Adaptive filtration of parameters of the movement UAV according to sensor networks based on measurements of the received signal strength


Abstract

Introduction. In modern conditions increasingly important begin to play unmanned aerial vehicles (UAVs), which give rise to a new class of threats. This leads to the need to develop security systems that solve tasks of detection, location and motion parameters of the UAV. At the radiation of the UAV of signals, his location can be defined by wireless sensor networks using the method of RSS (received-signal strength). Changing the type of UAV motion occurs at random times. At intervals of hovering and motion of the UAV without a maneuver it is possible to significantly improve the accuracy of estimation of its coordinates. Thus, in practice, it is often of interest to determine the types of UAV motion.
Statement of the problem. UAV movement with different types of maneuver in a rectangular coordinate system is described by a stochastic dynamical system with random structure in discrete time. For definition of location of the UAV on the plane the wireless sensor network has to consist of three or more sensors. When using the RSS method, the model of direct distribution of a signal which considers only his attenuation is used. Required to synthesize an adaptive algorithm of a filtration of parameters of the movement UAV according to the sensor network.
The main part. The optimum algorithm of the adaptive filtering is recurrent and describes evolution of posteriori probability density of the expanded mixed Markov process including a continuously valued vector of parameters of movement of the UAV and the discrete valued variable of switching describing type of its movement. The optimum device realizing an algorithm is multichannel with number of channels M and belongs to the class of devices with feedback between channels. Existence of feedback between channels is caused by Markov property of a discrete component.
In obtained by linearization of the equation of measurements of the quasi-optimal algorithm of adaptive filter are calculated first and second moments aposteriori conditional distributions of the vector of motion parameters of the UAV and it allows to keep the representation of the a posteriori probability density of the continuous component as a sum of M Gaussian densities of probabilities. It implements a parallel procedure perform calculations when entering measurements from sensors of a sensor network. The quasioptimum device realizing an algorithm also is multichannel with number of channels M and generally keeps the structure and feedback inherent in the optimum device.
Analysis of the effectiveness of the algorithm. Analysis of the effectiveness of the developed algorithm for estimating the parameters of motion of the UAV with the discovery of the maneuver was conducted using the statistical modeling. The sensor network is composed of eight sensors. For descriptive reasons works of an algorithm the test trajectory of the movement UAV has been created. A comparison of the accuracy characteristics of the considered algorithms with the lower bound of Rao-Cramer is carried out.
For the considered model example, the use of trajectory filtering allows to reduce the MSD error of the positioning of the UAV compared with the MSD error of the positioning method RSS in 2 – 4 times. Compared with Kalman filter based on the model, motion of the UAV to maneuver, developed an adaptive algorithm allows to improve the location accuracy in areas hovering and motion without maneuver more than 2-3 times to avoid systematic errors estimates. At the same time the adaptive filter allows to recognize a freeze and nearly uniform motion of a UAV with a probability close to one.
Conclusions. On the basis of the mixed Markov processes in discrete time optimum and quasioptimum adaptive algorithms a filtration of parameters of the movement UAV according to sensor network on the basis of measurement of power of the accepted signal are synthesized. Realizing their devices, are multichannel and belongs to the class of devices with feedback between channels. At the same time in them the parallel procedure of performance of calculations at receipt of measurements from sensors network is realized. The analysis of a quasioptimum algorithm is made by means of statistical modeling on the computer.

How to cite:

 
GOST 7.1:2006 Harvard
 
Товкач И. О. Адаптивная фильтрация параметров движения БПЛА по данным сенсорной сети на основе измерения мощности принимаемого сигнала / И.О. Товкач, С.Я. Жук // Вестник НТУУ «КПИ». Серия Радиотехника. Радиоаппаратостроение. – 2017. – № 69. – с. 41-48. Tovkach, I. O., Zhuk, S. Ya. (2017) Adaptive filtration of parameters of the movement UAV according to sensor networks based on measurements of the received signal strength. Visn. NTUU KPI, Ser. Radioteh. radioaparatobuduv., no. 69, pp. 41-48. (in Russian)
 

Full Text:


References


Список источников

Nonami K. Autonomous Flying Robots Unmanned Aerial Vehicles and Micro Aerial Vehiclesх / K. Nonami, F. Kendoul, S. Suzuki, W. Wang, D. Nakazawa. - N. Y. : Springer, 2010. - 329 p.

Wallace R. J. Examining Unmanned Aerial System Thr eats & Defenses: A Conceptual Analysis / R.J. Wallace, J.M. Loffi // International Journal of Aviation, Aeronautics, and Aerospace. - 2015. - Vol. 2, Is. 4. - 33 p.

Защита от дронов – небесный рубеж вашей безопасности [Електронний ресурс]. - Режим доступу: http://dronebouncer.com

Tarrno P. An RSS Localization Method Based on Parametric Channel Models / P. Tarrno, A. M. Bernardos, J. R. Casar // IEEE International Conference on Sensor Technologies and Applications. - 2007. - pp. 265-270.

Mao G. Wireless Sensor Network Localization Techniques / G. Mao, B. Fidan, B.D.O. Anderson // Journal Computer Networks: The International Journal of Computer and Telecommunications Networking archive. - 2007. - Vol. 51, Iss. 10. - pp. 2529-2553.

Black T. J. Position Estimation and Tracking of an Autonomous Mobile Sensor using Received Signal Strength / T. J. Black, P. N. Pathirana, S. Nahavandi // IEEE Intelligent Sensors, Sensor Networks and Information Processing. - 2008. - pp. 19-24.

Zhang L. Tracking Mobile Robot in Indoor Wireless Sensor Networks / L. Zhang, C. Lim, Y. Chen, H. R. Karimi // Mathematical Problems in Engineering. - 2014. - Vol. 2014. - 8 p.

Masieroa A. UAV positioning and collision avoidance based on RSS measurements / A. Masieroa, F. Fissorea, A. Guarnieria, F. Pirottia, A. Vettorea // The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. - 2015. - Vol. XL-1/W4. - pp. 219-225.

Жук С. Я. Методы оптимизации дискретных динамических систем со случайной структурой / С.Я. Жук. - НТУУ "КПИ", 2008. - 232 с. - Режим доступа : http://rtps.kpi.ua/ru/node/171

Жук С. Я. Синтез цифровых обнаружителей-измерителей смешанных марковских процессов / С.Я. Жук // Известия вузов СССР. Радиоэлектроника. - 1989. - Т. 32, № 11. - с. 31-37.

Евланов П.А. Комплексирование измерителей с отказами / П.А. Евланов, С.Я. Жук // Известия вузов СССР. Радиоэлектроника. - 1990. - Т. 33, №3. - с. 15-21.

Товкач И. О. Рекуррентный алгоритм пассивной локации в сенсорной сети на основе измерения мощности принимаемого сигнала / И.О. Товкач, С.Я. Жук // Вестник НТУУ "КПИ". Серия Радиотехника. Радиоаппаратостроение. - 2016. - № 66. - с. 46-55.

References

Nonami K., Kendoul F., Suzuki S., Wang W. and Nakazawa D. (2010) Autonomous Flying Robots Unmanned Aerial Vehicles and Micro Aerial Vehicles, N. Y., Springer, 329 p. DOI: 10.1007/978-4-431-53856-1_8

Wallace R. J. and Loffi J. M. (2015) Examining Unmanned Aerial System Threats & Defenses: A Conceptual Analysis, International Journal of Aviation, Aeronautics, and Aerospace, Vol. 2, Iss. 4, 33 p. DOI: 10.15394/ijaaa.2015.1084

Protection from drones - the heavenly boundary of your safety. Available at: http://dronebouncer.com.

Tarrno P., Bernardos A. M. and Casar J. R. (2007) An RSS Localization Method Based on Parametric Channel Models, International Conference on Sensor Technologies and Applications, SensorComm 2007, pp. 265-270. DOI: https://doi.org/10.1109/SENSORCOMM.2007.4394932

Mao G., Fidan B. and Anderson B. D.O. (2007) Wireless Sensor Network Localization Techniques, Computer Networks, Vol. 51, Is. 10, pp. 2529-2553. DOI: 10.1016/j.comnet.2006.11.018

Black T. J., Pathirana P. N. and Nahavandi S. (2008) Position Estimation and Tracking of an Autonomous Mobile Sensor using Received Signal Strength, IEEE Intelligent Sensors, Sensor Networks and Information Processing, pp. 19-24. DOI: 10.1109/issnip.2008.4761956

Zhang L., Lim C., Chen Y. and Karimi H. R. (2014) Tracking Mobile Robot in Indoor Wireless Sensor Networks, Mathematical Problems in Engineering, Vol. 2014, Article ID 837050, 8 p. DOI: 10.1155/2014/837050

Masieroa A., Fissorea F., Guarnieria A., Pirottia F. and Vettorea A. (2015) UAV positioning and collision avoidance based on RSS measurements, The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XL-1/W4, pp. 219-225. DOI: 10.5194/isprsarchives-XL-1-W4-219-2015

Zhuk S. Ya. (2008) Metody optimizatsii diskretnykh dinamicheskikh sistem so sluchainoi strukturoi [Methods for optimization of discrete dynamical systems with random structure]. NTUU ``KPI'', 232 p.

Zhuk S.Ya. (1989) Synthesis of digital detector-meters for mixed Markovian processes, Izvestiya VUZ. Radioelektronika, Vol. 32, No. 11, pp. 31-37.

Yevlanov P.A. and Zhuk S.Ya. (1990) Kompleksirovanie izmeritelei s otkazami [Metering with Malfunctions], Izvestiya VUZ. Radioelektronika, Vol. 33, No. 3, pp. 15-21.

Tovkach I. O. and Zhuk S. Ya. (2016) Recursive algorithm of the passive location in sensor networks based on measurement of the received signal strength, Visn. NTUU KPI, Ser. Radioteh. radioaparatobuduv., no. 66, pp. 46-55.






Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.