The objective of the term paper is to:
- Conduct a literature review from a range of pre-selected topics
- To write a 4500 word survey paper to summarize the related research works
The term paper should fulfill the following criteria:
- Topic selected: To be significant, timely and impactful
- Content of the paper: Scope to be well defined, and contents should be comprehensive in terms of broadness and depth
- Analytical and critical information: Paper to provide insightful analysis
- Presentation: Paper should be clear and logically organized, while easy to understand
In this submission folder, the following files are available:
Term Paper - LPWAN.pdf
: Submitted Term Paper based on the current Low-Power Wide Area Network technologies.
The objective of the project is to:
- Host a Virtual Machine with convolutional neural network capability to do animal detection
- Create Android application ("StarSearch") to communicate with the Virtual Machine
In this submission folder, the following files are available:
Image_Recognition.ipynb
: Google Colab Notebook used for training.- Dataset used: Animals-10 dataset. Source: Dataset
- Statistics:
- Training Loss: 0.2541
- Training Accuracy: 91.55%
- Validation Loss: 0.5997
- Validation Accuracy: 81.16%
/Android App Code/starsearch_iot
: Folder containing the source code for the Flutter Applicationpubspec.yaml
: File used to import modules in the Flutter Application.AndroidManifest.xml
: File required for permission controls in Android.- To be stored in "/android/app/build/src/main" folder
/lib
: Folder containing all the dart files to build the Flutter applicationREADME.md
: File containing the instruction for installation of the Flutter application
/Server Code
: Folder containing the source code to be implemented in the Virtual Machineapp.py
: Code used to run the Flask Application. TensorFlow Lite is imported in this application as well./model/imagereg.tflite
: TensorFlow Lite Model used for inference. Exported from the Image_Recognition.ipynb file./templates/index.html
: HTML file to allow users to upload the image to the server through the web browserrequirements.txt
: Text file containing the list of dependencies for the server application.
README.md
: Readme file.
Available for reference (until 31 October 2022):
- OneDrive for Business:
- Includes the following files:
- PowerPoint Slides
- Presentation Video
- TensorFlow Model
- StarSearch Application
- Folder Name: CE4171 IoT Communications and Networking Submission
- OneDrive Link (Requires NTU Account): OneDrive
- Includes the following files:
- Run the Flask server in the root folder by executing the command:
$ python3 app.py
- Record the IP address and port allocated by the Flask server. It will be used for connection to the client.
Note: Firewall permissions and port forwarding must be allowed for incoming connections. Otherwise, the client cannot connect even with the correct settings.
- Download and install the "app-debug.apk" file.
Link in Additional Materials section.
Note: The application will only work when the StarSearch Server is operational.
Additionally, while iOS, macOS and Web files exist in the folder, the application was tested only on Android.
- Alternatively, follow the
README.md
in/Android App Code/starsearch_iot
to install into your device.
Note: Please replace the StarSearch URI with the IP address allocated by your service provider.
Modify the
MainScreen.dart
in/Android App Code/starsearch_iot/lib/Main
.