, which is instrumental to discuss the fundamental trade-offs arising in the system parameters settings, when both observation noise and coupling noise are considered in the performance analysis. ©2008 IEEE.
Distributed decision in sensor networks based on local coupling through Pulse Position Modulated signals / Pescosolido, Loreto; Barbarossa, Sergio. - (2008), pp. 1445-1448. (Intervento presentato al convegno 2008 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP tenutosi a Las Vegas, NV nel 31 March 2008 through 4 April 2008) [10.1109/icassp.2008.4517892].
Distributed decision in sensor networks based on local coupling through Pulse Position Modulated signals
PESCOSOLIDO, Loreto;BARBAROSSA, Sergio
2008
Abstract
In this work we propose a physical layer design, based on Pulse Position Modulated (PPM) signals, for a decentralized wireless sensor network implementing an iterative consensus algorithm. The proposed scheme does not require any MAC protocol to avoid or resolve collisions, and is also suitable for a half-duplex implementation. The considered network model assumes only local coupling among the nodes, thus allowing for low transmit power even in large scale networks. Furthermore, we show how to remove the effect of propagation delays, multipath, and non perfect synchronization among the nodes, without requiring any channel parameter estimate. As an example of application, we consider a simple parameter estimation problem, which is instrumental to discuss the fundamental trade-offs arising in the system parameters settings, when both observation noise and coupling noise are considered in the performance analysis. ©2008 IEEE.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.