**5. Conclusions**

In the wireless sensor network, hundreds of sensors are dispatched in the desired area and sensors are continuously operating. Various previous studies were conducted, including the proper placement of sensors and the selection of cluster headers to transmit the collected data to the desired sink nodes. However, wireless sensors are inherently energy-limited, thus the appearance of the sensing hole cannot be avoidable and sensing hole recovery to maintain network operation is necessary. In particular, since the charging system of the sensor is not always available, it is essential to consider energy-efficiency from the time the sensor is relocated to recover the sensing hole.

While there has been vigorous study of hopping mobility models suitable for rough areas recently, the majority of past studies have been based on not practical classic theory (i.e., relocation schemes based on source-based routing). In this paper, overcoming these problems, a new hopping sensor relocation protocol is proposed to suit the real distributed environment. To reflect the real situations, obstacles of a given area (stones, mud, etc.) have been considered. Since the status of obstacles around the sensing hole is unknown, the condition of the surrounding environment is predicted through the success rate of the hopping sensor relocation.

We would strongly emphasize that the relocation protocol proposed by this research team is a protocol in a distributed environment and the successful implementation of OMNeT++ simulation in the hopping sensor relocation protocol is our unique skill in this field of research. Various obstacles were considered for performance evaluation and analysis. Finally, superior energy efficiency with delaying sensing hole appearance of the proposed scheme was demonstrated.

**Author Contributions:** S.P. and M.K. are the co-first authors and contributed equally. S.P., M.K., and W.L. designed the protocol and the simulation process. S.P., M.K., and W.L. coordinated the gran<sup>t</sup> funding, conducted the study, analyzed the data, and drafted the first version of the manuscript. The simulation was conducted by S.P. and M.K., and W.L. supervised the software development. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was supported by Research Assistance Program (2019) in the Incheon National University and the National Research Foundation of Korea (NRF) gran<sup>t</sup> funded by the Ministry of Science, ICT & Future Planning (No. NRF - 2019R1G1A1007832).

**Conflicts of Interest:** The authors declare no conflict of interest.
