本文選題：井下跟蹤　切入點：無線傳感器網(wǎng)絡　出處：《計算機應用研究》2017年05期

【摘要】：井下環(huán)境復雜多變,射頻信號易受到陰影效應、多徑衰落等因素的影響。采用傳統(tǒng)的粒子濾波跟蹤方法誤差較大,研究了一種基于改進粒子濾波的井下跟蹤算法。初始化階段利用第一次指紋匹配算法的定位結(jié)果來設計初始化概率分布函數(shù);采用核函數(shù)法與指紋匹配技術(shù)相結(jié)合的算法,在采樣數(shù)據(jù)中搜索與目標節(jié)點指紋特征相匹配的位置并加權(quán)得到位置坐標作為跟蹤中的觀測值;最后利用粒子濾波將觀測值與目標運動狀態(tài)相融合以跟蹤目標運動軌跡。實驗結(jié)果表明,粒子濾波算法較優(yōu)化卡爾曼濾波算法更適用于井下跟蹤;改進的算法有效增強了跟蹤系統(tǒng)的可靠性,提高了跟蹤精度,滿足了井下的跟蹤要求。
[Abstract]:The underground environment is complex and changeable, and the radio frequency signal is easily affected by the shadow effect and multipath fading. In this paper, an improved particle filter based downhole tracking algorithm is studied. In the initialization stage, the initial probability distribution function is designed by using the location result of the first fingerprint matching algorithm, and the kernel function method is combined with the fingerprint matching technique. The position matching the fingerprint feature of the target node is searched in the sampled data and the position coordinate is obtained as the observation value in the tracking. Finally, the particle filter is used to track the moving trajectory of the target by combining the observed values with the moving state of the target. The experimental results show that the particle filter algorithm is more suitable for underground tracking than the optimized Kalman filter algorithm. The improved algorithm can effectively enhance the reliability of the tracking system, improve the tracking accuracy and meet the requirements of underground tracking.
【作者單位】： 內(nèi)蒙古科技大學信息工程學院;
【基金】：內(nèi)蒙古自治區(qū)科技計劃資助項目(201502013-1) 內(nèi)蒙古自治區(qū)自然基金資助項目(2015MS0623)
【分類號】：TD76;TN713

【相似文獻】

相關(guān)期刊論文 前1條

1 高廣飛;姚軍;;基于Hadoop云平臺的礦井指紋定位算法研究[J];金屬礦山;2013年12期

，

本文編號：1656324