CAJ | 학술논문

针对无线局域网室内定位系统中，因参考点密集布设而带来的数据采集、更新及定位匹配运算量增加的问题，提出了一种新的基于半监督流形学习的降维判别嵌入定位算法。该算法利用少量已标记数据和部分未标记数据，通过求解目标函数最优化，对高维接收信号进行维数约减，保留最具判别力的定位特征，然后采用确定性定位算法找到定位特征与位置坐标的映射关系。实验结果表明，算法定位精度高于传统的定位算法，降低了离线阶段的数据采集工作量，便于后期数据库的实时更新。
침대무선국역망실내정위계통중，인삼고점밀집포설이대래적수거채집、경신급정위필배운산량증가적문제，제출료일충신적기우반감독류형학습적강유판별감입정위산법。해산법이용소량이표기수거화부분미표기수거，통과구해목표함수최우화，대고유접수신호진행유수약감，보류최구판별력적정위특정，연후채용학정성정위산법조도정위특정여위치좌표적영사관계。실험결과표명，산법정위정도고우전통적정위산법，강저료리선계단적수거채집공작량，편우후기수거고적실시경신。
A new positioning algorithm based on semi-supervised discriminant embedding manifold learning is proposed to resolve problems deriving from dense reference point deployment,such as tremendous time on lo-cation fingerprints collection,calibration and online computation in wireless local area network.The proposed algorithm utilizes a small amount of labeled data and partial unlabeled data to reduce the dimensionality of re-ceived signals.Its strong discriminative features are then retained in the low-dimensional forms through solving the objective function optimization.The reduced signals are taken as inputs to the deterministic positioning algo-rithm and the mapping between localization features and position coordinates is established.The experimental results show that the new algorithm decreases the labor cost to collect fingerprints in the offline stage and cali-brate on time.Compared to the traditional methods,the proposed algorithm shows a considerable accuracy im-provement in the same positioning environment.