SOMU VENKATESWARLU* et al.

[IJESAT] INTERNATIONAL JOURNAL OF ENGINEERING SCIENCE & ADVANCED TECHNOLOGY

ISSN: 22503676
Volume-2, Special Issue-1, 105 110

ONLINE MOBILE BACKUP AND POWER OPTIMIZED RETRIEVAL THROUGH MESSAGING

Somu Venkateswarlu 1, P Tanuj Prabhakar2, M Ravi Teja 3,Sk. Khaja Moin 4, K. Venkateswara Rao 5 1: Professor & Head of the Department, 2, 3, 4, 5: Student, Computer Science Dept, K L University, Andhra Pradesh, India 1: somu23@yahoo.com, 2: tanujprabhakar@gmail.com, 3: tejamunnangi@gmail.com, 4: khajamoin1991@gmail.com 5: venki.1113@gmail.com

Abstract
Backup has gained an important role as the best strategy for data preservation in case of equipment failure or other catastrophes. However the backup applications that are available do not contribute much to the power optimization. This poses a serious challenge to the otherwise ubiquitous methodologies of backup. The technique presented in this paper mainly concentrates on optimizing the power consumption of mobile devices while backing up the data. The existing backup strategies enable users to back up the data either in their virtual space or online and retrieve the data through internet. And also, the existing system utilizes the internet connectivity for both backing up the data and retrieval of the data as well which ultimately results in more consumption of power and processing. To save power utilization, one of the remedy is to avoid the dependence on Wi-Fi and internet connectivity to a maximum extent. This paper also discusses a measurement study of the energy consumption characteristics of three widespread mobile networking technologies: 3G, GSM, and Wi-Fi. The measurements pertain to tail energy and ramp energy of the technologies.

Index Terms: Backup, Mobile Computing, Wi-Fi, power optimization, 3G, GSM, tail energy, ramp energy --------------------------------------------------------------------- *** -----------------------------------------------------------------------1. INTRODUCTION
Mobile Phone Backup is a service that safeguards all data of your mobile phone including Contacts, Calendar, SMS, Photos, Videos and Music. Approximately 10 million mobile phones were being lost every year and this staggering figure specifies the importance that backup has earned especially in the case of mobile devices. The problem with the existing backup applications is that the user is made to retrieve the data of the mobile only through internet. The effects of the sole dependence on internet for both backup and retrieval of the data are:(i) There is danger of data loss while retrieving the data in the case when the signal is lost in the middle of the transaction. (ii) The power utilization is considerably augmented. The scarce resources in mobile phones are: Power Processing Speed Storage To optimize the power utilization, one of the remedy is to avoid the use of Wi-Fi and internet connectivity to a maximum extent. Hence in designing backup application, this factor is needed to be highlighted. The existing system utilizes the internet connectivity for both backing up of data and retrieval of the data as well. This ultimately results in more power consumption and since mobile devices are more power sensitive, its main resource i.e., power is needed to be utilized at a minimum extent.

2. ENERGY CONSUMPTION IN MOBILE PHONES: A MEASUREMENT STUDY

Mobile phones are ubiquitous today with an estimated cellular subscription of over 4 billion worldwide [2]. Most phones today support one or more of 3G, GSM, and Wi-Fi for data transfer. Below is a summary of the key ndings of measurement study, which remain consistent across three different cities, diurnal variation, mobility patterns, and devices. 1. In 3G, a large fraction (nearly 60%) of the energy, referred to as the tail energy, is wasted in high-power states after the completion of a transfer. In comparison, the ramp energy spent in switching to this high-power state before the transfer is small. Tail and ramp energies are constants that amortize over larger transfer sizes or frequent successive transfers.

IJESAT | Jan-Feb 2012

Available online @ http://www.ijesat.org 105

SOMU VENKATESWARLU* et al. [IJESAT] INTERNATIONAL JOURNAL OF ENGINEERING SCIENCE & ADVANCED TECHNOLOGY 2. In GSM, although a similar trend exists, the time spent in the high-power state after the transfer, or the tail time, is much smaller compared to 3G (6 vs. 12 secs). Furthermore, the lower data rate of GSM implies that more energy is spent in the actual transfer of data. 3. In WI-Fi, the association overhead is comparable to the tail energy of 3G, but the data transfer itself is signicantly more efficient than 3G for all transfer sizes. Two factors determine the energy consumption due to network activity in a cellular device. First, is the transmission energy that is proportional to the length of a transmission and the transmit power level. Second is the Radio Resource Control (RRC) protocol that is responsible for channel allocation and scaling the power consumed by the radio based on inactivity timers.

ISSN: 22503676
Volume-2, Special Issue-1, 105 110

3G and GSM measurements quantify the: 1) Ramp energy: energy required to switch to the high-power state, 2) Transmission energy, and 3) Tail energy: energy spent in high-power state after the completion of the transfer.

2.1 3G Measurements:
Figure 2(a) shows the average energy consumption for a typical 50KB download over 3G [2]. We find that the Tail energy is more than 60% of the total energy. The Ramp energy is significantly small compared to the tail energy, and is only 14% of the total energy.

(a) 3G: Energy components

(a) (b)

Figure 1: (a) The radio resource control state machine [1] for 3GPP [10] networks consisting of three states: IDLE, DCH and FACH (b) Instantaneous power measurements for an example transfer over 3G showing the transition time between high to low power state Figure 1(b) shows the instantaneous power measurements for an example transfer. The graph shows the time taken to transition from a high power to a low power state. Instead of transitioning from the high to the low power state immediately after a packet is transmitted, the device transitions only when the network has been inactive for the length of the inactivity timer. This mechanism serves two benefits: 1) it alleviates the delay incurred in moving to the high power state from the idle state, and 2) it reduces the signaling overhead incurred due to channel allocation and release during state transitions. Since lingering in the high power state also consumes more energy, network operators set the value of the inactivity timer based on this performance/energy trade-off with typical values being several seconds long.

(b) 3G: Varying inter-transfer times

Figure 2: 3G Measurements: (a) Average ramp, transfer and tail energy consumed to download 50K data. The lower portion of the stacked columns show the proportion of energy spent for each activity compared to the total energy spent. (b)

IJESAT | Jan-Feb 2012

Available online @ http://www.ijesat.org 106

SOMU VENKATESWARLU* et al. [IJESAT] INTERNATIONAL JOURNAL OF ENGINEERING SCIENCE & ADVANCED TECHNOLOGY Average energy consumed for downloading data of different sizes against the inter-transfer time

ISSN: 22503676
Volume-2, Special Issue-1, 105 110

2.3 Wi-Fi MEASUREMENTS

Figure 4(a) shows the average energy consumption in Wi-Fi composed of scanning association and transfer, for a 50 K download [3]. It is also observed that the scanning and association energy is nearly five times the transfer energy. Figure 4(b) shows that the energy consumption of Wi-Fi increases with the time between successive transfers. Interestingly, the energy consumption does not plateau after a threshold inter-transfer time like in 3G. The reason for this monotonic increase is the high maintenance energy in Wi-Fi.

2.2 GSM MEASUREMENTS

Figure 3(a) shows the average energy consumption in GSM networks as a proportion of the Tail energy; Ramp energy and transfer energy for a 50K download [2]. Unlike in 3G, the Tail energy only accounts for 30% of the transfer energy. However, similar to 3G, the Ramp energy in GSM is small compared to the Tail energy and the transfer energy. Due to the small tail-time in GSM (unlike 3G), data sizes dominate energy consumption rather than the inter-transfer times. Figure 3(b) shows the average energy consumed when varying the time between successive transfers.

(a) Wi-Fi: Energy components (a) GSM: Energy components

(b) GSM: Varying inter-transfer times Figure 3: GSM Measurements: (a) Average ramp, transfer and tail energy consumed to download 50K data. The lower portion of the stacked columns show the proportion of energy spent for each activity compared to the total energy spent. (b) Average energy consumed for downloading data of different sizes against the inter-transfer time.

(b) Wi-Fi: Varying inter-transfer times Figure 4: Wi-Fi Measurements: (a) Average scan/associate and transfer energy consumed to download 50K data. The lower portion of the stacked columns show the proportion of energy spent for each activity compared to the total energy spent. (b) Average energy consumed for downloading data of different sizes against the inter-transfer time.

IJESAT | Jan-Feb 2012

Available online @ http://www.ijesat.org 107

SOMU VENKATESWARLU* et al. [IJESAT] INTERNATIONAL JOURNAL OF ENGINEERING SCIENCE & ADVANCED TECHNOLOGY Figure 5 shows that the transfer energy for Wi-Fi grows nearly three times slower compared to the cellular networks. For a download of size 10K, Wi-Fi consumes one-sixth of 3Gs energy and one-third of GSMs energy. With increasing data sizes Wifes efficiency increases dramatically. The graph shows that when the cost of scan and transfer is included (marked in the graph as Wi-Fi + SA), Wi-Fi becomes inefficient for small sized transfers compared to GSM. Surprisingly, when compared to 3G, Wi-Fi is energy efficient even when the cost of scanning and association is included. 1.

ISSN: 22503676
Volume-2, Special Issue-1, 105 110

2.

When internet connectivity is lost in the middle of retrieving the data, the transaction is said to be lost and there is no way to resume the retrieval process. Sole dependence on internet eventually leads to more energy consumption.

Figure 6: Framework of the existing backup applications There are some firm statistics confirming the excess power consumption while data related accesses through internet. Figure 5: Wi-Fi versus 3G versus GSM measurements: Average energy consumed for downloading data of different sizes against the inter-transfer time To achieve traffic scenarios, we have obtained an experimental environment. The environment basically consists of three core components. 1. An Apache web server running on Windows 7 platform 2. Nokia Energy Profiler measuring software running on Nokia 5530 handset 3. Bandwidth Controller software [6] running in the server machine The studies collectively specify that internet accessing via mobile ultimately leads to over utilization of power.

3 BACKGROUND AND RELATED ALTERNATIVES AND

THEIR DRAWBACKS

The framework of the existing system consists of a mobile backup application which may provide the facility of user authentication in the case when the backup is made online, an abstract intermediate internet layer, and the server which is responsible to handle the user data being backed up. The architecture of the existing system has a serious flaw in it. The exclusive concentration on applications performance has led to the compromise on minimizing the power utilization. As illustrated in figure 6, the unfettered dependence on internet for backup and retrieval causes hindrance to the power optimization which unconditionally yields to poor performance in terms of power management. There are nevertheless, few drawbacks of the existing applications:

Figure 7: Power Consumption for Smooth Traffic at 50KBps Data Rate

IJESAT | Jan-Feb 2012

Available online @ http://www.ijesat.org 108

SOMU VENKATESWARLU* et al. [IJESAT] INTERNATIONAL JOURNAL OF ENGINEERING SCIENCE & ADVANCED TECHNOLOGY

ISSN: 22503676
Volume-2, Special Issue-1, 105 110

all the messages that are to be received by default and send them when the mobile phone is intact with connection. Secondly, messaging capability of a mobile phone utilizes significantly less energy when compared to internet connectivity.

Figure 8: Power Consumption for Bursty Traffic at 50KBps Data Rate However, regardless of what the measurements signify the unanimous preference goes to the performance and hence priority to power optimization is continuously deteriorating.

ENERGY EFFICIENT METHODOLOGY AND

ARCHITECTURE

Figure 9: Proposed architecture where a gateway acts as a bridge to different networks. However, considering the performance issues, this architecture compromises certain aspects to compensate it with better power management.

Optimistically, though performance is of a paramount consideration, energy efficient methods needed to be developed to meet the requirements of a power sensitive device like mobile phone. The proposed architecture is depicted in figure 9. The new architecture utilizes the messaging capabilities of mobile phone to retrieve the information. Retrieving the data via messages significantly increases the power optimization and efficintly manages energy resource when compared to the sole dependence on internet for both backup and retrieval. A gateway which is an internetworking system capable of joining together two networks that use different base protocols is used in between the server and the mobile application. A network gateway can be implemented completely in software, completely in hardware, or as a combination of both. Depending on the types of protocols they support, network gateways can operate at any level of the OSI model. The proposed system addresses the problems that a backup application which is solely dependent on internet faces. Firstly, the problem of resuming the support of retrieval in case of a connection interruption, is not at all an issue while working with messages because, the service provider buffers

5 FUTURE ENHANCEMENTS
Despite handling the problems of power optimization, this model poses some serious limitations. One such is that retrieving considerably large data is a lethargic and tedious task with messaging as the option. For handling this issue effectively, a switch must be made available between messaging and internet based on the amount of data to be retrieved. Also it is better to utilize energy effective methodologies such as using mobile proxies [8], multiple interfaces [9], etc. of network access as an alternative to the traditional strategies. The capabilities of cloud can be employed in order to obtain omnipresent functionality of the application. And also, certain data which is regarded as sensitive can be protected using some data leakage detection and prevention techniques. Prioritization of data by the user helps unfolding the flexibility of backup methodology and the retrieval technique can be moulded as per the user request and this strategy helps handling the performance issues better.

IJESAT | Jan-Feb 2012

Available online @ http://www.ijesat.org 109

SOMU VENKATESWARLU* et al. [IJESAT] INTERNATIONAL JOURNAL OF ENGINEERING SCIENCE & ADVANCED TECHNOLOGY

ISSN: 22503676
Volume-2, Special Issue-1, 105 110

6. CONCLUSION
Energy on mobile phones is a precious resource. As phones with multiple wireless technologies such as 3G, GSM, and Wi Fi become commonplace, it is important to understand their relative energy consumption characteristics and to get to the best strategy of utilizing this power in the most efficient manner. By the application of the proposed strategy of the mobile content backup system which is utilizing the minimum power resource. By the effect of the above strategy the maximum of 65% of total energy was saved and as this also helps in controlling the radiation that affects the human intensely. [9] Reducing the Power Consumption of Wireless Mobile Devices with Multiple Radio Interfaces by Trevor Pering, Yuvraj Agarwal, Rajesh Gupta and Roy Want [10] Third generation http://www.3gpp.org [11] partnership project (3GPP)

3G: Wikipedia. Httpd://en.wikipedia.org/wiki/3G

[12] content backup of mobile devices and applications http://www.synchronoss.com/resources/industrytopics/connected-devices/content-backup.html

BIOGRAPHIES REFERENCES
[1] C.-C. Lee, J.-H. Yeh, and J.-C. Chen. Impact of inactivity timer on energy consumption in wcdma and cdma2000. In Proceedings of the Third Annual Wireless Telecommunication Symposium (WTS). IEEE, 2004 [2] Energy Consumption in Mobile Phones: A Measurement Study and Implications for Network Applications by Niranjan Balasubramanian, Aruna Balasubramanian and Arun Venkataramani [3] A. Rahmati and L. Zhong. Context-for-wireless: context-sensitive energy-efficient wireless data transfer. In MobiSys 07: Proceedings of the 5th international conference on Mobile systems, applications and services, pages 165178, New York, NY, USA, 2007. ACM. [4] Applicability of Different Models of Burstiness to Energy Consumption Estimation by Kazi Wali Ullah, Aalto University. [5] Apache web http://httpd.apache.org/. server. Website, 2011.

Professor Somu Venkateswarlu, KL University

P Tanuj Prabhakar, KLCE

M Ravi Teja, KLCE

[6] Bandwidth controller software, standard edition. ebsite, 2011. http://bandwidthcontroller.com/download.html. [7] Nokia energy profiler software. Website, 011.http://www.forum.nokia.com/Library/Tools_and_downloa ds/Other/Nokia_Energy_Profier/.

Sk. Khaja Moin, KLCE

K. Venkateswara Rao, KLCE

[8] Improving battery life for wireless web services through the use of a mobile proxy by Mandyam, G.D. in Personal Indoor and Mobile Radio Communications (PIMRC), 2010 IEEE 21st International Symposium

IJESAT | Jan-Feb 2012

Available online @ http://www.ijesat.org 110