This article offers a viable solution to address the "hot spot" problem in Wireless Sensor Networks ( \(\mathbb {WSN}\) s) where sensor nodes ( \(\mathbb{S}\mathbb{N}s\) ) located nearer to the sink node ( \(\mathbb{S}\mathbb{K}\) ) deplete its energy quickly, resulting in a network partition. The proposed solution is to deploy multiple mobile sinks ( \(\mathbb {MMS}s\) ) to extend the network’s lifespan. The target region is splitted among the subareas ( \(\mathbb{S}\mathbb{A}s\) ) by employing K-means clustering, and the count of \(\mathbb{S}\mathbb{A}s\) is configured according to the number of mobile sinks ( \(\mathbb{M}\mathbb{S}s\) ) available. Each \(\mathbb{M}\mathbb{S}\) is assigned with a specific region to collect data from. Within each \(\mathbb{S}\mathbb{A}\) , the Voronoi diagram is used to identify probable rendezvous points ( \(\mathbb{R}\mathbb{P}s\) ) for the \(\mathbb{M}\mathbb{S}s\) . These \(\mathbb{R}\mathbb{P}s\) are then optimized using several parameters. The final collection of \(\mathbb{R}\mathbb{P}s\) is used to create a trajectory for the \(\mathbb{M}\mathbb{S}s\) such that they can gather information through the nearby \(\mathbb{S}\mathbb{N}s\) under a permissiable time period. The Delaunay triangulation and Delaunay centroid are also used for determining the \(\mathbb{M}\mathbb{S}s\) ’ trajectory within the same \(\mathbb{S}\mathbb{A}\) and network scenario. The proposed approach is analysed on MATLAB and analyzed against existing algorithms. The suggested solution outperforms existing methods as it yields better output in terms of network lifetime, residual energy, and other assessment metrics.
