By Vikram Karwal, Rogers Communications; Jenny Panman, Rogers Communications
The need to guarantee Quality of Service (QoS) in modern telecommunication networks is more important than ever before with near-real time sensitive applications relying on low-latency DOCSIS® networks as backhaul. Proactive network maintenance (PNM) approach needs to be emphasized for DOCSIS networks with technology’s supporting virtual reality, online gaming, health care and trading use cases.
This paper proposes to group the major issues faced by the customer in group of symptoms. Thereafter, the impaired network KPIs that flag for each of these symptoms may be identified. These network KPIs can be tracked for the entire base of customers. When the identifying KPIs fall below the threshold it can be predicted that the customer is going to experience the symptom. Network Health Prediction for DOCSIS networks can be divided into two major parts comprising of head-end, access layer, node and tap taken care by maintenance technicians and the other part south of drop till customer equipment maintained by service technicians.