Vision and Challenges for Knowledge Centric Networking

Abstract

In the creation of a smart future information society, Internet of Things (IoT) and Content Centric Networking (CCN) break two key barriers for both the front-end sensing and back-end networking. However, we still observe the missing piece of the research that dominates the current networking traffic control and system management, e.g., lacking of the knowledge penetrated into both sensing and networking to glue them holistically. In this paper, we envision to leverage emerging machine learning or deep learning techniques to create aspects of knowledge for facilitating the designs. In particular, we can extract knowledge from collected data to facilitate reduced data volume, enhanced system intelligence and interactivity, improved service quality, communication with better controllability and lower cost. We name such a knowledge-oriented traffic control and networking management paradigm as the Knowledge Centric Networking (KCN). This paper presents KCN rationale, KCN benefits, related works and research opportunities.

Vehicular knowledge networking and application to risk reasoning

“…[10] explored content election strategies to allocate the cache resources near-optimally for edge caching. The parking lot monitoring system is investigated for end-to-end parking application to predict the parking availability in [21]. [12] proposed a course recommendation system where the most promising course according to students is suggested to increase the online course completion rate.…”

“…Clustering algorithm is used for knowledge extraction where a set of objects (e.g., human trajectories) is grouped in same cluster in terms of similarity (e.g., socioeconomic activities) in [23]. Knowledge, hidden relationship in data, is explored through the machine learning approaches in [10,12,21]. From the existing literature, we observe that there is a current interest within the scientific community in knowledge oriented applications.…”

“…We further utilize the knowledge to facilitate reduced volume and share it through the VKN for better traffic planning and safer driving. Purpose Knowledge Method(s) Application [13] Resource Utilization Network Selection Optimization Access Network Selection [24] Network Management Optimal Links and Paths Optimization UAV Network Management [23] Understanding Human Mobility Socioeconomic Activities Clustering Urban Area Planning [10] Caching Utilization Content Election Machine Learning Edge Caching [21] Parking Lot Monitoring Parking Availability Machine Learning End-to-End Parking [12] Increasing Online Course Completion Rate The Most Promising Course Machine Learning Course Recommendation…”

Adaptable and Data-Driven Softwarized Networks: Review, Opportunities, and Challenges

“…Indeed, there is an increasing consensus that the network operations should be supported by data-driven ML-based models revolving around high-level goals and a holistic view of the underlying network (see the surveys [68]- [70] and position papers [14], [71]- [76]). Interestingly, we note a particular lack: 1) of investigations on the adaptation potentials of the combinations of SDN and NFV in [14], [69], and [72], and 2) thorough investigations of the increasing complexity of the decision phase arising from SDN and NFV in [68], [73], and [75].…”

“…Wu et al [76] have extended the notion of knowledge-defined networking to Knowledge Centric Networking. Wu et al [76] considered an IoT scenario, where ML can extract useful information from sensors at the edge of the network to reduce the burden in the core of the network. They propose the usage of SDN together with ML to deploy and manage network slices.…”