Papers by Somanath Tripathy
GLOBECOM 2022 - 2022 IEEE Global Communications Conference, Dec 4, 2022
Bookmarks Related papers MentionsView impact
GLOBECOM 2022 - 2022 IEEE Global Communications Conference
Bookmarks Related papers MentionsView impact
Information Systems Security
Bookmarks Related papers MentionsView impact
2021 IEEE International Conference on Advanced Networks and Telecommunications Systems (ANTS)
Bookmarks Related papers MentionsView impact
2022 IEEE International Conference on Blockchain and Distributed Systems Security (ICBDS)
Bookmarks Related papers MentionsView impact
arXiv (Cornell University), May 15, 2019
Bookmarks Related papers MentionsView impact
2019 IEEE International Symposium on Smart Electronic Systems (iSES) (Formerly iNiS)
This paper presents an efficient method to classify an audio event automatically. The proposed me... more This paper presents an efficient method to classify an audio event automatically. The proposed method blindly detects audio events before feature extraction which can be considered a novelty. Once an audio event is detected; after that, every frame belongs to the event is passed through the feature extraction and classification stages. Mel-scaled filter bank energy features were extracted frame-wise. The extracted features were classified into eleven subsequent classes of audio events using deep neural network (DNN) technique. The proposed methodology was validated on the TUT DCASE-2016 task 2 database which is a synthesised office environment with a controlled level of background noises, i.e., 6dB, 0dB and -6dB SNR. The proposed methodology reported an accuracy of over 96% in detecting audio events blindly whereas classification accuracy is reported to be over 80%. Obtained results were compared with the state-of-art methodologies reported in the literature.
Bookmarks Related papers MentionsView impact
Journal of Information Security and Applications
Bookmarks Related papers MentionsView impact
2021 22nd International Symposium on Quality Electronic Design (ISQED)
Internet of Things (IoT) devices have made their presence felt across the domain, society, and in... more Internet of Things (IoT) devices have made their presence felt across the domain, society, and individuals. These technologies have played a prominent role in shaping the digital world. However, they bring their own set of security and privacy threats. These devices being resource-constraint, are unable to mitigate the challenges with the existing traditional solutions. Various software and hardware solutions have been tuned for IoT security. Physical unclonable function (PUF) and random number generator (RNG) are the most useful for building security applications, especially in the resource restrained devices like IoT. The security protocols, including identification, authentication, and key-agreement, can be developed using PUFs, while RNGs can produce ephemeral keys and nonces. TRNG and PUF as a reconfigurable circuit, reduce the hardware cost and prove to be an effective solution. A memristor being the fundamental circuit element offers certain unique advantages, including the possibility of hybridization with complementary metal-oxide-semiconductor (CMOS) circuits. This paper proposes a low-cost re-configurable TRNG-PUF named (TRGP), which can harness the benefits of both the TRNG and PUF. We evaluate their performance against various parameters and compare them with some of the exiting PUF and TRNG architectures.
Bookmarks Related papers MentionsView impact
Network and System Security, 2017
HTTP cookie plays an important role in web applications, as it is used for session authentication... more HTTP cookie plays an important role in web applications, as it is used for session authentication without using the login information repeatedly. On the other hand, such technique introduces several security vulnerabilities allowing an attacker, to have the complete control of a session by extracting the corresponding cookie. Therefore, HTTPS is recommended to prevent the exposure of cookie. Unfortunately, cookie can be extracted by different techniques even if HTTPS is employed. This work proposes a simple but effective solution called CookiesWall to prevent session hijacking. CookiesWall is implemented as a client side proxy using Python. The proposed mechanism imposes negligible overhead. False positive and false negative of this mechanism is observed to be much lesser.
Bookmarks Related papers MentionsView impact
IEEE Transactions on Computational Social Systems, 2021
The tremendous popularity of online social network (OSN) services in recent years has offered a n... more The tremendous popularity of online social network (OSN) services in recent years has offered a new way of information sharing. OSN services allow the spread of information like fire in a forest, among the target audience. On the other hand, these services raise serious concerns about the privacy of their users. In this article, we propose a novel exponential model to measure the visibility of tweets exploiting the trust and interest of the followers. Furthermore, we develop a polynomial regression model and a deep neural network (DNN) model for visibility prediction. The experimental results with the real data show that the exponential model achieves better visibility prediction accuracy in comparison to the regression model and DNN model for a specific value of hop count.
Bookmarks Related papers MentionsView impact
ACM Transactions on Multimedia Computing, Communications, and Applications, 2021
Clinical trials and drug discovery would not be effective without the collaboration of institutio... more Clinical trials and drug discovery would not be effective without the collaboration of institutions. Earlier, it has been at the cost of individual’s privacy. Several pacts and compliances have been enforced to avoid data breaches. The existing schemes collect the participant’s data to a central repository for learning predictions as the collaboration is indispensable for research advances. The current COVID pandemic has put a question mark on our existing setup where the existing data repository has proved to be obsolete. There is a need for contemporary data collection, processing, and learning. The smartphones and devices held by the last person of the society have also made them a potential contributor. It demands to design a distributed and decentralized Collaborative Learning system that would make the knowledge inference from every data point. Federated Learning [21], proposed by Google, brings the concept of in-place model training by keeping the data intact to the device. T...
Bookmarks Related papers MentionsView impact
2018 International Conference on Advances in Computing, Communications and Informatics (ICACCI), 2018
The emergence of various Online Social Network (OSN) services has revolutionized the way people e... more The emergence of various Online Social Network (OSN) services has revolutionized the way people express themselves among their social connections and to the world. Twitter is one of the most popular OSNs, which allows its users to share ideas with their followers and public, in the form of tweets. Visibility prediction of a tweet is an interesting issue that might be useful in estimating privacy risk caused by the tweet. In this paper, we propose a technique inspired by epidemic models to predict the visibility of a tweet. Our model exploits user interest and relationship strength to predict the visibility of a tweet. The evaluation results show that one can predict the total number of likes and re-tweets of a tweet with the accuracy of approximately 89%.
Bookmarks Related papers MentionsView impact
Distributed Computing and Internet Technology, 2017
Implementing Peer to Peer (P2P) system on a cellular network is an interesting idea to provide di... more Implementing Peer to Peer (P2P) system on a cellular network is an interesting idea to provide distributed storage, that has caught attention of many researchers. Leaving aside the legal issues, overcoming technical challenges are key to the development of cellular based P2P business applications. In this paper, we propose a mobile P2P file sharing system called GMP2P on GSM-GPRS network. GMP2P integrates distributed hash table (DHT) mechanism into GSM mobile stations and base transceiver stations. The proposed solution addresses the issues related to efficient P2P file sharing over GSM-GPRS network without requiring a centralised server. The communication cost involved in searching and downloading of the shared files is also analysed and the results are compared with existing mobile P2P schemes. GMP2P is found to be more efficient and scalable.
Bookmarks Related papers MentionsView impact
Proceedings of the 19th International Conference on Distributed Computing and Networking, 2018
Content sharing in cellular infrastructure including 3G, 4G or LTE through mobile phones or PDAs ... more Content sharing in cellular infrastructure including 3G, 4G or LTE through mobile phones or PDAs is now at a premature stage. The users need to pay a fee to the service providers per content access. A number of schemes, assisted by centralized web servers, try to reduce the cost of data usage through Bluetooth or WiFi mechanisms. However, a user still has to pay a fee in order to find the location of the host server and then to download the contents. The cost could be reduced by leveraging P2P file sharing through Vicinity Chord ring (V-Chord) proposed in this paper. V-Chord is a Chord [13] like data structure based on the locations of the mobile users. The analysis of V-Chord establishes that the proposed scheme lead to a reduced cost for data usage compared to other existing schemes. An implementation of a P2P file sharing system over mobile phones, based on V-Chord, has been designed and tested using J2ME midlets.
Bookmarks Related papers MentionsView impact
TENCON 2019 - 2019 IEEE Region 10 Conference (TENCON), 2019
Intrusion Detection Systems (IDS) has become an indespensive tool to protect the network by detec... more Intrusion Detection Systems (IDS) has become an indespensive tool to protect the network by detecting the attacks. As each attack is associated with a large number of attributes, it is challenging to select a good set of attributes for achieving better classification accuracy. This work proposes an effective unsupervised feature selection technique using Genetic algorithm (GA) for analyzing IDS data. The search capability of GA has been utilized for optimizing different unsupervised feature quality measures including Pearson correlation, mutual information, and entropy. Different combinations of these features are utilized as fitness functions of the proposed GA based framework. The algorithm is able to find out that subset of features which are uncorrelated and mutually exclusive to each other. Finally, the optimal feature subset obtained is utilized for developing classification systems using some popular machine learning models like decision trees, support vector machines, k-nearest neighbor classifier on the KDD-Cup 99 dataset. The experimental results show that decision tree produces better results than other classifiers. The result confirms 99.62% accuracy, 98.78% detection rate and 0.25% false alarm rate. The most attractive feature of the proposed scheme is that it does not require any labeled information during the feature selection process.
Bookmarks Related papers MentionsView impact
Information Systems Security, 2017
Bookmarks Related papers MentionsView impact
2019 IEEE Region 10 Symposium (TENSYMP), 2019
Nowadays, Software Defined Networking (SDN) is an emerging paradigm which facilitates easy manage... more Nowadays, Software Defined Networking (SDN) is an emerging paradigm which facilitates easy management with enhanced reliability. OpenFlow enables the network controller to direct the packets across the switches in the network, so has many advantages. But, OpenFlow SDN networks are exposed to Denial-of-Service (DoS) attacks in which attacker consume the bandwidth of the the control plane and overload the butter memory of OpenFlow switch by sending bogus packets with non existence destination (random) IP. In this work, we propose a mitigation technique called SDNGuard to address this issue. SDNGuard extends the existing OpenFlow by adding a dataplane cache and a simple packet migration module. We evaluate SDNGuard and the results show that this model is effective in reducing the saturation attack.
Bookmarks Related papers MentionsView impact
Computers & Security, 2021
Abstract In today’s cryptocurrency, the Payment Channel Network (PCN) is noticed as one of the mo... more Abstract In today’s cryptocurrency, the Payment Channel Network (PCN) is noticed as one of the most gifted off-chain solutions for scalability issues. Besides this, it consumes lesser transaction fees and low transaction confirmation time. However, security and privacy issues need to be addressed appropriately to make the solution even more effective. Most of the existing HTLC (Hashed Time-Lock Contract) protocols revealed the sender’s information to the intermediate users in the payment route. In this work, we propose an effective secure and privacy-preserving Payment Channel Network protocol, named Neo Hashed Time-Lock Commitment ( n -HTLC) protocol. ( n -HTLC) does not require the sender to send any information to each intermediate user along the payment route, thus preserves the identity of the sender. But, ( n -HTLC) is not compatible with Sphinx onion packet format. Therefore, a symmetric key encryption-based protocol called k-TLC has been proposed. k-TLC is compatible with the Sphinx onion packet format, which is used in the current Lightning network atop of the Bitcoin blockchain network. The security of both n -HTLC and kTLC are proved using the Universal Composability (UC) framework. It is observed that both ensure that no attacker can extract information on the payment route if at least one of the users in the path is honest. To analyze the performance of both n -HTLC and kTLC payment protocol, we conduct experiments using the snapshots of Ripple network 1 , Lightning network 2 , and synthetic network of Mazumdar and Ruj (2020) . Our experimental results show that both n -HTLC and kTLC outperform state-of-the-art off-chain payment protocols in terms of computational and communication overhead.
Bookmarks Related papers MentionsView impact
Medical data held in silos by institutions, makes it challenging to predict new trends and gain i... more Medical data held in silos by institutions, makes it challenging to predict new trends and gain insights, as, sharing individual data leaks user privacy and is restricted by law Meanwhile, the Federated Learning framework [11] would solve this problem by facilitating on-device training while preserving privacy However, the presence of a central server has its inherent problems, including a single point of failure and trust Moreover, data may be prone to inference attacks This paper presents a Distributed Net algorithm called DNet to address these issues posing its own set of challenges in terms of high communication latency, performance, and efficiency Four different networks have been discussed and compared for computation, latency, and precision Empirical analysis has been performed over Chest X-ray Images and COVID-19 dataset The theoretical analysis proves our claim that the algorithm has a lower communication latency and provides an upper bound © Springer Nature Switzerland AG ...
Bookmarks Related papers MentionsView impact
Uploads
Papers by Somanath Tripathy