International Journal of Image Mining (10 papers in press)
LEVEL CONTROL OF THREE TANK SYSTEM USING INTELLIGENT TECHNIQUES
by Srinivasan Karuppanan, Juno Devassy, Sindhiya Dhanapal
Abstract: In process control industries, different physical parameters like flow, pressure, temperature, level etc. are important parameters to be controlled. In that, liquid level control in process tanks is one of the complex problems in process control industries. A three tank level control system can help control system engineers to analyze both interacting and non-interacting cases in a system. In the proposed work, three tank systems are intended to maintain the level of the third tank at some predefined value irrespective of changes in inflow to first tank. In this paper neural network controller and PI based fuzzy controller are used for optimizing the output. The performances of both the controllers are compared with that of the conventional PI controller.
Keywords: Three tank system; process control; neural network controller; PI based fuzzy controller.
Performance Analysis of Colour Coded Multiple Access Orthogonal Frequency Division Multiplexing Visible Light Communication System
by Jim Godwin R S, Veena K, Sriram Kumar D
Abstract: Performance analysis of Colour Coded Multiple Access (CCMA) Orthogonal Frequency Division Multiplexing (OFDM) Visible Light Communication (VLC) System is discussed in this paper. First, characteristics of Optical OFDM are studied. Secondly, CCMA is achieved using three Laser Diodes (LD): red, green and blue LDs. Finally OFDM modulated signals are optically modulated with Red Green and Blue LDs (RGB LDs). Then these three optical signals are combined together to form white light which illuminates as well as provides high speed data transmission. The performance of the proposed CCMA OFDM VLC system is studied using constellation diagrams, signals and spectrum comparisons. The system is simulated for the data rate of 10 Gbps.
Keywords: Colour Coded Multiple Access; Laser diodes; High data rate; Orthogonal Frequency Division Multiplexing; Visible Light Communication; Red Green Blue Laser Diodes.
Special Issue on: Innovative Mobile Technology
Dynamic Salt Generation for Mobile Data Security using Elliptic Curves against Precomputation Attacks
by Padma Bh., Raj Kumar G.V.S.
Abstract: Pattern Lock is a mechanism that most of the people set to their Android smart phones. As the built-in pattern lock system adds an extra layer of protection and is one of the innovative unlock methods supported by Android, a fair quantity of study has been done about its (in)security. Pattern locks are not difficult to crack and are vulnerable to precomputation attacks such as brute forcing, dictionaries and rainbow tables. Older versions of Android produce SHA-1 signatures for authentication process which are not salted hashes. However the newer versions of Android pattern locks utilize scrypt hash function that generate random salt value which needs to be stored in the database to withstand to such attacks. But for pattern passwords attaching a salt value is still found not to be enough and susceptible to brute force. This research, therefore, proposes a method where it helps to produce and append a salt value to a password dynamically by representing the pattern using points of an elliptic curve. After the implementation and analysis, the results show this method exhibits SAC(Strict Avalanche Criterion) and passwords will become more tolerant to brute forcing, and other precomputation attacks which makes it more difficult to compromise.
Keywords: android; smart phones; dictionaries; brute force; rainbow tables; salt; elliptic curves; SAC.
Analysis on Spectrum Sensing Schemes for Cognitive Radio Networks
by Perarasi Sambantham, Nagarajan G
Abstract: Spectrum is a significant resource in RF environment and sensing is an essential function of cognitive radio to prevent harmful interference pertaining to the primary users. Spectrum sensing identifies the unutilised spectrum in order to improve the spectrum utilization. Cooperative spectrum sensing is a technique which incorporates multiple secondary users. It paves way in analysing the primary user information to achieve more exact detection performance. In this paper, Soften hard decision fusion mechanism is used to solve the sensing failure as it provides fine sensing and increases the detection probability with trade-off between probability of detection and complexity. Binary hypothesis testing rule is applied to verify the sensing concept. Censoring policy can be chosen in the case of good signal to noise ratio which provides energy efficient sensing scheme. Results conclude that Soften hard decision performs better than the other types of Sensing schemes. Receiver operating characteristics are analysed for the spectrum sensing schemes.
Keywords: Censoring sensing; Cyclostationary detector; Energy detector; Log Likelihood ratio test; Matched filter; Spectrum sensing.
Efficient Routing For Low Rate Wireless Network A Novel Approach
by Ambeth Kumar Devadoss
Abstract: In recent years, Wireless device networks have achieved associate degree attention on a world level. This comprises small sensors with restricted power and restricted resources. The applicability of Wireless Sensor Networks (WSNs) in various fields such as medical research and civilian applications is proliferating enormously. Every routing protocol designed for WSNs should be energy-efficient, reliable and prolong the network lifetime. Due to the resources and energy constraints in WSNs, routing can be considered as a matter of contention in networks. This research proposes a beacon-enabled least-time and energy-efficient routing protocol with single-level data-aggregation using an IEEE 802.15.4. Considering the low complexness and low power consumption it is appropriate for Low-Rate Wireless Personal Area Networks as WSNs. On comparison of the proposed protocol with the popular ad hoc and other WSN routing protocols, the results show that the proposed protocol outperforms the routing protocols in the literature in terms of latency, throughput, and average energy consumption.
Keywords: Wireless Sensor Network; cygwin; greedy; PRR; BR.
Active Resource Allocation Design for QoS Improvement in Wireless Broadband Interworking Networks
by Kokila Subramanian, Sivaradje Gopalakrishnan
Abstract: The key factor to achieve guaranteed Quality of Service (QoS) and increased networks capacity, without increasing the complexity of the network layers in Interworking Heterogeneous Broadband Wireless Network is through Radio Resource Management (RRM) mechanism. In this paper, an Active Resource Allocation (ARA) scheme is developed to organize the available resource among the interworking Ultra Mobile Broadband (UMB) - Worldwide Interoperability for Microwave access (WiMAX) - Wireless Local Area Network (WLAN) network. The resource provisioning scheme deals with complete partitioning of the available radio channels based on prioritizing the type of applications, such as real-time and non-real time services. The proposed ARA Scheme is validated through OPNET results obtained related to the defined set of applications such as VoIP, FTP, HTTP and Email. The obtained results were compared with the proposed ARA implemented in Universal Mobile Telecommunication System (UMTS)-WiMAX-WLAN Interworking network and without resource allocation algorithm, which unveils the ability of the UMB network and the proposed scheme to provide high QoS satisfaction level with adequately exploiting the overall available resource.
Keywords: Radio Resource Management; Complete partitioning; Real-time services; Non-real time services; Heterogeneous wireless network; Broadband Interworking Network; Active Resource Allocation.
Spectrum Sensing for 4G LTE OFDM Signals in Heterogeneous Network Using Neural Network
by Praveen Thakur, Mamta Madan, Sushila Madan
Abstract: Spectrum sensing is an important requirement in cognitive radios deployed in advanced 4G wireless communication systems. The cognitive radio has to classify the signal of interest from many such signals in its vicinity. There has been wide spread acceptance of multicarrier signal like Orthogonal Frequency Division Multiplexed (OFDM) due to its better protection characteristics against channel degradations. Many recent and upcoming wireless standards thus employs OFDM signal and a CR radio has to classify the OFDM signal operating in a heterogeneous operating environment. In this paper we have used Artificial Neural Network based classification of OFDM signal of third-Generation Partnership Project Long Term Evolution (3GPP LTE) signals that is used on 4G wireless networks. We used reference signal induced cyclostationarity and Cyclic Prefix property as feature for classification. The 3GPP LTE OFDM signal classification is done in a heterogeneous network environment in which other OFDM signal from IEEE WiMAX network and other single carrier digital modulation signal presence is considered. Comparison of classification performance for Multi layer Perceptron and Radial Basis Function neural network is presented. Effect of two training algorithm, Levenberg - Marquardt (LM) and Back Propagation with momentum , on the convergence rate for training the neural network is presented.
Keywords: WiMAX IEEE 802.16 ; MultiCarrier Signals ; orthogonal frequency division multiplexing (OFDM); signal classification; Neural Networks.
Optimal Network Selection Algorithms in Next Generation Heterogeneous Wireless Networks
by Harsha Bhute
Abstract: In the next generation heterogeneous wireless networks it is obligatory to consider the integration of various networks for providing the seamless connectivity to the mobile node roaming. While providing the seamless connectivity, a suitable network to choose from the given set of networks is the crucial issue. During vertical handoff it is required to select the best network not only according to user preference but also the application, network and terminal contexts. The network selection algorithm plays a very important role in ensuring quality of service and seamless connectivity in such networks. Network selection in heterogeneous wireless networks has been widely studied by using various mathematical models, in which MADM is one of the most popular. This paper surveys the different network selection techniques available in the literature and explains a method for network selection which takes into consideration of multiple decision making criteria for selecting the optimal network.
Keywords: Network selection; vertical handoff; decision algorithm; MADM.
PERFORMANCE OF LANDWEBER ITERATION ALGORITHM IN TOMOGRAPHIC IMAGE RECONSTRUCTION
by Sathish Kumar
Abstract: The rapid growth of Computer Tomography has been accompanied by equally advancing in image reconstruction algorithm also. The fundamental inverse problem is the reconstruction of a function from finitely many measurements, so pertaining to that function. The measured data is limited, it cannot serve to determine one single correct solution. The iterative reconstruction (IR) image reconstruction algorithms have stable solution to the limited projection. In this paper landweber based iteration image reconstruction is simulated and its performance is compared with different algorithm. Then, the quality of the reconstructed image is expressed interms of mean absolute error and correlation coefficient as compared to the original image. The entire simulations are performed in Matlab tool.
Keywords: Computed Tomography Image Reconstruction Iterative Reconstruction
Landweber Algorithm Limited Projection.
Multilevel Watermarking Technique For Securing Multibiometric Templates Using DTCWT & SVD
by Manonmani Panneerselvam, V.Vijayalakshmi , R.Divya
Abstract: Multibiometric authentication is the usage of more than one physiological or behavioral characteristic to identify an individual. It involves the fusion of two or more biometrics such as fingerprint, facial recognition, iris scanning, hand geometry, signature verification, or speech recognition. The limitations of unimodal biometrics include noise in sensed data, non universality, upper bound on identification accuracy and spoofing attacks. Multimodal biometrics is more advantageous over unimodal biometrics as it is resilience to spoofing and has low False Acceptance Rate (FAR). Existing multibiometric technique uses two biometric traits namely fingerprint and face. This technique is invasive in nature and is prone to spoofing attack. The proposed multibiometric technique implements finger vein as an additional biometric trait along with fingerprint and iris. Finger vein is non-invasive in nature and increases security. In addition, in order to overcome spoofing attack on system database of multibiometric system template, the existing system further focuses on securing the Biometric template by implementing Digital watermarking technique. Watermarking is carried out by using Singular Value Decomposition (SVD) and Discrete Wavelet Transform (DWT) techniques. The performance metrics such as PSNR, NCC and SSIM is simulated using MATLAB tool. The Existing technique is not robust against JPEG compression, median filtering and noise addition like Gaussian noise, salt & pepper noise etc., and also the Discrete Wavelet Transform has Lack of Shift Invariance problem. To overcome this problem, this project is implemented with Dual Tree Complex Wavelet Transform (DTCWT) in multilevel watermarking technique. The advantage of Dual Tree Complex Wavelet Transform improves denoising performance. The Authentication is implemented to prove the security. The performance metrics such as PSNR, NCC and SSIM is simulated using MATLAB for the Dual Tree Complex Wavelet Transform (DTCWT) and the security measures parameters like FAR, FRR, System Accuracy and the ROC comparison curve of the DWT and DTCWT are obtained.
Keywords: Multibiometric; Singular Value Decomposition; Discrete Wavelet Transform; Dual Tree Complex Wavelet Transform; Digital Watermarking; Multibiometric Authentication; template security.