在現代社會中，因為各種網路技術快速的演進，各種新的技術不停的研發，在現在的頻譜管理策略下，造成可以使用的頻譜資源越趨匱乏，但是頻譜的管理策略卻導致有些頻譜的使用率很低，而近期興起的感知無線電（Cognitive radio）將是解決無線頻帶缺乏（spectrum scarcity problem）的一項關鍵技術，其允許我們在不干擾合法使用者的情況下使用已被註冊的頻譜，增加頻譜的使用率。在感知無線電（Cognitive radio）網路中，因為合法使用者會在任意的時間點下出現，所以可能會造成網路中的節點隨時必須中斷連線，這樣的狀況會使得網路的效能不佳，但只要各個網路中的節點，可以使用的頻道資源越多，不管任何合法使用者出現，都可以隨時轉換到另一個可以使用的頻道上，整體網路就可以繼續傳輸資料，效能就不會因為合法使用者出現而低落。本論文的主要目標是希望在跳頻的感知網路系統中，讓相遇的多樣性的特性可以在一個跳頻序列的長度下達到，以避免突然出現的合法使用者造成整個網路的通訊中斷，在我們設計的跳頻序列中，他可以使用到整個網路頻道總數的一半，且不需要交換相關的資訊，來達到相遇多樣性的需求，在模擬中，我們比較了鄰居節點數，合法使用者的數量以及可以使用的頻道數，我們都有較佳的整體網路效能。In recent years, there are more advanced wireless communication techniques be published. However, the policy of spectrum management make that available spectrum resource is rare than before. Besides, the utilization ratio of some spectrum resources is lesser. In Cognitive radio networks (CRNs), this technique is addressing spectrum scarcity problem. That enables sharing the licensed spectrum without affecting the user of licensed. The licensed user would appear any time. When they appear, the whole network nodes which are using the channel of licensed users will stop transmission. This will cause the throughput of whole network is downgrade. It is crucial for users in CRNs to communicate with each user via rendezvous process, establish links, and achieve rendezvous diversity. In this thesis, we focus on the design of algorithm for blind rendezvous, and the rendezvous diversity is about half channels. This will reduce the influence that is the long time block problem by PUs. The basic idea is to let each frame in a sequence have different length. Our algorithm achieve better work under the impact factors which is includes numbers of PU, numbers of channels, and numbers of neighbor compared to previous work.
