| 要約: | Focused on the issues that large positioning errors produced by the inconsistency of received signal strength (RSS)sample population distributions under indoor wireless local area network (WLAN),an indoor positioning algorithm based on skewness-kurtosis testing was proposed.By using the testing method of skewness and kurtosis,whether the RSS samples come from the normal population or not was checked.The distribution functions of the samples accepting null hypothesis were approximated with normal distribution,and the probability density functions of the samples refusing null hypothesis were estimated by kernel function.Experimental results show that the proposed algorithm leads to a 15 percent improvement over the previous methods.Moreover,the proposed algorithm can significantly reduce the workload of the off-line phase at the same positioning accuracy.
|
|---|
