By Shlomo Ovadia, Ph.D., Deependra Rawat & Dan Lynch, Charter Communications
In the last decade, home network architectures have become more complicated due to advances in wireless technology, and the explosion of different types of wirelessly connected devices such as Internet-of-Thing (IoT) devices, cell phones, tablets, laptops, gaming devices, etc. In fact, the U.S. smart home market is expected to show a compound annual growth rate of 16.9%, reaching 46.6B by 2024. In this home network architecture, some cable Multiple System Operators (MSOs) deploy a two-box solution for both residential and Small and Medium-size Business (SMB) customers where one box is an access CPE device (i.e., cable modem or an ONU), and the second box is a wireless router. Cable operators typically have limited information about the access CPE device’s health status, and no information about the customer’s home network, including what type of clients are connected to the home network, and their bandwidth usage vs. time. Such home network health and traffic information would be very useful to the Cable operators in order to enhance their customers’ experience by optimizing the customer’s home network traffic, prevent potential field issues, and be able to introduce new services such as customer’s home network management and security. Furthermore, some Cable operators have already begun the transition towards cloud-based management of wireless routers and other devices, which is not supported by the currently field-deployed access CPE devices.
In this paper, the challenges with the current home network architecture are first explained. Then, the proposed home network architecture with an agile software stack on the access CPE device is discussed.
This includes the agile software stack and components with the cloud-based management of the access CPE device to enable streaming of all the telemetry data. Third, the streaming telemetry data path, including the components and operation of the cable MSO’s Streaming and Analytics platform are explained. Fourth, the organized hierarchical Grafana dashboard design with all of the different types of streamed data telemetry, including home network traffic metrics, D3.1 eMTA health metrics, DOCSIS RF info, event alarm and notifications is explained. A comparison between the OpenWrt-based streaming telemetry method in the paper and other streaming telemetry methods is then reviewed. Finally, the benefits to Cable operators using the OpenWrt-based streaming data telemetry method for access CPE devices are summarized.