Evaluation of Energy-Efficient Heuristics for ACO-based Routing in Mobile Wireless Sensor Networks

Arliones Hoeller Jr.,Antônio Augusto Fröhlich

DOI: https://doi.org/10.48550/arXiv.1512.02114

2015-12-08

Abstract:Mobile ad-hoc networks demand routing algorithms able to adapt to network topologies subject to constant change. Moreover, with the advent of the Internet-of-Things (IoT), network nodes tend not only to show increased mobility, but also impose further restrictions to these mobile network systems. These restrictions include non-functional requirements of low-power consumption, small-size, and low-cost. ADHOP (Ant-based Dynamic Hop Optimization Protocol) is a routing algorithm based on ant-colony optimizations that target such small-size and low-cost platforms, consuming little amounts of memory and processing power. This paper elaborates on ADHOP to investigate the use of energy-related heuristics to guide routing decisions. The goal is to minimize network usage without jeopardizing network operation. We replace the original ADHOP heuristic of network latency by two energy-related heuristics based on the battery charge and the estimated node lifetime. Simulations compare the energy-aware versions of ADHOP to its original version, to AODV and to AOER. Results show that the proposed approaches are able to balance network load among nodes, resulting in a lower number of failures due to battery depletion. The energy-aware versions of ADHOP also deliver more packets than their counter-parts in the simulated scenario, delivering 2x more packets than the original ADHOP, and, respectively, 5x and 9x more packets than AOER and AODV.

Networking and Internet Architecture