MSAR: A metric self-adaptive routing model for Mobile Ad Hoc Networks

This paper proposes a metric self-adaptive routing scheme for Mobile Ad Hoc Networks (MANET). By applying the proposed model, each node is able to detect whether the mobility states of the network is relatively static or mobile without the support of the Global Positioning System (GPS). The mobility state detection model is designed based on an indicator named MSI (for proactive routing) or GMSI (for reactive routing) computed at each node. Based on MSI/GMSI, an adaptive algorithm is then designed to employ the appropriate routing metric, i.e., either Expected Transmission Count (ETX) or Path encounter Rate (PER), for each detected state in order to achieve the optimum routing performance for different network conditions (i.e., static or mobile).