Papers

1. Enhancing IIoT Security Using Hybrid CNN-BiLSTM Models with Blockchain Integration (
Status: Submitted to ICCAIC.

2. A Novel Web-Based FL Platform for Smart Grid Contract Optimization
Status: Experiments ongoing; approximately 60 progress toward journal submission.

3. Robust Intrusion Detection in Edge IoT Using Blockchain and Trimmed Mean Aggregation
Status:Initial framework completed; simulations of Byzantine attacks are underway to validate robustness.

Last Week's Progress

1. A Novel Web-Based FL Platform for Smart Grid Contract Optimization
Last Week's Progress :Outsourced new datasets

2.Robust Intrusion Detection in Edge IoT Using Blockchain and Trimmed Mean Aggregation
Last Week's Progress: Initial framework was established, and simulations of Byzantine attacks commenced to evaluate system resilience.

This Week Todo's

1. A Novel Web-Based FL Platform for Smart Grid Contract Optimization
Last Week's Progress : Perform experiments on the new datasets

2. Robust Intrusion Detection in Edge IoT Using Blockchain and Trimmed Mean Aggregation
This Week's Progress: Continued Byzantine attack simulations, focusing on analyzing the trimmed mean aggregation's effectiveness in maintaining model integrity.

Project Progress

1. Digital Twin
Overall Progress: 25% complete.
2. Purepet
Overall Progress: 80% complete.

Last Week's Project

1. Digital Twin
Last Week’s Efforts: Designed a section of the bridge

2. Purepet
Last Week’s Efforts: Finalized backend integration for server-client interactions.

This Week's Project

1. Digital Twin
This Week: Designed a section of the bridge

2. Purepet
This Week: Integrate app with more of the web application features

Monthly Goals

Make the federated learning (FL) web application more robust by improving functionality and security.
Tasks:
- Refine the user interface for better client management and real-time progress tracking.
- Enhance client-server communication by implementing secure authentication, better error handling, and logging.
- Improve system scalability and performance to handle more clients efficiently, including testing for workload management.
- Integrate security measures like encryption and potentially blockchain for secure model weight updates.

This will result in a more reliable and efficient federated learning platform, ready for larger-scale applications.

Annual Goals

1. Submit at least two articles to scientific journals.
2. Attend one NSL-approved domestic conference.
3. Attend one NSL-approved international conference.