A PNM System Using Artificial Intelligence, HFC Network Impairment, Atmospheric and Weather Data to Predict HFC Network Degradation and Avert Customer Impact (2018)

By Larry Wolcott, Michael O'Dell, Peter Kuykendall, Vishnu Gopal, Jason Woodrich & Nick Pinckernell, Comcast

Proactive network maintenance (PNM) has become a cornerstone technology within the Data Over Cable Service Interface Specification (DOCSIS®), providing tremendous benefit to cable operators and their customers. From adaptive equalization to full band capture, a rich and extensive data model exists to proactively maintain our valuable networks and reduce operational costs. However, due to the complexity, financial and cultural barriers, many operators have been unable to gain traction with implementing such systems. This paper will examine PNM capabilities, weather information, artificial intelligence, machine learning, operational practice and financial implications to provide a meaningful approach for implementation.

Physical networks that are exposed to the environment are subject to environmental influences which can affect their performance, especially, but not limited to, coaxial and HFC networks and also including fiber optic, satellite, and wireless networking solutions. This is due in part to physics and the simple fact that physical things respond to environmental conditions in predictable ways. Some of the most common factors that influence these networks are atmospheric conditions such as heat, cold, wind, rain, snow, humidity, and freezing. The core premise of this work is to correlate the predictable nature of weather and seemingly unpredictability of weather related outages. While it is generally accepted that these things affect the network, very little work has been done to quantify in an empirical way. Thus, having predictable correlation with faulty network elements allows for operators to proactively prioritize and repair the problems before they impact customers.

Further, by understanding the environmental influence throughout the life span of a network, better decisions can be made about design and construction. This allows operators to evaluate the true cost of ownership over the entire life of the network including support and maintenance, which are generally very expensive.

By clicking the "Download Paper" button, you are agreeing to our terms and conditions.

Similar Papers

Can Future Networks Survive Without Artificial Intelligence?
By Dr. Claudio Righetti, Emilia Gibellini, Carlos Germán Carreño Romano & Gabriel Carro; Telecom Argentina S.A.
Proactive Network Maintenance Evolution to the Optical Domain in Coherent Optics
By L. Alberto Campos, Ph.D. & Zhensheng (Steve) Jia, Ph.D., CableLabs & Larry Wolcott, Comcast
Using Historical Traffic Data to Schedule Service Interruptions for Minimum Customer Impact
By Jason Rupe & Colin Justis, CableLabs
Characterizing Network Problems Using DOCSIS® 3.1 OFDM RxMER Per Subcarrier Data
By Ron Hranac, Cisco Systems; James Medlock, Akleza,Inc.; Bruce Currivan, JJP Development; Roger Fish & Tom Kolze, Broadcom; Jason Rupe & Tom Williams, CableLabs; Larry Wolcott, Comcast
Profile Management Informed Proactive Network Maintenance
By Jason Rupe & Jingjie Zhu, CableLabs
Operationalizing Streaming Telemetry and Machine Learning Model Serving: Customer Experience Automation
By Nick Pinckernell & Scott Rome, Comcast
Training Machines to Learn From Signal Meter Readings
By Gary Ventriglia, Jack Birnbaum, Robert Gonsalves, Anastasia Vishnyakova, Michael Kreisel & Larry Wolcott, Comcast Corporation
Convolutional Neural Networks for Proactive Network Management
By Jude Ferreira, Maher Harb, Karthik Subramanya, Bryan Santangelo & Dan Rice, Comcast
Augmented Intelligence: Next Level Network and Services Intelligence
By Dr. Claudio Righetti, Mariela Fiorenzo, Omar Hurtado & Gabriel Carro, Telecom Argentina S.A.
Software Revolution of Field Meters Using a Field-Capable Measurement Device
By Anthony Curran & Andy Martushev, Comcast
More Results >>