Category: CSE Seminar Reports

This paper discussed about response mechanisms in Intrusion Detection System (IDS) and also discussed about few IDS based on intelligent agents.

Overview:

Intrusion Detection is a main objective of Intrusion Detection Agent System.  The other objectives of this system are responding to the attacks and tries to tracks down the attacker’s machines.  Intruders are tracked efficiently by looking at MLSI (Mark Left by Suspected Intruder).  Manager, Information gathering agent and Tracing agent are three types of mobile agents. Information about intrusions can be gathered by IDS sensors from system logs.

 IDS sensors inform the managers whenever they detect MLSI.  Then a tracing agent, which can activate the information-gathering agent and employs a message box upon arrival, will get dispatched in target host by the manager.  After getting the additional information, information gathering agent returns back to the manager and informs about the additional information. 

To response to intrusion, a system must use large number of agents and have very elaborative mechanism. In AAIRS, IDS sensors give the output to interface agents who can keep track of IDS reports credibility. Here IDS middle tier components are implemented as mobile agents. ADEPTS system contains Intrusion Directed Acyclic Graph (I-DAG). Every node in I-DAG has Effect Service Set and Cause Service Set. Cause Service Set contains all the services which compromised to achieve the goal and Effect Service set contains all the services which will get compromised only after the goal is achieved.  ADEPTS system accepts multiple alerts simultaneously and for each alert, a separate I-DAG will get initiated. Confidence will assign to the node, if there exists a node in I-DAG corresponding to the intrusion.

Conclusions:

IDs mostly have response as its secondary activity.  For every response in IDA there are more than 100 responses in ADEPTS.  All models have provision for response and consider mostly parallel attacks.  All the models fit in distributed infrastructure, based on three mobile agents. There is a possibility of the response to failed in ADEPTS case. 

Download  B.tech CSE Seminar Topic with Report for Response Mechanisms in Ids  On Intelligent Agents  .

Introduction to Data Mining Technique Helps To Reduce Test Suite Seminar Topic:

This paper discussed about data mining technique which helps to reduce test suite and thereby facilitating the knowledge extraction from test cases. Both functional and non-functional requirements from SRS document transformed into state charts. These state charts will helpful to generate different test cases automatically for a single planning problem. Test cases generated from state charts are used for manual software testing or automated testing at system level.

Overview:

There is a need to have an effective test automation methods or tools to address many issues in testing like Test ware creation and its inconsistent maintenance. Data mining is a powerful technique which helps to create patterns from an existing database. Data mining techniques helps to manage the project by maintaining project related valuable data and produce higher quality software systems. The three steps involved in applying data mining technique to the current project is generation of classification rules, create the UML state machine, then generate test cases from it and applying data mining techniques to the generated test cases to reduce the test suite size.

The WEKA classifier system generates classification rules and those rules will get apply to both functional and non-functional requirements. State machines which can specify the current state\behavior of the system and these states machines can be created based on non-functional requirements.

  In a software testing world, state models are used to derive test cases from state machines. The main target of model-based testing is to automate the generation of test cases process from test models. Predicate selection, Predicate transformation and Test data generation are the three steps involved in generating the test cases from state machines.

Conclusions:

The process of covering patterns from data is termed as Data mining. Verification and validation of patterns on data samples is one of the important part in Data mining process. For mining different kinds of data, there are many methods available including clustering, characterization etc. 

Download  B.tech Seminar Topic for CSE Students on Data Mining Technique Helps To Reduce Test Suite .

Introduction to Captcha Seminar Topic:

This paper is about a challenge-response test or a Turing test, Captcha which is administered by a machine and targeted to a human.  

 Overview:

Captcha have simple problems to solve by users rather than having difficult problems in artificial intelligence. Captcha helps in developing artificial intelligence of computers by creating incentives. It has the capability of distinguish humans from computers. Captcha wont depends on difficult problems in artificial intelligence.  

Captcha provides distorted image by showing letters or digits to the users and then allow the users to type them. Captcha can be in the form of distorted images, visual puzzle, photo recognition, speech captcha and 3D captcha. Development of 3D captcha is still in progress. Content-based video labeling as a Captcha task, addresses many problems which are encountered by using existing character recognition captcha. Public video databases generate many challenges.

A video captcha is a task in which user watches a video and then provides three tags.   If, once a video is selected, then there is a need to generate challenges using a function with four parameters. In order to enhance usability and maintain security against a video captcha frequency-based attack there is a need to extend the set of ground truth and user supplied tags.

Conclusions:

Captcha helps to prevent comment spam in public blogs. It protects the website registration. It helps to record the votes and managing in a way to avoid duplicate votes in online polls. In password systems, dictionary attacks can be prevented by captcha. Captcha truly guarantees that bots won’t enter or even read a website. A plausible solution against email worms and spam are offered by captcha. Can create customize captcha to address the problem directly and make the resistant implementation. Captcha can prevent unwanted macro-scripting of a standalone application.  Video captcha security is verified by a frequency based attack. 

Download  BE CSE Seminar Topic on Captcha with Report  .

Introduction to Intel vPro Technology Seminar Topic:

This paper discussed about the latest emerging vPro technology by Intel, which helps to achieve maximum benefits to users by minimizing IT efforts.  VPro technology is not a PC itself but it is a set of features built in hardware and motherboard.

Overview:

Intel vPro technology is a combination of many enhancements in hardware, management, maintenance and security of a PC irrespective of its OS or its power consumption. New hardware innovations with compatible software solutions, high manageability and security are the main features of this technology.  Energy efficiency is one of the major benefits of this technology. PC with this technology is running with less power consumption. This technology provides a feature of accessing and managing the PC even it is powered off. This technology helps to identify the threats more quickly before it affects the OS, with the help of third-party security software and later it updates PC irrespective of its power state.  This security check on PCs can be carried out by vPro’s virtualization technology and its virtual appliances.

While roaming, this technology supports encrypted communication. It also supports both wireless and wired network connections. Intel’s Execute Disable bit functionality is also provided by this technology and remained as one of the powerful hidden strength of this technology. This technology helps to run various operating systems on the same machine at the same time. This technology supports the controlling of host network access based on system’s health and this process considers as Network access protection technology which proposed by Microsoft. 

Conclusions:

VPro technology is a powerful technology which helps to rapid enterprises growth around the globe.  This technology helps to provide qualitative service by meeting growing needs with less cost. The investment on this technology is apt and tolerable when considers the benefits offered by implement it.  To use vPro technology, Intel is providing technical support to many enterprises as required.  All the drawbacks and pits of older versions of this technology are overcome by its new versions.

Download  B.tech CSE Seminar Topic on Intel vPro Technology for Final Year CSE Students .

Introduction to Energy Management in Wireless Sensor Networks Using Energy-Hungry Sensors Seminar:

This paper discussed about the energy management schemes which addresses the energy saving issues at the sensor level.  There are some strategies which help to reduce the power consumption at the radio level by considering data aggregation, data compression, topology management and predictive monitoring. 

Overview:

Duty cycling and adaptive sensing are the two main approaches which can reduce the energy consumed by a sensor. The process of waking up the sensorial system only, when a new set of samples are needed and powering it off immediately after the usage is called Duty cycling.   By this strategy, power saving is possible but periodic sensing is typically considered only in few applications, where the process dynamics are stationary.  There is a need of adaptive sensing strategy to adapt the sensor activities dynamically to the real dynamics of the process.  Both adaptive sensing and Duty cycling are complementary approaches.

                 Operating system provides a set of primitives for powering on and off the sensors in order to support duty cycle mechanisms. Later, the application uses such primitives to get data as per the (adaptive) sensing strategy it implements. The general framework of Wireless sensor Networks allows the designers to focus on selecting best adaptive sensing strategy by leaving low-level duty cycling aspects to the operating system. Adaptive sensing approach has been further classified into three approaches namely hierarchical sensing, adaptive sensing and model-based active sensing.

Hierarchical sensing: The main assumption of this technique is multiple sensors are installed on sensor nodes and observe different resolutions and power consumption.

Adaptive sensing: Adaptive sampling techniques include the adapting of sensor sampling rate dynamically based on the correlation among acquired data.

Model-based active sensing: Model based active sensing technique works by building a sensed phenomenon abstraction by using a forecasting model.

Conclusions:

Power saving is possible with an efficient sensing strategy, by reducing the number of samples and also need to reduce the data to be processed. 

Download  Energy Management in Wireless Sensor Networks Using Energy-Hungry Sensors Seminar Report .

This paper discussed about the Wearable computing which is the process of recognizing human activities from body-worn sensors and about the methods which enhances the robustness of activity-recognition systems.  Wearable computers are miniature electronic devices which can be worn by the bearer under or on top of clothing. These computers are useful for applications which require complex computational support.  The main feature of these computers is its consistency.

Overview:

As there is a rapid growth in mobile sensing and computing has developed few methods for recognizing human activities with the help of sensor data.  Wearable computers have the ability to perform multi-tasking.  Wearable computers can attach to the clothes of a human or can directly attach to the body by inserting it into chains or wristbands. The user can wear this computer and can use its power and functionality virtually in their environment.  These computers are ready to use for the user at any time and there is no need to turn on or off these devices.  These computers help to augment the senses.  Wearable computers provide a ubiquitous sensing and computational capabilities. Wearable system helps for the applications which require activity based computing or interaction-based computing.

Annotation or Labeling and Recognition or Spotting are few of the terms which commonly used in wearable activity recognition process. Sensors help to acquires user’s activities or gestures through signals but these Sensors must be small devices which are invisible to outside and should not provide discomfort to the users. These sensors should select based on its wearability, power usage and computational needs.  Sensor modalities are body-worn accelerometers, which are very small and have low power.

Conclusions:

The large collection of human behaviors by Wearable computing helps to bootstrap humanoid robot behaviors. The main object of wearable computing is Human activity recognition. 

There are many challenges by using activity-recognition systems in robotics. The recognition of human activities for HRI must consider that the activities recognized by machines are not perfectly accurate. 

Download  BE CSE Seminar Topic with Report for Wearable computing  .

Introduction to Security on Digital Library Seminar Topic:

This paper discussed about the meaning of quality in digital libraries (DLs). DLs are differentiated as bad or good based on its services quality and contents. 5S is a standard terminology which discusses the issues in formal framework for digital libraries. This paper also discussed about the quality indicators of Digital libraries.

Overview:

Library RFID Management systems consists of four phases like library security system only, patron self check-in\check-out, support library circulation and quick inventory. The action of library security system only phase is using RFID Tag which replaces EM Security Stripes. The target of this phase is Radio Frequency Detection.  The action of support library circulation phase is using RFID Tag which replaces barcode in circulation process. 

The target of this phase is allowing multiple check-outs and check-ins simultaneously.  The action of patron self check-in\check-out are allowing patron to do check-out and check-in process of books. The target of this phase is providing benefits to librarian and patrons. The action of smart and quick inventory is to search and inventory by using RFID Tag. The target of this phase is quick inventory and searching the misplace material.

            The anti-theft part of the library is LibBest RFID EAS Gates. RFID Tags in the library items are used by RFID Management Systems. When an un-borrowed item passed through any lane then lanes triggers the alarm system.  When a patron passes through with the un-borrowed library material then the alarm will sounds and lights on the gate which will flash at that instance. RFID Tags will get detected by Electronic Article Surveillance Anti-theft Gates.  Whenever an armed RFID tags detected then the alarm will sound on the gate. 

Theft detection is a standalone technology which works independently of the library database. Library RFID Management System can handle material without exception for video and audio tapes. RFID system is an effective library detection system. Automated materials handling is the application of RFID technology. 

Download  Computer Science(CSE) Seminar Report on Security on Digital Library for CSE Students .

Introduction to Face Recognition Technology Seminar Topic:

This paper discussed about “Face recognition” technology which helps to secure applications. In today’s networked world, there is a severe need of providing security to both information and physical properties.  Even though there are many existing technologies which provide security but still it have its own drawbacks.

As it is difficult to forge using biometrics there is a need to have another alternative for providing security to the applications and human.  Usually people identify others by their face and voice, so people feels more comfort with systems that use face and voice recognition. As face recognition is people’s primary method of person identification there is a major focus of research carrying out on this technology because of its non-invasive nature.

Overview:

Geometric approach and pictorial approach are the two approaches of face recognition technology. Geometric approach focuses on the main geometrical features of the face also focuses on geometrical distances and angles between the features. Pictorial approach mainly focuses on the templates of the major facial features.  This technology has a strong relation with the science of neural networks.

There is no need for a user to freeze in order to get captured by this technology. Face recognition is technique of detecting faces and search for an exact match through a dataset whereas face detection is a technique of searching for any matches and stops the search after found. But it is difficult to maintain large databases with plenty of data in this face recognition technique. Many aspects need to consider in the researches of this technology like biology, psychology, physics and computer science.

Conclusions:

Now-a-days, applications developed by face recognition technology are more reliable, non-intrusive and highly accurate. Face recognition technology is one of the most challenging recognition technologies. This paper discussed about the face recognition framework and the factors which affect the recognizer performance. 

Download  Full Seminar Report on Face Recognition Technology for Final Year CSE Students .

Introduction to Computer Science Seminar Topic on Body Sensor Networks:

This paper discussed about a Body Sensor Networks, which are wireless sensor networks designed to operate in a pervasive manner for on-body applications.  Also discussed about the issues with wireless sensors like power optimization, radio design and battery life performance.

Overview:

Generation of ‘hard data’ which shows clinical benefits and cost benefits and this data helps to promote the use of wireless sensor systems in healthcare applications.  But due to lack of ‘hard data’ in assisted living area many of the results reported are having lack of specific details. To demonstrate the automatic properties of sensor network there is a need of end-to-end pilots. In healthcare data must be secure and reliable. Body sensor networks are different from general wireless sensor networks based on its requirements.  The environment for body sensor networks is developed in a way that it can be used by non-technical users to quickly test prototype networks.

The designers of body sensor networks are facing many design challenges, out of which many were inversely related. The parameters need to be considered for designing these Body sensor networks are type and protocol used in wireless transmission, better power consumption, type of sensor device and operations, operation system and  security requirements

To achieve communications and to manage trusted data with much accuracy and authenticity there is a need of Reliable and Robust Body Sensor Networks.  From a wireless sensor mote Rene, wireless sensor operating systems are evolved.  ISO/IEEE 11073, IEEE P1451, Continua Alliance, IHE Initiative, Health Level 7 are the main standards which initiatives pertaining to body sensor networks.  There are many barriers which slowed down the adoption of body sensor networks which includes lack of reimbursement, privacy and security and broadband proliferation. Wireless body sensor networks are used in Tele medicine and tele monitoring.  Many organizations have conducted tests in the development of wearable body sensor technologies.  Researches are carried out to know more science aspects of body sensor networks.

 Download  Computer Science Seminar Report for Body Sensor Networks Topic .

This article is for the awareness of controlling the engineering in a better way. The article tells that instrumentation of the detailed information continuously due to the changes in the degree of the critical confusion state. If any person is able or is capable to understand refer and read the symbols and the short hands that are used in the stated diagrams then is capable of understanding the ANSI instrumentation related symbols and the methods that are used to identify the standard cases of the system and the diagrams too.

Here each every single symbol is used to create a diagram which has objects like the graphic attributes, abbreviations, functioning blocks and also the connecting wire lines that is most necessary object here.

Each and every alphabet in the alphabet series has some or the other functions allocated to it by the system. Here the graphical elements that are needed are divided into four more sub types like discreet instruments, sharing display, functions of the computer and also the programming language that is used to develop the source code for the system. While making this a control called the PLC called as the Programmer Logic Controller is used to show or mention the diagram names properly which has the symbols like the square the triangle etc…  here the letters and the alphabets are combined together and a standard ISA is developed for the  development of the proper system.

Here the first letter of the diagram names the Analysis for A then Flow for F and Temperature for T etc… and in this way each alphabet has some or the other operations. These alphabets which are allocated by each and every operation are standard named as the General instrument or the function symbols. This diagram also has the lines as a diagram which represent the hydraulic signals and sonic signals etc…

 Download  Control Engineering Reference Material .