Hareketli sensörlerle pasif işbirliksiz yayıcı konumlandırma

Localization of radio frequency (RF) emitters with moving sensors is a multidimensional problem which requires trajectory planning, robust estimation systems and efficient localization algorithms etc. This study provides a complete framework to achieve a successful localization system with moving sensors. For trajectory planning, this study introduces a new concept called Fisher continuous information matrix (FCIM) which assumes continuous observations through continuous trajectories. Using FCIMs, this study proves that the best direction of a limited linear trajectory is only a function of the ratio between the total travel length and the initial distance to the emitter. Moreover, for received signal strength (RSS) based localization, the best trajectory is found to move towards the emitter, if the moving sensor is able to reach it. Next, a new powerful geometrical solution called Direction of Exponent Uncertainty (DEU) is proposed for RSS based localization when path loss exponent (PLE) and transmit power are both unknown. DEU is a basis to move towards the emitter without estimating the emitter location. Therefore, DEU is proposed as an efficient route planning tool for moving sensors and an effective localization scheme which attains Cramer Rao Lower Bound (CRLB) with increased computational efficiency. Exploiting moving sensors in emitter localization inevitably results in imprecise sensor positions. Therefore, this study proposes a new search strategy, namely Circular Uncertainty which safely finds the global minimum of Maximum Likelihood estimation (MLE) by searching for the emitter along a special circle. Circular Uncertainty attains CRLB, where other competing methods partly fail.