College of Engineering, Pune

About: College of Engineering, Pune is a based out in . It is known for research contribution in the topics: Sliding mode control & Control theory. The organization has 4264 authors who have published 3492 publications receiving 19371 citations. The organization is also known as: COEP.

Advanced Encryption Standard (AES) implementation on FPGA with hardware in loop

Abstract: In cryptography, Advanced Encryption Algorithm (AES) is a computer security standard defined by National Institute of Standards & Technology (NIST). It is a symmetric key algorithm which uses block of 128 bits of input data & key with different sizes. The size of key can be 128,192 or 256 bits. This paper explains a hardware platform for implementation of AES algorithm with key size 128 bits. Field Programmable Gate Array (FPGA) Artix 7 Nexys 4 kit is used to perform the algorithm by configuring it with the help of Software Development Kit (SDK) in Xilinx ISE design suite. Microblaze is the soft core processor used for the interface between hardware & software. The data is transmitted to the FPGA with the help of MATLAB.

Rethinking Decentralised Identifiers and Verifiable Credentials for the Internet of Things

Abstract: In digital era of Internet of Things (IoT), internet-connected devices are becoming smart and autonomous. In this distributed world IoT device operate not only with the owner of the device but also need to operate with third parties, which leads to have identification for each IoT device. In the coming years, deploying IoT solutions at large scale will have to address complex issues of managing digital identity, security and identity ownership. Most of the current IoT solutions rely upon centralized client-server mechanism for providing connectivity and access to devices and services. However, when billions of devices will begin connecting simultaneously, then existing centralized solution will not suffice. Now days, we rely on federated identities however in this user need to depend on third party as users are not owner of their own identity, this leads to have Self-Sovereign Identities (SSI). In the sequel, there is a need of decentralized method for managing identities in near future. In this chapter, various identification methods are analyzed. Further, Decentralized Identifiers and Verifiable Credentials are discussed, the discussion focus on whether DID and VC are applicable to the IoT as IoT includes resource constrained devices at base level. It also presents smart home use case utilizing DID and VCs in order to evaluate its applicability. In the same line threat analysis is presented to analyze applicability of the DID and VC for the IoT applications.

Multimodal biometrics by integrating fingerprint and palmprint for security

Abstract: Multimodal Biometrics there is a combination of different biometric modalities. In most biometric systems from real world applications uses a single source of biometric modality for authentication that is only a single fingerprint, palm print, face, voice or iris which is known as unimodal biometrics. Some of the drawbacks from unimodal biometrics are eliminated by combining information from multiple biometrics systems for unique personal identity. This paper gives a multimodal biometrics system that combines features of fingerprint and palm print to overcome several limitations of unimodal biometrics. Preprocessing is a heart of image processing in which feature enhancement is prior step, in this system feature enhancement of input image that is fingerprint and palm print are performed by applying a series of preprocessing techniques. Modified Gabor filter is used to independently extract a fingerprint and palmprint feature which provides more accuracy as compared to traditional Gabor filter In addition to this Short Time Fourier transformation is applied for better quality of resultant images. In the later step the resultant images are combined in feature level fusion method and finally using Euclidean distance method the features are classified to match the resultant image with database template. We conclude that proposed methodology has better performance as compared to unimodal approaches using individually only a fingerprint or a palm print. The multiple biometrics helps to reduce the system error rate.

Electric Vehicle Li-Ion Battery State of Charge Estimation Using Artificial Neural Network

Abstract: State of Charge is an important parameter of the Battery Management System. It is an indication of the remaining battery capacity. It is very important to have an accurate estimate of the State of Charge, to avoid over charging and over discharging of Lithium ion batteries used in the Electric Vehicle. In this project, we have developed a novel algorithm for the estimation of SOC using Artificial Neural Networks. A comparison of the SOC calculations from the current Battery Management System used in the Electric Vehicle and SOC estimation from the Artificial Neural Network has been presented.

Face Detection and Tracking Using Image Processing on Raspberry Pi

Abstract: Image processing is nowadays used as a piece of various applications. Confront area and its following is one of the basic frameworks used as a piece of the uses of image processing. in this paper we have shown an application for the improvement of the auto applications with the help of face revelation. The paper shows the examination how the assurance of the camera impacts the edge each seconds and respectably the perfect open door for stand up to acknowledgment. The image processing using OpenCV for face detection, face tracking and its recognition for automotive application is done here. We have used processor controller and common usb camera for controlling the electrical motor which in this way controls the motorized auto application. the viola-jones estimation for stand up to acknowledgment using adaboost library in this way structure relies upon adaboost computation and changed works faces haar-like highlights [5] in the item part we have used python vernacular with opencv reinforce for the face distinguishing proof using haar course records gave by intel open source. we have shown the results on layout consistently impact on stand up to distinauishing proof.