| 摘要: | 自由空間光通訊（FSO）鏈路需仰賴高指向精度之定位與追蹤（PAT）以維持穩定耦合與可用的通訊品質；然而在 PAT 閉迴路條件下，光束偏離、光型非理想與環境擾動會增加誤差訊號與角度換算的不確定性，進而影響對位效率與可重複性。本研究以自由空間光通訊之定位與追蹤方法為核心，針對在偏離光軸與動態擾動條件下，如何在地面環境中建立可系統性量化與可重複驗證之追蹤性能評估流程，提出並驗證一套同時整合四象限光電二極體（QD）定位量測與光電二極體（PD）通訊調變接收之實驗平台，於不同距離與環境條件量化 QD 誤差訊號特性、角度換算可靠性與對位效能，並以 PD 調變接收結果檢視長距離鏈路中耦合與環境擾動之影響。
模擬與實測結果顯示，環形光斑在 QD 上形成典型 S 形誤差訊號，且光斑尺寸需在定位靈敏度與線性可用範圍之間取捨，此外光型橢圓化與亮暗不均會造成誤差曲線偏移與變形；當距離延伸導致接收光強降低時，量測訊雜比（SNR）下降使該非理想效應的影響相對增大。本研究亦證實 QD 的各個象限背景訊號不一致會影響歸一化位置訊號並降低角度換算可靠性。為提升閉迴路對位效能，本研究在長距離戶外條件下改以實測掃描建立角度轉換係數並導入回授流程，使對位收斂速度與穩定度相較中距離條件明顯提升，
顯示實測校正可有效降低模型失配並改善長距離對位表現。
在 PD 通訊量測方面，本研究以時域波形之有效調變振幅作為接收強度指標，並在一致的量測鏈路下比較不同距離結果。量測顯示中距離相對短距離的強度差異小且未呈現明顯的頻率選擇性；當距離延伸至長距離戶外時，接收振幅變動主要受耦合效率與天氣擾動造成的光束漂移和閃爍所主導，而非單調地距離衰減。
整體而言，本研究建立一套可由室內延伸至戶外之 PAT 地面驗證方法，並指出長距離 FSO 鏈路的可用性與穩定度主要受角度換算模型適配性、耦合狀態與環境擾動所共同限制；因此為提升自由空間光通訊鏈路之可重現性與可靠度，關鍵在於實測導向的模型校正與耦合監測。;Free-space optical communication (FSO) links rely on high pointing accuracy in pointing, acquisition, and tracking (PAT) to maintain stable optical coupling and acceptable communication quality. However, under closed-loop PAT operation, beam misalignment, non-ideal beam profiles, and environmental disturbances
increase the uncertainty of the error signal and the angle conversion, thereby degrading alignment efficiency and repeatability. Focusing on positioning and tracking methods for FSO, this study aims to establish a ground-based performance evaluation procedure that enables systematic quantification and repeatable verification under off-axis conditions and dynamic disturbances. We propose and validate an experimental platform that integrates quadrant photodiode (QD) position sensing and photodiode (PD) reception of modulated communication signals. Across different link distances and environmental conditions, the platform is used to quantify the characteristics of QD error signals, the reliability of angle conversion, and alignment performance, while PD reception results are used to examine the impacts of coupling efficiency environmental disturbances in long-range links.
Both simulation and measurement results show that an annular beam spot produces a typical S-shaped error response on the QD, and that spot size involves an inherent trade-off between positioning sensitivity and the usable linear range. In addition, beam ellipticity and non-uniform irradiance lead to shift and deform of the error curve. As the link distance increases and the received optical power decreases, the signal-to-noise ratio (SNR) degrades, making these non-ideal effects comparatively more pronounced. This study further confirms that inconsistent background signals across the quadrants of the QD can affect the normalized position signal and reduces the reliability of angle conversion.
To improve closed-loop alignment performance, this work adopts, experimentally measured scan data obtained under long-range outdoor conditions to derive the angle-conversion coefficients and incorporates them into the feedback procedure. Compared with the medium-range case, the convergence speed and stability are significantly improved, demonstrating that measurement-driven calibration can effectively reduce model mismatch and enhance long-range alignment performance.
For PD-optical communication measurements, the effective modulation amplitude in the time domain is used as an indicator of received signal strength, and results at different distances are compared using a consistent testing links. Measurements show that the signal difference at medium range relative to short range is small and does not exhibit evident frequency selectivity. When extended to long-range outdoor links, fluctuations in received amplitude are dominated by coupling efficiency and weather-induced beam wander and scintillation, resulting in higher uncertainty rather than a monotonic distance-dependent attenuation trend.
Overall, this study establishes a PAT ground-verification approach extensible from indoor to outdoor environments, and indicates that the usability and stability of long-range FSO links are jointly constrained by the suitability of the angleconversion model, the coupling state, and environmental disturbances. Therefore, improving repeatability and reliability in FSO links hinges on measurement-driven
model calibration and coupling-state monitoring. |
