Asee peer logo

Wireless Sensor Network for Data Mining in Engineering Projects

Download Paper |

Conference

2023 ASEE North Central Section Conference

Location

Morgantown, West Virginia

Publication Date

March 24, 2023

Start Date

March 24, 2023

End Date

March 25, 2023

Page Count

12

DOI

10.18260/1-2--44939

Permanent URL

https://peer.asee.org/44939

Download Count

143

Paper Authors

author page

Wesley Thomas Noble

biography

Zachary Owen Dickinson

visit author page

Zachary Dickinson is a Cyber engineering student at Gannon University, Erie, PA, and expected to graduate in May 2024. His areas of research interests include embedded systems and hardware security.

visit author page

biography

Ramakrishnan Sundaram Gannon University

visit author page

Dr. Sundaram is a Professor in the Electrical and Cyber Engineering Department at Gannon University. His areas of research include computational architectures for signal and image processing as well as novel methods to improve/enhance engineering education.

visit author page

Download Paper |

Abstract

This paper describes the setup of a wireless sensor network to collect and analyze data associated with engineering projects in the real world. The sensor network comprises WiFi modules, configured in a grid, to transmit and receive radio frequency signals. These signals are recorded at a command center and analyzed for activity within the grid. Typical radio tomographic imaging systems are based on wireless sensor networks comprising transmitting and receiving nodes. Radio frequency signals are used to perform non-invasive and device-free localization of objects or entities in space. The image of each object or entity is formed from the attenuation in the received signal strengths along each path or link between the transmitting node and the receiving node. The attenuations characterize the loss of or drop in the radio frequency signal strength within the space being sensed and monitored. The radio tomographic imaging system determines the location information from the reconstructed image by solving the inverse problem using the attenuations. Accurate localization of the entities in indoor and outdoor environments is one of the critical requirements of the radio tomographic imaging system. The inverse problem of image reconstruction from the changes in the received signal strength or link values is ill-conditioned due to the extremely small singular values of the weight matrix. Regularization is included to offset the non-invertible nature of the weight matrix. This is achieved by adding a regularization term such as the matrix approximation of derivatives in each dimension based on the difference operator. This operation yields a smooth least-squares solution for the measured data by suppressing the high energy or noise terms in the derivative of the image. Instances of the engineering projects discussed in this paper relate to (a) room surveillance (b) environmental studies (c) vehicle identification and classification.

Noble, W. T., & Dickinson, Z. O., & Sundaram, R. (2023, March), Wireless Sensor Network for Data Mining in Engineering Projects Paper presented at 2023 ASEE North Central Section Conference, Morgantown, West Virginia. 10.18260/1-2--44939

ASEE holds the copyright on this document. It may be read by the public free of charge. Authors may archive their work on personal websites or in institutional repositories with the following citation: © 2023 American Society for Engineering Education. Other scholars may excerpt or quote from these materials with the same citation. When excerpting or quoting from Conference Proceedings, authors should, in addition to noting the ASEE copyright, list all the original authors and their institutions and name the host city of the conference. - Last updated April 1, 2015