Yapay Sinir Ağları İle Kablosuz Yerel Alan Ağlarında Veri Trafiği Optimizasyonu

Gelişen teknoloji ile birlikte kablosuz yerel alan ağlarında (Wireless Local Area Network-WLAN) hizmet kalitesi (Quality of Service-QoS) istekleri de artmaktadır. Yapılan birçok çalışmada WLAN’larda hizmet kalitesini artırmak için çeşitli metotlar ve farklı algoritmalar kullanılmıştır. WLAN’larda ortama erişim kontrol (Medium Access control- MAC) katmanında hizmet kalitesini etkileyen RTS Eşik Değeri (RTSED), Parçalama Eşik Değeri (PED) ve Arabellek Boyutu (AB) hizmet kalitesini direk etkilemektedir. WLAN’larda hizmet kalitesini artırmak için kullanılan kanal kullanım durumu, alınan veri trafiği ve gönderilen veri trafiği ağdaki performansını etkileyen önemli parametrelerdir. Bu çalışmada WLAN’larda Yapay Sinir Ağları (Artificial Neural Networks-ANN) kullanılarak RTSED, PED ve AB parametreleri optimize edilip, alınan veri trafiği ve gönderilen veri trafiğinin ideal değerleri elde edilmiştir. Riverbed Modeler simülasyon aracı kullanılarak 11 düğümlü ve 27 farklı giriş değerleri seçilerek kanal kullanım durumu elde edilmiştir. Alınan veri trafiği ile gönderilen veri trafiği sonuç değerleri ANN Modellemesi ile WLAN’larda performansın tahminini sağlanmıştır.

Anahtar Kelimeler:

Kablosuz yerel alan ağları, yapay sinir ağları, alınan veri trafiği

Data Traffic Optimization in Wireless Local Area Networks with Artificial Neural Networks

In recent years, quality of service (QoS) has been rapidly increasing in wireless local area networks (WLANs) with the increasing technology. In many studies, various methods and different algorithms are used to improve the quality of service in WLANs. RTS Threshold Value (RTSTV), Fragmentation Threshold Value (FTV) and Buffer Size (BS) are affect service quality directly at MAC (Medium Access Control) layer in WLAN. Channel utilization, data traffic received and data traffic sent parameters are important improve quality of service in WLANs. In this study, RTSED, PED and AB parameters were optimized by using Artificial Neural Networks (ANN) in WLAN and ideal values of received data traffic and received data traffic were obtained. Using the Riverbed Modeler simulation tool, 11 nodes and 27 different input values were selected to obtain channel utilization status. With the ANN Modeling of the results of the data traffic sent and data traffic received provides the estimation of the performance in the WLAN. It is observed that the average squared error value of 1000 epoch training result is less than 10-6, and that the test and estimation abilities are larger than 10-6. According to this value, it means that the improved YSA model can not memorize, it can establish a relation between input and output data. Thus, it is proved that the best learning values are obtained with these results obtained by the developed model.

Keywords:

Wireless local area network artificial neural networks, data traffic received, data traffic sent,

___

Referans1 Doğancı Y. U., “802.11 Standartlarını Kullanarak Pozisyon Tespiti”, Yüksek Lisans Tezi, TOBB Ekonomi ve Teknoloji Üniversitesi Fen Bilimleri Ensitiüsü, (2008).
Referans2 Stallings, W., “Wireless Communications and Networks”, 2nd ed., Prentice Hall, Englewood Cliffs, NJ, (2005).
Referans3 Singh H., Singh T., Kaur M., “Improving the quality of Service of EDCF over DCF For Real Time Applications Using Probability Algorithm”, International Journal of Advanced Research in Computer and Communication Engineering, Cilt 3:4, 6330- 6333, (2014).
Referans4 Dalvi A., Svamy P., Meshram B.B., “DCF Improvement for Satisfactory Throughput of 802.11 WLAN”, International Journal on Computer Science and Engineering, 3:7, 2862-2868, (2011).
Referans5 Borsuk B, Koçak C., “Kablosuz Ağlarda Gizli Düğüm Probleminde IEEE 802.11 MAC Katmanı RTS/CTS Mekanizmasının Çoklu Ortam Uygulamalarında Performansa Etkisi”, Bilişim Teknolojileri Dergisi, 9:2, 187-195, (2016).
Referans6 Karakurt H. B., Kocak C., “On wireless network PCF, DCF and EDCF with fragmentation threshold”, XVII. Academic Informatics Conference, Eskisehir/Turkey, (2015).
Referans7 Biçen M., Çalhan A., Yücedağ İ., “Kablosuz Heterojen Algılayıcı Ağlarda Bulanık Mantık Tabanlı Ağ Geçidi Seçimi”, Düzce Üniversitesi Bilim ve Teknoloji Dergisi, 4: 655-660, (2016).
Referans8 Inderjeet K., Manju B., Harpreet B., “Performance Evaluation of Wlan by Varying PCF, DCF and Enhanced DCF Slots To Improve Quality of Service.”, IOSR Journal of Computer Engineering (IOSRJCE), 2:5, 29-33, (2012).
Referans9 Hussein H., Elsayed H.A., “A Predictive Model For Mac Protocol (P-Mac) Of Cognitive Radio Network Using Exponential Smoothing Model.”, MIC-WCMC 2012 MOSHARAKA International Conference, Palma de Mallorca, Spain, (2012).
Referans10 Kaplan Y., Saray U., Emeksiz C., Yeşilnacar Y.O., Önal S., Karaca V., “Yapay Sinir Ağı Geri Yayınım Algoritması kullanılarak Rüzgar Hızı Tahmini.”, IX. Clean Energy Symposium, UTES’13, (2013).
Referans11 Erdem S., “YSA ve GA Temelli Yeni Bir Algoritma İle Doğrusal Olmayan Optimizasyon.”, İşletme Fakültesi Dergisi, 8:2, 219-231, (2007).
Referans12 Ulas H.B., Ozkan M.T., Malkoc Y.,”Vibration Prediction İn Drilling Processes With HSS And Carbide Drill Bit By Means Of Artificial Neural Networks.”, Neural Computing And Applications, First Online, (2018).
Referans13 Yigit T., Ersoy M., “Testing and Design of Indoor WLAN Using Artificial Intelligence Techniques.”, Elektronika Ir Elektrotechnika. 20:6, 154-157, (2014).
Referans14 Karakurt H.B. “Kablosuz Yerel Alan Ağlarında Kontrol Fonksiyonları İçin Parçalama Eşik Değeri İle Performans Geliştirme”, Yüksek Lisans Tezi, Gazi Üniversitesi, Fen Bilimleri Enstitüsü, Bilgisayar Mühendisliği Anabilim Dalı, Ankara, (2015).
Referans15 Isizoh A. N., Anazia A.E., Okide S.O., Okwaraoka C.A.P.,”Effects Of Different Fragmentation Thresholds On Data Dropped And Retransmission Attempts In A Wireless Local Area Network”, Onyeyili T.I./ International Journal of Engineering Research and Applications (IJERA), 3:2, 76-79, (2013).
Referans16 Zhai H., Kwon Y., Fang Y., “Performance analysis of IEEE 802.11 MAC protocols in wireless LANs”, Wireless Communications And Mobile Computing, 4: 917-931, (2004).
Referans17 Li T.,“Improving Performance for CSMA/CA Based Wireless Networks”, Doctoral Thesis, Hamilton Institute National University of Ireland, Maynooth Maynooth, Co. Kildare, Ireland, (2007).
Referans18 Malone D., Clifford P., Leith D.J., “On Buffer Sizing for Voice in 802.11 WLANs”, IEEE Communications Letters, 10:10, 701-703, (2006).
Referans19 Sun, H., Chen, X., Shi, Q., Hong, M., Fu, X., Sidiropoulos, N. D., “Learning to optimize: Training deep neural networks for wireless resource management.”, In Signal Processing Advances in Wireless Communications (SPAWC), 2017 IEEE 18th International Workshop on, 1-6, (2017).
Referans20 C. Jiang, H. Zhang, Y. Ren, Z. Han, K. C. Chen, and L. Hanzo., “Machine learning paradigms for next-generation wireless networks,” IEEE Wireless Communications, 24:2, 98–105, (2017).
Referans21 Kato, N., Fadlullah, Z. M., Mao, B., Tang, F., Akashi, O., Inoue, T., Mizutani, K., “The deep learning vision for heterogeneous network traffic control: Proposal, challenges, and future perspective.”, IEEE wireless communications, 24:3, 146-153, (2017).
Referans22 S. Bi, R. Zhang, Z. Ding, and S. Cui., “Wireless communications in the era of big data.”, IEEE Communications Magazine, 53:10, 190–199, (2015).
Referans23 T. J. O’Shea and J. Hoydis., “An introduction to machine learning communications systems,” available online arXiv:1702.00832, (2017).
Referans24 Ozkan, M. T., “Experimental and artificial neural network study of heat formation values of drilling and boring operations on Al 7075 T6 workpiece.”, Indian Journal of Engineering and Materials Sciences, 20:4, 259-268, (2013).
Referans25 M. T. Ozkan, H. B. Ulas, M. Bilgin., “Experimental design and artificial neural network model for turning the 50crv4 (sae 6150) alloy using coated carbide/cermet cutting tools”, Materials and Technology., 48:2, 227–236, (2014).