我們在本論文中將探討如何利用壓縮式感測技術以解決頻譜感測之問題。壓縮式感測技術可以幫助我們利用較少的觀測點以還原一完整的訊號，因此在近年來受到各方矚目。在現有文獻中，壓縮式感測被用來估測接收訊號的頻譜，接著再計算此一估測頻譜中各個通道中的能量以進行頻譜感測。然而，值得注意的是我們僅需要各個通道的平均能量即可完成頻譜感測。因此，我們將在此本論文中提出一種嶄新的技術，利用壓縮式感測技術為基礎而直接估測出各個通道中的訊號平均能量，以直接完成頻譜感測之目的。和傳統方法相比，本論文所提出方法將具有更低的複雜度。此外，藉由模擬進行驗證，所提之方法比傳統方法可以更準確的完成頻譜感測。

In this investigation, we address the problem of spectrum sensing by using compressive sensing. Compressive sensing (CS) is an attractive technique which is capable of recovering the signal by tilizing a relatively few observation results. In the existing literatures, CS technique is exploited to estimate the spectrum. Subsequently, the spectrum sensing is performed by calculating the average power over each channel from the estimated pectrum. However, it is noted that the information of the average power over each channel is sufficient for spectrum sensing. Therefore, we will develop in this work a novel compressive spectrum sensing method, which estimates directly the power in each channel. The developed method has lower complexity than conventional methods. And moreover, the simulation results are conducted to show that the proposed method perform spectrum sensing more accurately than the conventional method.