Self-Organized Connected Objects: Rethinking QoS Provisioning for IoT Services

Abstract

The proliferation of connected objects has revolutionized the traditional Internet, giving rise to the emerging Internet of Things (IoT). The IoT ecosystem is very large, and it includes smart interconnections among sensors and devices with applications in both the industrial world and customers' daily lives. As a unified standard for IoT is still under development, many challenges related to IoT must be discussed and addressed, especially those related to energy efficiency. This article tackles the challenge of energy efficiency in IoT from a novel perspective. It shows that instead of maximizing the QoS, which is generally energy costly, better energy efficiency can be achieved by targeting satisfactory QoS levels only. The approach aims to enhance energy efficiency while ensuring a desired QoS threshold. This is supported by a game theoretical solution concept referred to as the satisfaction equilibrium. Moreover, as IoT objects require self-configuring techniques to maintain the network scalability and flexibility, this article introduces fully distributed schemes in order to reach efficient satisfaction equilibria in both slow- and fast-fading channel contexts. The proposed schemes can also be adapted to achieve the maximum performance of IoT applications that desire the highest QoS levels. The performance of these algorithms is illustrated through a smart home use case scenario.