如何提出有效率的網路選擇機制(network selection)在網路上一直扮演很重要的角色，目前有用數學模式來設計出網路選擇機制讓網路更有效率，例如:遊戲理論(game theory)、多屬性決策方法(MADM)、馬可夫鏈(Markov chain) 和模糊邏輯理論(fuzzy logic)，上述所提到的方法因為設計功能和演算法的不同，會產生不一樣的結果。因此有學者研究建議將這些不同方法所利用融合的方式(fusion)來設計選得網路的機制，但利用組合融合處理這類問題時，要決定何時和如何將這些不同方法融合是一項挑戰。
在此篇論文中，我們使用三種不同的測量網路效率的指標(metrics) 利用組合融合(combinatorial fusion) 方法分別處理 (a) 垂直接續互通 (vertical handoff) 和 (b) 負載平衡 (load balancing) 問題。
我們所提出的方法是將模糊邏輯理論和組合融合整合一起，由實驗結果顯示出我們提出的方法可以讓網路選擇的機制更簡單且更有效率。

Network selection in Heterogeneous Wireless Networks (HWN) aims to select the best network for a variety of communication tasks at any time and anywhere. Due to its importance and necessity, it has been studied using various mathematical models including game theory, multiple attribute decision making (MADM), Markov chain, and fuzzy logic. Since these models have different features and functions to produce different results, it has been suggested to combine these models in order to harness the benefit of individual model. However, it remains a challenge to decide when and how to combine these models or systems.
In this thesis, we propose a new approach to study the network selection problem in HWN using combinatorial fusion. More specifically, we investigate: (a) vertical handoff decision to fuse three metrics: received signal strength (RSS), data rate, and network latency using fuzzy logic and combinatorial fusion; and (b) load balancing using combinatorial fusion on three metrics: RSS, accumulated message queue length, and channel utilization. Experimental results demonstrated that our method can make network selection much simpler and more effective. Our work provides a novel way to fuse these metrics for network selection. It is also the first method to integrate fuzzy logic and combinatorial fusion in solving the network selection problem in heterogeneous wireless network.

[1] A. Barua, L.S. Mudunuri, and O. Kosheleva. Why trapezoidal and triangular membership functions work so well: Towards a theoretical explanation. Journal of Uncertain Systems, 8(3):164–168, 2014.
[2] A. Batallones, K. Sanchez, B. Mott, C. Coffran, and D.-F. Hsu. On the combinationoftwovisualcognitionsystemsusingcombinatorialfusion. Brain Informatic, 2(1):21–32, 2015.
[3] Y. Bejerano, S.-J. Han, and L. Li. Fairness and load balancing in wireless LANs using association control. IEEE/ACM Trans. Networking, 15(3):560–573, 2007.
[4] A. Calhan and C. Ceken. Case study on handoff strategies for wireless overlay networks. Computer Standards & Interfaces, 35(1):170 – 178, 2013.
[5] K.-H. Chi and L.-H. Yen. Load distribution in non-homogeneous wireless local area networks. Wireless Personal Commun., 75(4):2569âĂŞ–2587, 2014.
[6] Y.-S. Chung, D. F. Hsu, C.-Y. Liu, and C.-Y. Tang. Performance evaluation of classifier ensembles in terms of diversity and performance of individual systems. Int. J. Pervasive Computing and Communications, pages 373–403, 2010.
[7] Y.-S. Chung, D. F. Hsu, and C.Y. Tang. On the diversity-performance relationship for majority voting in classifier ensembles. Proceedings of the 7th international conference on Multiple classifier systems, pages 407–420, 2007.
[8] S. Dhakal, M. M. Hayat, J. E. Pezoa, C. Yang, and D. A. Bader. Dynamic load balancing in distributed systems in the presence of delays: A regeneration-theory approach. IEEE Trans. Parallel and Distributed Systems, 18(4):485–497, Apr. 2007.
[9] R. Engle. Anticipating Correlations: A New Paradigm for Risk Management. Princeton University Press, 2009.
[10] Y. Freund and R. Schapire. Boosting: Foundations and Algorithms. The MIT Press, 2012.
[11] H. Gong and J. W. Kim. Dynamic load balancing through association control of mobileusersinwifinetworks. IEEE Trans. Consumer Electronics, 54(2):342–348, 2008.
49
[12] T. Hey, S. Tansley, and K. Tolle. Jim Gray on e-science: A transformed scientific method. Microsoft Research, 2009.
[13] D. F. Hsu, Y.-S. Chung, and B. Kristal. Combinatorial fusion analysis: Methods and practices of combining multiple scoring systems. Advanced Data Mining Techologies in Bioinformatics, pages 32–62, 2006.
[14] D. F Hsu, T. Ito, C. Schweikert, T. Matsuda, and S. Shimojo. Combinatorial fusionanalysisinbraininformatics: gendervariationinfacialattractivenessjudgment. Proceedings of the 2011 International Conference on Brain Informatics, pages 2–20, 2011.
[15] D. F. Hsu, B. Kristal, and C. Schweikert. Rank-score characteristics (RSC) function and cognitive diversity. Proceedings of the 2010 International Conference on Brain Informatics, pages 42–54, 2010.
[16] D. F Hsu and I. Taksa. Comparing rank and score combination methods for data fusion in information retrieval. Inf. Retr., 8(3):449–480, 2005.
[17] L. I. Kuncheva. Combining Pattern Classifiers: Methods and Algorithms. John Wiley & Sons, Inc., 2004.
[18] I.KustiawanandK.-H.Chi. Handoffdecisionusingakalmanfilterandfuzzylogic in heterogeneous wireless networks. IEEE Communications Letters, 19(12):2258– 2261, 2015.
[19] Y. Le, L. Ma, H.-J. Yu, X.-H. Cheng, Y. Cui, M. A. Al-Rodhaan, and A. AlDhelaan. Load balancing access point association schemes for ieee 802.11 wireless networks. Lecture Notes in Computer Science, 6843:271–279, 2011.
[20] M.-T. Lee, L.-T. Lai, and D. Lai. Enhanced algorithm for initial ap selection and roaming. US patent 20040039817 A1, Feb 2004.
[21] K.-C Li, H. Jiang, L. T. Yang, and A. Cuzzocrea. Big Data: Algorithms, Analytics, and Applications. Chapman & Hall/CRC, 1st edition, 2015.
[22] W. Li, S. Wang, Y. Cui, X. Cheng, R. Xin, M. A. Al-Rodhaan, and A. AlDhelaan. AP association for proportional fairness in multi-rate WLANs. IEEE/ACM Trans. Networking, 22(1):191–202, 2014.
[23] K.-L. Lin, C.-Y. Lin, C.-D. Huang, H.-M. Chang, C.-Y. Yang, C.-T. Lin, C.-Y. Tang, and D. F. Hsu. Feature selection and combination criteria for improving accuracyinproteinstructureprediction. IEEE Transactions on Nano Bioscience, 6(2):186 –196, 2007.
[24] C.-Y Liu, C.-Y Tang, and D. F Hsu. Comparing system selection methods for the combinatorial fusion of multiple retrieval systems. Journal of Interconnection Networks, 14(01):135–151, 2013.
50
[25] H.-Z. Liu, Z.-H Wu, and D. F Hsu. Combination of multiple retrieval systems using rank-score function and cognitive diversity. IEEE 26th International Conference on Advanced Information Networking and Applications, pages 167 –174, 2012.
[26] H.-Z. Liu, Z.-H. Wu, X. Zhang, and D. F Hsu. A skeleton pruning algorithm based on information fusion. Pattern Recognition Letters, 34(10):1138 – 1145, 2013.
[27] D. M. Lyons and D. F. Hsu. Combining multiple scoring systems for target tracking using rank-score characteristics. Information Fusion, 10(2):124 – 136, 2009.
[28] C. McMunn-Coffran, E. Paolercio, H.-Z. Liu, R. Tsai, and D. F. Hsu. Joint decision making in visual cognition using combinatorial fusion analysis. In 10th IEEE International Conference on Cognitive Informatics Cognitive Computing, pages 254–261, 2011.
[29] S. Mohanty. A new architecture for 3G and WLAN integration and inter-system handover management. Wirel. Netw., 12(6):733–745, November 2006.
[30] K. B. Ng and P. B. Kantor. Predicting the effectiveness of nave data fusion on the basis of system characteristics. Journal of American Society for Information Science, 51:1177–1189, 2000.
[31] B. V. Quang, R. V. Prasad, and I. Niemegeers. A survey on handoffs lessons for 60 ghz based wireless systems. IEEE Communications Surveys Tutorials, 14(1):64–86, First 2012.
[32] B. Rengarajan and G. De Veciana. Practical adaptive user association policies for wireless systems with dynamic interference. IEEE/ACM Trans. Networking, 19(6):1690–1703, Dec. 2011.
[33] C. Schweikert, S. Brown, Z.-J Tang, P. R. Smith, and D. F Hsu. Combining multiple ChIP-seq peak detection systems using combinatorial fusion. BMC Genomics, 13(8):S12, 2012.
[34] C. Schweikert, S. Shimojo, and D. F. Hsu. Detecting preferences based on eye movement using combinatorial fusion. In IEEE 15th International Conference on Cognitive Informatics Cognitive Computing, pages 336–343, 2016.
[35] A. J. Sharkey. Combining Artificial Neural Nets: Ensemble and Modular MultiNet Systems. Springer-Verlag New York, Inc., Secaucus, NJ, USA, 1st edition, 1999.
[36] A. Singhrova and N. Prakash. Vertical handoff decision algorithm for improved quality of service in heterogeneous wireless networks. IET Communications, 6(2):211–223, 2012.
51
[37] W. Song, W. Zhuang, and Y. Cheng. Load balancing for cellular/WLAN integrated networks. IEEE Network, 21(1):27–33, 2007.
[38] D. J. Spiegelhalter. The future lies in uncertainty. Science, 345(6194):264–265, 2014.
[39] H.D. Vinod, D. F Hsu, and Y. Tian. Combinatorial fusion for improving portfolio performance. Lecture Notes in Statistics, 196:95–105, 2010.
[40] L. Wang and G.-S. Kuo. Mathematical modeling for network selection in heterogeneous wireless networks-a tutorial. IEEE Communications Surveys & Tutorials, 15(1):271–292, 2013.
[41] X.-H. Yan, Y. A. Şekercioğlu, and S. Narayanan. A survey of vertical handover decision algorithms in fourth generation heterogeneous wireless networks. Comput. Netw., 54(11):1848–1863, 2010.
[42] J.-M. Yang, Y.-F Chen, T.-W. Shen, B. Kristal, and D. F. Hsu. Consensus scoring criteria for improving enrichment in virtual screening. Journal of Chemical Information and Modeling, 45(4):1134–1146, 2005.
[43] Z.-H. Zhou. Ensemble Methods: Foundations and Algorithms. Chapman & Hall/CRC, 1st edition, 2012.
52