Considering Power Consumption of Wireless Stations Based on 802.11 Mode Control Bachelor Thesis Paper Ramadhan Praditya Putra 1 , Sujoko Sumaryono 2 , Sri Suning Kusumawardani 2 1 Student at Department of Electrical Engineering and Information Technology, UGM 2 Lecturer at Department of Electrical Engineering and Information Technology, UGM Jl. Grafika No. 2, Kampus UGM Yogyakarta, Indonesia 55281 [email protected], [email protected], [email protected]Abstract—The high bit rate 802.11n consumes more power compared to other 802.11 modes during data transmission, reception, and idle (not transmitting or receiving high volume data) [1][2] and may lead to a shorter battery life in mobile devices [3]. This study investigates the transmit, receive, and idle energy consumption of 802.11 modes then forms a dynamic 802.11 mode scheme that uses different energy saving modes for different STA (Wireless Station) operations. The dynamic scheme energy consumption was compared against static schemes using a test bed that realizes seamless data communication during mode transitions. The results show that 802.11n is the most energy efficient mode to use in transmissions and receptions whereas 802.11g is the most energy efficient to use while idle. Therefore, the 802.11g-802.11n-802.11g dynamic mode scheme was formed and it was proven to be the most energy efficient scheme in ’30-second idle – 100 MB transmit – idle’ experiments whereas the static 802.11g scheme was proven to be the most energy efficient scheme in ’30-second idle – 100 MB receive – idle’ experiments. Keywords-STA energy consumption, STA power consumption, 802.11 mode, bit-rate, Mann-Whitney U I. INTRODUCTION Nowadays, the use of WLANs for high quality video streaming and large file transfer has increased. As a result, higher bit-rate wireless connections are required to enable fast data transfer. To achieve higher bit-rates, the IEEE 802.11n standard can be used in APs (Access Points) and STAs (Wireless Stations) [4]. However, the 802.11n standard consumes more power compared to other 802.11 modes during data transmission and idle [1][2]. The excess of power consumption in STAs due to 802.11n usage during idle conditions will lead to a shorter battery life [3]. A study by Putra has proposed the use of 802.11n only when the transmission of high volume data takes place and 802.11g when AP enters idle condition [1]. However, Putra’s research did not incorporate the validation of the 802.11 modes energy consumption data significance using statistical calculations and did not implement seamless data communication during 802.11 mode transitions resulting in a connection termination during mode transitions. Furthermore, the power consumption of STAs was also not considered yet. Therefore, it is important to validate the significance between STAs’ 802.11 modes energy consumption data and realize seamless data communication when mode transitions occur. Based on a white paper by Texas Instruments that describes the low-power advantage of 802.11a/g compared to 802.11b, this study considers the bit-rate of every 802.11 mode as an important factor in energy consumption. The advantage of 802.11a/g is its high bit rate (up to 54 Mbps) while 802.11b has a bit-rate only limited to 11 Mbps [5]. The higher bit-rate of 802.11a/g is what causes the data transmission time for a certain amount of bytes to become shorter thus more energy is saved. This research involved measurements of STA current consumption during transmission, reception, and idle. Current values were used in power and energy calculations. The energy calculation results of every mode were tested for significance against the other modes’ using the Mann-Whitney U-test in SPSS. The Mann-Whitney U-Test is a nonparametric statistical test for comparing two unrelated and independent variables therefore it was used in this study [6]. II. RESEARCH METHOD To implement the test bed used in this study, the following hardware are used: • A computer capable of running Parallels Desktop 6 virtualization software with two Ubuntu 11.10 operating systems as guest operating systems (one functions as the Linux Server while the other as the Linux Client). • Three PCI MZK-RP150N USB powered 802.11b/g/n APs. • Three TP-LINK TL-WN7200ND 802.11b/g/n STAs (USB Wireless Adapters). • Two USB hubs (one with an external voltage source via AC adapter). • A Buffalo WHR-HP-G300N, which functions as a router. • A digital multi meter capable of measuring current up to 10 A. Besides the hardware, several software programs are used to implement the test bed in this study: • The bwm-ng program, which functions as a bit- rate usage detector in the Linux Client [7]. ISSN: 2088-6578 Yogyakarta, 12 July 2012 CITEE 2012 240 DEEIT, UGM – IEEE Comp. Soc. Ind. Chapter
13
Embed
Bachelor Thesis Paper - repository civitas UGM · 2014-07-08 · Considering Power Consumption of Wireless Stations Based on 802.11 Mode Control Bachelor Thesis Paper Ramadhan Praditya
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Considering Power Consumption of Wireless Stations Based on 802.11 Mode Control
Bachelor Thesis Paper
Ramadhan Praditya Putra1, Sujoko Sumaryono2, Sri Suning Kusumawardani2 1Student at Department of Electrical Engineering and Information Technology, UGM
2Lecturer at Department of Electrical Engineering and Information Technology, UGM Jl. Grafika No. 2, Kampus UGM Yogyakarta, Indonesia 55281
Abstract—The high bit rate 802.11n consumes more power compared to other 802.11 modes during data transmission, reception, and idle (not transmitting or receiving high volume data) [1][2] and may lead to a shorter battery life in mobile devices [3]. This study investigates the transmit, receive, and idle energy consumption of 802.11 modes then forms a dynamic 802.11 mode scheme that uses different energy saving modes for different STA (Wireless Station) operations. The dynamic scheme energy consumption was compared against static schemes using a test bed that realizes seamless data communication during mode transitions. The results show that 802.11n is the most energy efficient mode to use in transmissions and receptions whereas 802.11g is the most energy efficient to use while idle. Therefore, the 802.11g-802.11n-802.11g dynamic mode scheme was formed and it was proven to be the most energy efficient scheme in ’30-second idle – 100 MB transmit – idle’ experiments whereas the static 802.11g scheme was proven to be the most energy efficient scheme in ’30-second idle – 100 MB receive – idle’ experiments.
Keywords-STA energy consumption, STA power consumption, 802.11 mode, bit-rate, Mann-Whitney U
I. INTRODUCTION Nowadays, the use of WLANs for high quality video
streaming and large file transfer has increased. As a result, higher bit-rate wireless connections are required to enable fast data transfer. To achieve higher bit-rates, the IEEE 802.11n standard can be used in APs (Access Points) and STAs (Wireless Stations) [4]. However, the 802.11n standard consumes more power compared to other 802.11 modes during data transmission and idle [1][2]. The excess of power consumption in STAs due to 802.11n usage during idle conditions will lead to a shorter battery life [3].
A study by Putra has proposed the use of 802.11n only when the transmission of high volume data takes place and 802.11g when AP enters idle condition [1]. However, Putra’s research did not incorporate the validation of the 802.11 modes energy consumption data significance using statistical calculations and did not implement seamless data communication during 802.11 mode transitions resulting in a connection termination during mode transitions. Furthermore, the power consumption of STAs was also not considered yet. Therefore, it is important to validate the significance between STAs’ 802.11 modes energy consumption data and realize seamless data communication when mode transitions occur.
Based on a white paper by Texas Instruments that describes the low-power advantage of 802.11a/g compared to 802.11b, this study considers the bit-rate of every 802.11 mode as an important factor in energy consumption. The advantage of 802.11a/g is its high bit rate (up to 54 Mbps) while 802.11b has a bit-rate only limited to 11 Mbps [5]. The higher bit-rate of 802.11a/g is what causes the data transmission time for a certain amount of bytes to become shorter thus more energy is saved.
This research involved measurements of STA current consumption during transmission, reception, and idle. Current values were used in power and energy calculations. The energy calculation results of every mode were tested for significance against the other modes’ using the Mann-Whitney U-test in SPSS. The Mann-Whitney U-Test is a nonparametric statistical test for comparing two unrelated and independent variables therefore it was used in this study [6].
II. RESEARCH METHOD To implement the test bed used in this study, the
following hardware are used:
• A computer capable of running Parallels Desktop 6 virtualization software with two Ubuntu 11.10 operating systems as guest operating systems (one functions as the Linux Server while the other as the Linux Client).
• Three PCI MZK-RP150N USB powered 802.11b/g/n APs.
• Three TP-LINK TL-WN7200ND 802.11b/g/n STAs (USB Wireless Adapters).
• Two USB hubs (one with an external voltage source via AC adapter).
• A Buffalo WHR-HP-G300N, which functions as a router.
• A digital multi meter capable of measuring current up to 10 A.
Besides the hardware, several software programs are used to implement the test bed in this study:
• The bwm-ng program, which functions as a bit-rate usage detector in the Linux Client [7].
ISSN: 2088-6578 Yogyakarta, 12 July 2012 CITEE 2012
240 DEEIT, UGM – IEEE Comp. Soc. Ind. Chapter
• Iperf, for sending TCP packets between the Linux Client and Server [8].
In general, the workings of the hardware implementation in Fig. 1 are as follows. At first, no data transmission is occurring and only one STA is active. Both the currently active AP and STA are using the most energy efficient mode during idle. Suppose that the mode is 802.11g. A few moments later, the user through Linux Client (Guest OS 2) downloads a large file from Linux Server (Guest OS 1). Linux Client will detect a bit-rate increase on the bonded interface port at 172.29.144.4. The interface bonds three STAs (Wireless USB Adapters), which are connected to a USB hub. Because a bit-rate increase is detected, the Linux Client will activate the STA with the most energy efficient 802.11 mode during transmission (e.g. 802.11n) and then include it into the bonded interface.
Once the newly activated 802.11n STA has associated with the 802.11n AP, the previously used 802.11g STA will be deactivated. Deactivation is done by excluding the previously used STA from the bonded interface.
The procedure in the previous paragraph creates a seamless data communication during mode transitions due to the soft handoff between the STAs and APs [9]. The soft handoff procedure relies on the active-backup scheme in the Linux interface bonding of the three STAs. In short, interface bonding is logically merging two or more interfaces in order to have a single MAC address if seen from other network devices [10]. The term active-backup means that if one interface is disabled, then the connection will be seamlessly replaced by another interface [11].
The power consumption of every 802.11 mode was measured during transmission, reception, and idle operations, each involving sixteen repeated measurements. In the transmission operation, power measurements were performed during 30 MB TCP data transmission whereas in the reception operation, power measurements were performed during 30 MB TCP data reception. In the idle operation, measurements were performed during 20-second idle conditions.
Power values were calculated by multiplying the measured current values by five, which is the measured USB port voltage. Afterwards, energy consumption values were calculated by multiplying the power values by two, which is the period between current measurements in seconds. After the energy values of every operation are obtained, the significance of energy consumption between
modes in every operation was tested using the Mann-Whitney U-test.
After the most energy-saving mode during transmit/receive and idle has been determined, a dynamic mode scheme that is a combination of the most energy-saving modes in every operation was formed. The energy consumption of the dynamic scheme was measured and compared with the static schemes (802.11b only, 802.11g only, and 802.11n only) in a ’30-second idle – 100 MB transmit – idle’ and ’30-second idle – 100 MB transmit – idle’ operation pattern. The ’30-second idle – 100 MB transmit/receive – idle’ pattern was arbitrarily chosen based on the author’s subjective personal operation pattern during web browsing and file downloading or uploading. Mode scheme energy consumption measurements were conducted sixteen times for every operation pattern and the significance between each mode scheme’s energy consumption was tested using the Mann-Whitney U-test.
III. RESULTS AND DISCUSSIONS The following sub-sections describe and discuss the
power measurement results.
A. Transmit Energy Consumption Fig. 2 shows the transmit power consumption of
802.11b, 802.11g, and 802.11n and shows that 802.11n at approximately 85 Mbps has the highest power consumption. However, due to the high-bit rate, transmission time is shortest thus consuming less energy. The 802.11b, which transmits at 4.66 Mbps, has relatively high power consumption and the longest transmission time. The 802.11g was able to transmit in a shorter time than 802.11b due to its 16.7 Mbps and at a lower power consumption.
The high power consumption of 802.11n is a result of its high bit-rate, which caused the STA baseband processor to increase its clock speed and due to its use of 40 MHz channel width [3][12]. The 802.11g mode achieved low power consumption as a result of its relatively low bit rate compared to its maximum bit-rate (16.7 out of 25 Mbps) and its use of 20 MHz channel width. On the other hand, 802.11b consumed higher power due to its relatively high bit-rate compared to its maximum bit-rate (4.66 out of 5.17 Mbps) and mostly due to its relatively high RF power level [3].
Each mode was unable to transmit at more than half of its maximum specified bit-rate (11 Mbps for 802.11b, 54 Mbps for 802.11g, and 300 Mbps for 802.11n). The
Figure 2. Test bed hardware assembly. Figure 1. 802.11 modes 30 MB transmit power consumption.
CITEE 2012 Yogyakarta, 12 July 2012 ISSN: 2088-6578
DEEIT, UGM – IEEE Comp. Soc. Ind. Chapter 241
reason is the overhead caused by the use of CSMA/CA access control, which requires STA to wait for a random amount of time before determining that the medium is free for transmission. Another reason is the RTS/CTS sequence used to overcome the hidden node problem [13].
Fig 3. shows the transmit energy consumption of every mode. Fig. 3 shows that the high bit-rate 802.11n is the least energy consumptive despite its high power consumption whereas the low bit-rate 802.11b is the most energy consumptive.
B. Receive Energy Consumption Fig. 4 shows the receive power consumption of
802.11b, 802.11g, and 802.11n and shows that 802.11n at approximately 13.2 Mbps has the highest power consumption. However, due to the high-bit rate, reception time is shortest thus consuming less energy. The 802.11b, which receives at 3.37 Mbps, has a higher power consumption compared to 802.11g and the longest transmission time. The 802.11g was able to receive in a shorter time than 802.11b due to its 3.7 Mbps and at a lower power consumption.
The high power consumption of 802.11n is due to its high receive rate, which caused the STA baseband processor to increase its clock speed and due to its use of 40 MHz channel width [3][12]. The 802.11g mode achieves low power consumption as a result of its relatively low receive rate compared to its maximum bit-rate (3.7 out of 25 Mbps) and its use of 20 MHz channel width. On the other hand, 802.11b consumes higher power due to its relatively high receive rate compared to its maximum bit-rate (3.37 out of 5.17 Mbps).
Fig 5. shows the receive energy consumption of every mode. Fig. 5 shows that the high bit-rate 802.11n is the least energy consumptive despite its high power consumption whereas the low bit-rate 802.11b is the most energy consumptive.
C. Idle Energy Consumption Fig. 6 shows the idle power consumption of 802.11b,
802.11g, and 802.11n and shows that 802.11n has the highest power consumption. On the other hand, 802.11g and 802.11b have lower power consumption than 802.11n but have quite similar power consumption values among each other.
The high power consumption of 802.11n while both 802.11b and 802.11g have the same power consumption shows that channel bandwidth does have an effect on power consumption [12]. In idle mode, the RF power amplifier is idle therefore signal strength does not affect power consumption [3]. In addition, the baseband processor is still turned on to scan for RF signals. Since no bits are transmitted or received, the baseband processor speed does not count for power consumption. Only the processing required for powering the 40 MHz channel is accounted.
Fig 7. Shows the idle energy consumption of every
Figure 4. 802.11 modes 30 MB transmit energy consumption.
Figure 5. 802.11 modes 30 MB receive power consumption.
Figure 6. 802.11 modes 20-second idle power consumption.
Figure 3. 802.11 modes 30 MB receive energy consumption. Figure 7. 802.11 modes 20-second idle energy consumption.
ISSN: 2088-6578 Yogyakarta, 12 July 2012 CITEE 2012
242 DEEIT, UGM – IEEE Comp. Soc. Ind. Chapter
mode. The 802.11n has the highest energy consumption while it has not been determined yet which among 802. 11b and 802.11g has the lowest energy consumption.
D. Mode Scheme Formation In order to determine which mode is the most energy
saving in each operation, significance tests have been conducted using the Mann-Whitney U-test. The level of significance (α) used was 0.05 and the following hypotheses were used:
• HO : Both modes X and Y have identical energy consumption.
• HA : Both modes have significantly different energy consumption.
Table 1 shows the energy consumption ranks of the 802.11 modes based on the conducted significance tests. It was found that the idle energy consumption of 802.11b and 802.11g are not significantly different, therefore to decide which mode to use during idle and low bit-rate transmission/reception, a small experiment has been conducted to investigate the power consumption of every mode in all possible bit-rates. Fig. 8 shows the experiment results and shows that 802.11g is more power-efficient compared to 802.11b during transmission and reception. Based on the significance tests, Table 1, and Fig. 8, the dynamic ‘802.11g-802.11n-802.11g’ mode scheme was formed.
E. ’30-second idle – 100 MB transmit – idle’ Energy Consumption In this experiment, the power consumption of three
static mode schemes (802.11b, 802.11g, and 802.11n) and one dynamic scheme (‘802.11g-802.11n-802.11g’) were measured during a ’30-second idle – 100 MB transmit – idle’ operation pattern. Fig. 9 shows that 802.11b has the longest transmit time due to its 4.66 Mbps bit-rate. The 802.11g at 16.7 Mbps took shorter time and lower power consumption compared to 802.11b. The 802.11n at approximately 85 Mbps was able to achieve the shortest transmit time but high power consumption during idle.
The proposed dynamic ‘802.11g-802.11n-802.11g’ scheme achieved low power consumption during idle as well as a shorter transmit time compared to 802.11g and 802.11b. However, the soft handoff process during mode transitions consumed a significant amount of power due to the requirement of two STAs to be simultaneously powered on during the handoff process.
Fig. 10 shows the energy consumption of the schemes in the ’30-second idle – 100 MB transmit – idle’ experiments. Based on Fig. 10 and the Mann-Whitney U-tests performed between mode scheme energy consumption values, 802.11b was proven to be the most energy consumptive whereas ‘802.11g-802.11n-802.11g’ was proven to be the most energy-saving scheme despite the high power consumption during mode transitions.
F. ’30-second idle – 100 MB receive – idle’ Energy Consumption In this experiment, the power consumption of three
static mode schemes (802.11b, 802.11g, and 802.11n) and one dynamic scheme (‘802.11g-802.11n-802.11g’) were measured during a ’30-second idle – 100 MB receive – idle’ operation pattern. Fig. 11 shows that 802.11b has the longest receive time due to its 3.37 Mbps bit-rate. The 802.11g at 3.7 Mbps takes shorter time and lower power consumption compared to 802.11b. The 802.11n at approximately 13.2 Mbps was able to achieve the shortest receive time but high power consumption during idle. The proposed dynamic ‘802.11g-802.11n-802.11g’ scheme achieved low power consumption during idle as well as shorter receive time compared to 802.11g and 802.11b. However, the soft handoff process during mode transitions consumed a significant amount of power due to the requirement of two STAs to be simultaneously
Figure 8. 802.11 modes power consumption vs. bit-rate.
CITEE 2012 Yogyakarta, 12 July 2012 ISSN: 2088-6578
DEEIT, UGM – IEEE Comp. Soc. Ind. Chapter 243
powered on during the handoff process.
Fig. 12 shows the energy consumption of the schemes in the ’30-second idle – 100 MB receive – idle’ experiments. Based on Fig. 12 and the Mann-Whitney U-tests performed between mode scheme energy consumption values, 802.11b was proven to be the most energy consumptive whereas 802.11g was proven to be the most energy-saving scheme.
IV. CONCLUSION The 802.11n mode is the most energy-efficient mode
to use during transmission and reception whereas 802.11g is the most energy efficient during idle. Consequently, the ‘802.11g-802.11n-802.g’ dynamic mode scheme was formed. The dynamic scheme was proven to be the most energy-efficient in the ’30-second idle – 100 MB transmit – idle’ experiment whereas the 802.11g scheme was proven to be the most energy-efficient in the ’30-second idle – 100 MB receive – idle’ experiment. Seamless communication during mode transitions was implemented using the soft handoff method between STAs. However, the soft handoff method was proven to consume a significant amount of power.
V. FUTURE STUDIES In this study, the ’30-second idle – 100 MB
transmit/receive – idle’ operation pattern was arbitrarily chosen based on the author’s personal experience in web browsing and file downloading or uploading. In future studies, experiments based on an objective determination of user’s operation pattern should be conducted.
ACKNOWLEDGMENT The author wishes to express his gratitude to his
bachelor thesis supervisors, Sujoko Sumaryono and Sri Suning Kusumawardani who was abundantly helpful and offered invaluable assistance, support, and guidance. Deepest gratitude is also due to Prof. Koji Okamura and Prof. Takumi Miyoshi without whose knowledge this study would not have been successful.
REFERENCES [1] Putra, Ramadhan P. and Miyoshi, Takumi (2012). “Considering
Power Consumption of Access Points Based on 802.11 Mode Control”. Paper Presented at IEICE General Conference, March 20-23, 2012 in Okayama.
[2] --------- (2008). ”Practical Considerations for Deploying 802.11n”. White Paper. Siemens.
[3] Ebert, J.P., Burns, B., and Wolisz, A. (2002). “A Trace-based Approach for Determining The Energy Consumption of a WLAN Network Interface”. Proceeding. Appeared in Proceeding of European Wireless 2002, pp. 230-236.
[4] --------- (2009). ”IEEE Std 802.11n-2009”. White Paper. The Institute of Electrical and Electronic Engineers, Inc. http://standards.ieee.org/getieee802/download/802.11n-2009.pdf. Accessed on January 25, 2012.
[5] --------- (2003). ”Low Power Advantage of 802.11a/g vs. 802.11b”. White Paper. Texas Instruments Incorporated. http://focus.ti.com/pdfs/bcg/80211_wp_lowpower.pdf. Accessed on January 28, 2012.
[6] Corder, Gregory W. and Foreman, Dale I. (2009). ”Nonparametric Statistics for Non-Statisticians: A Step-by-Step Approach”. Hoboken: John Wiley & Sons, Inc.
[7] Gropp, Volker (2011). ”Bwm-ng (Bandwidth Monitor NG)”. Homepage of Volker Gropp. http://www.gropp.org/?id=projects&sub=bwm-ng. Accessed on December 5, 2011.
[8] --------- (2011). ”Iperf”. SourceForge.Net. http://iperf.sourceforge.net/. Accessed on November 29, 2011.
[9] Ramachandran, K., Rangarajan, S., and Lin, J.C. (2006). “Make-Before-Break MAC Layer Handoff in 802.11 Wireless Networks”. in IEEE International Conf. on Communications. Istanbul. 2006. pp. 4818-4823.
[10] --------- (2003). ”Bonding (Port Trunking)”. LINUX Horizon. http://www.linuxhorizon.ro/bonding.html. Accessed on January 29, 2012.
[11] Davis, T. et al (2011). ”Linux Ethernet Bonding Driver HOWTO”. The Linux Kernel Archives. http://www.kernel.org/doc/Documentation/networking/bonding.txt. Accessed on January 29, 2012.
[12] Chandra, R. et al. (2008). ”A Case for Adapting Channel Width in Wireless Networks”. Microsoft Research, Redmond. Presented in SIGCOMM’08 August 17-22, 2008 in Seattle.
[13] Cook, Diane J. and Das, Sajal K. (2005). “Smart Environments: Technologies, Protocols, and Applications”. Hoboken: John Wiley & Sons, Inc.
ISSN: 2088-6578 Yogyakarta, 12 July 2012 CITEE 2012
244 DEEIT, UGM – IEEE Comp. Soc. Ind. Chapter
CITEE2012
PROCEEDINGS OFINTERNATIONAL CONFERENCE ON
INFORMATION TECHNOLOGYAND
ELECTRICAL ENGINEERING
Yogyakarta, IndonesiaJuly 12, 2012
Inte
rna
tion
al C
on
fere
nce
on
Info
rma
tion
Tech
no
log
y an
d E
lectrica
l En
gin
ee
ring
DEPARTMENT OF ELECTRICAL ENGINEERINGAND INFORMATION TECHNOLOGY
FACULTY OF ENGINEERING
GADJAH MADA UNIVERSITY
20
12
Department of Electrical Engineering and Information TechnologyFaculty of Engineering, Gadjah Mada UniversityJalan Grafika no. 2, Kampus UGMYogyakarta, 55281, Indonesia
ISSN: 2088-6578
CITEE 2012 ISSN: 2088-6578
PROCEEDINGS OF INTERNATONAL CONFERENCE ON
INFORMATION TECHNOLOGY AND ELECTRICAL ENGINEERING
Yogyakarta, 12 July 2012
DEPARTMENT OF ELECTRICAL ENGINEERING AND INFORMATION TECHNOLOGY
FACULTY OF ENGINEERING GADJAH MADA UNIVERSITY
ISSN: 2088-6578 Yogyakarta, 12 July 2012 CITEE 2012
ii DEEIT, UGM – IEEE Comp. Soc. Ind. Chapter
ORGANIZER 2012
Technical Program Committee Chair (cont.)
Andreas Timm-Giell (Universität Hamburg-
Harburg Germany)
Ryuichi Shimada (Tokyo Institute of
Technology, Japan)
Ismail Khalil Ibrahim (Johannes Kepler
University Linz, Austria)
Kang Hyun Jo (University of Ulsan, Korea)
David Lopez (King’s College London, United
Kingdom)
Martin Klepal (Cork Institute of Technology,
Ireland)
Tamotsu Nimomiya (Nagasaki University,
Japan)
Ekachai Leelarasmee (Chulalongkorn
University, Thailand)
Marteen Weyn (Artesis University College,
Belgium)
Chong Shen (Hainan University, China)
Haruichi Kanaya (Kyushu University, Japan)
Ramesh K. Pokharel (Kyushu University,
Japan)
Ruibing Dong (Kyushu University, Japan)
Kentaro Fukushima (CRIEPI, Japan)
Mahmoud A. Abdelghany (Minia University,
Egypt)
Sunil Singh (G B Pant University of
Agriculture & Technology, India)
Abhishek Tomar (G B Pant University of
Agriculture & Technology, India)
Lukito Edi Nugroho (Universitas Gadjah
Mada, Indonesia)
Umar Khayam (Institut Teknologi Bandung,
Indonesia)
Anton Satria Prabuwono (Universiti
Kebangsaan Malaysia, Malaysia)
Eko Supriyanto (Universiti Teknologi
Malaysia, Malaysia)
Kamal Zuhairi Zamli (Universiti Sains
Malaysia, Malaysia)
Sohiful Anuar bin Zainol Murod (Universiti
Malaysia Perlis, Malaysia )
Advisory Board
F. Danang Wijaya
Risanuri Hidayat
General Chair
Widyawan
Chair
Eka Firmansyah
Indriana Hidayah
Eny Sukani Rahayu
Avrin Nur Widyastuti
Bimo Sunarfri Hantono
Sigit Basuki Wibowo
Budi Setiyanto
Ridi Ferdiana
Yusuf Susilo Wijoyo
Adhistya Erna Permanasari
Prapto Nugroho
Muhammad Nur Rizal
Selo Sulistyo
Sunu Wibirama
Lilik Suyanti
Indria Purnamasari
Reviewer
Adhistya Erna Permanasari
Agus Bejo
Avrin Nur Widyastuti
Bambang Sugiyantoro
Bambang Sutopo
Bimo Sunarfri Hantono
Bondhan Winduratna
Budi Setiyanto
Danang Wijaya
Eka Firmansyah
Enas Duhri Kusuma
Eny Sukani Rahayu
Harry Prabowo
Indriana Hidayah
Insap Santosa
Isnaeni
Iswandi
Litasari
Lukito Edi Nugroho
Noor Akhmad Setiawan
Prapto Nugroho
Ridi Ferdiana
Risanuri Hidayat
Rudy Hartanto
Samiadji Herdjunanto
Sarjiya
Sasongko Pramono Hadi
Selo
Sigit Basuki Wibowo
Silmi Fauziati
Suharyanto
Sujoko Sumaryono
Sunu Wibirama
T. Haryono
Teguh Bharata Adji
Wahyu Dewanto
Wahyuni
Warsun Nadjib
Widyawan
Yusuf Susilo Wijoyo
CITEE 2012 Yogyakarta, 12 July 2012 ISSN: 2088-6578
DEEIT, UGM – IEEE Comp. Soc. Ind. Chapter iii
FOREWORD
Welcome to this year’s CITEE 2012 in Yogyakarta.
Peace be upon you. First of all, praise to Allah, for blessing us with healthy and ability to come
here, in the Conference on Information Technology and Electrical Engineering 2012 (CITEE 2012).
If there is some noticeable wisdoms and knowledge must come from Him.
This conference is the fourth annual conference organized by the Department of Electrical
Engineering and Information Technology, Faculty of Engineering, Universitas Gadjah Mada. It is
expected that CITEE 2012 can serve as a forum for sharing knowledge and advances in the field of
Information Technology and Electrical Engineering, especially between academic and industry
researchers.
On behalf of the committee members, I would like to say thank you to all of the writers, who come
here enthusiastically to share experiences and knowledge. I also would like to say thank you to the
keynote speakers for the participation and contribution in this conference.
According to our record, there are 150 papers from 15 countries are being submitted to this
conference and after underwent reviewing process there are 78 papers that will be presented. It is a
52% acceptance rate. There are 15 papers in the field of Power Systems, 26 papers in the area of
Signals System and Circuits, 11 papers in Communication System and 26 papers in Information
Technology. Furthermore, the proceedings of this conference is expected to be used as reference for
the academic and practitioner alike.
Finally, I would like to say thank you to all of the committee members, who worked tirelessly to
prepare this conference. Special thank to IEEE Computer Society Indonesian Chapter, Department
of Electrical Engineering and Information Technology UGM and LPPM UGM for the support,
facility and funds.
Thank you and enjoy the conference, CITEE 2012, and the city, Yogyakarta
12 July 2012
Widyawan
ISSN: 2088-6578 Yogyakarta, 12 July 2012 CITEE 2012
DEEIT, UGM – IEEE Comp. Soc. Ind. Chapter iv
Schedule of CITEE 2012
Yogyakarta, 12 July 2012
07.30 – 08.00 Registration
08.00 – 08.10 Opening Speech
1. Chairman of the Organizing Committee
2. Head of Department of Electrical Engineering and Information Technology of Gadjah Mada
University
PLENARY SESSION (at Room 1): Keynote Speech
08.10 – 08.50 The Development Trend of a Next Generation Vehicle and its Propulsion Motor
Professor Jin Hur, Ph.D., University of Ulsan, Korea