在這篇論文，我們考慮一個無線網路規劃問題，並且把焦點放在AP位置的選擇上。我們用網路流量演算法來求解這個問題，把結果拿來和用CPLEX求解這個無線網路規劃問題的IP模型的結果做比較。一開始我們定義一個我們要提供服務地方的地圖，這地圖包含了， (1) 候選存取點 (candidate access points)的位置， (2) 需求點(demand points)的位置， (3) 障礙物的位置。我們也需要知道AP和DP之間的訊號強度值，這個值能從傳遞模型(propagation model)獲得。當我們得到這些資訊之後，一個無線網路規劃問題就被建立完成。這問題會被轉換成整數規劃模型(integer programming)，交由CPLEX求解；我們也把問題轉換成網路流量問題(network-flow-type problem)，透過我們的演算法求解。之後把這兩個結果拿來做比較分析，結果顯示，我們的演算法用極短的時間就能獲得總體平均訊號最大的配置，雖然所需要的存取點數目比用CPLEX求解的結果還要多，但兩者間數目差距在大的例子下相當小。結果顯示我們的演算法和IP模型間的AP數目差距在20個DPs下是51.72\%，50個DPs下是32\%，100個DPs下是11\%，150個DPs下是9.33\%，200個DPs下是5.50\%，300個DPs下是3.67\%，400個DPs下是2.75\%，500個DPs是2\%，600個DPs下是1.67\%。

In this paper, we consider a wireless network planning problem, and focus on the choice of AP (access point) locations. We use a network-flow-type algorithm to solve this problem approximately and compare it with the IP model solved by CPLEX. In the beginning, we define a map where we want to provide wireless service. The map includes (1) the location of candidate access points (AP), (2) the location of demand points (DP), (3) the location of walls and obstacles. We also need to know the value of signal power between the AP and the DP obtained by the propagation model. After we get the above information, a wireless network planning problem can be established. This problem will be transformed into an IP model and solved by CPLEX. It also will be transformed into a network-flow-type problem and solved by our algorithm. Then we make a comparison between them, and the results show that our algorithm can get the result in a quite short time, at the same time its total average signal power is maximal. Although the AP numbers of our algorithm are more than the AP numbers of IP model, the difference between them in a large instance is quite small. We obtain that the gaps of AP numbers between our algorithm and IP model are 51.72\% in 20 DPs, 32\% in 50DPs, 11\% in 100 DPs, 9.33\% in 150 DPs, 5.50\% in 200 DPs, 3.67\% in 300 DPs, 2.75\% in 400 DPs, 2\% in 500 DPs, and 1.67\% in 600 DPs.

Kaveh Pahlavan, Prashant Krishnamurthy, Principles of Wireless Networks, ISBN 986-7727-75-4.
Kotchakorn Maksuriwong, Vara Varavithya, and Nachol Chaiyaratana,Wireless LAN Access Point Placement using a Multi-Objective Genetic Algorithm, Systems, Man and Cybernetics, 2003. IEEE International Conference on Volume 2, 5-8 Oct. 2003 Page(s)：1944-1949.
Rong-Hou Wu, Yang-Han Lee, and Shih-An Chen, Planning System for Indoor Wireless Network, Consumer Electronics, IEEE Transactions on Volume 47, Issue 1, Feb. 2001 Page(s)：73-79. Digital Object Identifier 10.1109/30.920422.
Zhong Ji, Tapan Kumar Sarkar, and Bin-Hong Li, Methods for Optimizing the Location of Base Stations for Indoor Wireless Communications, IEEE Transactions on Antennas and Propagation 50(10)(2002)1481-1483.
Youngseok Lee, Kyoungae Kim, and Yanghee Choi, Optimization of AP Placement and Channel Assignment in Wireless LANs, Local Computer Networks, 2002. Proceedings. LCN 2002. 27th Annual IEEE Conference on 6-8 Nov. 2002 Page(s)：831-836.
Ricardo C. Rodrigues, Geraldo R. Mateus, and Antonio A. F. Loureiro, Optimal Base Station Placement and Fixed Channel Assignment Applied to Wireless Local Area Network Projects, Networks, 1999. (ICON '99) Proceedings. IEEE International Conference on 28 Sept.-1 Oct. 1999 Page(s):186 - 192 Digital Object Identifier 10.1109/ICON.1999.796177.
Can Fang and Chor Ping Low, On the wireless local are network design problem with performance guarantees}, Computer Networks 48 (2005) 856-866.
Byoung-Seong Park, Jong-Gwan Yook, and Han-Kyu Park, The Determination of Base Station Placement and Transmit Power in an Inhomogeneous Traffic Distribution for Radio Network Planning, IEEE Communications Magazine 39 (11) (2001) 98-104.
Abderraouf Bahri, Steven Chamberland, On the wireless local area network design problem with performance guarantees, Computer Networks 48 (2005) 856-866.
Cisco Systems Inc., Cisco Aironet 1200 Series, 2003. Available from:(www.cisco.com).