測試一:

原始數(shù)據(jù):                              [-61,-68,-63,-61,-68,-63,-61,-68,-63,-80,-80,-81]
平滑,歷史權(quán)重0.7:                       [-61,-63,-63,-62,-64,-64,-63,-64,-64,-69,-72,-75]
卡爾曼,q:0.04,r:0.1:                   [-61,-65,-64,-63,-65,-64,-63,-65,-64,-71,-75,-78]
高斯加權(quán),$windowSize:5,$stdDev:1.0:     [-64,-64,-64,-64,-64,-64,-64,-65,-69,-75,-79,-81]

測試二:

原始數(shù)據(jù):                              [-61,-68,-63,-61,-68,-63,-61,-68,-63,-80,-61,-61,-68,-63]
平滑,歷史權(quán)重0.7:                       [-61,-63,-63,-62,-64,-64,-63,-64,-64,-69,-66,-65,-66,-65]
卡爾曼,q:0.04,r:0.1:                   [-61,-65,-64,-63,-65,-64,-63,-65,-64,-71,-67,-64,-66,-65]
高斯加權(quán),$windowSize:5,$stdDev:1.0:    [-64,-64,-64,-64,-64,-64,-64,-65,-68,-70,-66,-64,-64,-65]

總結(jié): 卡爾曼和高斯加權(quán)相對于簡單的平滑都可以減少瞬間波動,但是當rssi正在發(fā)生有效變化的時候,卡爾曼和高斯加權(quán)相對于簡單的平滑可以更加實時的跟隨變化.

卡爾曼和高斯加權(quán)對比,高斯加權(quán)在發(fā)生瞬時波動的時候,下一次的數(shù)據(jù)返回影響較大,而卡爾曼濾波沒有這個問題.

總結(jié)對于藍牙rssi數(shù)據(jù)到處理,卡爾曼濾波更加適合.