An Efficient QoS-aware Services Selection in IoT using A Reputation Improved- Social Spider Optimization Algorithm

Abstract

Internet of Things (IoT) has been grown rapidly over the last years to connect a considerable number of spatially distributed objects or actuators. The connected objects create new functionality and provide various services to enhance and satisfy End-users daily lives. The issue is to provide the End-users with optimal services based on their requirements. The critical challenge is to select the optimal service from similar services functionally and various services non-functionality requirements (Quality of services). To achieve this challenge, this paper proposed a services selection model under QoS constraints in the IoT environment. The introduced model implements a meta-heuristic optimization algorithm with a friendly Likert scale measurement method. It aims to improve the performance of bio-inspired optimizing algorithms, called a Social Spider Optimization (SSO) Algorithm, by adding a reputation value to member's weight. The proposed model used a Likert scale measurement to evaluate the reputation of the services from the End-users. In the experiments, a comparative study was done between an original SSO and the proposed RI-SSO model. The results show the efficiency of the proposed RI-SSO model against the original SSO, in both maximization and minimization problems. It obtains a better outperform in terms of fitness values. ​​

An Ontology Based Knowledge Representation for Coordinated SS-bots

“…We envision the collective behaviour of SS-bots as based on individual decision making and local interactions policies (James et al, 2015;Li et al, 2017;Bouriot and Chevrier, 2020). Collective behaviour is only visible at swarm level (Saffre et al, 1999;Abosaif and Elrofai, 2020) though it is generated at individual SS-bot level. Little has been documented on the nature of the low-level actions and the information SS-bots share.…”

“…SS-bots inspired swarm systems function under uncertainty (Abosaif and Elrofai, 2020). They equally rely on the environment, SS-bots, data for decision-making, action planning, and interaction among the SS-bots.…”

“…The design and capabilities of the SS-bot are essential. Most designs of robotic devices imitate social colonies (Abosaif and Elrofai, 2020). For example, ant colony systems mimicked the foraging behaviour of ants (Li et al, 2017).…”

“…Intriguing is the interest by researchers to investigate even the extension of such knowledge to the coordination of swarms of heterogeneous members (Saffre et al, 1999). Swarms of heterogeneous members can deal with complex mission too challenging for human to perform (Saffre et al, 1999;Tan and Zheng, 2013), tasks that may present a mammoth assignment to a swarm of homogeneous members (Abosaif and Elrofai, 2020). For example, what form of swarm intelligence can achieve a search and rescue mission in a collapsed mine?…”

“…For example, such swarms may comprise ant-like members for robust and faulttolerant search (Melaine et al, 2020), as well as bee-like members for a waggle-dance inspired recruitment (Cheraghu et al, 2021;Carrillo-Zapata et al, 2020) of other members when it becomes necessary. Swarms of heterogeneous members are better when homogeneous swarms or human may fail to endure (Abosaif and Elrofai, 2020) because such members' working efficiency may be much better (Carrillo-Zapata et al, 2020).…”