Preventing Cyber Attacks using Artificial Intelligence
Exploring Advantages and Challenges of Chatbot Integration in Human-Robot Interaction
Counterfeit Currency Detection using Machine Learning
Artificial Intelligence use in Lung Cancer Screening to Improve Patient Recovery
An Analysis of Various Crypto Coins and their Suitability for Real-Time Applications
Design and Evaluation of Parallel Processing Techniques for 3D Liver Segmentation and Volume Rendering
Ensuring Software Quality in Engineering Environments
New 3D Face Matching Technique for an Automatic 3D Model Based Face Recognition System
Algorithmic Cost Modeling: Statistical Software Engineering Approach
Prevention of DDoS and SQL Injection Attack By Prepared Statement and IP Blocking
Control unit, in digital systems is one of the vital problems of optimal design. Several approaches of developing minimized modules for digital systems were studied. Algorithmic State Machine (ASM) has been analyzed by different approaches as coding the Excitation Function, coding the Input Logical Word and coding the Output Variables. This paper presents a modified method to synthesize the State Table (ST) of a digital system based on modifying the State Table Strings - rows. This new method results in minimizing the hardware used. Hardware Description Language (HDL) is the modern way for hardware design implementation, used in Programmable Logical Devices (PLD) such as Generic Array Logic (GAL) and Field-Programmable Gate Array (FPGA).The suggested method was tested and analyzed using Verilog HDL.
Users on the distributed web servers over the Internet demand for high scalability, availability, throughput and better response time for their requests. This is achieved by using load balancing techniques, which uniformly distributes the client requests to the web servers organized into a cluster. A cluster is a group of physical servers with identical content, networked together to form a single virtual server. A distributed web servers is a collection of replicated servers located at same or different sites and associated by the communication media. Web servers located at same site are organized into a cluster is distributed LAN web servers where as web servers are deployed in different geographical locations are structured as distributed WAN web servers. Load balancing strategies on distributed web servers are implemented in two ways namely message passing and mobile agent paradigm. In message passing paradigm, web servers have to exchange numerous messages on load periodically in order to finalize the processing of a request in any one of the server in a cluster. Alternative approach for message passing is the mobile agent based approaches. A mobile agent is a software component that can move freely from one host to another on a heterogeneous network and transport its state and code from home host to other host and execute various operations on the site. The advantages of using mobile agents over message passing paradigm are low network traffic, supports asynchronous and autonomous operations. This paper discusses about the various techniques of load balancing based on mobile agents.
Network Intrusion detection systems aim is to detect attacks against computer systems and networks. This paper focuses on the development of Genetic Algorithm in Intrusion Detection System (GAIDS). Genetic Algorithms can provide appropriate heuristic search methods. However, balancing the need to detect all possible attacks found in network with the need to avoid false positives is a challenge, given the scalar fitness values required by Genetic Algorithms. This study discusses a fitness function independent of variable parameters to overcome this problem. This fitness function allows the IDS to significantly reduce both its false positive and false negative rate.
This paper presents new techniques for off-line signature recognition. The proposed systems are based on invariant moment, texture and global features. For each one of these features sets neural network classifiers which are based on MLP (Multilayer Perceptron), RBF (Radial Basis Function) and Elman neural network classifier are implemented. These nine signature recognition systems are tested then the optimized system that has the highest recognition rate is selected. The optimized system is evaluated with rotation and scaling effects.
Security has become a primary concern in order to provide protected communication between mobile nodes in a hostile environment. Unlike the wire-line networks, the unique characteristics of mobile computing pose a number of nontrivial challenges to security design, such as shared wireless medium, stringent resource constraints, and highly dynamic network topology. These challenges clearly make a case for building multicast security solutions that achieve both broad protection and desirable network performance. In this article, the fundamental security problem of protecting the multicast routing between mobile nodes in a mobile computing is focused. The security in network layer issues related to protecting routing and forwarding protocols Challenges and Solutions are also identified.
TCP is a prominent protocol used by Internet community. Among the TCP implementations, TCP Reno is widely used, and TCP Vegas is a protocol under active research and is expected to replace TCP Reno. TCP Vegas performance is better than other TCP implementations when is in operation separately, independent of other protocols. But its performance is affected when interoperated with TCP Reno. In this paper modifications to congestion avoidance procedure of TCP Vegas are proposed to solve the interoperability problem. Modifications proposed allow TCP Vegas to use its share of bandwidth efficiently. For simulation NS2 is used.
It has recently been reported that elliptic and hyper-elliptic curve cryptography are the two public key cryptographic techniques used to implement the cryptosystems more efficiently and effectively. Many researches are being done to implement these in both hardware and software fields. In this paper, the basic concepts of elliptic and hyper-elliptic curves are discussed. The algorithms for private key, public key generation, encryption and decryption for both elliptic curve and hyper-elliptic curve cryptography are described. Also, the performance of both ECC and HECC are demonstrated. Finally, the comparative study between these two is analyzed.
In this paper, different encryption methods based on hash function and stream ciphering are proposed and implemented using Field Programmable Gate Array (FPGA). PANAMA module is chosen to be implemented on FPGA technology since it has high amount of parallelism and simple operation. Several modifications on PANAMA are proposed that makes the complexity of the collision increase and the randomness of its output is improved. The two proposed methods are called PANAMA2 and PANAMA3. Characteristics (the diffusion factor, the branch number, the nonlinearity, and the correlation) of the functions used in these methods are evaluated. It is found that the time required for PANAMA2 is 11ns and for PANAMA3 is12ns. The area occupied by each method on FPGA Virtex2 is calculated. It is found that PANAMA2 and PANAMA3 need 81% of the available slices. Randomness test using five basic security tests are performed. It is found that the randomness of all methods is acceptable with PANAMA3 being the best and then PANAMA2. Complexity of collision proposed by Rijmen on PANAMA is also calculated when it is applied to PANAMA2 and PANAMA3. It is found that the complexity of this collision increases. It is also found that PANAMA, PANAMA2 and PANAMA3 have resistance against attacks on stream cipher.
The high expectation of performance and availability for communication systems has presented great challenges in the modeling and design of fault tolerant wireless systems. The proper modeling methodology to study the degradation of such systems is so-called performability modeling. In this paper, an overview of approaches for the construction and the solution of performability model spectral approximation method is given. In the existing exact solutions for Markovmodulated queues are computationally intensive and prone to numerical problems when the number of states of the Markovian environment becomes large. To address this problem, a simple technique, spectral approximation method is proposed. It uses the dominant eigenvalue of the characteristic matrix polynomial, together with the associated left eigenvector. This approximation is shown to be asymptotically exact in heavy arrival.
The combination of wavelet theory and neural networks has lead to the development of wavelet networks (wavenets). Wavenets are feed-forward neural networks which used wavelets as activation functions, and the basis used in wavenet has been called wavelons. Wavenets are successfully used in identification problems. The strength of wavenets lies in their capabilities of catching essential features in “frequency-rich” signals. In wavenet, both the translation and the dilation of the wavelets (wavelons) are optimized beside the weights. The wavenet algorithm consist of two processes: the selfconstruction of networks and the minimization of errors. In the first process, the network structure is determined by using wavelet analysis. In the second process, the approximation errors are minimized. The wavenets with different types of frame wavelet function are integrated for their simplicity, availability, and capability of constructing unknown nonlinear function. Thus, wavelet can identify the localization of unknown function at any level. In addition, genetic algorithms (GAs) are used successfully in training wavenets since GAs reaches quickly the region of the optimal solution. Tests show that GA obtains best weight vector and produces a lower sum square error in a short period of time.
