A Hybrid Multicast Routing for Large Scale Sensor Networks with Holes

Abstract

In this article, we present RE $^2$ MR, the first hybrid multicast routing protocol that builds on the strengths of existing topology-based, hierarchical and geographic multicast solutions, while addressing their limitations. In RE $^2$ MR, the multicast path search problem is formulated as the capacitated concentrator location problem (CCLP) which yields the network topology that minimizes the sum of path lengths from the multicast root to multicast members. Furthermore, its trajectory-based lightweight hole detection (THLD) discovers deployment area irregularities (i.e., network holes) that affect its solution and autonomously take them into account to generate updated routing paths, and its Energy-efficient Packet Forwarding (EPF) and Multi-level Facility Computation (MFC) reduce computational and communication overheads. We implement RE $^2$ MR in TinyOS and evaluate it extensively using TOSSIM for relatively large-scale simulations (400 nodes); we also implement RE $^2$ MR on real-hardware and perform experiments on a testbed consisting of 42 TelosB motes. Through the simulations and experiments on real-hardware, we demonstrate that RE $^2$ MR reduces the energy consumption by up to 57 percent and the end-to-end delay by up to 8 percent, when compared with the state-of-the-art multicast routing protocols.