Computer Networks 98 (2016) 57–71 Contents lists available at ScienceDirect Computer Networks journal homepage: www.elsevier.com/locate/comnet A practical cross layer cooperative MAC framework for WSNs M. Sarper Gokturk a, , Ozgur Gurbuz b , Murat Erman c a AirTies Wireless Networks, Istanbul 34394, Turkey b Faculty of Engineering and Natural Sciences, Sabanci University, Istanbul 34956, Turkey c Vodafone Turkey, Istanbul, Turkey article info Article history: Received 22 June 2015 Revised 17 January 2016 Accepted 27 January 2016 Available online 3 February 2016 Keywords: Cooperative MAC Medium access control Energy efficiency Wireless sensor network WSN for IoT abstract The evolving Internet of Things is expected to enable realization of wireless sensor net- works (WSNs) for a variety of applications. Energy efficiency and reliability are the key criteria for the success of WSNs of IoT. In this article, a cooperative medium access con- trol (MAC) framework is proposed for improving the performance and energy efficiency of WSNs, while satisfying a given reliability constraint. The energy-reliability trade off is achieved through a relay selection and power assignment algorithm, which is implemented within the COMAC cooperative MAC protocol that enables the coordination of candidate relays, calculation of the decision metrics, selection and actuation of the relay nodes with optimal power levels for cooperation. The proposed cross-layer MAC framework is evalu- ated in terms of energy costs as well as network performance metrics, in terms of through- put, delay and overhead. It is shown that the network throughput can be improved signif- icantly, while the energy consumption is reduced by at least two orders of magnitude as compared to standard Zigbee WSNs, at negligibly small overhead and computational costs. © 2016 Elsevier B.V. All rights reserved. 1. Introduction The term Internet of Things (IoT) refers to uniquely identifiable objects and their virtual representations in an internet-like structure. These objects can include any kind of goods, such as buildings, cars, trains, planes, ma- chines, industrial plants, human beings, animals and plants or their body parts, all connected to form a smart envi- ronment [1]. While IoT does not assume a specific com- munication technology, in particular, wireless sensor net- works (WSNs) will proliferate many applications and many industries [2,3]. The current WSN technology has matured on Zigbee, WirelessHART and ISA100.11a standards, which involve the basic wireless communication and networking functions This work was supported by TUBITAK Career grant no: 105E093. Corresponding author. Tel.: +90 212 318 62 00. E-mail addresses: sarper.gokturk@airties.com (M.S. Gokturk), ogurbuz@ sabanciuniv.edu (O. Gurbuz), murat.erman@vodafone.com (M. Erman). from IEEE 802.15.4 [4]. However, the success of IoT will depend on the improved performance, mainly in relia- bility and energy efficiency performance of WSNs [5,6]. The reliable delivery of sensor data plays a crucial role in the WSNs of smart environments and IoT. The reliability requirements may vary depending on the application or the content of the data itself. For instance, for tempera- ture monitoring in a smart home, a certain percentage of data loss can be tolerated for delivering temperature data within normal range; while a high temperature measure- ment must be delivered at very high reliability, since it can be the sign of a fire. In a health monitoring application, de- livery of all the sensory data may require high reliability. Reliability, as one of the Quality of Service (QoS) param- eters, can be measured in terms of Bit Error Rate (BER), Frame Error Rate (FER) or packet loss rate metrics. QoS provisioning in wireless networks is achieved via resource allocation [7], and resource allocation considering reliabil- ity has been an important problem in all types of wireless networks, such as third generation cellular systems [8]. For http://dx.doi.org/10.1016/j.comnet.2016.01.013 1389-1286/© 2016 Elsevier B.V. All rights reserved.