Abstract:
The present invention relates to a locating method using an AOA/TDOA combination and a locating apparatus using the same. The locating method comprises (a) setting a location range of a target using an AOA (Angle of Arrival) of each receiver; (b) randomly selecting N locations within the set location range of the target; (c) computing a delay time due to a distance between the receivers at each of the N selected locations and extracting N TDOA (Time Difference of Arrival) pairs at a reference receiver using the computed delay time; (d) comparing the N TDOA pairs with K candidate TDOA pairs to measure error values; (e) selecting M minimum error values from the measure error values and choosing M TDOA pairs having the selected minimum error values; and (f) selecting a location corresponding to the M selected TDOA pairs when repeating steps (b) through (e) reaches a predetermined number of times. According to the present invention, AOA (Angle of Arrival) information is used to minimize pulse ambiguity that may be caused upon using TDOA schemes, thus enhancing locating accuracy. [Reference numerals] (AA) Start; (BB) YES; (CC) NO; (DD) End; (S100) Set target location range using AOA of each receiver; (S200) Select N random locations; (S300) Extract TDOA pair at N random locations; (S400) Compare N TDOA pairs with K candidate TDOA pairs and select M ones; (S500) Select M ones over TDOA having smaller error value; (S600) Repetition count=L; (S700) Reset range from location corresponding to M TDOA pairs; (S800) Select location with respect to TDOA pair having minimum error value
Abstract:
The present invention relates to a radar signal recognition technique and, more specifically, to a method of recognizing a complex pulse repetition interval (PRI) modulation type through a support vector machine (SVM) by generating an exponential moving average signal using time of arrival (TOA) information of collected pulses and extracting a distinguishing factor after generating a signal by layer based on the exponential moving average signal.
Abstract:
The present invention relates to a method for recognizing a user-adaptive cursive script in a smart device based on state transition probability. The method is realized in a manner of allowing a recognizer to train a cursive script inputted by a user, making a chain code, generated from the cursive script, into a state transition probability matrix, providing the recognizer with each of cursive script discriminating factors (P1-P9) extracted from the state transition probability matrix and then outputting the recognition result, terminating the recognition of the cursive script, when the recognition result is normal, and receiving a request for characters intended by the user and allowing the recognizer to retrain the characters, when the recognition result is abnormal. Thus, a writing style of a user can be recognized quickly with a high accuracy, a fast writing style (cursive script) with stronger user-specific font characteristics can be recognized as an exact writing style and, in particular, the poor performance and use of current smart devices can be extended by improving the character recognition of the smart devices.
Abstract:
The present invention relates to a method and an apparatus for synchronizing time of a separate operation system without disconnection by using GNSS common view information and precision time protocol information. The method and the apparatus for dual mode time synchronization of the present invention can perform time synchronization between separate operation systems continuously regardless of GNSS signal jamming and a failure of a GNSS receiver, by performing time synchronization between the separate operation systems by being converted to a precision time protocol information mode using the precision time protocol, if the GNSS signal jamming and the failure of the GNSS receiver occur, as time synchronization between the separate operation systems is performed in a GNSS common view information mode using the GNSS common view information.