دانلود رایگان مقاله استفاده از وسایل نقلیه هوشمند برای برای محلی سازی اشیا ایزوله شده

Abstract

Elementary to the success of the Internet of Things (IoT) is the capability to accurately and efficiently localize its network components, information, and processes. In this paper, we focus on enabling localization of Things that have limited capabilities deployed in isolated areas. Specifically, we explore the scenario where the deployment or the utilization of dedicated anchor nodes becomes costly or practically unfeasible, and where the dependence on multi-hop localization techniques becomes inevitable. We further advocate the use of emerging IoT components such as smart vehicles, capable of self-localization and short-range communication. The proposed scheme thus illustrates the feasibility of a multi-hop wireless localization scheme dependent on mobile anchors (reference points). A key advantage of the proposed scheme is overcoming collinear trajectory (flip-ambiguity) problem, which arises whenever the smart vehicle moves in a straight trajectory. A Kalman Filter (KF) is used to decrease the location error introduced from the multi-hopping during the localization process. Through simulation, we show that the use of our localization scheme with KF reduces errors by 31% compared to localization using anchors from a single direction and 16% compared to a weighted means approach. Moreover, our scheme with KF consistently outperforms the typical range-based DV-Distance scheme with fixed anchors.

6. Conclusion

Our intent in this work is to demonstrate the viability of localizing isolated Things through the use of a combination of multihop wireless localization and mobile anchors (or mobile reference points). We advocate the high feasibility of this proposal in IoT through the emergence of smart vehicles. A novel multi-hop localization scheme was introduced for Sensor Nodes (SNs) with limited capabilities, with the scheme capable of utilizing location information from multiple mobile anchor nodes. The scheme operates in two stages. In the first stage, the scheme estimates the location for each SN from multiple directions using the estimated distance between SNs and the flow direction of the message. In the second stage, we apply Kalman filtering to improve localization accuracy. Simulation results illustrated our scheme’s superiority over other mechanisms that rely on location information from a single direction. As well, and unlike existing schemes, our proposal was shown to accurately perform position estimation regardless of mobile anchor collinearity.