Title: Soft Computing Assisted Cryptographic Technique based on Tuning of Grouped Triple Hidden Layer Perceptron
Dr. J. K. Mandal
Professor, Department of Computer Science & Engineering,
Ex-Dean, Faculty of Engineering Technology & Management
Director, IQAC and Chairman, CIRM, Kalyani University
Honorary Chairman, CSI Kolkata Chapter
University of Kalyani, Kalyani, Nadia 741235, West Bengal, India.
Telefax:(033) 25809617, M:(91) 9434352214, (R) (033) 25736302
M. Tech.(Computer Science, University of Calcutta), Ph. D.(Engg., Jadavpur University) in the field of Data Compression and Error Correction Techniques, Professor in Computer Science and Engineering, University of Kalyani, India. Former Dean Faculty of Engineering, Technology & Management 2008-2012(two consecutive terms), 29 years of teaching and research experiences. Served as Professor, Computer Applications, Kalyani Govt. Engineering College for two years. Served as Associate and Assistant Professor at University of North Bengal for sixteen years. Life Member of Computer Society of India since 1992 and life member of Cryptology Research Society of India. Member of AIRCC. Honarary Vice Chairman of CSI Kolkata Chapter. Chairman elect CSI Kolkata Chapter 2016-17. Working in the field of Network Security, Steganography, Remote Sensing & GIS Application, Image Processing, Wireless and Sensor Networks. Domain Expert of Uttarbanga Krishi Viswavidyalaya, Bidhan Chandra Krishi Viswavidyalaya for planning and integration of Public domain networks. Chief Editor, Advanced Computing: An International Journal, Associate Editor (Guest), Microsystem Technologies, Springer, Editor, Journal of Data Science, Inderscience, Editor of Proceedings of ETCS 2012, NIDS-98 and ERC-95 of CSI. Seventeen Scholars awarded Ph.D., Two submitted till January 2016 and 8 scholars are pursuing for their Ph. D. degree. Published five books from LAP- Lambert Academic Publishing, Germany and one book from IGI Global publishers, Indexed by Thomson Royter. Total number of publications 360 including 135 publications in various International Journals. Edited fifteen volumes as volume editor from Science Direct, Springer, CSI etc., Organizing various international Conferences of Springers and Science Direct.
In this paper a novel soft computing assisted cryptographic technique, based on synchronization of Grouped Triple Hidden Layer Perceptron (GTHLP), has been proposed. The most important hazard of cryptography is how to firmly swap over the shared secrets between the parties. As a result, key exchange protocols are mandatory for transferring keys in a protected manner. As the same time as key exchange protocols are developed for exchanging key between two parties, many applications do necessitate the need of swapping over a secret key among group of parties. In existing technique each communicating party has to synchronize with other. So, if there are n parties then total number of synchronizations needed is O(n2). This is quite computationally complicated especially in wireless communication where the computational power and the resource constrain is a major issue. GTHLP technique of this paper eliminates all the above stated drawbacks of the existing technique. GTHLP introduces Grouped Triple Hidden Layer Perceptron (GTHLP) synchronization mechanism for offering synchronization of group of parties. Here a key swap over by synchronization among cluster of THLP has been proposed which is a fresh addition to the field of cryptography. The proposed technique implements the key swap over technique with the help of complete binary tree framework which makes the technique scales logarithmically with the number of parties participating in the key swap over protocol. Two parties can swap over a common key using synchronization between their own CTHLP. But the problem crop up when group of n parties desire to swap over a key. Using proposed technique a set of n parties can be able to share a common key with only O(log2 n) synchronization steps. This is logarithmic complexity and feasible in wireless communication with limited amount of resources. Here, CGTHLP based synchronization is performed for tuning of group of parities by placing on the complete binary tree framework. Synchronization initiated between nodei and nodej to construct a common seed value at both sides. The synchronized identical seed value is used to generate the common input vector for nodei and nodej. Two GTHLPs one at nodei and another at nodej start with identical input vector and anonymous random weight vector. In each time both THLPs compute their final output based on input and weight vector, and communicate to each other. If both are be in agreement on the mapping between the present input and the output, their weights are updated according to an appropriate learning rule. After synchronization procedure of all parties (nodes in a complete binary tree) in the group weight vector of the group THLPs become identical. These indistinguishable weight vector forms the session key for a particular session. Authentication steps also get performed parallel to the synchronization steps. This synchronized network can be used for transmitting message using any light weight encryption/decryption technique with the help of session key of the synchronized network.
Title: Approximation for Minimum Connected Dominating Sets in Wireless Ad hoc and Sensor Network
Dr. Anil K Yadav
Assistant Professor Computer Application Department
University Institute of Engineering & Technology
C.S.J.M University Kanpur -208024 India
Dr. Anil K Yadav is Asst. Professor and Head Computer Application Department at University Institute of Engineering & Technology, C.S.J.M University, Kanpur, India since October 2005. He is a member of Board of Studies of Computer Application, MTech (CSE) and Academic Council of C.S.J.M University, Kanpur. He obtained MTech from GGSIP University Delhi and Ph.D. in Computer Science and Engineering from MNNIT Allahabad. He has authored one Book entitled “Connected Dominating Set: Theory and Experimental Study” by LAMBERT Academic Publishing, Germany. His research interests are graph algorithm for ad hoc and sensor networks, combinatorial optimization and big data analytics. He is a recipient of various Travel Fellowship to present, participant in National and International Conferences. He is a reviewer of International Journals like IEEE Sensor Networks and International Journal of Electronics (Taylor and Francis) and TPC of various International Conferences. He is life member of Computer Society of India and serving in the capacity of Vice Chairman CSI Kanpur chapter. For more information, please visit www.uietkanpur.org.
Researcher and engineers always want faster and faster computing devices, and in general faster devices require more energy. With increasing energy costs, there is an urgent need for energy-efficient computing in Wireless ad hoc and sensor networks at many different levels. This talk focuses on energy-aware resource management in heterogeneous centralized, parallel and distributed computing systems. We address the problem by approximation, generating efficient minimum Connected Dominating Set in wireless ad-hoc networks. The execution time and energy consumption of each node is based on how the task’s computational requirements interact with the node’s capabilities. We have designed models for defining, deriving, and quantifying the degree of robustness of a node behavior using stochastic (probabilistic) information about the execution times of tasks on different nodes. We provide an analysis framework that will allow a system to investigate the tradeoffs between minimum and maximum dominating set and maximizing the computing performance (utility) achieved by a system. The CDS approaches presented can be applied to a variety of computing and communication system environments, including centralized, parallel, distributed, cluster, wireless ad hoc and sensor networks. Furthermore, the approaches can be used with many different system performance metrics and constraints. This talk is intended for undergraduate, postgraduate students, research scholar, faculty and scientists who want to learn how to model and manage resources in parallel and distributed computing systems with energy-aware. In particular, energy can be used as a constraint when trying to optimize a system computing performance metric, or energy can be optimized while meeting a computing performance constraint goal.