Optimal Placement of Fire Detection Sensors for Reduced Interference in Wireless Sensor Networks

The optimal deployment of wireless sensor networks for forest fire detection is crucial for the formulation of cost-effective networks and reliable communication with authorities. There are some traditional methods for detecting forest fires, which are not efficient due to their low scan period and high infrastructural costs. While contemporary approaches have improved aspects like localization, they often overlook a critical factor: the impact of signal interference on network performance and reliable data transmission. This paper addresses this gap by proposing an optimal sensor deployment method that mitigates interference, thereby ensuring robust and efficient communication of critical fire alarm messages. We propose optimal placement of the fire detection sensors based on the signal-to-interference-plus-noise ratio criterion. Data collected by sensors/nodes are transmitted to their respective cluster head using carrier sense multiple access with capture effect, that is, the signal to interference plus noise ratio (SINR) of each sensor is compared with a predefined threshold set by the cluster head. The signal will be decoded if SINR exceeds the predefined threshold; otherwise, it will be discarded and require retransmission. This SINR evaluation is conducted by each cluster head only at every geo-location point rather than by every node, to make this approach scalable and practical for large-scale implementation. From this method, we can find the optimal positions at which sensor nodes should be deployed for maximum throughput, thus reducing the problem of interference. The proposed approach is analytically evaluated, and the results show that the throughput is 53.28% compared to the random deployment of sensors, where the throughput is 45.21%. Benchmarking is also conducted with contemporary approaches, where our proposed algorithm gives stable performance in terms of delay, energy efficiency, and network lifetime.