California State University (CSU) is the most extensive four-year public university system in the U.S., recognized for its commitment to inclusive, applied education. CSU campuses conduct research in energy, technology, and environmental science to serve regional communities. NEC Labs America and California State University work together on educational data mining, adaptive learning environments, and AI-supported student engagement. We research and enhance personalized education technologies. Please read about our latest news and collaborative publications with California State University.

Posts

Accelerating Distributed Machine Learning with AllReduce Reconfiguration Based on Optical Circuit Switching

We propose to apply optical circuit switching to enable dynamic AllReduce reconfiguration for accelerating distributed machine learning. With simulated annealing-based optimization, theproposed AllReduce reconfiguration approach achieves 31% less average training time than existing solutions.

Accelerating Distributed Machine Learning with an Efficient AllReduce Routing Strategy

We propose an efficient routing strategy for AllReduce transfers, which compromise of the dominant traffic in machine learning-centric datacenters, to achieve fast parameter synchronization in distributed machine learning, improving the average training time by 9%.

Field Implementation of Fiber Cable Monitoring for Mesh Networks with Optimized Multi-Channel Sensor Placement

We develop a heuristic solution to effectively optimize the placement of multi-channel distributed fiber optic sensors in mesh optical fiber cable networks. The solution has beenimplemented in a field network to provide continuous monitoring.

Optical Network Anomaly Detection and Localization Based on Forward Transmission Sensing and Route Optimization

We introduce a novel scheme to detect and localize optical network anomaly using forward transmission sensing, and develop a heuristic algorithm to optimize the route selection. The performance is verified via simulations and network experiments.

Seamless Service Handover in UAV-based Mobile Edge Computing

Unmanned aerial vehicles (UAVs), such as drones, can carry high-performance computing devices (e.g., servers) to provide flexible and on-demand data processing services for theusers in the network edge, leading to the so-called mobile edge computing. In mobile edge computing, researchers have already explored how to optimize the computation offloading and the trajectory planning of UAVs, as well as how to perform the service handover when mobile users move from one location to another. However, there is one critical challenge that has been neglected in past research, which is the limited battery life of UAVs. On average, commercial-level drones only have a battery life of around 30 minutes to 2 hours. As a result, during operation, mobile edge computing carriers have to frequently deal with service handovers that require shifting users and their computing jobs from low-battery UAVs to new fully-charged UAVs. This is the first work that focuses on addressing this challenge with the goal of providing continuous and uninterrupted mobile edge computing service. In particular, we propose a seamless service handover system that achieves minimum service downtime when handling the duty shift between low-battery UAVs and new fullycharged UAVs. In addition, we propose a novel UAV dispatchalgorithm that provides guidelines about how to dispatch new fully-charged UAVs and where to retrieve low-battery UAVs, with the objective of maximizing UAVs’ service time. The effectiveness of the proposed service handover system and the proposed UAV dispatch algorithm is demonstrated through comprehensive simulations using a time-series event-driven simulator.

An Efficient Approach for Placing Distributed Fiber Optic Sensors with Concurrent Sensing Capability

We propose an efficient approach for placing distributed fiber optic sensors (DFOS) with concurrent sensing capability. It consumes 5.7% to 9.5% fewer sensors than that using DFOS without concurrent sensing, for covering the same network.