NES (Networked-Embedded-Systems) Lab

--------------------------------------------------------------------------------------------------------

I have listed couple of projects that we are interested to investigate. Please contact me at mohammad.mozumdar@csulb.edu for details.

Development of Micro Wireless Sensor Platforms for Collecting Data of Passenger-Freight Interactions

Traditionally, pavement inductive loop sensors are used to collect real time traffic data for passenger-freight movement in roadways. This method, however, is expensive to install and maintain, and also requires an electronic control unit connected to the induction loop. In the last decade, significant improvements have been achieved in Micro-Electro-Mechanical System (MEMS) sensors domain with respect to size, cost and accuracy. Moreover, extreme miniaturization of RF transceivers and low power micro-controllers motivated the development of small and low power sensors and radio-equipped modules, which are now replacing traditional wired sensor systems. These modules can communicate with other similar modules to build an intelligent sensing network. Due to process miniaturization and low power consumption, these “sensor nodes” can potentially remain functional years. Motivated by these novel advances, we propose a wireless MEMS sensor based passenger-freight interaction detection framework for Intelligent Transportation Systems (ITS). The proposed system is mainly composed of two parts: the sensor nodes that contain wireless Magneto-Resistive (MR) sensors to detect passenger-freight vehicles and an Electronic Control Unit (ECU) to collect and generate the traffic data from sensor nodes’ data such as vehicle detection, speed, and most importantly, classification. The sensor’s primary function is to classify vehicles, which inherently includes detection. Through the use of machine learning algorithms, ECU can produce correct classification rates nearing 100%. Through the use of multiple sensors, the ECU can calculate and extrapolate the speed and level of congestion of the area where the sensors are installed. Our proposed solution will be significantly more cost effective than the traditional induction loop approach and is scalable to cover millions miles of roadways all over the US.

Smart sensing interface for driver’s drowsiness detection

This project will focus on the development and design of a system to make transportation safer for society as a whole through the detection of distracted driving. For the purpose of this project, distracted driving is defined as fatigue or drowsiness behind the wheel either from acute or chronic health conditions. According to the National Highway Traffic and Safety Administration (NHTSA), over 100,000 vehicular crashes occur annually due to distracted driving. Our suggested solution is the implementation of a smart drowsiness detection interface that will be used as a continuous monitoring device to alert drivers about any significant changes in their physiological health which leads to distracted driving. The first objective will be to design an embedded system which will collect electrocardiogram (ECG) signals from the driver through state-of-the-art, non-contact, and non-invasive ECG sensors. Our next objective will be to design a system that will efficiently filter out all undesirable noise from the collected ECG signals. Lastly, our final objective will be to develop a drowsiness detection algorithm using features extracted from the ECG with the explicit purpose of detecting driver drowsiness behind the wheel. The algorithm will detect driver drowsiness in real-time and the embedded system will provide safety indications to the driver should he or she be in danger of falling asleep. Ultimately, we hope this interface will be adopted as a paradigm shift for safety behind the wheel, specifically amongst drivers who are at the higher risk due to fatigue or pre-existing health conditions.