2013-12-012024-05-18https://scholars.lib.ntu.edu.tw/handle/123456789/713937摘要：在傳統的感知無線電(cognitive radio)概念中，在傳送端的頻譜感知(spectrum sensing)功能是不可或缺的：透過對授權頻帶(licensed spectrum)的頻譜感知與量測，智慧型次要傳送端(secondary transmitter)得以知道授權主要用戶(primary user)使用的頻帶，進而在未被使用的「頻帶空隙」(spectrum holes)中進行傳輸，避免干擾授權用戶。然而，頻譜感知中最大的問題在於其延遲性，特別是當授權用戶的狀態劇烈變動時：頻譜感知的結果可能只反應了先前的授權用戶狀態，而當次要用戶開始傳送時，授權用戶可能已經改變其狀態。例如，在原先被認為未被使用的頻帶上，次要用戶決定開始傳輸，而此時主要用戶突然決定開始使用此頻帶，造成干擾；這是由於頻譜感知的延遲，造成次要用戶無法得知即時狀況。 在本計畫中，我們提出了一個新的感知無線電設計概念，使得解決頻譜感知中的延遲性問題得以解決。關鍵的創新是基於新的傳輸方法，可以使次要用戶的近似傳輸速率等同於當次要用戶事先知道授權用戶的狀態時所能達到的傳輸速率，同時不影響授權用戶的傳輸速率。值得注意的是，由於在此方法中沒有使用任何的預測(prediction)，所以即使次要用戶無法從過去的觀察來判斷下一個時間點授權用戶的狀態，此方法依舊可以提供同樣的效能增益。而所需額外付出的，僅是授權用戶接收端些許的更動。根據上述的初步成果，我們計畫審視更複雜的感知無線網路，並設計實際的感知無線電架構，從而解決頻譜感知的延遲所造成的問題。 <br> Abstract: Conventional cognitive radio requires spectrum sensing at the cognitive secondary transmitter to learn whether or not the primary user is occupying a particular band in the licensed spectrum, so that it can communication over “spectrum holes” without user intervention. However, delay is the most critical issue in spectrum sensing, especially when the activity of the primary user is dynamic – the result of spectrum sensing may only reflect the state of the primary user in the past, which may have already changed when the secondary user starts transmitting. For example, the primary user may start to occupy the band which was sensed unused, while the secondary user decides to transmit on that particular band based on the outcome of its spectrum sensing, which does not reflect the instantaneous situation. In this project, we propose a new paradigm of cognitive radio system design that can resolve the issue of spectrum sensing delay. The key innovation is based on a new strategy that can achieve the same asymptotic capacity for the secondary user as if the secondary transmitter learned the activity of the primary user beforehand, without degrading the primary user`s transmission rate. Remarkably, even if the secondary transmitter cannot predict the state of the primary user in the future from the observation in the past, the capacity benefit for the secondary user remains, since there is no prediction in the proposed scheme. The price we have to pay, on the other hand, is a slight modification of the decoding architecture of the primary receiver. Based on the above preliminary findings in a canonical system, we aim to further investigate more complicated cognitive radio networks and design a practical architecture following this new paradigm, so that the fundamental issue of delay in spectrum sensing can be resolved in cognitive radio networks.感知無線電干擾管理Cognitive radiointerference management