A heuristic and analytical framework for routing protocol optimization in wireless sensor networks: A performance metric-based analysis

Wireless Sensor Networks (WSNs), composed of numerous spatially dispersed, low-power, and computationally constrained sensor nodes, are instrumental in acquiring real-time data from dynamic environments. These miniature embedded systems, though economically viable, demand judicious protocol design due to their intrinsic limitations in energy, bandwidth, and processing capacity. In this study, a rigorous evaluation of transport and MAC layer routing protocols is undertaken to ascertain their performance efficacy under resource-constrained conditions. Utilizing the NS2 (v2.35) simulation platform, three transport protocols-AODV, DSDV, and DSR-and three MAC protocols-IEEE 802.11, IEEE 802.15.4, and S-MAC-are comparatively analyzed. Key performance indices such as Packet Delivery Ratio (PDR), mean end-to-end delay, and network throughput are employed to assess their operational characteristics under controlled testbed scenarios. The empirical findings reveal that AODV and DSR both attained an ideal PDR of 100%, whereas DSDV, though relatively efficient, registered a slightly diminished delivery rate of 97.16%. DSDV demonstrated the shortest average delay at 0.3105 seconds, outperforming both AODV (0.4921 s) and DSR (0.5102 s). Throughput-wise, AODV emerged superior with 263.34 kbps, while DSR closely followed at 258.77 kbps. At the MAC layer, S-MAC delivered impeccable reliability, achieving full packet delivery without any loss, albeit at the cost of increased latency (2.516 seconds) owing to periodic sleep cycles. WSN performance and careful optimisation-guided by simulation and metric- driven analysis-are essential for tailoring protocol behavior to meet specific application demands.