NDN4VKN: A Simulation Framework for Supporting Vehicular Knowledge Networks

Vehicular Knowledge Networking (VKN) is a paradigm where vehicles exchange knowledge instead of data. Named Data Networking (NDN) provides a resilient architecture for data sharing between vehicles without requiring details of the hosting vehicle. However, NDN as well as state-of-art NDN simulation environments lack knowledge representation and reasoning (KRR) support to process, understand and optimize knowledge exchange between vehicles. In this paper, we present an open-source co-simulation framework tailored to VKN, integrating NDN, traffic, perception, and control simulators with an AI-as-a-Service (AIaaS) platform for KRR. This paper describes the APIs between various simulators and tools and demonstrates on a proof-of-concept the benefits of KRR for NDN by dividing NDN response time by a factor of five, while reducing NDN storage size by a factor of two.