A novel holistic approach for performance evaluation in Internet of Things. (2nd August 2020)

• Record Type: Journal Article
• Authors: Martinez‐Caro, Jose‐Manuel; Cano, Maria‐Dolores
• Other Names: Park Jong Hyuk guestEditor.; Younas Muhammad guestEditor.; Arabnia Hamid R. guestEditor.; Chilamkurti Naveen guestEditor.
• Is Part Of: International journal of communication systems. Volume 34:Number 2(2021)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2020-08-02
• Subjects: IoT -- LoRa/LoRaWAN -- performance -- quality -- simulation; Telecommunication systems -- Periodicals; 621.382
• DOI: 10.1002/dac.4454
• Languages: English
• ISSNs: 1074-5351

Abstract

Summary: The deployment of Internet of Things (IoT) solutions in smart cities or industrial environments (Industrial IoT, IIoT) demands careful consideration in terms of user‐centric or system‐centric target metrics. A better monitoring system able to transform performance outputs into decision‐making and intelligent actions requires less restrictive performance evaluation methods. Classic approaches to performance evaluation in telecommunication networks rely on Quality of Service (QoS) and/or Quality of user Experience (QoE) assessment models. However, the new IoT paradigm establishes a completely different scenario, where, for instance, consumers might no longer be users but machines. In this paper, we propose the evaluation of the performance of IoT services and applications comprising the combination of four quality measures, namely, Quality of Data (QoD), Quality of Information (QoI), Quality of user Experience (QoE), and Quality Cost (QC). The proposal is analyzed using computing simulations. Specifically, we improve the simulation tool FLoRa (Framework for LoRa) incorporating additional LoRa Wide Area Network (LoRaWAN) features. LoRaWAN is a Low‐Power Wide Area Network (LPWAN) technology that operates on the LoRa modulation scheme. Long coverage ranges, low power consumption, and support for a massive number of IoT devices using a limited infrastructure are the main assets of LoRaWAN, establishing it as one of the top IoT technologies. Results show that it is easier and more efficient to disassociate metrics into different dimensions in order to provide a clear vision of the performance of IoT services. This performance evaluation method can be customized and applied to different IoT markets. Abstract : A novel performance evaluation model for IoT is presented, based on the combination of four quality dimensions, namely, Quality of Data (QoD), Quality of Information (QoI), Quality of user Experience (QoE), and Quality Cost (QC). The proposal is evaluated through intensive computer simulations in a LoRa/LoRaWAN environment with real datasets. Results show that it is easier and more efficient to disassociate quality metrics into different dimensions in order to provide a clear vision of the performance of IoT services.