In the edge-based Internet of Things (IoT) era, wireless sensor networks (WSNs) are the prime source for data collection. In such WSNs, mobile edge nodes such as mobile sinks (MSs) are the superior means to collect sensed data by visiting rendezvous points (RPs). However, WSNs are often obstacle-ridden, which creates hurdles to the movement of the MSs. Most of the existing path planning works dealing with obstacles do not address optimal and smooth path construction. In other words, they have not considered a) optimizing the number of RPs and constructing a feasible path and b) smoothing the constructed path by considering sharp edges and convexity of the obstacle perimeter. In this paper, we address all such issues and develop an efficient scheme for determining an optimal number of RPs using a greedy approach to the set-cover problem and optimized path construction, both in polynomial time. Then, we apply the modified BUG2 algorithm to construct an obstacle-free path, which is then smoothed using the concept of the Bezier curve. Extensive simulations show the superiority of our proposed scheme over the existing algorithms in terms of energy consumption, latency, and so on.

With the ever-increasing demand for Internet of Things (IoT) applications, wireless sensor networks (WSNs) have become the central means to disseminate data for analysis in the era of mobile edge computing. Mobile sinks (MSs) as edge nodes have emerged as an efficient solution to the performance enhancement of WSNs. One important task of the MSs is to collect data in a sustainable and optimized manner by visiting certain rendezvous points (RPs) inside the WSN. However, most existing works focus only on connected WSNs, while disjoint networks are the reality in many IoT applications. Moreover, none of them have considered a realistic propagation model. They have also ignored optimizing both the number of RPs and MSs. This paper proposes a novel data collection scheme while paying attention to all these issues. The scheme is specially designed for delay-harsh applications. First, we propose a convex hull-based algorithm to determine RPs for constructing an optimal tour of a MS. Then using the resulting set of RPs, we present another algorithm based on the Jaya metaheuristic to determine an optimal number of MSs and their balanced tours. Rigorous simulations show that our scheme outperforms existing algorithms in terms of various performance metrics.

Exploiting mobile sink (MS) for data gathering in the wireless sensor networks has been extensively studied in the recent researches to address energy‐hole issues, thereby facilitating balanced energy consumption among nodes and so prolonging network lifetime. However, such approaches suffer from an extended data collection delay causing buffer overflow problem. In this regard, finding the optimal number of locations (i.e. rendezvous points (RPs) where the MS sojourns for data collection), is not only of utmost importance, but also a challenging task. A novel scheme for trajectory design of MS for data collection is presented in this study. The authors' primary goal is to optimise the number of RPs and their locations to minimise the travelling length of the MS. First, they reduced the problem size by using a combination of breadth‐first search and Tarjan's algorithm and then applied spectral clustering to find the optimal set of RPs to plan the tour for the MS. They have performed extensive simulations, and the results are compared with relevant existing schemes. The comparative results confirm the effectiveness of their approach in terms of the number of RPs, path length, the variance of RPs, and energy consumption per round.