Conventional IoT applications rely on seamless data collection from the distributed sensor nodes of Wireless Sensor Networks (WSNs). The energy supplied to the sensor node is limited and it depletes after each cycle of data collection. Therefore, data flow from the network to the base station may cease at any time due to the nodes with a dead battery. A replacement of the battery in WSNs is often challenging and requires additional efforts. To ensure the robust operation of WSNs, many fault recovery routing mechanisms have been proposed. Most of the previous fault recovery routing methods incur considerable delays in recovery and high overhead in either energy consumption or device cost. We propose an energy-efficient fail recovery routing method that is aimed to operate over a data aggregation network topology using a TDMA media access control (MAC). This paper introduces a novel fault recovery routing algorithm for TDMA-based WSNs. It finds an optimal neighbor backup parent (NBP) for each node in a way that reduces the energy consumption. The proposed method allows the NBPs to utilize the time slot of the faulty parent nodes, so it eliminates the overhead of TDMA rescheduling for NBPs. To evaluate the fault recovery performance and energy efficiency of the proposed method, we implemented it in C++ simulation program. Simulation experiments with an extensive set of network examples demonstrate that the proposed method can extend the network lifetime by 21% and reduce the energy consumption by 23% compared with the reference methods.Electronics 2018, 7, 444 2 of 21 collisions, which incur retransmissions of packets causing extra energy loss. A time division multiple access (TDMA) protocol is regarded as an effective alternative to CSMA, since it can ensure fair and collision-free data forwarding from all nodes, therefore reducing the energy loss [4,5]. Our proposed method is thus based on TDMA. Regardless of the choice of protocol, however, any WSN is susceptible to devise failure or battery depletion, and therefore it may lose network connectivity.Recent studies on WSNs have achieved considerable enhancement in network architecture and data forwarding protocols to reduce the energy consumption [2]. The primary goal of many WSNs is to maximize the network lifetime even under the event of node failures [6]. Hence, it needs a fail recovery method that operates the rest of the WSN to maintain the desired lifetime.For low-power WSNs, a tree structure topology is often adopted [7], since it permits simple routing paths from all the nodes towards the sink (root) node, which acts as a gateway collecting all the sensing data. In WSNs of tree structure topology, each child node at a lower level forwards its sensing data to its parent node at a higher level until all data are delivered to the sink node [7]. If any parent node fails, then, all nodes in the subtree under the failed parent lose their routing path towards the sink node. A large portion of the network, therefore, can be isolated, resulting in all their sensing ...