Improving gateway safety in wireless sensor networks using cognitive techniques

Abstract

In this paper, we present an innovative approach for gateway relocation in Wireless Sensor Networks (WSNs). We employ existing well-known artificial intelligence techniques to make smart gateway relocation based on many input factors from the environment. The gateway placement problem has been extensively researched. However, most of the proposed solutions are geared for boosting network-related performance metrics, such as throughput and energy consumption. The proposed approach for safe gateway relocation, named G-Safe, emphasises both increased protection of the gateway and improvements in overall network performance. We argue that relocating without taking safety concerns into consideration may cause the gateway to move dangerously close to one or multiple serious events in the environment. G-Safe utilises artificial neural networks to select the safest location accessible to the gateway. Experimental validation of G-Safe was conducted in two different environment setups; one that emphasises improvements in network longevity and the other strives to enhance the timeliness of the collected sensor readings. The validation results have confirmed the effectiveness of G-Safe in protecting the gateway while keeping the performance at an acceptable level.