Sunjiv Soyjaudah - Academia.edu (original) (raw)
Papers by Sunjiv Soyjaudah
Intelligent Automation and Soft Computing, 2018
Indonesian Journal of Electrical Engineering and Computer Science, Jun 1, 2017
H.264/AVCis currently the most widely adopted video coding standard due to its high compression c... more H.264/AVCis currently the most widely adopted video coding standard due to its high compression capability and flexibility. However, compressed videos are highly vulnerable to channel errors which may result in severe quality degradation of a video. This paper presentsa concealment aware Unequal Error Protection (UEP) scheme for H.264 video compression using Reed Solomon (RS) codes. The proposed UEP technique assigns a code rate to each Macroblock (MB) based on the type of concealment and a Concealment Dependent Index (CDI). Two interleaving techniques, namely Frame Level Interleaving (FLI) and Group Level Interleaving (GLI) have also been employed. Finally, prioritised concealment is applied in cases where error correction is beyond the capability of the RS decoder. Simulation results have demonstrated that the proposed framework provides an average gain of 2.96 dB over a scheme that used Equal Error Protection (EEP).
This paper proposes an enhanced framework for H.264 video transmission whereby prioritised delay ... more This paper proposes an enhanced framework for H.264 video transmission whereby prioritised delay constrained retransmissions and error concealment have been employed. With the prioritisation framework, the impact of all lost macroblocks on the reconstructed frame is first assessed and macroblocks with the highest impact are given higher priority to be retransmitted and concealed. Moreover, the concealment algorithm is adaptively selected for both I and P frames. Extensive simulations with different transmission schemes were performed. Results showed that the proposed framework provides an average gain of 2.2 dB in PSNR over a conventional scheme which employs neither prioritisation nor adaptive concealment.
Asymmetric Digital Subscriber Line (ADSL) is a broadband communications technology which is widel... more Asymmetric Digital Subscriber Line (ADSL) is a broadband communications technology which is widely used for transmitting high bandwidth data such as audio and video over the internet. However, ensuring a good Quality of Service (QoS) in the transmission of such data over ADSL is challenging due to channel impairments such as noise and crosstalk. Moreover, given that the total monthly fixed data traffic is increasing at a phenomenal rate, (QoS) management is becoming a crucial issue. This paper proposes an enhanced transmission scheme for audio over ADSL. The proposed scheme exploits the unequal importance of the subbands generated by an MP1 codec to offer varying protection to them during ADSL transmission. This is achieved by multiplexing the bit streams from the subbands before Discrete Multi-tone Modulation (DMT) so that bit streams corresponding to more important subbands are offered higher protection than the least important ones. Additionally, a sophisticated retransmission protocol is proposed which allows frames from more important subbands to benefit from a greater number of retransmissions than least important ones. The retransmission protocol uses eight different Negative Acknowledgement (NACK) packet types which allow adaptive multiplexing of frames at the transmitter. The proposed Unequal Error Protection (UEP) scheme provides an average gain of over 28 dB in Segmented Signal to Noise Ratio (SSNR) over a conventional Equal Error Protection (EEP) one but incurs a negligible loss in throughput.
International Journal of Engineering Trends and Technology
Turbo coded Quadrature Amplitude Modulated (QAM) systems have been adopted by standards such as C... more Turbo coded Quadrature Amplitude Modulated (QAM) systems have been adopted by standards such as Code Division Multiple Access (CDMA) 2000 and Long Term Evolution (LTE) to achieve high data rates. Although several techniques have been developed to improve the performance of Turbo coded QAM systems, combinations of these techniques to produce hybrids with better performances, have not been fully exploited. This paper investigates the performance of LTE Turbo codes with Joint Source Channel Decoding (JSCD), adaptive Sign Division Ratio (SDR) based scaling and prioritised QAM constellation mapping. JSCD exploits a-priori source statistics at the decoder side and SDR based scaling provides a scale factor for the extrinsic information as well as a stopping criterion. Additionally, prioritised constellation mapping exploits the inherent Unequal Error Protection (UEP) characteristic of the QAM constellation and provides greater protection to the systematic bits of the Turbo encoder. Simulation results show that with 16-QAM at Bit Error Rates (BERs) above 10-1 , the combination of these three techniques achieves an average gain of 1.7 dB over a conventional LTE Turbo coded 16-QAM system. For BERs below 10-1 , the combination of prioritized constellation mapping, JSCD and SDR scaling provides an average gain of 0.6 dB. The proposed scheme with 64-QAM gives a performance gain of 3 dB on average over the conventional LTE Turbo coded 64-QAM system over the whole E b /N 0 range.
2017 3rd Iranian Conference on Intelligent Systems and Signal Processing (ICSPIS), 2017
Keystroke dynamics, which distinguishes individual by its typing rhythm, is the most prevalent be... more Keystroke dynamics, which distinguishes individual by its typing rhythm, is the most prevalent behavior biometrie authentication system. Neural Network is the active research area where different area has been presented. This paper present a keystroke dynamics Biometric system using chaotic neural network as the dimensional reduction and pattern recognition of the individual. Biometric scheme are being extensively used as their security qualities over the prior authentication system based on their history, that is the records were easily lost, guessed or forget. Biometric is more complex than password and is unique for each individual. In this work, the focus is made on the dwell time and flight time of the users' typing to recognize or reject an imposter. For this paper, the recognition rate obtained for the application of chaotic neural network was 99.1%.
Int. Arab J. Inf. Technol., 2021
Laws are often developed in a piecemeal approach and many provisions of similar nature are often ... more Laws are often developed in a piecemeal approach and many provisions of similar nature are often found in different legislations. Therefore, there is a need to classify legislations into various legal topics to help legal professionals in their daily activities. In this study, we have experimented with various deep learning architectures for the automatic classification of 490 legislations from the Republic of Mauritius into 30 categories. Our results demonstrate that a Deep Neural Network (DNN) with three hidden layers delivered the best performance compared with other architectures such as Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs). A mean classification accuracy of 60.9% was achieved using DNN, 56.5% for CNN and 33.7% for Long Short-Term Memory (LSTM). Comparisons were also made with traditional machine learning classifiers such as support vector machines and decision trees and it was found that the performance of DNN was superior, by at least 10%, ...
International Journal of Information and Communication Sciences, 2018
To meet different time delay requirements of different data, three scheduling strategies based on... more To meet different time delay requirements of different data, three scheduling strategies based on event, time and event-time are studied in this paper. The network calculus method is used to derive the maximum delay upper bound of virtual link based on three scheduling strategies, the relationship between the maximum delay upper, virtual link parameters, time scheduleparameters and the frame length is compared, the difference between the alternate scheduling strategy and the Time-Triggered or event-triggered scheduling is studied. The AFDX (Avionics Full-Duplex Switched Ethernet) virtual network model is established toverifythetheory based on OPNET, thedeterminist of theTime-Triggered link, the maximum delay of the event-triggered link and the efficiency of the end system transmission are compared under different scheduling strategies. Also, the applicable occasions of different scheduling strategies are studied. The results show that the event and time alternate scheduling strategy not only ensures the certainty of Time-Triggered link transmission, but also reduces the event triggering. The maximum delay upper bound of the link effectively improves the data processing capability and transmission efficiency, ensures the stability and certainty of the periodic signal transmission, and is suitable for the aviation bus system with different kinds of data with different delay requirements.
Abstract: This paper describes an application which increases the overall efficiency of a neural ... more Abstract: This paper describes an application which increases the overall efficiency of a neural network classifier intended for authentication whilst using fewer biometric features. Normalization of the biometric data is generally performed to remove unwanted impurities. However, in this case, when performing normalization, the statistical property for each set of data has also been taken into consideration prior to the classification process. Combination of the normalized biometric features has been performed while comparing their standard ...
Int. Arab J. Inf. Technol., 2021
Laws are often developed in a piecemeal approach and many provisions of similar nature are often ... more Laws are often developed in a piecemeal approach and many provisions of similar nature are often found in different legislations. Therefore, there is a need to classify legislations into various legal topics to help legal professionals in their daily activities. In this study, we have experimented with various deep learning architectures for the automatic classification of 490 legislations from the Republic of Mauritius into 30 categories. Our results demonstrate that a Deep Neural Network (DNN) with three hidden layers delivered the best performance compared with other architectures such as Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs). A mean classification accuracy of 60.9% was achieved using DNN, 56.5% for CNN and 33.7% for Long Short-Term Memory (LSTM). Comparisons were also made with traditional machine learning classifiers such as support vector machines and decision trees and it was found that the performance of DNN was superior, by at least 10%, ...
With the rapid development of various multimedia technologies, more and more multimedia data are ... more With the rapid development of various multimedia technologies, more and more multimedia data are generated and transmitted in the medical, commercial, and military fields, which may include some sensitive information which should not be accessed by or can only be partially exposed to the general users. Therefore, security and privacy has become an important. Over the last few years several encryption algorithms have applied to secure image transmission. This paper is a review on the aspects and approaches of design an image cryptosystem. First a general introduction given for cryptography and images encryption and followed by different techniques in image encryption and related works for each technique surveyed. Finally, general security analysis methods for encrypted images
Turbo coded 64-QAM systems have been adopted by standards such as CDMA-2000 and Long Term Evoluti... more Turbo coded 64-QAM systems have been adopted by standards such as CDMA-2000 and Long Term Evolution (LTE) to achieve high data rates. Although several techniques have been developed to improve the performance of Turbo coded QAM systems, combinations of these techniques to produce hybrids with better performances, have not been fully exploited. This paper proposes a combination of Joint Source Channel Decoding (JSCD), adaptive Sign Division Ratio (SDR) based scaling and prioritised constellation mapping, to improve the performance of Turbo coded 64-QAM. JSCD exploits a-priori source statistics at the decoder side and SDR based scaling provides a scale factor for the extrinsic information as well as a stopping criterion. Additionally, prioritised constellation mapping exploits the inherent Unequal Error Protection (UEP) characteristic of the 64-QAM constellation and provides greater protection to the systematic bits of the Turbo encoder. Simulation results show that at Bit Error Rates...
When using modern Code Division Multiple Access (CDMA) in mobile communications, the user must be... more When using modern Code Division Multiple Access (CDMA) in mobile communications, the user must be able to vary the transmission rate of users to allocate bandwidth efficiently. In this work, Orthogonal Variable Spreading Factor (OVSF) codes are used with the same principles applied in a low-rate superorthogonal turbo code due to their variable-length properties. The introduced system is the Variable Rate Superorthogonal Turbo Code (VRSTC) where puncturing is not performed on the encoder’s final output but rather before selecting the output to achieve higher rates. Due to bandwidth expansion, the codes outperform an ordinary turbo code in the AWGN channel. Simulations results show decreased performance compared to those obtained with the employment of WalshHadamard codes. However, with OVSF codes, the VRSTC system keeps the orthogonality of codewords whilst producing variable rate codes contrary to Walsh-Hadamard codes where puncturing is usually performed on the final output. Keywor...
Turbo coded Quadrature Amplitude Modulated (QAM) systems have been adopted by standards such as C... more Turbo coded Quadrature Amplitude Modulated (QAM) systems have been adopted by standards such as Code Division Multiple Access (CDMA) 2000 and Long Term Evolution (LTE) to achieve high data rates. Although several techniques have been developed to improve the performance of Turbo coded QAM systems, combinations of these techniques to produce hybrids with better performances, have not been fully exploited. This paper investigates the performance of LTE Turbo codes with Joint Source Channel Decoding (JSCD), adaptive Sign Division Ratio (SDR) based scaling and prioritised QAM constellation mapping. JSCD exploits a-priori source statistics at the decoder side and SDR based scaling provides a scale factor for the extrinsic information as well as a stopping criterion. Additionally, prioritised constellation mapping exploits the inherent Unequal Error Protection (UEP) characteristic of the QAM constellation and provides greater protection to the systematic bits of the Turbo encoder. Simulat...
2015 Fifth International Conference on Digital Information and Communication Technology and its Applications (DICTAP), 2015
High Efficiency Video Coding (HEVC/H.265), the latest standard in video compression, aims at half... more High Efficiency Video Coding (HEVC/H.265), the latest standard in video compression, aims at half the bit rate while maintaining the same quality compared to its predecessor, AVC/H.264. However, the increase in number of modes (35) for intra prediction in HEVC has significantly impacted on the encoder complexity. In this paper, a reduced number of candidate modes is therefore proposed to improve the encoding time. The most probable modes (MPMs) are inserted in the candidate list based on their computed Rough Mode Decision (RMD) cost rather than appending them to the list. In addition, the number of modes selected for the 4×4 and 8×8 PUs are based on the analysis of the graph obtained by plotting the cost of each mode against the mode number. Using this approach, the number of candidate modes in the RDO process is further reduced. Experimental results show that the average time savings of the proposed method is 38.0% along with an acceptable loss in quality (0.059dB) and a negligible increase in bit rate (0.30%) compared to HM 10.
Temporal Key Integrity Protocol (TKIP) is a provisional solution for Wired Equivalent Privacy (WE... more Temporal Key Integrity Protocol (TKIP) is a provisional solution for Wired Equivalent Privacy (WEP) security loopholes present in already widely deployed legacy 802.11 wireless devices. In this work, we model and analyse the computational complexity of TKIP security mechanism and propose an optimised implementation, called LOTKIP, to decrease processing overhead for better energy efficient security performance. The LOTKIP improvements are based on minimising key mixing redundancy and a novel frame encapsulation with low overhead. We simulate and compare LOTKIP with baseline TKIP in terms of complexity and energy consumption for ad hoc wireless network security. From simulation results, we demonstrate that LOTKIP executes with lower computational complexity, hence, with faster encryption time and more energy-efficient.
88 Abstract—In this paper we present the Study on a Tiny Encryption Algorithm. There is a require... more 88 Abstract—In this paper we present the Study on a Tiny Encryption Algorithm. There is a requirement to specify cryptographic strength in an objective manner rather than describing it using subjective descriptors such as weak, strong, acceptable etc. Such metrics are essential for describing the characteristics of cryptographic products and technologies. Towards this objective, we use two metrics called the Strict Plaintext Avalanche Criterion (SPAC) and the Strict Key Avalanche Criterion (SKAC) mentioned in our study that the strength of popular ciphers such as DES and TEA. A related issue of significance in the context of cryptographic applications is the quality of random number generators which need to pass certain tests. In this Paper, we expose DES and TEA to some of the standard random number generator tests.
EURASIA Journal of Mathematics, Science and Technology Education
The number of students completing mathematics education at higher levels are in decreasing number... more The number of students completing mathematics education at higher levels are in decreasing numbers, despite good grades at primary level. This research study sought to explore the mathematics learning trajectory by collecting Mathematics grades obtained by 1652 students at the end of primary and secondary schooling. Data analysis showed a moderate association between these grades. But those who performed well at the primary level did not necessarily study the subject at the same learning performance at higher levels. This research study unveiled that girls, in general, tend to maintain their learning performance better than boys at higher levels. It also revealed that some students experienced a positive turning point in their learning, and the overarching conclusions from students' interviews included perceived usefulness of Mathematics education, intrinsic and extrinsic motivation to boost the learning process and the need to overcome past hurdles. Finally, a model was developed to monitor learning progress achievement.
Intelligent Automation and Soft Computing
Intelligent Automation and Soft Computing, 2018
Indonesian Journal of Electrical Engineering and Computer Science, Jun 1, 2017
H.264/AVCis currently the most widely adopted video coding standard due to its high compression c... more H.264/AVCis currently the most widely adopted video coding standard due to its high compression capability and flexibility. However, compressed videos are highly vulnerable to channel errors which may result in severe quality degradation of a video. This paper presentsa concealment aware Unequal Error Protection (UEP) scheme for H.264 video compression using Reed Solomon (RS) codes. The proposed UEP technique assigns a code rate to each Macroblock (MB) based on the type of concealment and a Concealment Dependent Index (CDI). Two interleaving techniques, namely Frame Level Interleaving (FLI) and Group Level Interleaving (GLI) have also been employed. Finally, prioritised concealment is applied in cases where error correction is beyond the capability of the RS decoder. Simulation results have demonstrated that the proposed framework provides an average gain of 2.96 dB over a scheme that used Equal Error Protection (EEP).
This paper proposes an enhanced framework for H.264 video transmission whereby prioritised delay ... more This paper proposes an enhanced framework for H.264 video transmission whereby prioritised delay constrained retransmissions and error concealment have been employed. With the prioritisation framework, the impact of all lost macroblocks on the reconstructed frame is first assessed and macroblocks with the highest impact are given higher priority to be retransmitted and concealed. Moreover, the concealment algorithm is adaptively selected for both I and P frames. Extensive simulations with different transmission schemes were performed. Results showed that the proposed framework provides an average gain of 2.2 dB in PSNR over a conventional scheme which employs neither prioritisation nor adaptive concealment.
Asymmetric Digital Subscriber Line (ADSL) is a broadband communications technology which is widel... more Asymmetric Digital Subscriber Line (ADSL) is a broadband communications technology which is widely used for transmitting high bandwidth data such as audio and video over the internet. However, ensuring a good Quality of Service (QoS) in the transmission of such data over ADSL is challenging due to channel impairments such as noise and crosstalk. Moreover, given that the total monthly fixed data traffic is increasing at a phenomenal rate, (QoS) management is becoming a crucial issue. This paper proposes an enhanced transmission scheme for audio over ADSL. The proposed scheme exploits the unequal importance of the subbands generated by an MP1 codec to offer varying protection to them during ADSL transmission. This is achieved by multiplexing the bit streams from the subbands before Discrete Multi-tone Modulation (DMT) so that bit streams corresponding to more important subbands are offered higher protection than the least important ones. Additionally, a sophisticated retransmission protocol is proposed which allows frames from more important subbands to benefit from a greater number of retransmissions than least important ones. The retransmission protocol uses eight different Negative Acknowledgement (NACK) packet types which allow adaptive multiplexing of frames at the transmitter. The proposed Unequal Error Protection (UEP) scheme provides an average gain of over 28 dB in Segmented Signal to Noise Ratio (SSNR) over a conventional Equal Error Protection (EEP) one but incurs a negligible loss in throughput.
International Journal of Engineering Trends and Technology
Turbo coded Quadrature Amplitude Modulated (QAM) systems have been adopted by standards such as C... more Turbo coded Quadrature Amplitude Modulated (QAM) systems have been adopted by standards such as Code Division Multiple Access (CDMA) 2000 and Long Term Evolution (LTE) to achieve high data rates. Although several techniques have been developed to improve the performance of Turbo coded QAM systems, combinations of these techniques to produce hybrids with better performances, have not been fully exploited. This paper investigates the performance of LTE Turbo codes with Joint Source Channel Decoding (JSCD), adaptive Sign Division Ratio (SDR) based scaling and prioritised QAM constellation mapping. JSCD exploits a-priori source statistics at the decoder side and SDR based scaling provides a scale factor for the extrinsic information as well as a stopping criterion. Additionally, prioritised constellation mapping exploits the inherent Unequal Error Protection (UEP) characteristic of the QAM constellation and provides greater protection to the systematic bits of the Turbo encoder. Simulation results show that with 16-QAM at Bit Error Rates (BERs) above 10-1 , the combination of these three techniques achieves an average gain of 1.7 dB over a conventional LTE Turbo coded 16-QAM system. For BERs below 10-1 , the combination of prioritized constellation mapping, JSCD and SDR scaling provides an average gain of 0.6 dB. The proposed scheme with 64-QAM gives a performance gain of 3 dB on average over the conventional LTE Turbo coded 64-QAM system over the whole E b /N 0 range.
2017 3rd Iranian Conference on Intelligent Systems and Signal Processing (ICSPIS), 2017
Keystroke dynamics, which distinguishes individual by its typing rhythm, is the most prevalent be... more Keystroke dynamics, which distinguishes individual by its typing rhythm, is the most prevalent behavior biometrie authentication system. Neural Network is the active research area where different area has been presented. This paper present a keystroke dynamics Biometric system using chaotic neural network as the dimensional reduction and pattern recognition of the individual. Biometric scheme are being extensively used as their security qualities over the prior authentication system based on their history, that is the records were easily lost, guessed or forget. Biometric is more complex than password and is unique for each individual. In this work, the focus is made on the dwell time and flight time of the users' typing to recognize or reject an imposter. For this paper, the recognition rate obtained for the application of chaotic neural network was 99.1%.
Int. Arab J. Inf. Technol., 2021
Laws are often developed in a piecemeal approach and many provisions of similar nature are often ... more Laws are often developed in a piecemeal approach and many provisions of similar nature are often found in different legislations. Therefore, there is a need to classify legislations into various legal topics to help legal professionals in their daily activities. In this study, we have experimented with various deep learning architectures for the automatic classification of 490 legislations from the Republic of Mauritius into 30 categories. Our results demonstrate that a Deep Neural Network (DNN) with three hidden layers delivered the best performance compared with other architectures such as Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs). A mean classification accuracy of 60.9% was achieved using DNN, 56.5% for CNN and 33.7% for Long Short-Term Memory (LSTM). Comparisons were also made with traditional machine learning classifiers such as support vector machines and decision trees and it was found that the performance of DNN was superior, by at least 10%, ...
International Journal of Information and Communication Sciences, 2018
To meet different time delay requirements of different data, three scheduling strategies based on... more To meet different time delay requirements of different data, three scheduling strategies based on event, time and event-time are studied in this paper. The network calculus method is used to derive the maximum delay upper bound of virtual link based on three scheduling strategies, the relationship between the maximum delay upper, virtual link parameters, time scheduleparameters and the frame length is compared, the difference between the alternate scheduling strategy and the Time-Triggered or event-triggered scheduling is studied. The AFDX (Avionics Full-Duplex Switched Ethernet) virtual network model is established toverifythetheory based on OPNET, thedeterminist of theTime-Triggered link, the maximum delay of the event-triggered link and the efficiency of the end system transmission are compared under different scheduling strategies. Also, the applicable occasions of different scheduling strategies are studied. The results show that the event and time alternate scheduling strategy not only ensures the certainty of Time-Triggered link transmission, but also reduces the event triggering. The maximum delay upper bound of the link effectively improves the data processing capability and transmission efficiency, ensures the stability and certainty of the periodic signal transmission, and is suitable for the aviation bus system with different kinds of data with different delay requirements.
Abstract: This paper describes an application which increases the overall efficiency of a neural ... more Abstract: This paper describes an application which increases the overall efficiency of a neural network classifier intended for authentication whilst using fewer biometric features. Normalization of the biometric data is generally performed to remove unwanted impurities. However, in this case, when performing normalization, the statistical property for each set of data has also been taken into consideration prior to the classification process. Combination of the normalized biometric features has been performed while comparing their standard ...
Int. Arab J. Inf. Technol., 2021
Laws are often developed in a piecemeal approach and many provisions of similar nature are often ... more Laws are often developed in a piecemeal approach and many provisions of similar nature are often found in different legislations. Therefore, there is a need to classify legislations into various legal topics to help legal professionals in their daily activities. In this study, we have experimented with various deep learning architectures for the automatic classification of 490 legislations from the Republic of Mauritius into 30 categories. Our results demonstrate that a Deep Neural Network (DNN) with three hidden layers delivered the best performance compared with other architectures such as Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs). A mean classification accuracy of 60.9% was achieved using DNN, 56.5% for CNN and 33.7% for Long Short-Term Memory (LSTM). Comparisons were also made with traditional machine learning classifiers such as support vector machines and decision trees and it was found that the performance of DNN was superior, by at least 10%, ...
With the rapid development of various multimedia technologies, more and more multimedia data are ... more With the rapid development of various multimedia technologies, more and more multimedia data are generated and transmitted in the medical, commercial, and military fields, which may include some sensitive information which should not be accessed by or can only be partially exposed to the general users. Therefore, security and privacy has become an important. Over the last few years several encryption algorithms have applied to secure image transmission. This paper is a review on the aspects and approaches of design an image cryptosystem. First a general introduction given for cryptography and images encryption and followed by different techniques in image encryption and related works for each technique surveyed. Finally, general security analysis methods for encrypted images
Turbo coded 64-QAM systems have been adopted by standards such as CDMA-2000 and Long Term Evoluti... more Turbo coded 64-QAM systems have been adopted by standards such as CDMA-2000 and Long Term Evolution (LTE) to achieve high data rates. Although several techniques have been developed to improve the performance of Turbo coded QAM systems, combinations of these techniques to produce hybrids with better performances, have not been fully exploited. This paper proposes a combination of Joint Source Channel Decoding (JSCD), adaptive Sign Division Ratio (SDR) based scaling and prioritised constellation mapping, to improve the performance of Turbo coded 64-QAM. JSCD exploits a-priori source statistics at the decoder side and SDR based scaling provides a scale factor for the extrinsic information as well as a stopping criterion. Additionally, prioritised constellation mapping exploits the inherent Unequal Error Protection (UEP) characteristic of the 64-QAM constellation and provides greater protection to the systematic bits of the Turbo encoder. Simulation results show that at Bit Error Rates...
When using modern Code Division Multiple Access (CDMA) in mobile communications, the user must be... more When using modern Code Division Multiple Access (CDMA) in mobile communications, the user must be able to vary the transmission rate of users to allocate bandwidth efficiently. In this work, Orthogonal Variable Spreading Factor (OVSF) codes are used with the same principles applied in a low-rate superorthogonal turbo code due to their variable-length properties. The introduced system is the Variable Rate Superorthogonal Turbo Code (VRSTC) where puncturing is not performed on the encoder’s final output but rather before selecting the output to achieve higher rates. Due to bandwidth expansion, the codes outperform an ordinary turbo code in the AWGN channel. Simulations results show decreased performance compared to those obtained with the employment of WalshHadamard codes. However, with OVSF codes, the VRSTC system keeps the orthogonality of codewords whilst producing variable rate codes contrary to Walsh-Hadamard codes where puncturing is usually performed on the final output. Keywor...
Turbo coded Quadrature Amplitude Modulated (QAM) systems have been adopted by standards such as C... more Turbo coded Quadrature Amplitude Modulated (QAM) systems have been adopted by standards such as Code Division Multiple Access (CDMA) 2000 and Long Term Evolution (LTE) to achieve high data rates. Although several techniques have been developed to improve the performance of Turbo coded QAM systems, combinations of these techniques to produce hybrids with better performances, have not been fully exploited. This paper investigates the performance of LTE Turbo codes with Joint Source Channel Decoding (JSCD), adaptive Sign Division Ratio (SDR) based scaling and prioritised QAM constellation mapping. JSCD exploits a-priori source statistics at the decoder side and SDR based scaling provides a scale factor for the extrinsic information as well as a stopping criterion. Additionally, prioritised constellation mapping exploits the inherent Unequal Error Protection (UEP) characteristic of the QAM constellation and provides greater protection to the systematic bits of the Turbo encoder. Simulat...
2015 Fifth International Conference on Digital Information and Communication Technology and its Applications (DICTAP), 2015
High Efficiency Video Coding (HEVC/H.265), the latest standard in video compression, aims at half... more High Efficiency Video Coding (HEVC/H.265), the latest standard in video compression, aims at half the bit rate while maintaining the same quality compared to its predecessor, AVC/H.264. However, the increase in number of modes (35) for intra prediction in HEVC has significantly impacted on the encoder complexity. In this paper, a reduced number of candidate modes is therefore proposed to improve the encoding time. The most probable modes (MPMs) are inserted in the candidate list based on their computed Rough Mode Decision (RMD) cost rather than appending them to the list. In addition, the number of modes selected for the 4×4 and 8×8 PUs are based on the analysis of the graph obtained by plotting the cost of each mode against the mode number. Using this approach, the number of candidate modes in the RDO process is further reduced. Experimental results show that the average time savings of the proposed method is 38.0% along with an acceptable loss in quality (0.059dB) and a negligible increase in bit rate (0.30%) compared to HM 10.
Temporal Key Integrity Protocol (TKIP) is a provisional solution for Wired Equivalent Privacy (WE... more Temporal Key Integrity Protocol (TKIP) is a provisional solution for Wired Equivalent Privacy (WEP) security loopholes present in already widely deployed legacy 802.11 wireless devices. In this work, we model and analyse the computational complexity of TKIP security mechanism and propose an optimised implementation, called LOTKIP, to decrease processing overhead for better energy efficient security performance. The LOTKIP improvements are based on minimising key mixing redundancy and a novel frame encapsulation with low overhead. We simulate and compare LOTKIP with baseline TKIP in terms of complexity and energy consumption for ad hoc wireless network security. From simulation results, we demonstrate that LOTKIP executes with lower computational complexity, hence, with faster encryption time and more energy-efficient.
88 Abstract—In this paper we present the Study on a Tiny Encryption Algorithm. There is a require... more 88 Abstract—In this paper we present the Study on a Tiny Encryption Algorithm. There is a requirement to specify cryptographic strength in an objective manner rather than describing it using subjective descriptors such as weak, strong, acceptable etc. Such metrics are essential for describing the characteristics of cryptographic products and technologies. Towards this objective, we use two metrics called the Strict Plaintext Avalanche Criterion (SPAC) and the Strict Key Avalanche Criterion (SKAC) mentioned in our study that the strength of popular ciphers such as DES and TEA. A related issue of significance in the context of cryptographic applications is the quality of random number generators which need to pass certain tests. In this Paper, we expose DES and TEA to some of the standard random number generator tests.
EURASIA Journal of Mathematics, Science and Technology Education
The number of students completing mathematics education at higher levels are in decreasing number... more The number of students completing mathematics education at higher levels are in decreasing numbers, despite good grades at primary level. This research study sought to explore the mathematics learning trajectory by collecting Mathematics grades obtained by 1652 students at the end of primary and secondary schooling. Data analysis showed a moderate association between these grades. But those who performed well at the primary level did not necessarily study the subject at the same learning performance at higher levels. This research study unveiled that girls, in general, tend to maintain their learning performance better than boys at higher levels. It also revealed that some students experienced a positive turning point in their learning, and the overarching conclusions from students' interviews included perceived usefulness of Mathematics education, intrinsic and extrinsic motivation to boost the learning process and the need to overcome past hurdles. Finally, a model was developed to monitor learning progress achievement.
Intelligent Automation and Soft Computing