Muthucumaru Maheswaran

Amir Hossein Estiri, Muthucumaru Maheswaran

Machine learning (ML) is expected to play a major role in 5G edge computing. Various studies have demonstrated that ML is highly suitable for optimizing edge computing systems as rapid mobility and application-induced changes occur at the edge. For ML to provide the best solutions, it is important to continually train the ML models to include the changing scenarios. The sudden changes in data distributions caused by changing scenarios (e.g., 5G base station failures) is referred to as concept drift and is a major challenge to continual learning. The ML models can present high error rates while the drifts take place and the errors decrease only after the model learns the distributions. This problem is more pronounced in a distributed setting where multiple ML models are being used for different heterogeneous datasets and the final model needs to capture all concept drifts. In this paper, we show that using Attention in Federated Learning (FL) is an efficient way of handling concept drifts. We use a 5G network traffic dataset to simulate concept drift and test various scenarios. The results indicate that Attention can significantly improve the concept drift handling capability of FL.

Richard Olaniyan, Muthucumaru Maheswaran

Real-time artificial intelligence (AI) applications mapped onto edge computing need to perform data capture, process data, and device actuation within given bounds while using the available devices. Task synchronization across the devices is an important problem that affects the timely progress of an AI application by determining the quality of the captured data, time to process the data, and the quality of actuation. In this paper, we develop a fast edge-based synchronization scheme that can time align the execution of input-output tasks as well compute tasks. The primary idea of the fast synchronizer is to cluster the devices into groups that are highly synchronized in their task executions and statically determine few synchronization points using a game-theoretic solver. The cluster of devices use a late notification protocol to select the best point among the pre-computed synchronization points to reach a time aligned task execution as quickly as possible. We evaluate the performance of our synchronization scheme using trace-driven simulations and we compare the performance with existing distributed synchronization schemes for real-time AI application tasks. We implement our synchronization scheme and compare its training accuracy and training time with other parameter server synchronization frameworks.

Peter Henderson, Muthucumaru Maheswaran

Embedded systems permeate through nearly all aspects of modern society. From cars to refrigerators to nuclear refineries, securing these systems has never been more important. Intrusions, such as the Stuxnet malware which broke the centrifuges in Iran's Natanz refinery, can be catastrophic to not only the infected systems, but even to the wellbeing of the surrounding population. Modern day protection mechanisms for these embedded systems generally look only at protecting the network layer, and those that try to discover malware already existing on a system typically aren't efficient enough to run on a standalone embedded system. As such, we present a novel way to ensure that no malware has been inserted into an embedded system. We chaotically randomize the entire memory space of the application, interspersing watchdog-monitor programs throughout, to monitor that the core application hasn't been infiltrated. By validating the original program through conventional methods and creating a clean reset, we can ensure that any inserted malware is purged from the system with minimal effect on the given system. We also present a software prototype to validate the possibility of this approach, but given the limitations and vulnerabilities of the prototype, we also suggest a hardware alternative to the system.