18th International Symposium of Electrical Engineering
SIE 2019
Problematic: Local positioning systems are interesting for applications where the Global Positioning System (GPS) is considered unsuitable. Received Signal Strength Indicator (RSSI)-based systems are desirable as they represent an inexpensive way to provide localization, although their limitations. For validation, network simulators are required, being ns-3 one of the most noted because of its characteristics. However, ns-3 includes no classes specifically developed to simulate positioning algorithms. Objectives: The objective of this paper is to report the development of new classes and methods allowing local positioning systems simulation, discussing various methodological/practical issues present. Methodology: We present a literature study on various localization techniques, briefly presenting architectural and modeling concepts on RSSI-based positioning systems. Then, we describe the implementation of software classes for ns-3. Simulation results are also shown for two implemented algorithms, literature versions of Multilateration and Fingerprinting, in various different scenarios. Results and discussion: Experimental results with the implemented classes are satisfactory. The simulation results confirm the known problems of RSSI for providing precise estimations. On the other side, the obtained results are coherent with those reported in other publications obtained by real experimentation. Conclusion: As ns-3 is widely used, the developments reported in this paper could benefit people working on related subjects. Allowing RSSI-based positioning is not trivial. The simulations allow to realize the existing practical problems, such as those related to the variations of RSSI, coverage, and other. Also, simulation is a useful resource to make further contributions on the subject.
Problematic: Local positioning systems are interesting for applications where the Global Positioning System (GPS) is considered unsuitable. Received Signal Strength Indicator (RSSI)-based systems are desirable as they represent an inexpensive way to provide localization, although their limitations. For validation, network simulators are required, being ns-3 one of the most noted because of its characteristics. However, ns-3 includes no classes specifically developed to simulate positioning algorithms. Objectives: The objective of this paper is to report the development of new classes and methods allowing local positioning systems simulation, discussing various methodological/practical issues present. Methodology: We present a literature study on various localization techniques, briefly presenting architectural and modeling concepts on RSSI-based positioning systems. Then, we describe the implementation of software classes for ns-3. Simulation results are also shown for two implemented algorithms, literature versions of Multilateration and Fingerprinting, in various different scenarios. Results and discussion: Experimental results with the implemented classes are satisfactory. The simulation results confirm the known problems of RSSI for providing precise estimations. On the other side, the obtained results are coherent with those reported in other publications obtained by real experimentation. Conclusion: As ns-3 is widely used, the developments reported in this paper could benefit people working on related subjects. Allowing RSSI-based positioning is not trivial. The simulations allow to realize the existing practical problems, such as those related to the variations of RSSI, coverage, and other. Also, simulation is a useful resource to make further contributions on the subject.
About The Speaker
Dr. Cristian Duran Faundez