Kalgi Shah
1232746283
API version 36
Question 1: Imagine you are new to the programming world and not proficient enough in coding. But, you have a brilliant idea where you want to develop a context-sensing application like Project 1. You come across the Heath-Dev paper and want it to build your application. Specify what Specifications you should provide to the Health-Dev framework to develop the code ideally.
If I were not very skilled in coding but wanted to build a context-sensing app like Project 1 with the Health-Dev framework, my main task would be to provide detailed system specifications. Since Health-Dev can take high-level descriptions and generate both sensor and smartphone code, the specifications should cover three areas: sensors, networking, and phone interface.
For the sensors, I would list what type of data they should capture (such as heart rate or breathing rate) and include details like sampling rate, device type (for example Arduino-based), and sensitivity. I would also note the algorithm needed for processing like peak detection for heart rate and define how the sensors will communicate (Bluetooth, ZigBee, etc.).
For the network, I would describe the overall structure, how messages are routed, and any energy-saving techniques like duty cycling instead of constant transmission.
Finally, for the smartphone interface, I would outline the user-facing features: buttons to start or stop data collection, areas to show measured values, charts to track changes, and possibly some lightweight analytics such as averaging or anomaly detection. With these specifications, Health-Dev could generate platform-specific code, letting me implement a working system without needing advanced programming expertise.
Question 2: In Project 1 you have stored the user’s symptoms data in the local server. Using the bHealthy application suite how can you provide feedback to the user and develop a novel application to improve context sensing and use that to generate the model of the user?
In Project 1, symptom data was saved only on the local device, which enabled collection but did not give the user much in terms of feedback. The bHealthy system shows how adding physiological feedback can make these apps more interactive and supportive. It integrates signals like ECG or EEG with applications that encourage healthier behaviors.
Using the same idea, my app could connect symptoms with feedback. For example, if the data shows both breathing difficulty and higher-than-normal respiratory rate, the system could send a prompt suggesting rest or even consulting a doctor. If patterns show signs of stress, the app might recommend relaxation exercises.
Over time, the app could build a personal model of each user. This model would track trends, notice deviations from the user’s normal state, and personalize recommendations. For instance, if someone often reports fatigue after exercise with an elevated heart rate, the app could suggest lighter workouts or better recovery methods. To keep engagement high, feedback could be made interactive, like using a game element or virtual companion (similar to bHealthy’s BrainHealth or PETPeeves). This way, the app does more than store data—it becomes adaptive, supportive, and motivating for the user’s health journey.
Question 3: A common assumption is mobile computing is mostly about app development. After completing Project 1 and reading both papers, have your views changed? If yes, what do you think mobile computing is about and why? If no, please explain why you still think mobile computing is mostly about app development, providing examples to support your viewpoint
Before this project, I mostly thought of mobile computing as writing and publishing apps: building the interface, adding features, and putting them on app stores. After completing Project 1 and reviewing the Health-Dev and bHealthy papers, my understanding has expanded.
Now I see mobile computing as the broader field of building context-aware systems that merge sensors, networks, algorithms, and user interfaces. Health-Dev, for example, shows that code can be automatically generated from specifications, focusing on reliability and scalability rather than just coding. Similarly, bHealthy illustrates how physiological signals can be processed and turned into real-time, adaptive feedback to improve wellbeing.
This means mobile computing is not just about an app’s UI or codebase, it also involves energy efficiency, signal processing, network design, and how devices fit into people’s daily lives. Apps are only the visible layer, true challenge is designing systems that connect hardware, software, and human needs to deliver meaningful, real-time insights.
Portions of the code in this project were generated with assistance from ChatGPT, an AI tool developed by OpenAI.
Reference:
OpenAI. (2024). ChatGPT [Large language model]. Available at: https://openai.com/chatgpt
Estimated percentage of code influenced by Generative AI: 21%