Henry Chiapa, Jarod Klinefelter, Leland Ralston
Completed at Penn State Behrend as a Senior Design Project Sponsored by Dr. Myung Cho
September 2020 - May 2021
In signal processing, there exists a problem known as subspace tracking that affects a variety of applications in radar, sonar, and others. In high-dimensional settings, subspace tracking becomes an increasingly difficult task to complete without highly incomplete observations. This project is motivated by the need to recover missing data that is collected by arrays of sensors. The goal of this project is to implement a proposed algorithm by the sponsor to recover missing data in a two-dimensional space. The proposed algorithm, Shift-Invariant Subspace Tracking (SIST), already exists to perform data recovery in a one-dimensional space. It utilizes a folded Hankel matrix structure that performs recursive least-squares algorithm to update and construct the folded Hankel matrix. Once data is recovered, a common algorithm known as ESPRIT is run to extract the frequency locations of the signals. The SIST algorithm is to be tested against two other contemporary subspace tracking algorithms, GROUSE, and PETRELS. It is expected that SIST will greatly outperform the other algorithms with less observations available to work with.
Project whereas signal processing, mathematical, and project management skills were utilized.
Expanding upon current research in moving target tracking (MTT) with missing data.
Developing an algorithm, Enhanced Shift-Invariant Subspace Tracking, that handles high-dimensional arrays of sensors that may be implemented for radar/sonar technologies.
Utilizing a folded Hankel matrix, a specialized mathematical structure, we are recovering incomplete/inaccurate sensor data.
Henry Chiapa and Timothy Volkar
Completed at Penn State Behrend as a Digital Design Project
November 2019
We intend on producing a home security system that uses an infrared motion sensor to detect intruders in one’s home. Since home security is becoming much more commonplace, a primary method of monitoring a home is using motion detection to detect intruders. To do this, we are using a hexadecimal pin for our user input and an infrared motion detector for triggering the alarm. There is set of modes the security system may be in, similar to current on-market security systems. These include idle, alert, intruder, arming, and an armed mode for the system. The actual architecture of the system is comparable to a control & datapath architecture, utilizing a control finite state machine and datapath with components thereof. It is our hope that we can develop a simple security system that then can be expanded in the future to include additional home security measures.
Project whereas digital design and computer architecture skills were utilized.
Artix-7 FPGA system written in VHDL utilizing general purpose input-output (GPIO) hardware completing a datapath & control design to perform home security tasks.
Expandable and adaptable design to perform for larger scale purposes.
Henry Chiapa, Bryan Bentz, and James Ferrare
Completed at Sterling Technologies Inc. as an IT Inventory Specialist
May 2021 - July 2021
Made modifications to the Global Shop ERP software to delineate warehouse bin locations.
Developed business workflows for warehouse organization.
Integrated wireless communication systems with the central business operation systems.