Accurate and timely mobility support in Internet of Things (IoT) applications is a challenging issue, considering the inherent scarce resources of IoT devices. However, the computational, memory and communication burden may be pushed into more ""muscled"" Software Defined Network (SDN) controllers. A centralised controller can exploit its global view of the network to predict and support seamless handovers. However, it requires the controller to be enhanced with extra link quality information. In this work, we present SDMob, an SDN-based mobility management solution that lifts the burden of computation intensive filtering algorithms from resource constrained nodes and achieves accurate and fast handovers upon nodes' mobility under Routing Protocol for Lossy Low-power Networks (RPL) and IPv6 over Low-Power Wireless Personal Area Networks (6LoWPAN). We show that SDMob improves the baseline RPL and the state-of-the-art mRPL in terms of packet delivery ratio leveraging more reliable routing and applying Particle filter and variations of Kalman filter on radio signal strength data enables more accurate localization for complex real world trajectories.