http://repository.iiitd.edu.in/xmlui/handle/123456789/889 2026-04-10T23:21:50Z http://repository.iiitd.edu.in/xmlui/handle/123456789/1105 Data modeling of financial crises Eshita; Kanjilal, Kiriti (Advisor) The COVID-19 crisis has made it clear that even for developed economies, like the USA, severe nancial disruptions can drag down the whole economy. It is considered the worst nancial crisis since the Great Depression. As the stock market has come to the attention of increasing numbers of researchers, an idea that has emerged to develop a mathematical theory of stock market crashes. This thesis is primarily concerned with data modeling of such a theory. The COVID-19 crisis has been used as a case study to analyse and study nancial crises. Stock market data has been taken from Yahoo Finance website and has been graphed for this analysis. 2020-12-01T00:00:00Z http://repository.iiitd.edu.in/xmlui/handle/123456789/1005 Reconfigurable architectures for online machine learning Santosh, Siripurapu Venkata Sai; Darak, Sumit Jagdish (Advisor) Multi-armed bandit (MAB) algorithms are designed to identify the best arm among several arms in an unknown environment. They guarantee optimal balance between exploration (select all arms sufficient number of times) and exploitation (select the best arm as many times as possible). They are widely used in applications such as website advertisement, robotics, healthcare, finance, and wireless radios. Robotics and radio applications need integration of MAB algorithms with the PHY on the hardware to meet the stringent area, power and latency constraints. Moreover, a single MAB algorithm may not be suitable for various scenarios and hence, the application needs to switch between MAB algorithms on-the-y. We effciently map the MAB algorithms on Zynq System on Chip (ZSoC) and make it reconfigurable such that the number of arms, as well as type of algorithm, can be changed on-the-y. We also exploit the proposed reconfigurable architecture to switch MAB algorithms on-the-y, after initial learning and obtain at least a 10-factor improvement in latency and throughput. Since learning duration depends on the unknown arm statistics, we offer intelligence embedded in architecture to decide the switching instant. To further improve the intelligence of the proposed dynamically reconfigurable architecture, we also propose an efficient aggregation algorithm to adaptively switch between various bandit algorithms in unknown environments. We have also validate the functional correctness and usefulness of the proposed architecture via a realistic wireless application and detailed complexity analysis demonstrates its feasibility in realizing intelligent radios. 2020-12-01T00:00:00Z http://repository.iiitd.edu.in/xmlui/handle/123456789/996 Design of synchronization burst transmitter and detector for 5G physical layer Lodhi, Khalid; Darak, Sumit Jagdish (Advisor) The continuously increasing demand for wireless communication technology has enabled the advancement from 1G to 5G and even beyond 5G. Synchronization is a time-frequency domain communication network that enables the UE ( user equipment ) to transmit and receive the signals efficiently. Therefore, synchronization has used to establish communication in 5G-NRuser equipment. Due to the wide spectrum utilization and less interference, the synchronization in 5G has gained much importance. Hence, in this study, we mainly focus on the development of the synchronization signal burst, which comprises signals such as PSS, SSS, PBCH and PBCH-DMRS. OFDM is the modulation scheme which has been used for the transmission of the synchronization signal Burst. OFDM supports multiple sub-carrier spacing which provides the broad spectrum utilization. In the proposed work, we have implemented the PSS and SSS signals along with the OFDM ( IFFT + CP) which completes the transmission of the synchronization signal burst using Vivado HLS. The OFDM IP has been developed and simulated using the Vivado Design Suite. 2020-05-01T00:00:00Z http://repository.iiitd.edu.in/xmlui/handle/123456789/983 Misinformation in public health Jain, Drishti; Sethi, Tavpritesh (Advisor) Following the tsunami of misinterpreted, manipulated and malicious information growing on the Internet, the misinformation surrounding COVID-19 has taken centre stage. In the context of the current COVID-19 pandemic, publications and social media platforms are particularly vulnerable to rumors and misinformation given the acute uncertainty surrounding the virus itself. At the same time, the uncertainty and new nature ofCOVID-19 means that what may appear to be a "rumor" - yet another piece of unverified information - may be an important indication of the behavior and impact of this new virus. We attempt to tackle this phenomenon by applying different Machine Learning models and Natural Language Processing techniques with a focus on Twitter and web articles. A thorough review of the data and its metrics has also been presented. 2020-12-01T00:00:00Z