For any emerging smart city with an Internet of Things (IoT) network, edge computing platforms equipped with artificial intelligence are crucial to managing high-bandwidth technologies and processing massive amounts of data in real time. However, many of the more than 10 billion active IoT devices — expected to produce more than 73 zettabytes of data by 2025 — contain safety and privacy-critical components. Edge computing is rarely used in these instances due to concerns about lower computational power, unpredictable latency and potential leakage of sensitive information when transmitting to the cloud.
“While edge computing enables devices to process data closer to where it is generated, the performance bottleneck of the whole system is transferred from the edge-cloud communication to the on-chip communication,” said Zheng Dong, assistant professor of computer science in the Wayne State University College of Engineering.
Facing architectural, scalability and security challenges of a network-on-chip system, Dong is developing a potential solution that was funded by the National Science Foundation (NSF). The three-year, $499,861 grant through NSF’s Computer and Network Systems: Core program supports a hardware and software co-design that enhances real-time collaborative computing on the edge.
Weisong Shi, professor of computer science and director of The CAR Lab at Wayne State, is a co-principal investigator on the project.
More From Wayne State Engineering
Recent Comments