The proliferation of sensors in consumer mobile phones has made them a powerful data collection platform. We have become more interested in collecting data to document and learn about our own lives, our relationships with the community, and the environment we live in. However, current research is mostly focused on analysis of quantitative data, and a lot of important information that can be extracted from qualitative data such as images and text annotations is being ignored. The focus of this dissertation was to research and build techniques based on narratology to help users analyze and make sense of the qualitative data they have collected and to communicate the information and knowledge they extract from the data though narratives. More specifically, our main contribution was in the creation and evaluation of two techniques that were fundamental to narration tools: clustering and conflict discovery. We used these techniques to build the Narreme toolset that enables user-assisted creation of compelling narratives with participatory sensing data. The toolset consists of three tools: (a) Narreme clustering tool which helps a user organize her thoughts and identify supporting qualitative data consisting of images and text for her narratives, (b) Narreme conflict discovery tool helps the user better understand where conflict could be introduced in her narratives, and, (c) Narreme video editing tool which is used to create a video slideshow narrative and adjust the order based on a narrative ten- sion curve. We evaluated the effectiveness of our techniques and toolset through user studies.

The ARIANNA experiment is an ultra-high energy neutrino detector designed to detect cosmogenic neutrinos. The experiment uses the Ross Ice Shelf to detect RF signals which are generated when the high energy neutrinos interact with the ice. The experiment plans to use a massive deployment of hundreds of antenna array stations to capture neutrino events over a large area. The stations use a data acquisition system to capture and store the data and subsequently transmit the data back to UCI. This thesis describes the design, functioning and results of the newest generation system which was deployed in the field in the 2014 season. The new system utilizes a new data digitization chip developed by the ARIANNA engineering group led by Dr. Stuart Kleinfelder. The new SST chip is a largely simplified and streamlined design compared to the previous generation chip and utilizes advanced circuitry to achieve fully synchronous operation while reducing the power usage by many factors. The new system makes substantial improvements over the previous generation system and introduces a number of new features, while providing a possible design solution for the final design requirements of the ARIANNA array systems.

We are deploying a metropolitan scale Wi-Fi mesh network near downtown Los Angeles to support the design and development of a data-centric network-fabric for urban participatory sensing. Participatory sensing employs software and network technology to enable people’s everyday mobile devices to act as credible sensors of the natural, built, and cultural environments. Current research focuses on how to make it easy and secure for both the public and professional users to define sensing ‘campaigns,’ recruit participants to collect data, to help ‘make a case’ with data they collect, and digitally publish the results. To further research in this area, our architecture will enable embedding network–attested location and time context in sensor readings. The network will also provide a research framework for developing policy-based privacy, and related security mechanisms for participatory sensing.

Extensive literature demonstrates that podcasts can support the process of learning a second language (Goodwin-Jones 2005; McBride, 2009; Phillips, 2017; Stanley, 2006). However, a small number of studies were found in the literature review investigating the implementation of podcasts through computer-assisted task-based language teaching (TBLT) to foster listening comprehension (Mulyadi et al., 2021; O’Bryan & Hegelheimer, 2007).

The present dissertation, firstly, illustrates how computer-assisted TBLT listening lessons can be designed and implemented in an online intermediate Spanish course, and, secondly, investigates the development of students’ listening performance using computer-assisted TBLT lessons, the impact of this type of instruction on students’ metacognitive awareness in listening strategies, as well as their opinions towards podcast tasks. In order to analyze these components, second language learners of Spanish who were enrolled in an online intermediate Spanish course at the University of California, Santa Barbara, participated in this study over the course of one quarter. The technologies used in this study are synchronous Zoom classes and an online module containing six podcast tasks, questionnaires, and open-ended questions.

A total of 51 second language learners of Spanish participated in this study. For the analysis of the development of listening performance, participants completed six computer-assisted TBLT listening lessons. In order to investigate the impact of the listening lessons on participants’ metacognitive awareness in listening comprehension, they completed the Metacognitive Awareness Listening Questionnaire (MALQ) at three different points during the ten-week quarter (Vandergrift et al., 2006). Finally, participants answered open-ended questions in which they self-reflected on the podcasts, their listening process, and listening strategies immediately after each listening lesson.

The results of the study found that participants’ listening comprehension significantly improved at the beginning of the experiment, whereas there was no statistically significant growth at the end of the experiment. Additionally, the results of the MALQ provided empirical evidence for the benefits of the computer-assisted TBLT lessons to raise participants’ metacognitive awareness in listening. Participants demonstrated statistically significant growth in four of the five MALQ categories. The results indicated that this type of listening instruction can be useful for the development of the planning and evaluation, directed attention, person knowledge, and problem-solving categories. Instead, the results of this investigation suggested that there was no statistically significant difference between pretest and posttest responses for the mental translation category. Finally, the written comments section showcased that the majority of participants enjoyed the TBLT listening lessons presented in this study and expressed their desire to continue listening to podcasts in the future. Some of the major trends throughout the responses were that participants increased their confidence, their metacognitive awareness in listening, and enhanced their language skills.

Background

Methods

Results

Conclusions

We present the Ping-Pong mote, a new light sensing module for the Mica mote platform. The Ping-pong mote achieves performance comparable to a commercial light intensity meter, while conforming to the size and energy constraints imposed by its application in wireless sensor networks. The Ping-pong mote was developed to replace the Mica sensor board (MTS310) whose slow response time and narrow dynamic range in light intensity capture is unsuitable to many applications, including media production. The Ping-pong mote features significantly improved SNR due to its adoption of high-end photo sensors, amplification and conversion circuits coupled with active noise suppression, application-tuned filter networks, and a noise-attentive manual layout. Unlike the MTS310, the Ping-pong mote can capture RGB color intensity (for color temperature calculation) and incident light angle (which discerns the angle of ray arrival from the strongest source). Our prototype demonstrated significantly faster response time (> 6x) and a much wider dynamic range (> 10x) in light intensity measurement as compared with the MTS310. The light-angle estimation results were well correlated with an average error of just 2.63 degree.

We present the Ping-Pong mote, a new light sensing module for the Mica mote platform. The Ping-pong mote achieves performance comparable to a commercial light intensity meter, while conforming to the size and energy constraints imposed by its application in wireless sensor networks. The Ping-pong mote was developed to replace the Mica sensor board (MTS310) whose slow response time and narrow dynamic range in light intensity capture is unsuitable to many applications, including media production. The Ping-pong mote features significantly improved SNR due to its adoption of high-end photo sensors, amplification and conversion circuits coupled with active noise suppression, application-tuned filter networks, and a noise-attentive manual layout.