在本論文中提出一個在無線個人通訊服務網路中實際預測行動通訊使用者移動行為的演算法，此演算法為一個適應性的PPM（Prediction by Partial Matching）演算法，而搭配此演算法的通訊協定及系統架構也詳細的被定義。此演算法能有效地和通訊協定整合，達到準確的預測及有效的通訊資源保留。
未來的個人行動通訊網路需要處理更複雜、更多樣的應用程式，所以對於行動通訊的使用者而言，服務品質（QoS）是非常重要的。為了支援服務品質，我們需要一個可以預估使用者移動行為的演算法，及相對應的通訊協定，來做到適當並有效的資源分配及保留。在這篇論文中，我們從系統的觀點出發，利用部分字串預測演算法（PPM）及預測準確率（PAR）來設計一套有效的適應性使用者行為預測系統。我們介紹多重協定標籤交換（MPLS）、行動IP（Mobile IP）、及其整合。並提出以前有關使用者移動行為預測方法的相關研究。提出了未來PCS網路的系統架構及假設，我們採用階層式的設計及MPLS和行動IP的整合，並提出搭配預測演算法的通訊協定流程，包括註冊（Registration），換手(Handoff)，預測及容錯的通訊協定流程。探討如何利用使用者停留時間及移動行為產生預估所需的序列，並討論如何利用部分字串預測演算法（PPM）及預測準確率（PAR）來設計一套有效的適應性使用者行為預測系統。利用史丹福大學的行動通訊使用者移動行為資料庫來分析實際使用者移動行為及呈現適應性預測演算法在實際個人行動通訊網路中的效能，並比較各個不同參數變化對準確率的影響。我們提出了預測PCS網路中使用者移動性適應性演算法，根據實際的使用者資料模擬顯示，本演算法有相當優異的效能，且能保持網路資源分配的彈性，更能根據使用者需求調整準確率。

A practical user mobility prediction mechanism with its
background operating networking system and protocol flow is introduced in this thesis.

This prediction mechanism is based on the prediction by partial matching (PPM) algorithm

and is able to adjust the resource reservation according to the classification of MN's

mobility behavior and service level agreement.

[1] E. Brewer et al., "A network architecture for heterogeneous mobile computing," IEEE Person. Commun., pp. 8-24, Oct. 1998.
[2] A. Campbell et al., "Design, implementation and evaluation of cellular IP," IEEE Person. Commun., pp.42-49, Aug. 2000.
[3] E. Gustafsson et al., "Mobile IP regional registration," Internet draft, Jul. 2000.
[4] E. Rosen, A. Viswanathan, and R. Callon, "Multiprotocol label switching architecture,"
RFC 3031, Jan. 2001.
[5] A. Ghanwani, B. Jamoussi, and D. Fedyk, "Traffic engineering standards in IP networks
using MPLS," IEEE Commun. Mag., pp. 42-47, Dec. 1999.
[6] R. Zhong, et al. "Integration of mobile IP and MPLS," Internet draft, Jul. 2000.
[7] G. Liu and G. Maguire Jr., "A class of mobile motion prediction algorithms for wireless mobile computing and communications," ACM/Baltzer MONET, 1(2), pp. 113-121,
1996.
[8] T. Liu, P. Bahl, and I. Chlamtac, "Mobility modeling, location tracking, and trajectory prediction in wireless ATM networks," IEEE JSAC, 16(6), pp. 922- 936, Aug. 1998.
[9] J.G. Cleary and I.H. Witten, "Data compression using adaptive coding and partial string
matching," IEEE Trans. Commun., vol. 32, no. 4, pp. 396-402, Apr. 1984.
[10] T. Bell and I.H. Witten, "The zero frequency problem: estimating the probabilities of novel events in adaptive text compression,"\textit{IEEE Trans. Inform. Theory}, vol.37,
no.4, pp. 1085-1094, Jul. 1991.
[11] S.W. Seng and H.S. Kim, "QoS provisioning in cellular networks based onmobility prediction techniques," IEEE Commun. Mag., vol. 41, pp. 86
-92, Jan. 2003.
[12] "Stanford University Mobile Activity TRAces," available at
http://www-db.stanford.edu/sumatra.
[13] C. Perkins, "IP mobility support for IPv4", RFC3344, Aug. 2002.
[14] J. Chan, S. Zhou, and A. Seneviratne, "A QoS adaptive mobility prediction scheme for wireless networks," IEEE GLOBECOM'98, pp. 1414-1419, Nov. 1998.