盲蔽式非同調頻譜偵測之研究

Abstract

傳統的頻譜使用方式，由政府代理機構分配給已註冊的使用者，該使用者只能使用所分配到的特定頻帶。然而，在頻譜需求日益成長的今天，新的感知網路概念被提出來︰在不干擾已註冊的使用者之下將閒置中的頻譜提供給其他使用者，以提高頻譜的使用效率。 頻譜偵測為感知網路系統中最基本的功能方塊，在本文中，我們提出二種能適用於各種不同感知網路運用的頻譜偵測演算法，利用比值的概念，在訊號能量、型態、通道頻率響應、甚至雜訊能量皆未知的情況下都能成功達成偵測。根據模擬的結果，在IEEE 802.22的規格標準下，我們所提出的演算法，可以確實避免雜訊能量的不確定性對偵測結果所帶來的影響。

Traditional, the frequency spectrum is licensed to users by government agencies in a fixed manner where the licensee has exclusive right to access the allocated band. However, with increasing demand for the spectrum and scarcity of vacant bands, a spectrum policy reform seems inevitable. Therefore, recent measurements suggest the probability of sharing spectrum subject to interference-protection constrains. In order to enable access to an unused unlicensed spectrum, a secondary user has to monitor licensed bands and opportunistically transmit whenever no primary user signal is detected. Spectrum sensing would be the basic functional block for a cognitive radio system. We want to propose a robust spectrum sensing algorithm which could be used in various applications even without any information from the sensed channel and signal of primary user. The traditional methods to solve this problem is to use the energy detector. However, an energy detector would suffer form noise uncertainty seriously. At this report, we propose the FBCME algorithm for multi-tone detection which means using received samples of multiple frequency bands to make decision at at one time, and the modified ratio detector for single-tone detection which means making detection with just one frequency band. The feature of ratio tests is used to avoid noise uncertainty in both proposed algorithms. Numerical analysis and simulation results do verify the proposed algorithms both prevent the detection performance from noise uncertainty successfully.