Seite nur auf Englisch verfügbar
As wireless sensor networking technology decreases in both cost and size making it viable for a wide range of applications, the softwarerequired to achieve energy-efficient, robust and flexible data dissemination remains an open research area with many competingsolutions. Clustering and data aggregation inherently reduce energy expenditure while simultaneously maintaining sufficient quality data.Traditional approaches, however, both spend extensive communication energy to identify the cluster heads and are inflexible to network dynamics such as sink mobility, node failure, or dwindling battery reserves.
In this talk Anna Förster will be presenting Clique, a role-free clustering approach for WSNs based on reinforcement learning. Its main goal is to avoid all-together the cluster head selection overhead and to enable the nodes to independently decide whether or not to act as a cluster head on a per-packet basis. The protocol is highly flexible in case of mobility and failures and increases the system lifetime by approximately 25% when compared to traditional role-assigning schemes.