**7. Conclusions**

In this paper, we proposed a method to reduce power consumption according to the protocol usage by introducing fine-grain power modes to the WSN node. The conventional sensor system has three kinds of power modes: sleep, idle, and run. We divide the sleep mode into three states: a deep sleep mode (State 0), a medium-idle state of the sensor (State 1), and a medium-idle state of the wake-up Rx channel (State 2). Thus, the proposed WSN platform has five states of power mode, including idle and run. Even in sleep mode, power consumption is very different depending on WSN status. For example, the mediumidle mode for the sensor consumes eight times more power than the medium-idle state for wRx. We can manage the scheduling of power modes at the protocol level using a frame pending bit in the header of data packets. This power mode can be applied to the scenario of node communication, event detection, and standby according to the environment. The sleep state, which is a standby state, is an event that has the most impact on the battery life of a sensor board and has the longest time occupancy in power mode. Event recognition in sleep mode is not continuous, but periodic sensing. Thus, the minimum time to operate the sensor was calculated and we obtained the energy consumption. Besides, periodic

radio signal receiving consumes the most energy in sleep mode. However, the proposed system saves energy during radio communication because the node is not used until the idle mode. The experimental results are simulations based on the architecture in Figure 1, using parameters in Tables 2 and 4, and Figure 5. We assumed the use of gas sensors on the Mica2 platform in WSNs. The node module senses gas at state 1 of a node. The sensor node is in sleep mode and the event sensing period is set to 1 s. The proposed FGPM controls the node status of the sleep mode finely that offers energy savings of 74.2% compared to the conventional approach. This can be seen as a significant contribution to battery saving since the sensor nodes are idle in sleep mode for majority of time. However, when events occur consecutively without a sleep time, the power reduction is less than 2%. As a result, the proposed method can be expected to save power more effectively in a wireless sensor network with a low event probability or a small number of events in which most networks operate.

**Author Contributions:** Conceptualization, S.Y. and K.C.; Validation, S.Y. and J.K.E.; formal analysis, S.Y. and K.C.; writing—original draft preparation, S.Y. and K.C.; writing—review and editing, J.K.E. and H.C.I.; supervision, H.C.I. and K.C.; All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was supported by the National Research Foundation of Korea gran<sup>t</sup> funded by the Korean governmen<sup>t</sup> (MSIT) (No. 2020R1F1A1069381) and Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (No. 2020R1A6A1A12047945)

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Data sharing not applicable.

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