1. 程式人生 > >【 SIMULATION 】RMSE Comparison of Linear Approaches for TOA - Based Positioning

【 SIMULATION 】RMSE Comparison of Linear Approaches for TOA - Based Positioning

前面的博文對TOA定位的線性方法給予了模擬實驗,這裡將這些RMSE模擬結果放到一起,比較它們的定位效能。

Repeat the test of  the linear approaches; that is, compare the MSPE performance of LLS, WLLS, two - step WLS, and subspace methods for SNR ∈ [ − 10, 60] dB.

 Figure 1 shows the RMSEs of different linear schemes at SNR ∈ [ − 10, 60] dB. It is seen that the two - step WLS scheme, which exploits the constraint of Equation 2.76 , gives the highest localization performance, followed by WLLS, LLS, and subspace estimators.

注:the two - step WLS scheme和WLLS在博文:WLLS Algorithm of TOA - Based Positioning (include the two - step WLS estimator)

響應的模擬可見我的其他博文,這裡就不貼出來了。

至於式2.76的限制為:R = \sqrt{x^2+y^2}

下面是這四種線性演算法的定位均方根誤差模擬圖:

從圖中可以看出,在信噪比為10之前,孰優孰劣,一看便知,至於信噪比為10dB之後,就不太能分清楚了。

下面分別給出這四條曲線,並給出信噪比為20以及30dB時候的RMSE曲線值,僅比較這兩個點:

LLS:

WLLS:

two step WLLS:

subspace:

可見,當信噪比在20dB時候,two_step WLLS演算法的定位精度最高,30dB的時候,也是two_step WLLS定位精度最高。

最後我們不得不說,總體來說,two_step WLLS演算法的定位精度在這四種定位演算法中是最優的。