Research Cell: An International Journal of Engineering Sciences

Ishpreet Singh Virk ¹, Raman Maini ¹

Affiliations
1 University College of Engineering, Punjabi University, Patiala, IN

Abstract

This paper provides the comprehensive overview of techniques and tools in the research area of Image retrieval, especially Content-Bases Image Retrieval (CBIR), a vivid research area for the last two decades. However, there are still many challenges to be overcome. Image feature representation and extraction, multi dimensional indexing and retrieval system design are fundamental basis for Content- Bases Image Retrieval (CBIR).We reviews various visual features and their corresponding representation and matching techniques. To provide fast and effective search in large scale image collection, high dimension indexing techniques needs to be implement. State-of-art commercial and research systems and their characteristics are described. Finally, we look at browsing systems as an alternative approach to dealing with large image databases. The hue sphere browsing system organizes images on a spherical visualization space so that visually similar images are located close to each other, and provides powerful interaction tools that aid in exploring the visualized image collections.

Keywords

Content Based, Retrieval, Feature Extraction, Indexing, Hue Sphere.

Full Text

   

Sneha Chhabra ¹, Raman Maini ¹

Affiliations
1 University College of Engineering, Punjabi University, Patiala, IN

Abstract

The expanding gap between microprocessor and DRAM performance has necessitated the use of some other techniques designed to reduce or hide the latency of main memory accesses. Although large cache hierarchies have proven to be effective in reducing this latency for the most frequently used data, but it is still not so efficient.

This paper proposed a technique i.e. DataPrefetching Technique for hiding the access latency of data referencing patterns that defeat caching strategies. Rather than waiting for a cache miss to initiate a memory fetch, data prefetching anticipates such misses and issues a fetch to the memory system in advance of the actual memory reference. With data prefetching, memory system call data into cache before processor needs it, thereby reducing memory access latency. The following survey examines several alternative approaches and discusses the design tradeoffs involved when implementing a data Prefetch strategy.

Keywords

Prefetching, Memory Latency.

Full Text

   

Raman Maini ¹, Suruchi Mehra ¹

Affiliations
1 University College of Engineering, Punjabi University, Patiala - 147002, Punjab, IN

Abstract

Metrics (quantitative estimates of product and project properties) can, if defined from sound engineering principles, be a precious tool for both project management and software development The purpose of this paper is to evaluate if software metrics can be used to determine the object oriented design quality of a software system. Encapsulation, inheritance, polymorphism, reusability, Data hiding and message-passing are the major attribute of an Object Oriented system. In order to evaluate the quality of an Object oriented system the above said attributes can act as indicators. The metrics are the well known quantifiable approach to express any attribute. Here in this paper we will go though different type of metrics.

Keywords

Object Oriented, Software Metrics, Methods, Attributes, Cohesion, Coupling, Inheritance.

Full Text

   

Sukhdeep Kaur ¹, Raman Maini ¹

Affiliations
1 Department of Computer Engineering, Punjabi University, Patiala-147002, IN

Abstract

The term refactoring is a crucial part of software engineering that changes the internal structure of source code with less cost of software maintenance to improve overall design of software. Design or code problem in software indicates the bad smell. Basically, bad smell is a sign of badly written code by programmer that makes the code more difficult to change and maintain in future. Bad smells prioritizing the code refactoring to improve software readability, modularity and reusability. In this paper, before applying refactoring techniques, bad smells are detected using some object-oriented metrics such as source lines of code, cyclomatic complexity, Average lines of code and number of parameters. Metrics are used to evaluate the various aspects of complexity of software code. The graphical user interface (GUI) window based applications are developed to detect bad smells in c# code using visual studio ultimate tool 2012. From simulation of refactoring techniques, it has been observed that extract method refactoring technique is easier and best way to remove long method bad smell for reducing complexity and duplicate code.

Keywords

Software Refactoring, Bad Smell, Object-Oriented Metrics, Refactoring Techniques.

Full Text

   

Anmol Kaur ¹, Raman Maini ²

Affiliations
1 Department of Computer Engineering, Punjabi University, Patiala, Punjab, IN
2 Department of Computer Engineering, Punjabi University, Punjab, IN

Abstract

Web Mining can be defined as highly expressive tool which is used to fetch the required information. The distilled data can be used to ameliorate the Web Usage Mining. It helps in the prediction of next accessed by the user, financial data analysis, intrusion detection. There are different steps to clean the data and then it is analyzed through different techniques to find relevant results. To find the results Web SIFT tool is used. Web Usage Mining consists of three parts which Web Usage Mining, Web Content Mining and Web Structure Mining. From the studies it has been analyzed that Web Usage Mining deals with the how to predict the behavior of the users to increase efficiency of the website by collecting the log files from the server. Web Content Mining is used to find out the relevant information from the various documents of the web, while Web Structure Mining comprises of web pages as nodes and hyperlinks as edges.

Keywords

Pattern Analysis, Pattern Discovery, Log File, Web Serve, Client, Visualization and Data Preparation.

Full Text

   

Gurpreet Singh ¹, Raman Maini ²

Affiliations
1 Department of Computer, Punjabi University Patiala, Punjab, IN
2 Punjabi University, Patiala, Punjab, IN

Abstract

Mobile ad-hoc networks are known by their continuously changing topology, no permanent infrastructure, resource restrictions and multi-hop scenario. Owing to this, they are more exposed to security attacks. In these networks, one of the hazardous attacks is packet dropping attack differentiated into 2 types: 1. Black hole and 2. Gray hole. Within both of them, an attacker replies to source node incorrectly that it is encompassing the shortest path to the destination. In former, attackers drops the entire volume of packets sent through it and in latter, attacker drops packets selectively. The following study implement to detect and isolate these attacks in network. The performance of the network has been compared without black hole node and with black hole node on the basis of PDR, throughput and energy consumption of the network. It has been observed that the actual performance of the network without black hole node is better as compare with the network with black hole node.

Keywords

Mobile ad hoc Networks, Black Hole Attack, Grey Hole Attack, Throughput.

Full Text

   

Anupama Khatak ¹, Raman Maini ¹

Affiliations
1 Department of Computer Engineering, Punjabi University, Patiala, Punjab, IN

Abstract

Over the past decades Wireless Sensor Networks WSN have shown a great change due to its advancement in technology and adaptability nature. It is that breed of organization that can be found in number of small devices namely sensors nodes. These are usually arranged in remote and distant areas. Every sensor node consists of radio interface to communicate with each other, also these nodes are limited because of the batteries embedded in them. When various sensors mutually monitor large sensible environments, they present a wireless sensor network (WSN). They are adaptable to various environments so they maintain vast range of functions. In this work various issues, challenges, and characteristics related to wireless sensor network are discussed. Since wireless devices should be bandwidth limited and small, a few main challenges in wireless networks are Signal fading, mobility, data rate, power and energy, security and Quality of service (QoS). Also, it has been observed that Wireless sensor networks have motivated various operations in the plots like health, environment, human activity monitoring etc.

Keywords

WSN, Sensors, Networks, Nodes, Issues, Challenges.

Full Text