Academia.eduAcademia.edu

Figure 4 – uploaded by Journal of Computer Science IJCSIS

See full PDFdownloadDownload figure
properties can be presented as follows [23],[24]:

Figure 4 properties can be presented as follows [23],[24]:

Related Figures (97)

TABLE I: The Details of Two Databases
TABLE I: The Details of Two Databases
a, First Row Original Image b, Second Row - Computation of Facial Boundartes c. Third Row - Extracted Face d. Fourth Row - Extracted Peatues Fig. 3: A Sample Facial Images - Put Face Database.
a, First Row Original Image b, Second Row - Computation of Facial Boundartes c. Third Row - Extracted Face d. Fourth Row - Extracted Peatues Fig. 3: A Sample Facial Images - Put Face Database.
Fig. 2: Features in 2-D Geometrical Space [26],[27].
Fig. 2: Features in 2-D Geometrical Space [26],[27].
a. First Row -Original Imaye b. Second Row - Computation of Facial Boundaries c. Third Row- Extracted Face d. Fourth Row - Extracted Features Fig. 4: A Sample Facial Images - Indian Face Database
a. First Row -Original Imaye b. Second Row - Computation of Facial Boundaries c. Third Row- Extracted Face d. Fourth Row - Extracted Features Fig. 4: A Sample Facial Images - Indian Face Database
A Euclidean metric matrix Q is being developed based on the normalized spatial distances (i.e. spatial relation- ships between two points) between the pixels of the re- spective biometric features. Therefore, the according to Eq. (5) and Eq. (6), the Euclidean geometrical formula for the Consider two images that can be written as the vectors P = [p1 po p3...-Pn] and q = [qi g2 Q3...--Gn]. According to Section III-A, the distance between the two images in the Euclidean domain can be stated as follows:
A Euclidean metric matrix Q is being developed based on the normalized spatial distances (i.e. spatial relation- ships between two points) between the pixels of the re- spective biometric features. Therefore, the according to Eq. (5) and Eq. (6), the Euclidean geometrical formula for the Consider two images that can be written as the vectors P = [p1 po p3...-Pn] and q = [qi g2 Q3...--Gn]. According to Section III-A, the distance between the two images in the Euclidean domain can be stated as follows:
Fig. 5: Extraction of Facial Features - Put Face Database. Fig. 6: Searching and Matching Process
Fig. 5: Extraction of Facial Features - Put Face Database. Fig. 6: Searching and Matching Process
TABLE II: Performance Evaluation in (%)-CRR Compar- ison
TABLE II: Performance Evaluation in (%)-CRR Compar- ison
TABLE IV: Performance Evaluation in (%) - FAR, FRR and EER Comparison
TABLE IV: Performance Evaluation in (%) - FAR, FRR and EER Comparison
TABLE III: Performance Evaluation in (%) - FAR, FRR and EER Comparison
TABLE III: Performance Evaluation in (%) - FAR, FRR and EER Comparison
Fig. 8: Verification - Performance Evaluation. Fig. 7: Identification - Performance Comparison
Fig. 8: Verification - Performance Evaluation. Fig. 7: Identification - Performance Comparison
Fig. 10: Verification - Performance Evaluation. Fig. 9: Verification - Performance Evaluation.
Fig. 10: Verification - Performance Evaluation. Fig. 9: Verification - Performance Evaluation.
TABLE V: Performance Evaluation in (%)-EER Compar- ison TABLE VI: Average Execution Time in Seconds
TABLE V: Performance Evaluation in (%)-EER Compar- ison TABLE VI: Average Execution Time in Seconds
Figure 1. Structure of QR Code well as some information bits that will tell a QR decoder what type of QR Code it is. After generating the string of bits; the Reed-Solomon technique is used to generate Error Correction [21]. The resultant data from string of bits and the Error Correction is used to generate eight different QR Codes, Each of which uses a different mask pattern. A mask pattern controls and changes the pixels to black 0 or White 1. Which makes sure that the QR code doesn’t contain patterns that might be difficult for a QR decoder to read [21]. Finally, the QR Code which uses the best mask pattern is generated as shown in figure 1.
Figure 1. Structure of QR Code well as some information bits that will tell a QR decoder what type of QR Code it is. After generating the string of bits; the Reed-Solomon technique is used to generate Error Correction [21]. The resultant data from string of bits and the Error Correction is used to generate eight different QR Codes, Each of which uses a different mask pattern. A mask pattern controls and changes the pixels to black 0 or White 1. Which makes sure that the QR code doesn’t contain patterns that might be difficult for a QR decoder to read [21]. Finally, the QR Code which uses the best mask pattern is generated as shown in figure 1.
A. Initialization Phase The major aim of the proposed protocol the drawbacks of password guessing attack is to eliminate . The proposed protocol uses OTP encrypted by RC4-EA method, then hid- ing cipher-OTP in QR code. Also, it integrates a web-based applications and mobile devices for user authentication over multi-channels. The proposed protocol invo ves two parties : a server (.S) and a remote user (U). Each authorized U can request service from S with the granted access rights. In addition, each U got an electronic mail and hold a mobile device. The protocol consists of four phases : initialization phase, registration phase, login phase and phase. The notations employed throughout shown in table I. authentication this paper are In this phase, Internet Protocol Authentication ([PAuth) is a protocol suite for securing internet communications by authenticating each IP packet of a communication session. IPAuth takes place between two parties of a server and a user. The various steps of [PAuth will be explain below:
A. Initialization Phase The major aim of the proposed protocol the drawbacks of password guessing attack is to eliminate . The proposed protocol uses OTP encrypted by RC4-EA method, then hid- ing cipher-OTP in QR code. Also, it integrates a web-based applications and mobile devices for user authentication over multi-channels. The proposed protocol invo ves two parties : a server (.S) and a remote user (U). Each authorized U can request service from S with the granted access rights. In addition, each U got an electronic mail and hold a mobile device. The protocol consists of four phases : initialization phase, registration phase, login phase and phase. The notations employed throughout shown in table I. authentication this paper are In this phase, Internet Protocol Authentication ([PAuth) is a protocol suite for securing internet communications by authenticating each IP packet of a communication session. IPAuth takes place between two parties of a server and a user. The various steps of [PAuth will be explain below:
Figure: 1 Conceptual diagram of RFID Reader In its easiest structure, RFID [7, 8] is an idea like standardized identification innovation, yet without obliging an immediate perceivability of the checked substances as presented in Figure 1. Much the same as standardized tag frameworks oblige a legitimate optical peruser and unique labels connected on RFID’s, RFID needs a peruser hardware and exceptional labels or cards appended to the student’s RFID’s all together for the readers to be followed
Figure: 1 Conceptual diagram of RFID Reader In its easiest structure, RFID [7, 8] is an idea like standardized identification innovation, yet without obliging an immediate perceivability of the checked substances as presented in Figure 1. Much the same as standardized tag frameworks oblige a legitimate optical peruser and unique labels connected on RFID’s, RFID needs a peruser hardware and exceptional labels or cards appended to the student’s RFID’s all together for the readers to be followed
RFID’s have an extensive history for a piece to mechanical transformation together present besides historical [9, 10]. RFID empowers brisk installment for rings in addition speedy ID of things. Furthermore, RFID gives advantages, for example, following resources, checking conditions for wellbeing, and serving to anticipate duplicating. RFID has indispensable influence in the innovative insurgency alongside the Internet in addition cell phones that have been associating the world collected wholly RFID frameworks, which comprise three fundamental segments. The primary of RFID’s labels, which are appended to a benefit otherwise thing. The label (tag) encompasses data approximately, which advantage otherwise thing furthermore possibly will consolidate sensors and devices [11].
RFID’s have an extensive history for a piece to mechanical transformation together present besides historical [9, 10]. RFID empowers brisk installment for rings in addition speedy ID of things. Furthermore, RFID gives advantages, for example, following resources, checking conditions for wellbeing, and serving to anticipate duplicating. RFID has indispensable influence in the innovative insurgency alongside the Internet in addition cell phones that have been associating the world collected wholly RFID frameworks, which comprise three fundamental segments. The primary of RFID’s labels, which are appended to a benefit otherwise thing. The label (tag) encompasses data approximately, which advantage otherwise thing furthermore possibly will consolidate sensors and devices [11].
Figure 3: Internet of Things schematic demonstrating the end clients and application zones taking into account information.
Figure 3: Internet of Things schematic demonstrating the end clients and application zones taking into account information.
Figure 4: Future application scenario for school using RFID on loT
Figure 4: Future application scenario for school using RFID on loT
At the point when the students get the item with RFID hostile to forging name, they utilize the mark code for right of entry to school network against falsifying data administrations address through RFID-empowered cellular telephones or Internet-joined PCs furnished through read-compose labels of RFID, and afterward request administrations to achieve item correlated data for recognize this legitimacy on item. Compelling on RFID-empowered cell telephone to instance of demonstrate particular hostile to duplicating steps when the buyer needs to distinguish the realness of the item.
At the point when the students get the item with RFID hostile to forging name, they utilize the mark code for right of entry to school network against falsifying data administrations address through RFID-empowered cellular telephones or Internet-joined PCs furnished through read-compose labels of RFID, and afterward request administrations to achieve item correlated data for recognize this legitimacy on item. Compelling on RFID-empowered cell telephone to instance of demonstrate particular hostile to duplicating steps when the buyer needs to distinguish the realness of the item.
Figure: Process flow of RFID application
Figure: Process flow of RFID application
Table 1. Sample Profile
Table 1. Sample Profile
We distributed a bilingual questionnaire to study the potential of accepting biometrics as an authentication tool. Respondents to this study found it a good idea and indicated an intention to use it in the future if it happened to be available. Obviously, our findings confirm the previous results. Results re- veal that perceived ease of use and perceived of usefulness are significant drivers of the behavior of intention to use biometrics as an authentication ool. In addition, this study results revealed perceived usefulness to be the most crucial factor in the decision to adopt new technologies. According to his study, security remained as a significant factor to affect the behavior of users. Moreover, we found a tight relationship between acceptance and age; young people showed more apatite to accept new biometrics interface.
We distributed a bilingual questionnaire to study the potential of accepting biometrics as an authentication tool. Respondents to this study found it a good idea and indicated an intention to use it in the future if it happened to be available. Obviously, our findings confirm the previous results. Results re- veal that perceived ease of use and perceived of usefulness are significant drivers of the behavior of intention to use biometrics as an authentication ool. In addition, this study results revealed perceived usefulness to be the most crucial factor in the decision to adopt new technologies. According to his study, security remained as a significant factor to affect the behavior of users. Moreover, we found a tight relationship between acceptance and age; young people showed more apatite to accept new biometrics interface.
The main part of the system is the data preparation. In the first step, the weather data is collected consisted of Temperature, Sea level Pressure, Dew Point, Humidity, Wind Speed, Visibility and Precipitation. The author found from their analysis that instead of yearly data, the predictability is efficient if the data is prepared according to seasons [19]. So, in The proposed system steps are as follows: The system is proposed to utilize temporal association rule mining to associate temperature with “tithis’. The proposed system is as shown in the following Figure 1. For the temporal association rule mining, used is intertransactional association rule mining to mine the association across the transactions instead the intra transactions.
The main part of the system is the data preparation. In the first step, the weather data is collected consisted of Temperature, Sea level Pressure, Dew Point, Humidity, Wind Speed, Visibility and Precipitation. The author found from their analysis that instead of yearly data, the predictability is efficient if the data is prepared according to seasons [19]. So, in The proposed system steps are as follows: The system is proposed to utilize temporal association rule mining to associate temperature with “tithis’. The proposed system is as shown in the following Figure 1. For the temporal association rule mining, used is intertransactional association rule mining to mine the association across the transactions instead the intra transactions.
then by replacing the image equivalent of the feature on the face mask performing proper pasting and smoothening such that the image looks genuine..
then by replacing the image equivalent of the feature on the face mask performing proper pasting and smoothening such that the image looks genuine..
When an input is passed to the method the primary task is to use feature detection methodologies and detect the prominent facial features. G In order to generate patterns from time series data for classification purposes several feature extraction methods have been introduced. A statistical measure of the amplitude of the time series is provided by the well known kurtosis method. Another method displays construction of a feature vector using the spectrum, where the power spectral density and the wavelet coefficients are used along with PCA for feature extraction. For extracting the phase information Hilbert transform requires conversion of the real-valued signal into complex- valued analytic signal. Time series data is predefined in the SDF-based feature extraction, which is first converted into symbol sequences, and then probabilistic finite-state automata are constructed from these symbol equines for compressing the pertinent information into low-dimensional statistical patterns. DF-based feature extraction from (wavelet- transformed) time series has been proposed by Jin et al. (2012 for target detection and classification in border regions. The time-frequency localization and demonizing of the underlying sensor time series leads for the use of rationale wavelet- based methods. However, this method requires selection and tuning of several parameters (e.g., wavelet basis function and scales for signal pre-processing in addition to the size of the symbo alphabet that is needed for SDF. Use of Cascade object detector is done in order to detect and extract the features.
When an input is passed to the method the primary task is to use feature detection methodologies and detect the prominent facial features. G In order to generate patterns from time series data for classification purposes several feature extraction methods have been introduced. A statistical measure of the amplitude of the time series is provided by the well known kurtosis method. Another method displays construction of a feature vector using the spectrum, where the power spectral density and the wavelet coefficients are used along with PCA for feature extraction. For extracting the phase information Hilbert transform requires conversion of the real-valued signal into complex- valued analytic signal. Time series data is predefined in the SDF-based feature extraction, which is first converted into symbol sequences, and then probabilistic finite-state automata are constructed from these symbol equines for compressing the pertinent information into low-dimensional statistical patterns. DF-based feature extraction from (wavelet- transformed) time series has been proposed by Jin et al. (2012 for target detection and classification in border regions. The time-frequency localization and demonizing of the underlying sensor time series leads for the use of rationale wavelet- based methods. However, this method requires selection and tuning of several parameters (e.g., wavelet basis function and scales for signal pre-processing in addition to the size of the symbo alphabet that is needed for SDF. Use of Cascade object detector is done in order to detect and extract the features.
Application of principal component analysis is done on each image by the Eigen Object Recognizer class, the results of which will be an array of Eigen values which can be recognized by a Neural Network which is trained. PCA is a frequently used method of object recognition as its results, can be fairly accurate and resilient to noise. The method of which PCA is applied can vary at different stages so what will be demonstrated is a clear method for PCA application that can be followed. It is up for individuals to experiment in finding the best method for producing accurate results from PCA. To perform PCA several steps are undertaken:
Application of principal component analysis is done on each image by the Eigen Object Recognizer class, the results of which will be an array of Eigen values which can be recognized by a Neural Network which is trained. PCA is a frequently used method of object recognition as its results, can be fairly accurate and resilient to noise. The method of which PCA is applied can vary at different stages so what will be demonstrated is a clear method for PCA application that can be followed. It is up for individuals to experiment in finding the best method for producing accurate results from PCA. To perform PCA several steps are undertaken:
Fig. V — Approach for detecting location of components. The outcome of all will be a patched form of imag which will have all the image components in their appropriat locations but the facial appearance of the image may not be a pleasant as expected. Resizing the components and realignin them is thus a necessity, for which application of certai filtering and smoothening algorithms is conducted.
Fig. V — Approach for detecting location of components. The outcome of all will be a patched form of imag which will have all the image components in their appropriat locations but the facial appearance of the image may not be a pleasant as expected. Resizing the components and realignin them is thus a necessity, for which application of certai filtering and smoothening algorithms is conducted.
Fig. VI— Approach for registering of features The output is finalized by performing gradient smoothening on the image and image blending algorithms are executed for obtaining a perfect outcome, the output is a visually pleasant image form as shown below,
Fig. VI— Approach for registering of features The output is finalized by performing gradient smoothening on the image and image blending algorithms are executed for obtaining a perfect outcome, the output is a visually pleasant image form as shown below,
Font: https://www.owasp.org/index.php/Top_ 10 2010- Main, viewed at: 05/08/2015. It is also true when a password is requested, since the malicious code described herein can work with a SQL whose where has more parameters such as: "Select * from customers where username = ‘$user name' and password = '$password'." Figure 1 - Attacks on Web A pplications Source:
Font: https://www.owasp.org/index.php/Top_ 10 2010- Main, viewed at: 05/08/2015. It is also true when a password is requested, since the malicious code described herein can work with a SQL whose where has more parameters such as: "Select * from customers where username = ‘$user name' and password = '$password'." Figure 1 - Attacks on Web A pplications Source:
Post Graduate (Lato sensu ) Specialization: Security in Computer Networks
Post Graduate (Lato sensu ) Specialization: Security in Computer Networks
Fig.1 Basic biometrics surveillance system for triangle approach Result: Based on input, output is generated.
Fig.1 Basic biometrics surveillance system for triangle approach Result: Based on input, output is generated.
After reviewing different papers and work on these approaches, the correct classification rate (CCR) obtained by different researchers in field of triangle based gait can be summarized as following table which results are shown in the table on the basis of their CCR rates and number of people under surveillance.
After reviewing different papers and work on these approaches, the correct classification rate (CCR) obtained by different researchers in field of triangle based gait can be summarized as following table which results are shown in the table on the basis of their CCR rates and number of people under surveillance.
In this review paper, we present a comparison between different approaches the techniques which are based on triangle of different body parts. In this review paper, we discussed only lower body part analysis. Here we took those work which is based on common components. All techniques used triangle based method they construct triangle by feature extraction and then recognition is done by various methods. By this review paper, it has been observed that the angle based gait detection and novel method give the same results.
In this review paper, we present a comparison between different approaches the techniques which are based on triangle of different body parts. In this review paper, we discussed only lower body part analysis. Here we took those work which is based on common components. All techniques used triangle based method they construct triangle by feature extraction and then recognition is done by various methods. By this review paper, it has been observed that the angle based gait detection and novel method give the same results.
Fig.1.Various areas of Sanskrit addressed by computer Fig.1. shows various areas of Sanskrit Literature addressed by computer. The legendary Sanskrit grammarian of 5th century BC, Panini is the world's first computational grammarian. Panini wrote Ashtadhyayi (the Eight-Chaptered book) [15], which is considered to be the most comprehensive scientific grammar ever written for any language. Many approaches were proposed by scientists and grammarians world-wide, to extract the richness of Sanskrit language in various contexts. Thin Dawes bi wns Anes KE Ul nee ne ChlaAlas te alana fhaaA development. The well-knit (i. e. syntactically and semantically strong) structure of this language, has encouraged current research. Sanskrit is a very scientific language. Its entire grammatical mechanism is perfected. From this perspective, Sanskrit grammar studies have received serious attention about the truthful representation of communication worldwide. Fig 10> chowe various areas of Sanckrit 7] iterahire addrecced hv
Fig.1.Various areas of Sanskrit addressed by computer Fig.1. shows various areas of Sanskrit Literature addressed by computer. The legendary Sanskrit grammarian of 5th century BC, Panini is the world's first computational grammarian. Panini wrote Ashtadhyayi (the Eight-Chaptered book) [15], which is considered to be the most comprehensive scientific grammar ever written for any language. Many approaches were proposed by scientists and grammarians world-wide, to extract the richness of Sanskrit language in various contexts. Thin Dawes bi wns Anes KE Ul nee ne ChlaAlas te alana fhaaA development. The well-knit (i. e. syntactically and semantically strong) structure of this language, has encouraged current research. Sanskrit is a very scientific language. Its entire grammatical mechanism is perfected. From this perspective, Sanskrit grammar studies have received serious attention about the truthful representation of communication worldwide. Fig 10> chowe various areas of Sanckrit 7] iterahire addrecced hv
(IJ CSIS) International J ournal of Computer Science and Information Security, Vol. 13, No.6, June 2015 Fig.2. Block diagram of the system
(IJ CSIS) International J ournal of Computer Science and Information Security, Vol. 13, No.6, June 2015 Fig.2. Block diagram of the system
TABLE.1. LIST OF GANA WITH EXAMPLE
TABLE.1. LIST OF GANA WITH EXAMPLE
Figure 3 illustrates a method of finding a Guru in the Sanskrit Shloka
Figure 3 illustrates a method of finding a Guru in the Sanskrit Shloka
Figure 4 illustrates a method of finding a Laghu in the Sanskrit Shloka Fig.3. Rules of marking Guru The overall flow of system implementation can be understood from the block diagram in Fig.5
Figure 4 illustrates a method of finding a Laghu in the Sanskrit Shloka Fig.3. Rules of marking Guru The overall flow of system implementation can be understood from the block diagram in Fig.5
(J CSIS) International J ournal of Computer Science and Information Security Vol. 13, No.6, June 201! Fig.5. Implementation flow of the system The designed system accepts Sanskrit Shloka in Unicode format. Unicode provides a unique number for every character irrespective of the platform and the language. UTF- 8 encodes each Unicode character. Ones the Shloka is _ stored in UTF-8 format, following process will be carried on:
(J CSIS) International J ournal of Computer Science and Information Security Vol. 13, No.6, June 201! Fig.5. Implementation flow of the system The designed system accepts Sanskrit Shloka in Unicode format. Unicode provides a unique number for every character irrespective of the platform and the language. UTF- 8 encodes each Unicode character. Ones the Shloka is _ stored in UTF-8 format, following process will be carried on:
Fig.8. Musical Notations of Vrutta Hansagati (IJ CSIS) International J ournal of Computer Science and Information Security, Vol. 13, No.6, June 2015
Fig.8. Musical Notations of Vrutta Hansagati (IJ CSIS) International J ournal of Computer Science and Information Security, Vol. 13, No.6, June 2015
Analysis of the System
Analysis of the System
TABLE I: Proposed Connected E-H shape Patch Antenna Design Parameters
TABLE I: Proposed Connected E-H shape Patch Antenna Design Parameters
TABLE II: Design Parameters of the Proposed E-shape Patch Antenna
TABLE II: Design Parameters of the Proposed E-shape Patch Antenna
Figure 1. Top view of the Connected E-H shaped antenna The Connected E-H shaped microstrip patch antenna is simpler in construction. The geometry is shown in Fig. | with box wall port which is the most common types of port that use reference plane to removes the effects of the transmission line effect. Patch is designed and simulated over Sonnet Software is a planar 3D electromagnetic simulator. The proposed Connected E-H shape microstrip patch antenna design parameter is shown in Table I.
Figure 1. Top view of the Connected E-H shaped antenna The Connected E-H shaped microstrip patch antenna is simpler in construction. The geometry is shown in Fig. | with box wall port which is the most common types of port that use reference plane to removes the effects of the transmission line effect. Patch is designed and simulated over Sonnet Software is a planar 3D electromagnetic simulator. The proposed Connected E-H shape microstrip patch antenna design parameter is shown in Table I.
Figure 2. Top view of the E-shaped antenna The proposed Connected E-H shape microstrip patch antenna design parameter is shown in Table II.
Figure 2. Top view of the E-shaped antenna The proposed Connected E-H shape microstrip patch antenna design parameter is shown in Table II.
Figure 3. Simulated Return Loss curve of Connected E-H_ shaped patch antenna The first important parameter which is help ful to calculate the bandwidth of the antenna structure is its S11 in decibel versus frequency. During this antenna feeding has been done at the point where the return loss is minimized. The return loss curve of the designed antenna is presented in Fig. 3, and minimum S11 level of -30.39 dB is shown in m3 caption. The figure shows that the antenna resonates at 3.4GH z band.
Figure 3. Simulated Return Loss curve of Connected E-H_ shaped patch antenna The first important parameter which is help ful to calculate the bandwidth of the antenna structure is its S11 in decibel versus frequency. During this antenna feeding has been done at the point where the return loss is minimized. The return loss curve of the designed antenna is presented in Fig. 3, and minimum S11 level of -30.39 dB is shown in m3 caption. The figure shows that the antenna resonates at 3.4GH z band.
Figure 4. Input impedance curve of Connected E- H shaped patch antenna
Figure 4. Input impedance curve of Connected E- H shaped patch antenna
The physical meaning of current density distribution is that it is a measure how the antenna is producing a beam. Figure 5. Current density diagram of the Connected E-H shaped patch antenna
The physical meaning of current density distribution is that it is a measure how the antenna is producing a beam. Figure 5. Current density diagram of the Connected E-H shaped patch antenna
Figure 7. Input impedance curve of E-shaped patch antenna nearly 56.25% of the center frequency.
Figure 7. Input impedance curve of E-shaped patch antenna nearly 56.25% of the center frequency.
Figure 6. Simulated Return Loss of E-shaped patch antenna the designed antenna is presented in Fig. 6, and minimum S11 level of -33.3 dB is shown in m3 caption. The figure shows that the antenna resonates at 3.2GHz band.
Figure 6. Simulated Return Loss of E-shaped patch antenna the designed antenna is presented in Fig. 6, and minimum S11 level of -33.3 dB is shown in m3 caption. The figure shows that the antenna resonates at 3.2GHz band.
Figure-8: Current density diagram of the E-shaped patch antenna at 3.2 GHz
Figure-8: Current density diagram of the E-shaped patch antenna at 3.2 GHz
As authors mentioned at the introduction section, the target of this study is to gather baseline data of how illiterates recognize and react to the mobile interface. This empirical study involves five novice illiterate participants. All users have no previous background of using mobile phones. Some other empirical studies involve only seven novice participants [1]. For some empirical study investigations, the baseline data is more important than the number of participants. The baseline data will be used for further investigations that involve more participants. The profile of the five novice illiterate participants is shown in table 1. Figure 1. Smart phone mobile numbers call interface
As authors mentioned at the introduction section, the target of this study is to gather baseline data of how illiterates recognize and react to the mobile interface. This empirical study involves five novice illiterate participants. All users have no previous background of using mobile phones. Some other empirical studies involve only seven novice participants [1]. For some empirical study investigations, the baseline data is more important than the number of participants. The baseline data will be used for further investigations that involve more participants. The profile of the five novice illiterate participants is shown in table 1. Figure 1. Smart phone mobile numbers call interface
The results of the tasks for the five participants, see table 1, are shown in figure 2. The results in figure 2 show the dialing and calling (i.e. talking) times in seconds for every participant plus the average. It is clear, from figure 2, that the dialing time is substantially greater. Figure 2. The results of the tasks for the five illiterate participants plus the average of the dialing time and calling (i.e. talking) time.
The results of the tasks for the five participants, see table 1, are shown in figure 2. The results in figure 2 show the dialing and calling (i.e. talking) times in seconds for every participant plus the average. It is clear, from figure 2, that the dialing time is substantially greater. Figure 2. The results of the tasks for the five illiterate participants plus the average of the dialing time and calling (i.e. talking) time.
RESULTS OF THE SECOND EXPERIMENT
RESULTS OF THE SECOND EXPERIMENT
The system is composed mainly of two parts: the application server that contains all mobile resources (e.g., pages, games, database, etc.) and the mobile application which sends queries to the server to load the required resource elements, as shown in figure 4. Figure 4. System architecture
The system is composed mainly of two parts: the application server that contains all mobile resources (e.g., pages, games, database, etc.) and the mobile application which sends queries to the server to load the required resource elements, as shown in figure 4. Figure 4. System architecture
Figure 3. System components (need to add component of multimedia generator)
Figure 3. System components (need to add component of multimedia generator)
By clicking the icon of the boy playing football; on the top row in Snapshot 1, the screen shown in figure 6, will be displayed. The icon in the upper row can be clicked to go back to the previous screen. When the user clicked the icon “possible” (Arabic: Momken), it is possible for him to click he icon “drink” (Arabic: Ashrab) located in the middle, therefore, a collection of drink flavors will be displayed to allow him to select the flavor he wants, as shown in figure 7.
By clicking the icon of the boy playing football; on the top row in Snapshot 1, the screen shown in figure 6, will be displayed. The icon in the upper row can be clicked to go back to the previous screen. When the user clicked the icon “possible” (Arabic: Momken), it is possible for him to click he icon “drink” (Arabic: Ashrab) located in the middle, therefore, a collection of drink flavors will be displayed to allow him to select the flavor he wants, as shown in figure 7.
When the mobile application is launched, the following main screen will be displayed which allows the user to access the basic buttons and listen to the recorded voice associated with them, as shown in figure 5.
When the mobile application is launched, the following main screen will be displayed which allows the user to access the basic buttons and listen to the recorded voice associated with them, as shown in figure 5.
Figure 9. Snapshot-5 When the user enters the learning mode in the application, it will load the progress of the user stored in the mobile. The mobile application starts by reading available options by playing audio files and waits for the user to choose one: 1) study the Arabic numbers; 2) study the Arabic letters; 3) input Arabic text to retrieve multimedia elements and 4) play games to evaluate what the user had learned, as shown in figure 9.
Figure 9. Snapshot-5 When the user enters the learning mode in the application, it will load the progress of the user stored in the mobile. The mobile application starts by reading available options by playing audio files and waits for the user to choose one: 1) study the Arabic numbers; 2) study the Arabic letters; 3) input Arabic text to retrieve multimedia elements and 4) play games to evaluate what the user had learned, as shown in figure 9.
Various icons that have particular meaning will be displayed in the mobile application, if the user pressed and held any button for two seconds, a loading progress will appear on the button to play an explanation voice record which explains the usage of the button, as shown in figure 8.
Various icons that have particular meaning will be displayed in the mobile application, if the user pressed and held any button for two seconds, a loading progress will appear on the button to play an explanation voice record which explains the usage of the button, as shown in figure 8.
Before starting the game, the user can watch an animated cartoon lesson to learn about the specified topic. Each lesson does not exceed three minutes duration to keep the user’s attention. Snapshot 5, shown in figure 10, shows a mathematic lesson to perform basic objects enumeration. After watching the lesson, the user can play games related to the lesson for self evaluation. The following screens A and B , shown in figure 11, show two different game evaluation pages to evaluate what the user learned about the enumeration of objects. In screen A, the user is asked about selecting the appropriate number that is presented by the hand; in screen B, he is asked to sort numbers drawn on eggs in the right order.
Before starting the game, the user can watch an animated cartoon lesson to learn about the specified topic. Each lesson does not exceed three minutes duration to keep the user’s attention. Snapshot 5, shown in figure 10, shows a mathematic lesson to perform basic objects enumeration. After watching the lesson, the user can play games related to the lesson for self evaluation. The following screens A and B , shown in figure 11, show two different game evaluation pages to evaluate what the user learned about the enumeration of objects. In screen A, the user is asked about selecting the appropriate number that is presented by the hand; in screen B, he is asked to sort numbers drawn on eggs in the right order.
The following screens A and B, shown in figure 12, is another game pages to evaluate the user and what been learned about the order of the week days. In screen A, the user is asked to sort days by selecting the appropriate day written on the mushroom. In screen B, the user is required to see which day the boy is asking for, and select the appropriate day written on each leaf.
The following screens A and B, shown in figure 12, is another game pages to evaluate the user and what been learned about the order of the week days. In screen A, the user is asked to sort days by selecting the appropriate day written on the mushroom. In screen B, the user is required to see which day the boy is asking for, and select the appropriate day written on each leaf.
The screens A and B, shown in figure 13, are pages used to teach the user Arabic letters. The application recite Arabic letters, as shown in screen A, then several examples are given for each letter, as shown in screen B. The screens A and B, shown at figure 14, represents a game that evaluates the user through various questions in different styles about what had been learned about Arabic alphabetic letters. In screen A, the user is asked to select the image where its name starts with the presented letter. In screen B, the user is requested to connect the appropriate letter with its corresponding image that its name starts with the presented letters.
The screens A and B, shown in figure 13, are pages used to teach the user Arabic letters. The application recite Arabic letters, as shown in screen A, then several examples are given for each letter, as shown in screen B. The screens A and B, shown at figure 14, represents a game that evaluates the user through various questions in different styles about what had been learned about Arabic alphabetic letters. In screen A, the user is asked to select the image where its name starts with the presented letter. In screen B, the user is requested to connect the appropriate letter with its corresponding image that its name starts with the presented letters.
After watching the adding process lesson, the user can play games related to that lesson for self evaluation, as
After watching the adding process lesson, the user can play games related to that lesson for self evaluation, as
Snapshot 13, shown in figure 17, demonstrates an example of the feedback a user would get after entering Arabic text into the system that translates as, “A rabbit has brown or white fur. It eats carrots and moves around by jumping.”
Snapshot 13, shown in figure 17, demonstrates an example of the feedback a user would get after entering Arabic text into the system that translates as, “A rabbit has brown or white fur. It eats carrots and moves around by jumping.”
In the snapshot 12, shown in figure 16, an animated cartoon lesson that teaches a lesson about week days. It gives full explanation how to pronounce days in order and how they are ordered. The lesson explains how days are sorted by constructing and connecting train parts in order according to the written day on each part. shown at figure 15. The user is asked to select the correc! answer from the presented numbers.
In the snapshot 12, shown in figure 16, an animated cartoon lesson that teaches a lesson about week days. It gives full explanation how to pronounce days in order and how they are ordered. The lesson explains how days are sorted by constructing and connecting train parts in order according to the written day on each part. shown at figure 15. The user is asked to select the correc! answer from the presented numbers.
A number of researches concurred that the current mobile phone user interface design is not highly recommended or suitable for illiterate individuals [14]. They found that utilizing audio and graphic support in user interfaces is highly valuable for the enhanced cognition and usage friendliness for illiterate people [14].
A number of researches concurred that the current mobile phone user interface design is not highly recommended or suitable for illiterate individuals [14]. They found that utilizing audio and graphic support in user interfaces is highly valuable for the enhanced cognition and usage friendliness for illiterate people [14].
The major principle of the IaaS cloud computing system is that its user can make use of the resources to have good performance and economic benefits. With the support of virtualization innovation the resources can be conveyed to the users in the form of virtual machines hence an efficient virtual machine allocation policy and management process is required to avoid underutilization or overutilization of the physical machines which may affect the quality of services of the IaaS cloud. The under utilization of servers is a well known expenditure concern in cloud management. Low utilization of server resources leads to the usage of more physical machines, increasing expenses for machine power and capital and operational expenses for cooling systems. Moreover, surplus machines require more carbon footprint. Figure 1: Framework model for VM placement in a cluster of physical machines by virtual machine optimizer we used a modified version of genetic algorithm for load optimization. The above figure shows the framework model in which the proposed algorithm is implemented. Here the physical clusters can be formed by adding a set of physical servers each server contributing its own share of resources such as CPU cores, main memory, disk capacity and network bandwidth. The users can create virtual machine instances by giving their requirements for running the applications and the VM requests are submitted by the users to the computing system. As the submitted VMs enter to the cloud they are wait for their turn in the stack. The VM requests can be handled by the virtual machine scheduler and it finds the appropriate physical machine by estimating the VM size and checking for the availability and capacity of the physica machine when it finds the appropriate physical machine the VM scheduler immediately allocates the identified physica machine to the virtual machine instance request in queue and the required resource can be allocated to the virtua machine. Rebalancing of load in this environment is handled
The major principle of the IaaS cloud computing system is that its user can make use of the resources to have good performance and economic benefits. With the support of virtualization innovation the resources can be conveyed to the users in the form of virtual machines hence an efficient virtual machine allocation policy and management process is required to avoid underutilization or overutilization of the physical machines which may affect the quality of services of the IaaS cloud. The under utilization of servers is a well known expenditure concern in cloud management. Low utilization of server resources leads to the usage of more physical machines, increasing expenses for machine power and capital and operational expenses for cooling systems. Moreover, surplus machines require more carbon footprint. Figure 1: Framework model for VM placement in a cluster of physical machines by virtual machine optimizer we used a modified version of genetic algorithm for load optimization. The above figure shows the framework model in which the proposed algorithm is implemented. Here the physical clusters can be formed by adding a set of physical servers each server contributing its own share of resources such as CPU cores, main memory, disk capacity and network bandwidth. The users can create virtual machine instances by giving their requirements for running the applications and the VM requests are submitted by the users to the computing system. As the submitted VMs enter to the cloud they are wait for their turn in the stack. The VM requests can be handled by the virtual machine scheduler and it finds the appropriate physical machine by estimating the VM size and checking for the availability and capacity of the physica machine when it finds the appropriate physical machine the VM scheduler immediately allocates the identified physica machine to the virtual machine instance request in queue and the required resource can be allocated to the virtua machine. Rebalancing of load in this environment is handled
Table I: Properties required for the index table of physical machine and physical machine cluster
Table I: Properties required for the index table of physical machine and physical machine cluster
The objective function of our algorithm wants to minimize The objective function of our algorithm wants to minimize the standard deviation of the remaining CPU, RAM and Network Bandwidth in each host. As we consider that the load of the entire physical cluster instead of taking into consideration of the total number of virtual machines in each physical host as a load balance metric we developed an objective function that tries to balance the consumption of CPU, RAM and Network Bandwidth on each host, in view of a heterogeneous environment, which consists of different hosts with different configurations.
The objective function of our algorithm wants to minimize The objective function of our algorithm wants to minimize the standard deviation of the remaining CPU, RAM and Network Bandwidth in each host. As we consider that the load of the entire physical cluster instead of taking into consideration of the total number of virtual machines in each physical host as a load balance metric we developed an objective function that tries to balance the consumption of CPU, RAM and Network Bandwidth on each host, in view of a heterogeneous environment, which consists of different hosts with different configurations.
Fig 3: Comparision of the overall resource utilization rate Though the numbers of servers used by the first fit and proposed algorithms are comparatively stable from figure 3 we can see that the resource utilization rate of our algorithm is appreciably outperforms the other three algorithms. is appreciably outperforms the other three algorithms. Fig 4 shows that the response time of all the algorithms are comparatively stable our algorithm takes little bit more time to allocate VMs than the first fit algorithm because of its nature of allocating VMs based on the user provided information and past usage history of the VMs
Fig 3: Comparision of the overall resource utilization rate Though the numbers of servers used by the first fit and proposed algorithms are comparatively stable from figure 3 we can see that the resource utilization rate of our algorithm is appreciably outperforms the other three algorithms. is appreciably outperforms the other three algorithms. Fig 4 shows that the response time of all the algorithms are comparatively stable our algorithm takes little bit more time to allocate VMs than the first fit algorithm because of its nature of allocating VMs based on the user provided information and past usage history of the VMs
Fig 2: Comparision of the number of Physical Servers Though the numbers of servers used by the first fit and
Fig 2: Comparision of the number of Physical Servers Though the numbers of servers used by the first fit and
Fig 5: Comparision of the percentage of Load Imbalance Value Our proposed algorithm is effective in improving the resource utilization rate and load balancing with the help of live migrations. But one of our major aims is increasing the total revenue which requires cutting down the VM migration cost which can be achieved by reducing the percentage of VM migration rate. We use migration rate as the estimation metric which is VMs to the tota results in the defined as the percentage of the migrated number of VM instances. We showed the following Fig. 6.The proposed algorithm decreases the migrating rate from about 18%-20% to less than 13 % which leads to reduce the VM migration cost. Though the curve of our proposed algorithm indicates that only less number of VMs migrated from their original host
Fig 5: Comparision of the percentage of Load Imbalance Value Our proposed algorithm is effective in improving the resource utilization rate and load balancing with the help of live migrations. But one of our major aims is increasing the total revenue which requires cutting down the VM migration cost which can be achieved by reducing the percentage of VM migration rate. We use migration rate as the estimation metric which is VMs to the tota results in the defined as the percentage of the migrated number of VM instances. We showed the following Fig. 6.The proposed algorithm decreases the migrating rate from about 18%-20% to less than 13 % which leads to reduce the VM migration cost. Though the curve of our proposed algorithm indicates that only less number of VMs migrated from their original host
Fig 6: Comparision of the Percentage of VM Migarations for Load Balancing to a new host we achieved the better resource utilizatior benefit and balanced load among the physical hosts.
Fig 6: Comparision of the Percentage of VM Migarations for Load Balancing to a new host we achieved the better resource utilizatior benefit and balanced load among the physical hosts.
Fig 7: Comparision of the Percentage of VMs that violate their SLA IX. CONCLUSION AND FUTURE WORK We presented our novel algorithm that considers user constraints of VM along with physical host load factor to address the problem of mapping the VMs into PMs such that the number physical host used is minimized, the overutilization and underutilization of the resources of a host can be identified and resolved at the same time without violating any SLA agreements. Since we consider this as a multi potential bin packing problem we combined three different heuristics which considers load factor of hosts along with user provided information at the various stages of placing the VMs in physical hosts. Based on our analysis we
Fig 7: Comparision of the Percentage of VMs that violate their SLA IX. CONCLUSION AND FUTURE WORK We presented our novel algorithm that considers user constraints of VM along with physical host load factor to address the problem of mapping the VMs into PMs such that the number physical host used is minimized, the overutilization and underutilization of the resources of a host can be identified and resolved at the same time without violating any SLA agreements. Since we consider this as a multi potential bin packing problem we combined three different heuristics which considers load factor of hosts along with user provided information at the various stages of placing the VMs in physical hosts. Based on our analysis we
Related topics:
Information SystemsComputer ScienceSoftware EngineeringArtificial IntelligenceInformation TechnologyImage ProcessingInformation SecurityComputer NetworksSoftware
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved