A novel MAC scheduler to minimize the energy consumption in a Wireless Sensor Network

Abstract

The rising success of the Internet of Things has led the Wireless Sensor Networks to play an important role in many fields, ranging from military to civilian applications. However, since sensor nodes are battery powered, communication protocols and applications for these networks must be carefully designed in order to limit the power consumption. In this work, a new MAC protocol able to significantly reduce the power consumption and compatible with the IEEE 802.15.4 standard, is designed and validated. The defined protocol is based on an efficient setting of the node’s duty cycle as a function of the transmission times of the neighbor nodes. In a duty cycle period, each node wakes up once to transmit and N times to receive, where N is the number of neighbors, while it remains in sleep mode for the rest of the time. The defined protocol has been validated through both an analytical and a simulative approach. By using the first approach, the proposed solution is compared with another energy-efficient protocol, namely AS-MAC; then, the differences between the simulated scenario and the analytical one are analyzed. By using the second approach (through Omnet++ simulator), we carried out a performance comparison between our protocol and the current MAC protocol compliant with the ZigBee standard. All the results have shown the effectiveness of the proposed solution, which has proved to be flexible and efficient, since it is able to provide high energy savings at different date rate, without a negative impact on the packets delivery.

Introduction

The capability to sense key parameters from an environment is becoming more and more important in many application scenarios such as military operations, surveillance, building automation, healthcare, and logistics. This trend aims at creating smart environments able to capture, in a pervasive way, all useful information from the real world, contributing to assert the concept of the Internet of Things (IoT). The use of the wireless technology can facilitate this evolution process also leading to a growth of the Internet itself, which is no longer seen only as a tool for linking people to services, but as a means to allow the implementation of the new Machine-to-Machine (M2M) paradigm. Among all wireless technologies, WSNs are the ideal choice since their ability to self-configure and self-organize allows to overcome the typical weaknesses of infrastructured networks, in order to carry out pervasive environments such as smart city, smart farm, and smart hospital [1]. A WSN consists of a number of nodes distributed in an area of interest and equipped with some sensors (e.g., temperature, pressure, humidity). These nodes communicate each other through a radio transceiver that allows them to send the captured data to a special entity called sink. One of the main characteristics of WSNs is the use of multi-hop communication, which allows nodes farthest from the sink to reach the sink itself, but at the same time it requires that the network is never partitioned, i.e. a path to the sink should always be present. With regard to this aspect, it is important to note that the nodes of a WSN are battery-powered and, in most cases, such a power cannot be regenerated or replaced, therefore its exhaustion corresponds to a definitive shutdown of the node. In this situation, not only data collected by that node are lost, but also routes eventually handled by that node. Of course, a large number of node losses, due to battery depletion, could lead to network partitioning. That said, the energy consumption is, therefore, a key aspect in WSNs and it is crucial that sensor nodes optimize power consumption to extend the network lifetime in a consistent way with the real use cases.

The main operations that a WSN node can perform are data sensing, data processing and data communication. This last operation is certainly the most stressful operation from the point of view of power consumption, because it is associated with phenomena such as collision, overhearing (i.e. listening of messages addressed to another node), over-emitting (i.e. transmission of data to a node that cannot receive them) and idle listening (i.e. listening to the channel in absence of communications). For these reasons, many works in literature are focused on energy saving exploiting enhancements at various layers of the protocol stack of sensor nodes [2], [3], [4], with particular interest on the MAC layer. In fact, since the MAC layer is responsible for managing channel access control mechanisms, it provides directly access to the functionality of the radio transmitter and allows to modify its state without any modifications to the standard. The opportunity, for example, to exploit the existing MAC control packet to interact with the radio transmitter enables new protocols to optimize the power consumption while maintaining compatibility with the IEEE 802.15.4 standard. Then, in this field, a very interesting approach is based on an appropriate tuning of the node duty cycle, through which nodes switch between ON and OFF state according to a predefined scheduling. Such an approach must deal with synchronization problems among nodes, high overhead, hidden node especially in the initial setup phase, etc.

In this work, an asynchronous scheduler is defined that significantly reduces the power consumption of WSN nodes using an approach based on duty-cycle. Thanks to the defined solution, each node is able to know in advance the time instants in which its neighbors transmit. By exploiting this information, the local scheduler of each node regulates the activation and deactivation of the radio transmitter so that this transmitter is active only when nodes actually send or receive data. This approach consistently reduces power consumption connected with the phenomenon of idle listening and thus it causes an extension of network lifetime. To make the proposed solution as consistent as possible to the real nodes behavior, also network topology changes have been suitably managed, since the information stored in each node is updated as soon as the fall of a neighbor node or the presence of a new one are detected. In addition, the desynchronization problem caused by clock drift is adequately addressed. The effectiveness of the proposed solution has been evaluated through both an analytical and a simulative approach. In particular, from the mathematical point of view, we compared our solution with another energy-efficient protocol, namely AS-MAC. The results of this comparison showed that our approach has a better flexibility in addressing different application requirements. Then, in order to evaluate the accuracy of the mathematical model, we analyzed the differences between the simulated scenario and the analytical one. For the simulative approach, we used OMNET++ simulator [5], which is considered one of the best choice for evaluating protocols for WSNs in a simulated environment [6]. By using this tool, we carried out a performance comparison between our protocol and the current MAC protocol compliant with the ZigBee standard. Simulation results are very encouraging since the proposed solution allows to achieve a significant energy savings in a chain topology and, therefore, a considerable extension of network lifetime. Finally, the drawbacks of the defined approach in terms of packet delivery ratio are very negligible especially considering that WSNs are usually exploited in low traffic scenarios.

The paper is organized as follows. Section 2 summarizes the state of the art related to the minimization of power consumption in a WSN. The defined scheduler is described in Section 3. A detailed description of the mathematical model consistent with the proposed protocol is reported in Section 4. In Section 5, a simple mathematical model for AS-MAC protocol is presented. The simulation model used in Omnet++ simulator is given in Section 6, whereas in Section 7 the results are discussed. Conclusions are drawn in Section 8.