OFFICE WIRELESS NETWORK PERFORMANCE IMPROVEMENT BY CHANGING WIRELESS ROUTERS INSTALLMENT PATTERN AND RADIO CHANNEL SETTING

Transcription

1 OFFICE WIRELESS NETWORK PERFORMANCE IMPROVEMENT BY CHANGING WIRELESS ROUTERS INSTALLMENT PATTERN AND RADIO CHANNEL SETTING 1 RATCHANEPORN PANTHAI, 2 SUWAT PATTARAMALAI 1,2 Electronic and Telecommunication Engineering Department, King Mongkut s University of Technology Thonburi, Thungkru, Bangkok, Thailand 1 p.ratchaneporn@gmail.com, 2 suwat.patt@kmutt.ac.th Abstract In this paper, the performance of office wireless network is improved by installing the access points in different patterns, from linear pattern to the triangular pattern, and all access points frequency channels are also set to be different. The office wireless network is planned by the Air Magnet Planner and Survey tools. Then, three access points are installed in linear or triangular patterns on the office ceiling. Furthermore, the wireless signal is measured from every area in the office to show user efficiency by Ixia Ix Chariot Console and Endpoint tools. In addition, the transmission power of all access points is increased from the default value to improve signal quality and the using frequency band of each access points is set to be different for reducing interference signal. Finally, the triangular pattern routers installation with increased power and different radio channel setting has higher throughput efficiency and also lesser elapsed time than linear pattern installation with default power and default radio channel setting by 59.38Mbps and 505 seconds, respectively. Keywords Frequency Channel, Throughput, Transmission Power. I. INTRODUCTION Nowadays, the wireless network users are increasing exponentially everywhere. There are more variety wireless devices which can access the internet network in every area including office area. More wireless access points are installed all over area and introducing more interferences to each other s. In [1], the wireless network was planned by using Free Space model and installed with real equipment for experimentation in school building. The transmission power of deploying access points was adjusted to reduce interferences and also the signal overlap areas are planned to be avoid in [2]. Furthermore, the wireless user location effecting the throughput performance was studied by using cell identification and space segmentation for high signal quality with wireless access direction in [3]. The cochannel interference, SINR monitoring, and suitable channel selection were studied in [4]. Finally, UDP protocol packets were used to send for analyzing throughput efficiency and user behavior [5]. In this paper, the office wireless network improvement method by the triangular pattern installation of access points with increasing transmission power and selecting different radio channel for each access point is proposed. Next, this paper is organized as follows: In section 2, the office wireless network model is presented. Then, the simulation design and experimental method are detailed in section 3. Finally, results from simulation and experiments are in section 4. Finally, the conclusion is in the last section. II. SYSTEM MODEL 2.1 Network Model Fig. 1 Wireless coverage in office area and access points location. Fig. 1 shows office area about 900 square meters and the red color high light area is the target area for wireless coverage which the received signal strength indication (RSSI) at -65 dbm is required. Also, three wireless access points are installed as linear pattern or triangular pattern in Fig. 1 or 1, respectively, which also show six tester spots. A. Linear pattern: All access points are deployed by installing as the line in the middle of target coverage area as show in Figure 1. The approximate distance between access points is about 7.5 meters. B. Triangular pattern: All access points are deployed by installing in triangular form with 60 degrees inside angle and the center of triangular is in the middle of target coverage area 2.2 Received Signal Strength Indication (RSSI) The wireless coverage area and performance efficiency are two main objectives for installing access points design. In additions, the RSSI is the 19

2 indicator which should be in between -45 and 87 dbm. So the threshold value is set at -65 dbm. 2.3 Channel Model In Fig. 2, the channel frequency at 2.4 GHz with 20MHz bandwidth divided into 13 sub frequency channels is used for all access points. There can be only three non-overlap isolated channels, which are channel no. 1 ( GHz), no. 6 ( GHz) and no. 11 ( GHz) from [6]. Fig. 2 Access point 2.4 GHz sub channel frequency bands. 3.2 Experiment In experiment, all three access points are Cisco Aironet 1702i model and installed on the ceiling of office room as linear pattern or triangular pattern shown in Fig. 5. The signal power and access points sub frequency channel are set and adjusted by using Wireless Controller Cisco Model The wireless coverage area is measured by walking test following the blue dash line as shown in Fig. 5, from point A to point B with the Air Magnet Survey tool. The Ixia IxChariot Console and Endpoint software is used to measure the throughput of users [9]. Each user device is also installed with this software and standing at each test spot in Fig. 1. Fig. 3 show the linear pattern installation of access points which also use the same sub channel frequency band number 1. In this case, the users in overlap area of wireless coverage can receive co-channel interference (CCI) from other access point which causes collision in CSMA access and results to the poor throughputs [7]. In another figure, Fig. 3, all access points are installed as triangular form and also assigned with different sub channel frequency bands, (1, 6, and 11). There are more coverage area and also there is no CCI because of different frequency bands which results to higher throughput. A Fig. 3 Wireless coverage area of three access points. III. SIMULATION AND EXPERIMENT 3.1 Simulation The Air Magnet Planner software is used as a simulation tool. All three access points used Cisco Aironet 1702i model and put on the ceiling as linear pattern or triangular pattern. The floor plan of 30 square meters with 3 meters height is input in the program. Also, the types of wall material and all office furniture are also input in the program by using tool menu circled in Fig B Fig. 5 Office wireless coverage testing experiment Fig. 5 shows the test server on notebook computer at test spot. Finally, all system parameters of simulation and experiment systems are shown in the Table 1. Table 1 Simulation and Experimental system parameters Fig. 4 Simulation by Air Magnet Planner software 20

3 Office Wireless Network Performance Improvement By Changing Wireless Routers Installment Pattern And Radio Channel Setting 3.3 Simulation and Experiment Steps Table 2 shows three scenario steps by setting three access points transmission power and sub frequency channel for both simulation and experiment using linear pattern and triangular pattern access points installation. Table 2 Scenarios setting (d) IV. RESULTS 4.1 Wireless coverage area by simulation Fig. 6 shows simulation wireless coverage area for linear pattern and triangular pattern of three access points using the same sub frequency channel with transmission power at 13dBm (20mW) in - and 19dBm (80mW) in -(d). Triangular pattern installation has more coverage area in office target area than linear pattern one and more transmission power also increases the wireless coverage area. (d) Fig. 7 Coverage area of the real wireless network 4.3 Throughput of real wireless experiment The throughput is measured by using Ixia IxChariot Console and Endpoint software which transmitting and receiving 100Mbytes packet to sever simultaneously. Fig. 8 shows throughput versus elapsed time for all six testing users devices at all six testing spots which are tested at the same time. Each user device is competed for wireless channel access because all access points are set to use the same sub frequency channel number. In the period of 0 to 01:20 minute, the average throughput is about 15 Mbps because there are a lot of collisions. The average throughput is increasing at later time to 30 Mbps in the last duration since some devices has completed the transmission. From Fig. 8, elapsed time for the complete transmission time of all six users devices can be calculated also. Fig. 6 Wireless overage area by simulation with transmission power as 13 dbm in and and 19 dbm in and (d) Moreover, the sub frequency channel of access points are set to use different channel in each access point (1, 6 and 11) but the coverage area results are the same as in Fig. 6. Fig. 8 Throughput versus Elapsed time of six testing users by using Ixia IxChariot Console and Endpoint Table 3 shows all throughputs and elapsed times for all three scenarios testing with three rounds for each scenario. First scenario, all access points using default transmission power at 13dBm, the triangular pattern installation has higher throughput and shorter elapsed time in all three round of testing. The results are the same in the last two scenarios, increasing transmission power to 19dBm and set different sub frequency channel for all three access points. All three rounds throughput and elapsed time results in Table 3 are averaged and used to plot in Fig. 9 and 4.2 Wireless coverage area by experiment Fig. 7 shows experiment wireless coverage area for linear pattern or triangular pattern of three access point using the same sub frequency channel with transmission power at 13dBm in - and 19dBm in -(d). The results are also confirm that triangular pattern installation has more coverage area and also more transmission power increases the wireless coverage area. Proceedings of ISER 28th International Conference, Penang, Malaysia, 7th April 2016, ISBN:

4 Fig. 10. Fig. 9 shows the average throughput versus scenario for linear and triangular patterns access point installation. As the results, with the different sub frequency channel in scenario 3, throughput is increased more than one with increasing transmission power only in scenario 2. Overall, throughput in triangular pattern with increasing power and different sub frequency channel in scenario 3 is 75.01Mbps increasing from 15.63Mbps in linear pattern with default power and same sub channel in scenario 1 about 59.38Mbps. In addition, fig. 10 shows the average elapsed time versus scenario for linear and triangular patterns access point installation. Also, the elapsed time can be reduced about 505 seconds from 637 seconds to 132 seconds by using triangular pattern installation with increasing transmission power and using different sub frequency channels in all three access points. Table 3 Average throughput and Elapsed time of testing six user devices in office wireless network Scenario 1) Default parameter 2) Tx Power Up 3) Tx Power Up and Difference Channel Times Linear pattern (Mbps) Average Throughput Triangular pattern (Mbps) Elapsed Time Disparity Linear Triangular Disparity Mbps % pattern (sec) pattern (sec) sec % Round Round Round Round Round Round Round Round Round and using IxChariot Console and Endpoint to measure throughput and elapsed time. From both simulation and experiment results, the triangular patter has more coverage area than the linear pattern installation. Finally, the triangular pattern installation with increasing transmission power and using different sub frequency channels increases throughput about 59.38Mbps and elapsed time is reduced by 505 seconds. Fig. 9 The average throughput versus scenario CONCLUSION Fig. 10 The elapsed time versus scenario The nine hundred square meters office wireless coverage area is improved by installing three access points in triangular pattern on the ceiling. In addition, throughput is increased and elapsed time is reduced by increasing transmission power and using different sub frequency channel for all three access points. The planning is simulated by using Air Magnet Planner software. The experiment is done by installing three CISCO Aironet 1702i access points REFERENCE [1] Shoa-Yei Yeong, Wafaa Al-Salihy and Tat-Chee Wan Indoor WLAN Monitoring and Planning using Empirical and Theoretical Propagation Models in Second International Conference on Network Applications, Protocols and Services, [2] Oghennekome, Pengfei Xia, Frank LaSita and Robert Olesen Advanced Power Control Techniques for Interference Mitigation in Dense Networks in Wireless Personal Multimedia Communications (WPMC), 16th International Symposium, [3] Bojan R. Ignjatovic,, Bojan B. Andjelinic and Mirjana I. Simic Impact of the User Orientation on WLAN Positioning Based upon the Access Point with the Strongest Signal in TELSIKS 11th International Conference on Volume: 01, 2013, pp [4] Eddie C.L. Chan, George Baciu, S.C. Mak Effect of Channel Interference on Indoor Wireless Local Area Network Position in 6th International Conference on Wireless and Mobile Computing, IEEE, [5] Alexander L. Wijesinha, Yeong-tae Song, Mahesh Krishnan, Vijita Mathur, Jin Ahn, and Vijay Shyamasundar Throughput Measurement for UDP Traffic in an IEEE g WLAN in SNPD/SAWN 05, IEEE. [6] Damien Kim RF Management on Wi-Fi ON - Everything on Wi-Fi Wireless Network, April 8 th, 2011, pp [7] Co-Channel interference (CCI) Available from : 22

5 -site-survey-heatmap-visualizations-part-7-channeloverlap/?more=2448. [8] AirMagnet Planner & Survey Available from: [9] Ixia IxChariot Console & Endpoint Available from: 23