The aim of a routing protocol is to compute routes between every couple of nodes. When the topology evolves over time, routing decisions should be constantly reconsidered to ensure continuous valid routing. Conventional proactive routing protocols periodically re-compute their routing-tables. However, due to their inherent nature based on shortest-paths, they select longer links (victims of rapid breakages as nodes move). Increasing the updates' frequency certainly allows a better tracking of the topology changes; however, it induces higher signaling overhead.

The host identity protocol-based (HIP) machine-to-machine (M2M) overlay network is a virtual, self-organised and secure M2M network build on the top of the internet. M2M devices forming this overlay network may change their point of attachment to the overlay while having running sessions. HIP provides a native mobility support to these M2M devices. However, this support is not adapted for the M2M context. We propose in this paper a regular and lightweight mobility strategies to efficiently handle M2M devices mobility within this overlay.