Muhammad Imran Iqbal

— it is a software implementation that runs on computers to run different programs like a real physical machine. There are two classifications of Virtual Machines categorized by their use i.e. System Virtual Machine and Process Virtual Machine. Virtual Machine Monitor or Hypervisor is the software layer that provides virtualization. The operating system running on virtual machine is called guest operating system. From Operating System designs and development, Virtualization has great impact and importance. Different techniques are used in virtualization; each has its own implementations under certain circumstances. One technique is not truly efficient for other environment. Shared kernel, full kernel and hypervisor virtualization are different techniques that be used combined as well as independently. Nowadays the full virtualization with binary translation is the most effective and reliable technology used ever with great ease of use. VMware uses both full virtualization with binary translation and hardware assisted virtualization. We have discussed many types of virtualization techniques and their interrupt handling. How virtual machine works over host operating system is well explained in this review. In nested virtualization there is a lack of architecture support for multi-level virtualization. For this reason it has up to 10% of overload. It is possible to overcome the overhead i.e. 6-10% for nested virtualization and to find a better solution.

In this paper, we propose a framework for unequal error protection (UEP) of image and video streaming over a wireless channel. Our framework of allocating the parity symbols associated with error control coding to the image or video codestream takes advantage of the ...

In this paper, we propose a new unequal error protection (UEP) scheme for wireless imaging. Our technique is based on approximating the optimal parity distribution using offline and online processing. For a given parity budget and signal-to-noise ratio (SNR), the offline processing is responsible for obtaining the optimal parity distribution and an approximation of it using polynomial curve fitting. Subsequently, in the online processing, the obtained polynomial parameters are used to regenerate the parity distribution curve. Eventually, channel codes are assigned to the packets of the codestream based on the regenerated parity distribution curve. The complexity of the proposed technique is very low which makes it suitable for realtime wireless imaging, especially for mobile devices that have low processing power and operate with very low memory. Despite having very low complexity, the proposed UEP technique gives still close to optimal performance.

In this paper, we propose a number of error protection schemes for wireless imaging ranging from simple but less efficient equal error protection to very complex yet optimal error protection. The main objective is to facilitate trade-offs between performance and complexity in choosing an error protection scheme. Our technique provides the system designer with a number of solutions from which the application can choose those that best suits the available resources such as processing power and data rate. For this purpose, an image codestream is split into smaller cells each having a certain number of packets while the available parity symbols from the deployed error protection are optimally allocated to these cells. Subsequently, the parity symbols allocated to these cells are distributed among the packets that constitute each of these cell. Finally, the error control codes for every packet are determined based on the number of parity symbols allocated to each packet. Larger cells provide less complex but also less efficient error protection while smaller cells result in a better error protection performance but at the cost of increased complexity. Simulation results validate the effectiveness of the considered technique in providing the desired performances-complexity trade-off. To obtain better correlation with human perception, performance is evaluated in terms of objective perceptual quality metrics.

Designing wireless imaging services with acceptable quality imposes great challenges due to the severe impairments induced by wireless channels, high data rate demands and limited transmission bandwidth. In this paper, we examine different error protection schemes for the JPEG2000 codestream using the error control codes offered by the wireless JPEG2000 (JPWL) standard. In particular, two unequal error protection (UEP) schemes are considered. In the first scheme, only the initial and more important packets of the JPWL codestream are protected while the remaining packets are left without protection. In contrast, the second scheme uses a strong code for the initial packets while a weaker code is used for the tail packets. As quality of visual services may not be captured adequately with traditional link layer metrics, code selection and their performance assessment is based on perceptual quality metrics. Specifically, the LP-norm, structural similarity (SSIM) index and visual information fidelity (VIF) criterion are utilized. Numerical results are provided for a range of system conditions under which UEP outperforms conventional equal error protection in terms of perceptual quality metrics.

Designing wireless imaging services with acceptable quality imposes great challenges due to the severe impairments induced by wireless channels, high data rate demands and limited transmission bandwidth. In this paper, we examine different error protection schemes for the JPEG2000 codestream using the error control codes offered by the wireless JPEG2000 (JPWL) standard. In particular, two unequal error protection (UEP) schemes are considered. In the first scheme, only the initial and more important packets of the JPWL codestream are protected while the remaining packets are left without protection. In contrast, the second scheme uses a strong code for the initial packets while a weaker code is used for the tail packets. As quality of visual services may not be captured adequately with traditional link layer metrics, code selection and their performance assessment is based on perceptual quality metrics. Specifically, the LP-norm, structural similarity (SSIM) index and visual information fidelity (VIF) criterion are utilized. Numerical results are provided for a range of system conditions under which UEP outperforms conventional equal error protection in terms of perceptual quality metrics.

— Todays in computing environment, most of our critical activities depends on network services in a variety of domain. For trustworthy environment a secure mechanism is needed that ensures the security of all activities from any damage and unauthorized access. Several defending systems are introduced to face challenges that enforcing security policies. Intrusion detections and prevention systems are one of those that are used to define security policies, documenting threats and observe entities that violating security policies. Snort is one of the efficient and widely used signature based IDS (Intrusion Detection System/) that analyses real time network packets for secure computing. But IDS employs complex syntax that describe attacks description and required more processing power. Software and hardware based platforms tolerate this difficulties and faces challenges that can considerably reduce performance. In this research a pre-filtering model is presented that improves the overall performance of an IDS. It is infrequent for each incoming packet to match with a number of rules. With the advantage of this remake, a pre-filtering approach is used that selects a small part of rules from IDS ruleset. This small part of rules further applicant for full match. Second full match module provides high throughput with this small subset of rules. When a fixed number of rules cross its limit, priority analyser (PA) judge packet priority and provide it processing resources to complete its residual processing. As the number of rules increase, the probability of priority analyser increase according to it. In our experiment we used DefCon traces and snort IDS ruleset and select prefix up to ten character of payload part of packet. This model reduce a large overhead of processing, drop ratio of packets and improve scalability of system. Keywords— Intrusion detection system, network intrusion deteciotn system , intrusion prevention system, packet pre-filtering, processing enigne.