Article Preview
Top1. Introduction
Fog computing is a distributed computing model that acts as an intermediate layer between Cloud layer and IoT layer. Due to large extent use of IoT devices in recent days, the computational demand has grown a lot and this demand is fulfilled by integrating Fog computing with IoT devices. This is possible by having Fog computing placed close to IoT devices (Yousefpour et. al., 2019; Mahmud, Srirama, Kotagiri & Buyya, 2019; Atlam, Walters & Wills, 2018). Fog computing not only provides services to Internet of things, but also extends its application to other networking systems such as Mobile network/Radio access network, Power line communication, Content distribution network, healthcare applications and Vehicular network (Mahmud, Kotagiri & Buyya, 2018;Paul & Pinjari, 2018). This integration has facilitated real time and latency sensitive applications to be deployed. But due to lack of trust on IoT devices, the integration of fog computing and IoT devices is hindering.
Numerous security issues and attacks exist such as message forging, message tampering and replay attacks (Al-khafajiy et. al., 2019; Khalid et. al., 2019; Birje, Challagidad, Goudar & Tapale, 2015; Birje & Hallappanavar, 2019; Mouna Jouini & Latifa Ben Arfa Rabai, 2016), which disturbs the secured communication and trust among entities (Usha Divakaria & K. Chandrasekaran, 2016). Trust is the basis for the IoT devices to collaborate with each other (Yuan & Li, 2018; Firdhous, Hassan & Ghazali, 2013). It requires establishment of a trust that eliminates the fear among entities and provides required quality of services (Hou et. al., 2015; Gupta & Saini, 2017). The trust mechanism eliminates the malicious service providers and allows the true service providers to provide services. It also keeps a track on the quality of services and keeps a check each service provider to provide good services. Hence there is a need of a mechanism to compute the trustworthiness of IoT devices.
Most of the previous works have considered either subjective trust (Jian & Qin, 2015) or objective trust for their trust calculation. The trust computing mechanism based on subjective trust in many works lack the information fusion from multiple sources and proper weight assignment to the feedbacks given (Azzedin & Ghaleb, 2019; Challagidad, Birje & Goudar, 2020). Trust mechanism based on objective trust in many works lack the proper weight assignment to the quality of services parameters (Zennaro, Ivanovska & Jøsang, 2019). Also there are no incentive and punishment mechanisms to promote or warn the IoT devices to provide correct feedback (Wang, Huang & Zou, 2016; Tian, Yang, Zhong & Liu, 2014; Xiong et. al., 2018; Lu & Mengshu, 2005). So it requires an attention to focus on subjective and objective trust which provides a reliable and trustworthy trust computation mechanism.
Therefore this work proposes a reliable trust computing mechanism that works in following three steps:
- 1.
The subjective trust is computed through multiple source feedback fusion.
- 2.
Further the incentive and punishment scheme is applied to the computed subjective trust
- 3.
The objective trust is computed based on quality of services.
The contribution of this work is to design a reliable trust computing mechanism which combines the subjective and objective trust. Also, incentive and punishment mechanism is applied to the trust value to boost the IoT devices or punish the malicious devices.
The organization of the paper is as follows. Section II explains about the related work carried out so far. Section III covers architecture considered in the proposed work and trust computing mechanism. Section IV consists of results. The conclusion of the paper is covered in Section V.