電離層形成衛星導航系統（GNSS）最大的誤差及干擾來源。電離層的電漿會對GNSS衛星訊號的傳遞速度產生延遲的效果，導致距離測量誤差。電離層的電漿不規則體，如：電漿泡，亦會導致 GNSS 訊號強度不穩定閃爍。嚴重的訊號閃爍可導致接收器對導航訊號脫鎖，干擾定位功能。GNSS 訊號閃爍的觸發機制、時間變化及空間分佈對衛星導航科技的軍事、民間應用有重大的影響。本計劃分兩大主軸。第一個主軸是運用福衛三號及福衛七號所提供的 S4 閃爍指數觀察訊號閃爍的多年時間、空間分佈與變化。高層大氣科學界近年推測電漿泡可由中性大氣擾動、如：重力波、觸發。我們另外將分析福衛 S4 閃爍指數是否與會產生重力坡的中低層大氣現象有明確的關係，如：聖嬰南方震盪、平流層急速暖化、大氣潮汐及行星波。我們會根據觀測成果進行數值實驗以了解該現象對電漿泡形成條件的影響。本計劃第二個主軸將針對閃爍 GNSS 訊號進行統計學分析，以了解不同區域及觸動機制的閃爍訊號機率密度函數（probability density function，簡稱 PDF）。閃爍訊號 PDF 對於閃爍事件辨識及福衛三號GNSS RO 酬載的 S4 指數運算與資料壓縮法有直接的關連。我們將於台灣及新加坡架設地面 GNSS 接收站，並收集高採樣頻率的訊號強度及訊噪比（SNR）資料，計算個訊號閃爍事件的 PDF。兩地所收及的 PDF 將互相比對，以了解不同地理位置對訊號閃爍 PDF 分布是否有影響。我們另外將結合福衛五號先進電離層探測儀（AIP）所提供的電漿泡觀測，以判斷訊號閃爍產生機制是否對 PDF 有影響。最後，我們會運用我們所收集的 SNR 資料測試福衛三號GNSS RO 酬載的 S4 指數運算與資料壓縮法。為了減少下傳資料量，福衛三號GNSS RO 酬載會先計算、下傳接收訊號 SNR 的標準差及強度平均值，並於地面處理階段假設訊號閃爍PDF為高斯分布。這個方式有別於 S4 指數的標準定義：訊號強度變異量數與平均訊號強度的比例。我們將比對運用上述兩種方式計算的 S4 指數，並了解可能的差異對衛星 S4 觀測的影響。這些成果可協助改良未來的 GNSS RO 酬載及衛星 S4 觀測。 ;The ionosphere is one of the largest sources of error and interference for Global Navigation Satellite Systems (GNSS). Ionospheric electron density slows the propagation of GNSS signals, resulting in ranging errors. Irregularities in ionospheric plasma caused by plasma bubbles can distort GNSS signals to produce rapid fluctuations in signal intensity known as scintillation. In severe cases, scintillation can result in loss of signal lock and positioning ability. The triggers, variation, and distribution of scintillation are therefore of great importance for civil and military applications of GNSS technology. In the first component of this project, we will utilize satellite observations of the S4 amplitude scintillation index from FORMOSAT-3/COSMIC and FORMOSAT-7/COSMIC-2 to examine the variability of scintillation, and understand its relation to possible middle and lower atmospheric events such as the El Nino Southern Oscillation (ENSO), stratospheric sudden warmings (SSWs), atmospheric tidal, and planetary wave events that are known to produce vertically propagating gravity waves that can seed scintillation-causing plasma bubbles, while also affecting the vertical ion drift that is a necessary condition for equatorial plasma bubble formation. Numerical experiments based on these observation results will be conducted to verify their effect on preconditioning the ionosphere for plasma bubble formation.In the second component of this project, we will focus on understanding the statistical distribution of scintillating GNSS signals from different sources and regions. The probability density function (PDF) of scintillation is of interest both for the identification of scintillation, as well as data compression algorithms used by satellite GNSS radio occultation (RO) payloads such as that aboard FORMOSAT-3/COSMIC. High frequency GNSS intensity and Signal to Noise Ratio (SNR) observations will be logged at ground based GNSS receiver stations located in Taiwan and Singapore. The PDFs of scintillated signals at the two locations will be calculated and compared with each other, as well as with co-incident observations of plasma bubbles from the FORMOSAT-5 Advanced Ionosphere Probe (AIP) to determine the differences in distribution with geographic location and source. The collected SNR data will further be used to test the FORMOSAT-3/COSMIC S4 data compression algorithm, which assumes a Gaussian scintillation distribution to avoid having to downlink the larger raw data. S4 indices calculated using the COSMIC algorithm will be compared to that calculated using raw data, and discrepancies quantified. This will allow for the improvement of S4 data compression algorithms aboard future GNSS RO payloads.
