TY - GEN
T1 - Distributed fault-tolerant backup-placement in overloaded wireless sensor networks
AU - Oren, Gal
AU - Barenboim, Leonid
AU - Levin, Harel
N1 - Publisher Copyright:
© ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering 2019.
PY - 2019/1/1
Y1 - 2019/1/1
N2 - Wireless Sensor Networks (WSNs) frequently have distinguished amount of data loss, causing data integrity issues. Sensor nodes are inherently a cheap piece of hardware - due to the common need to use many of them over a large area - and usually contain a small amount of RAM and flash memory, which are insufficient in case of high degree of data sampling. An overloaded sensor can harm the data integrity, or even completely reject incoming messages. The problem gets even worse when data should be received from many nodes, as missing data becomes a more common phenomenon as deployed WSNs grow in scale. In cases of an overflow, our Distributed Adaptive Clustering algorithm (D-ACR) reconfigures the network, by adaptively and hierarchically re-clustering parts of it, based on the rate of incoming data packages in order to minimize the energy-consumption, and prevent premature death of nodes. However, the re-clustering cannot prevent data loss caused by the nature of the sensors. We suggest to address this problem by an efficient distributed backup-placement algorithm named DBP-ACR, performed on the D-ACR refined clusters. The DBP-ACR algorithm re-directs packages from overloaded sensors to more efficient placements outside of the overloaded areas in the WSN cluster, thus increasing the fault-tolerance of the network and reducing the data loss.
AB - Wireless Sensor Networks (WSNs) frequently have distinguished amount of data loss, causing data integrity issues. Sensor nodes are inherently a cheap piece of hardware - due to the common need to use many of them over a large area - and usually contain a small amount of RAM and flash memory, which are insufficient in case of high degree of data sampling. An overloaded sensor can harm the data integrity, or even completely reject incoming messages. The problem gets even worse when data should be received from many nodes, as missing data becomes a more common phenomenon as deployed WSNs grow in scale. In cases of an overflow, our Distributed Adaptive Clustering algorithm (D-ACR) reconfigures the network, by adaptively and hierarchically re-clustering parts of it, based on the rate of incoming data packages in order to minimize the energy-consumption, and prevent premature death of nodes. However, the re-clustering cannot prevent data loss caused by the nature of the sensors. We suggest to address this problem by an efficient distributed backup-placement algorithm named DBP-ACR, performed on the D-ACR refined clusters. The DBP-ACR algorithm re-directs packages from overloaded sensors to more efficient placements outside of the overloaded areas in the WSN cluster, thus increasing the fault-tolerance of the network and reducing the data loss.
KW - Data loss
KW - Distributed backup-placement
KW - Networks connectivity
KW - Wireless sensor networks
UR - https://www.scopus.com/pages/publications/85059752004
U2 - 10.1007/978-3-030-05195-2_21
DO - 10.1007/978-3-030-05195-2_21
M3 - Conference contribution
AN - SCOPUS:85059752004
SN - 9783030051945
T3 - Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, LNICST
SP - 212
EP - 224
BT - Broadband Communications, Networks, and Systems - 9th International EAI Conference, Broadnets 2018, Proceedings
A2 - Althunibat, Saud
A2 - Sucasas, Victor
A2 - Mantas, Georgios
PB - Springer Verlag
T2 - 9th International EAI Conference on Broadband Communications, Networks, and Systems, Broadnets 2018
Y2 - 19 September 2018 through 20 September 2018
ER -