-\begin{enumerate} [(I)]
-\item \textbf{Data reduction schemes} that deal with reducing the amount of data need to be transmitted to sink. They can be divided into stochastic approaches, time series forecasting and algorithmic approaches. In stochastic approaches, the physical phenomena are transformed using stochastic characterization. the aggregating by these protocols require high processing so it is feasible to work on a powerful sensor nodes with a big battery. In time series forecasting, the old values of periodic sampling can be used to forecast a future value in the same series. In algorithmic approaches, sensed phenomena are demonstrated using heuristic or state transition model.
-\item \textbf{Energy efficient data acquisition schemes} are concentrated on the energy consumption reduction in the sensing unit. These schemes are divided into adaptive sampling, hierarchical sampling and model based active sampling. In adaptive sampling, the amount of data that acquired from the transducer can be reduced by spatial or temporal correlation between data. These approaches are more efficient to be used in centralized fusion but it consumes a high energy due to requiring a high processing. In hierarchical sampling, are more efficient when there are different types of sensor are installed on the nodes. These approaches are more energy efficient and application specific. The model based approaches are similar to data prediction schemes. These approaches aim to decrease the data samples by using computed models and conserve the energy by means of data acquisition.
-\end{enumerate}
+%\begin{enumerate} [(I)]
+\subsubsection{Data Reduction Schemes}
+Data reduction schemes deal with reducing the amount of data to be transmitted to a sink. They can be divided into stochastic approaches, time series forecasting, and algorithmic approaches. In stochastic approaches, physical phenomena are transformed using stochastic characterization. The aggregation by these protocols requires high processing. Therefore, it is feasible only on powerful sensor nodes with a big battery. In time series forecasting, the old values of periodic sampling can be used to forecast a future value in the same series.
+%In algorithmic approaches, sensed phenomena is described using heuristic or state transition model.
+
+\subsubsection{Energy Efficient Data Acquisition Schemes}
+They concentrate on the energy consumption reduction in the sensing unit. These schemes are divided into adaptive sampling, hierarchical sampling, and model-based active sampling. In adaptive sampling, the amount of data acquired from the transducer can be reduced by spatial or temporal correlation between data. These approaches are more efficient to be used in centralized fusion, but they consume more energy due to requiring a high processing. Hierarchical sampling is more efficient when there are different types of sensors installed on the nodes. These approaches are more energy efficient and application specific. The model-based approaches are similar to data prediction schemes. These approaches aim to decrease the data samples by using computed models and to conserve the energy by means of data acquisition.
+%\end{enumerate}
+
+\subsection{Battery Repletion}
+
+\indent In the last years, extensive researches have been focused on energy harvesting and wireless charging techniques. These solutions represent alternate energy sources to recharge wireless sensor batteries without human intervention~\cite{ref91,ref59}.
+
+\begin{enumerate} [i)]
+\item{Energy Harvesting} In energy harvesting, several sources of environmental energy have been developed so as to enable the wireless sensors to acquire energy from the surrounding environment. These energy sources are solar, wind energy, vibration based energy harvesting, radio signals for scavenging RF power, thermoelectric generators, and shoe-mounted piezoelectric generator to power artificial organs~\cite{ref59}.
+
+\item{Wireless Charging}In wireless charging, the power can be transmitted between the devices without requiring a connection between the transmitter and the receiver. These techniques participate in increasing the availability of WSNs and prolonging the network lifetime. Wireless charging in WSNs can be performed in two ways: magnetic resonant coupling and electromagnetic radiation~\cite{ref22}.
+
+\end{enumerate}
+
+\subsection{Radio Optimization}
+
+\indent In wireless sensor node, the radio is the most energy-consuming unit. Extensive researches have been focused on decreasing the power depletion due to wireless communication by means of optimizing the radio parameters such as coding and modulation schemes; transmission power and antenna
+direction; and cognitive radio and cooperative communications schemes~\cite{ref22}.
+
+\subsection{Relay nodes and Sink Mobility}
+\indent The relay node placement and the mobility of the sink can be considered as energy-efficient strategies to minimize the energy consumption.
+%\begin{enumerate} [(I)]
+\subsubsection{Relay node placement}
+In WSN, some wireless sensor nodes in a certain region may die and this creates a hole in the WSN. This problem can be solved by placing the wireless sensor nodes in sensing field by using an optimal distribution or by deploying a small number of relay wireless sensor nodes with powerful capabilities. The major goal of relay nodes is the communication with other wireless sensor nodes or relay nodes~\cite{ref52}. This solution can enhance the power balancing and avoid overloaded wireless sensor nodes in a particular region of a WSN.
+
+\subsubsection{Sink Mobility}
+In WSNs including a static sink, the wireless sensor nodes which are near the sink drain their power more rapidly compared with other sensor nodes, and this leads to WSN disconnection and limited network lifetime~\cite{ref53}. Sending all the data to the sink maximizes the overload on the sensor nodes near to the sink. In order to overcome this problem and prolong the network lifetime, we can use a mobile sink which moves within the area of interest to collect the sensory data from the static sensor nodes over a single hop communication. A mobile sink avoids the multi-hop communication and conserves the energy at the static sensor nodes near to the base station, extending the lifetime of WSN~\cite{ref54,ref55}.
+
+
+
+%\end{enumerate}
