A Dynamick-Nearest Neighbor Method for WLAN-Based Positioning Systems

Optimal weighted K‐nearest neighbour algorithm for wireless sensor network fingerprint localisation in noisy environment

“…To date, certain algorithms have been proposed to improve the localisation accuracy of the WKNN. These algorithms are mainly of two types: the first type involves a change in the number of selected nearest fingerprints in the WKNN [26–28], and the other involves a redefinition of the weight for the calibration point [29–31]. As the number of nearest fingerprints plays a negligible role in the estimation result of the node position, the improvement effect of the former algorithm on the positioning precision of the WKNN is rather limited.…”

Multi‐channel fingerprint localisation algorithm for wireless sensor network in multipath environment

“…Some algorithms to improve the localisation accuracy of WKNN have been proposed. These algorithms are usually divided into two categories: ones that change the number of NN fingerprints selected by the WKNN [16–18] and ones that redefine the calculation of the weight of the calibration point [19–21]. Since the number of NNs has little effect on the resulting estimate of the node position, any improvement in the positioning precision of WKNN observed with the former category of algorithms is rather limited.…”

Fingerprint localisation algorithm for noisy wireless sensor network based on multi‐objective evolutionary model

“…In the present, some algorithms have been proposed to improve the localisation accuracy of the WKNN. These algorithms mainly are divided into two classes: the one is to change the number of selected NNs in the WKNN [26–28], and the other is to redefine the weight calculation for the calibration point [29–31]. Since the number of NNs plays little role on the estimation result of the node position, the improvement effect of the former algorithm on the positioning precision of the WKNN is rather limited.…”