Hi, I'm Pasan Chamikara.

AutoSOC: A low budget flexible security operations platform for enterprises and organizations

This study introduces the Secure Physical Drive, a digital vault, as a response to prevailing security vulnerabilities in current data transfer techniques. Incorporating a physically fortified enclosure, IoT-integrated sensors, and a robust tracking algorithm, this drive guarantees maximal security during large amounts of critical data transport physical drives. The proposed solution notably mitigates the likelihood of data breaches during transit, establishing itself as a reliable shield for safeguarding classified and sensitive information. Through an exhaustive assessment of existing data transfer methods, the paper analyzes the Secure Physical Drive's efficacy against diverse attack scenarios. This groundbreaking methodology holds substantial promise in bolstering data security during transit, thus providing a resilient defense against potential breaches and attacks.

https://ieeexplore.ieee.org/document/9107599

Co-Authors: MIM Cooray, LSAM Wickramasinghe, YMS Koshila, KY Abeywardhana, AN Senarathna

A-STEM: Advanced Storage Toolkit for Encryption and Migrations

This study introduces the Secure Physical Drive, a digital vault, as a response to prevailing security vulnerabilities in current data transfer techniques. Incorporating a physically fortified enclosure, IoT-integrated sensors, and a robust tracking algorithm, this drive guarantees maximal security during large amounts of critical data transport physical drives. The proposed solution notably mitigates the likelihood of data breaches during transit, establishing itself as a reliable shield for safeguarding classified and sensitive information. Through an exhaustive assessment of existing data transfer methods, the paper analyzes the Secure Physical Drive's efficacy against diverse attack scenarios. This groundbreaking methodology holds substantial promise in bolstering data security during transit, thus providing a resilient defense against potential breaches and attacks.

https://ieeexplore.ieee.org/document/10417489

Co-Authors: Ravindu Prasad Pathirana, Heshan Keshawa Basnayaka, Avishka Kanchana Bandara, Ravindu Manupriya Rajasinghe, Amila Nuwan Senarathne, Pradeep Abeygunawardhana

HoneyIQ: Deep Reinforcement Learning-based approach for Enhancing honeypot adaptability (Ongoing)

Honeypots could be utilized within a given network for several reasons. One is that it would waste resources of the attackers and the second being it would provide ample information about the behavior of the attacker. Threats do differ from one session to another. Therefore, for new threats to be efficiently identified and treated based on their severity, it would be required to utilize a more dynamic method. The deep Reinforcement Learning (Deep Q-Learning based) approach would be one way to solve this kind of issue. Most of the solutions existing within the present market treat the attackers without considering their severity. But to keep the attacker in the loop, it would be required to dynamically adjust the honeypots. To dynamically adjust the honeypots, a deep-Q-learning-based microservice is proposed. The service would take the inputs from the deployed collector agents in honeypots and do the processing which would send recommendations to the respective honeypots to adjust their behaviour. Through this research it is also expected to evaluate the efficiency of the solution against Zero-day attacks.

Co-Contributors: Dr. Thanuja Ambegoda