AIDA
Adaptive, Intelligent and Distributed Assurance Platform Computer Science and Engineering
Description
RAID is Wedo's flagship end-to-end real-time data management platform handling the entire risk management lifecycle of enterprises. RAID provides a highly flexible and robust processing pipeline consisting of data collection, monitoring, notification, discovery and actuation phases. Wedo's platform is currently applied around the world to handle, among others, revenue assurance, business assurance and fraud management.
The current and anticipated market demand poses two major challenges to the existing RAID: i) the need to scale-out to unprecedented levels and ii) provide privacy and confidentiality guarantees when handling most of the input data sources, exacerbated when these stem from different owners or administrative domains. Meeting these new requirements poses complex scientific and technological challenges, including in topics that until now did not represent major issues such as cybersecurity, but also open market opportunities through the exploration of emergent federated machine learning techniques that will reinforce the analytic capabilities of the platform.
The overarching goal of the AIDA project was to conceive a new version of the current Wedo platform in which some of the pipeline phases can be dynamically moved to the edges of the system. As of today, the platform is fully deployed in physically colocated servers, either on premises or in the cloud. While some phases are likely to always be kept under the platform's service owner, with AIDA data collection and monitoring, and even actuation phases should be prepared to run in diverse hardware architectures outside the platform's owner physical or even administrative control. AIDA provides for highly configurable data collection and monitoring while preserving the current real-time, security and dependability guarantees. A complete prototype of AIDA was deployed and demoed at the end of the project in a realistic scenario in the area of telecommunication services.
Scientific Advances
The project aimed at improving a platform used by Mobileum for integral risk management in companies. This platform ensures revenue, corporate conditions and fraud control for companies. Thanks to the newest version of the platform, developed by AIDA, companies will be able to collect and monitor data in an extremely flexible way, with real time guarantees, security and reliability.
AIDA enabled data collection, monitoring and even actuation phases of the RAID platform to run at the edge of the network, considering diverse hardware architectures, such as end-user devices, intermediaries such as routers and gateways, telcos base stations, and cloudified microservices. AIDA also provides highly-configurable and rich data collection and monitoring while preserving the current real-time, security and dependability guarantees of the RAID platform.
The main technological goals of the project were:
- Distribution of the RAID platform components leveraging edge computing and 5G: To delegate processing to the edge or use central servers, according to the nature of the computation and the type and localization of monitoring and reference data.
- Exploration of emergent federated machine learning techniques: To learn from local data and push incremental model updates to coordinator nodes that maintain global models based on the contribution of edge nodes and other relevant data sources.
- Resilience to intrusion or tampering: To require the research and application of intrusion detection techniques at multiple levels of the architecture, with the goal of enabling system-wide intrusion tolerance.
- Data privacy and confidentiality: To maintain the confidentiality of the operational data being monitored and analysed, and protecting the privacy of the entities to whom the data refers to are fundamental concerns in the platform's design.
Impact
Total number of publications: 51
Number of publications authored only by Portuguese faculty: 45
Number of publications authored only by CMU faculty: 3
Number of publications jointly-authored by Portuguese and CMU faculty: 3
The current and anticipated market demand poses two major challenges to the existing RAID: i) the need to scale-out to unprecedented levels and ii) provide privacy and confidentiality guarantees when handling most of the input data sources, exacerbated when these stem from different owners or administrative domains. Meeting these new requirements poses complex scientific and technological challenges, including in topics that until now did not represent major issues such as cybersecurity, but also open market opportunities through the exploration of emergent federated machine learning techniques that will reinforce the analytic capabilities of the platform.
The overarching goal of the AIDA project was to conceive a new version of the current Wedo platform in which some of the pipeline phases can be dynamically moved to the edges of the system. As of today, the platform is fully deployed in physically colocated servers, either on premises or in the cloud. While some phases are likely to always be kept under the platform's service owner, with AIDA data collection and monitoring, and even actuation phases should be prepared to run in diverse hardware architectures outside the platform's owner physical or even administrative control. AIDA provides for highly configurable data collection and monitoring while preserving the current real-time, security and dependability guarantees. A complete prototype of AIDA was deployed and demoed at the end of the project in a realistic scenario in the area of telecommunication services.
Scientific Advances
The project aimed at improving a platform used by Mobileum for integral risk management in companies. This platform ensures revenue, corporate conditions and fraud control for companies. Thanks to the newest version of the platform, developed by AIDA, companies will be able to collect and monitor data in an extremely flexible way, with real time guarantees, security and reliability.
AIDA enabled data collection, monitoring and even actuation phases of the RAID platform to run at the edge of the network, considering diverse hardware architectures, such as end-user devices, intermediaries such as routers and gateways, telcos base stations, and cloudified microservices. AIDA also provides highly-configurable and rich data collection and monitoring while preserving the current real-time, security and dependability guarantees of the RAID platform.
The main technological goals of the project were:
- Distribution of the RAID platform components leveraging edge computing and 5G: To delegate processing to the edge or use central servers, according to the nature of the computation and the type and localization of monitoring and reference data.
- Exploration of emergent federated machine learning techniques: To learn from local data and push incremental model updates to coordinator nodes that maintain global models based on the contribution of edge nodes and other relevant data sources.
- Resilience to intrusion or tampering: To require the research and application of intrusion detection techniques at multiple levels of the architecture, with the goal of enabling system-wide intrusion tolerance.
- Data privacy and confidentiality: To maintain the confidentiality of the operational data being monitored and analysed, and protecting the privacy of the entities to whom the data refers to are fundamental concerns in the platform's design.
Impact
Total number of publications: 51
Number of publications authored only by Portuguese faculty: 45
Number of publications authored only by CMU faculty: 3
Number of publications jointly-authored by Portuguese and CMU faculty: 3
