New Services & Models Mean New Challenges
Networks are becoming increasingly complex with the rollout of new and diverse services by Broadband Service Providers (BSPs) and others. Additionally, new business models, such as wholesale and open access, mean new touch points with third parties. To operate, optimize, and scale such network, network automation is necessary.
OB-CAS – Enabling Automation
This is where OB-CAS can make a difference. The Broadband Forum’s Open Broadband CloudCO Application Software Development Kit (OB-CAS) aims to make monitoring, analytics, and maintenance easier for BSPs by providing an open API framework through which applications can be easily integrated into their access network management platforms. These applications can play a key role in making network and service assurance more efficient and networks easier to operate.
From that perspective, OB-CAS complements the Automated Intelligent Management (AIM) specifications suite (TR 436/468) of BBF. This suite defines how network automation can be enabled via closed loop interactions; eventually assisted by artificial intelligence.
OB-CAS as a Broadband Forum Open Broadband software project launched in 2024. The team has developed a first SDK version, consisting of wrappers for a selection of open APIs such as KAFKA, OpenSearch, etc. Through these APIs, the applications can access data collected from access network devices via an access domain controller platform.
Also a sandbox environment was created for OB-CAS leveraging the reference SDN platform and network simulators developed in the OB-BAA project. Hence OB-CAS provides a framework through which third party software vendors can test and integrate their applications, applications that will improve and automate network management for BSPs.
Launched from the OB-CAS Development Sandbox – An Alarm Correlation Tool
As a first sample application, the team developed an alarm correlation tool, leveraging the OB-CAS SDK and sandbox environment.
The motivation for this application is that, often when something goes wrong in the network, a Network Operations Center (NOC) can be overwhelmed with alarms coming from different devices, different interfaces, and different network stacks. While often there is a single root cause for the storm of alarms it is difficult to find amidst all the noise. The alarm correlation tool listens to all incoming alarms and notifications using the OB-CAS APIs, captured via the access management platform, and uses correlation rules to identify possible root causes.
A classic example is when a PON fiber cable is accidentally damaged during a civil works project (i.e., the classic backhoe fade) forcing multiple ONTs offline. For each impacted ONT, the OLT issues alarms. Using topology information, the alarm correlator quickly finds the root cause.
This is just one example of an alarm correlation use-case demonstrated in the Broadband Forum’s Innovation Demonstrations at Network X. By making use of a simulated network of OLTs and ONTs, combined with the passive fiber network topology, various network failures can be simulated. The alarm correlation app visualizer then shows the locations of the alarms on a map and the likely root cause.
Furthermore, another Broadband Forum Innovation Demonstration, Fixed Access Network Sharing (FANS), leverages the same principles of OB-CAS. In this case the OB-CAS’s open APIs are applied to a commercial access controller solution (Wholesale Network Portal of Altiplano developed by Nokia and Condor Technologies).
OB-CAS – Open for New Business
Going forward, the OB-CAS team wants to add additional open APIs as part of the SDK, as well as further demonstrate its capabilities by means of an application realizing a closed loop in the assurance domain. The plan is to also leverage AI capabilities to enable proactive fault detection and/or faster troubleshooting of network issues.
The Broadband Forum encourages anyone interested in learning more about or wishing to contribute to the OB-CAS project to join us at
https://wiki.broadband-forum.org/pages/viewpage.action?pageId=338427993