Performance Evaluation of Nonlinear Optimization Methods for TOA Localization Techniques
Само за регистроване кориснике
2016
Конференцијски прилог (Објављена верзија)
,
OTEH 2016
Метаподаци
Приказ свих података о документуАпстракт
Source localization based on the time of arrival (TOA) measurements is very important in military and civil applications such as Wireless Sensor Networks (WSNs), sonar, radar, security systems, health monitoring, etc. Some of the localization measurements are corrupted by errors which always exist no matter which localization techniques are used. Therefore, TOA source localization model, in the presence of additive noise, can be formulated as an optimization model with the sum of squared residuals as the objective function. This paper presents the application of three nonlinear optimization methods: the Steepest Descent, the Newton-Raphson and the Gauss-Newton methods and their performance is compared with each other in terms of localization accuracy. Numerical simulation results illustrate the performance comparison of these different proposed nonlinear optimization methods with different initial values and signal-to-noise ratio (SNR). The corresponding Cramer–Rao Lower Bound (CRLB) o...n the localization errors is derived, which gives a lower bound on the variance of any unbiased estimator. Finally, the simulation results analysis of the performance of the proposed gradient-based optimization methods are evaluated and compared with the CRLB and the closed-form LLS method.
Кључне речи:
Wireless Seknsor Networks / Localization / Optimization / Time of Arrival / Signal-to-noise ratioИзвор:
Proceedings of the 7th International Scientific Conference on Defensive Technologies, OTEH 2016, 2016, 466-471Издавач:
- Belgrade : Military Technical Institute
Колекције
Институција/група
Mašinski fakultetTY - CONF AU - Rosić Vitas, Maja AU - Simić, Mirjana AU - Pejović, Predrag V. PY - 2016 UR - https://machinery.mas.bg.ac.rs/handle/123456789/6034 AB - Source localization based on the time of arrival (TOA) measurements is very important in military and civil applications such as Wireless Sensor Networks (WSNs), sonar, radar, security systems, health monitoring, etc. Some of the localization measurements are corrupted by errors which always exist no matter which localization techniques are used. Therefore, TOA source localization model, in the presence of additive noise, can be formulated as an optimization model with the sum of squared residuals as the objective function. This paper presents the application of three nonlinear optimization methods: the Steepest Descent, the Newton-Raphson and the Gauss-Newton methods and their performance is compared with each other in terms of localization accuracy. Numerical simulation results illustrate the performance comparison of these different proposed nonlinear optimization methods with different initial values and signal-to-noise ratio (SNR). The corresponding Cramer–Rao Lower Bound (CRLB) on the localization errors is derived, which gives a lower bound on the variance of any unbiased estimator. Finally, the simulation results analysis of the performance of the proposed gradient-based optimization methods are evaluated and compared with the CRLB and the closed-form LLS method. PB - Belgrade : Military Technical Institute C3 - Proceedings of the 7th International Scientific Conference on Defensive Technologies, OTEH 2016 T1 - Performance Evaluation of Nonlinear Optimization Methods for TOA Localization Techniques EP - 471 SP - 466 UR - https://hdl.handle.net/21.15107/rcub_machinery_6034 ER -
@conference{ author = "Rosić Vitas, Maja and Simić, Mirjana and Pejović, Predrag V.", year = "2016", abstract = "Source localization based on the time of arrival (TOA) measurements is very important in military and civil applications such as Wireless Sensor Networks (WSNs), sonar, radar, security systems, health monitoring, etc. Some of the localization measurements are corrupted by errors which always exist no matter which localization techniques are used. Therefore, TOA source localization model, in the presence of additive noise, can be formulated as an optimization model with the sum of squared residuals as the objective function. This paper presents the application of three nonlinear optimization methods: the Steepest Descent, the Newton-Raphson and the Gauss-Newton methods and their performance is compared with each other in terms of localization accuracy. Numerical simulation results illustrate the performance comparison of these different proposed nonlinear optimization methods with different initial values and signal-to-noise ratio (SNR). The corresponding Cramer–Rao Lower Bound (CRLB) on the localization errors is derived, which gives a lower bound on the variance of any unbiased estimator. Finally, the simulation results analysis of the performance of the proposed gradient-based optimization methods are evaluated and compared with the CRLB and the closed-form LLS method.", publisher = "Belgrade : Military Technical Institute", journal = "Proceedings of the 7th International Scientific Conference on Defensive Technologies, OTEH 2016", title = "Performance Evaluation of Nonlinear Optimization Methods for TOA Localization Techniques", pages = "471-466", url = "https://hdl.handle.net/21.15107/rcub_machinery_6034" }
Rosić Vitas, M., Simić, M.,& Pejović, P. V.. (2016). Performance Evaluation of Nonlinear Optimization Methods for TOA Localization Techniques. in Proceedings of the 7th International Scientific Conference on Defensive Technologies, OTEH 2016 Belgrade : Military Technical Institute., 466-471. https://hdl.handle.net/21.15107/rcub_machinery_6034
Rosić Vitas M, Simić M, Pejović PV. Performance Evaluation of Nonlinear Optimization Methods for TOA Localization Techniques. in Proceedings of the 7th International Scientific Conference on Defensive Technologies, OTEH 2016. 2016;:466-471. https://hdl.handle.net/21.15107/rcub_machinery_6034 .
Rosić Vitas, Maja, Simić, Mirjana, Pejović, Predrag V., "Performance Evaluation of Nonlinear Optimization Methods for TOA Localization Techniques" in Proceedings of the 7th International Scientific Conference on Defensive Technologies, OTEH 2016 (2016):466-471, https://hdl.handle.net/21.15107/rcub_machinery_6034 .