Respond to a Cisco request for a proposal

Project ID:

RFP-2007-023

Title:

Predictive Analytics and Pattern Matching Methods for Video Service Deployments

Summary:

Future video services will reach consumers through incumbent service providers, over-the-top providers, and even through enterprise corporate IT enabling video mobility applications. There are few tools that can predict or provide trending analysis of possible future degradation of end-user Quality of Experience (QoE) for video-related services. Many management systems, probe vendors, and deep packet inspection tools gather extensive information between the video headend and the end-user set top box (STB). This information is modeled in such a way to understand the relationships between model elements in the context of video service delivery. Much of the video application packet flow is encrypted. After populating these models with information, deeper analysis and trending is required to filter and highlight potential future problems. When deployments range from 100,000 users to 30 million users, taking statistical samples from end-user gateways or STBs may provide a baseline to make future comparisons for trend analysis. We seek to sponsor research on analyzing and identifying trend patterns in data to provide predictive indicators surrounding service degradation. The question is, can we develop methods and analytics that can be componentized into an overall management framework.

Full Description:

Service Providers are rapidly expanding their wireline networks to deploy consumer video services. Both unicast and multicast-based services are deployed across converged IP-based networks. Technical enablers that have become available include: flexible netflow, deep packet inspection, control-plane path construction and monitoring, and active and passive measurement. The DSL Forum has developed the TR-069 standard to begin to address how to manage and monitor home router gateways and video set-top-boxes. There are many methods available today to gather information from Layer 2 , Layer 3, or even application layers. Many management systems gather data from the end-to-end service flow between headends and end-user STBs. Service providers struggle to integrate this data in a meaningful way with their management systems. Over-the-top providers may need to rely on more innovative ways to determine service quality levels. The enterprise may need to partner with service providers to enable corporate mobility architectures for managed video services.

There are methods today that propose filtering and correlation of fault data to identify service-level problems. There are very few methods today that provide an analytical capability to predict future IPTV service problems before they occur. Techniques include trend analysis from baseline data of a "known" service state when there were no known problems, and tracking multiple predictors and understanding how the combination of those predictors can affect the end-user Quality of Experience. Predictive analysis could result in the visualization of an IPTV service to understand if the service provider is in a "green, yellow, or red" operational state. Second, the analysis could trigger threshold alarms to warn of an impending degradation of IPTV service. Third, the analysis could be used to help troubleshoot an IPTV service where users are complaining about poor video or audio quality. Tools are also required to provide predictive capabilities to understand how the launch of a future video service may impact existing services.

We specifically invite proposals that result in meaningful definition of analytical methods or tools that could be used to provide predictive analysis for the degradation of a video service. Proposals should include developing recommendations for technologies, methods, or analytics that could be implemented within the framework of a management system for the visualization, alarming, and troubleshooting of an IPTV service.

Constraints and other information:

IPR will stay with the University. Cisco expects customary scholarly dissemination of results, and hopes that promising results would be made available to the community without limiting licenses, royalties, or other encumbrances.

Proposal submission:

Please use the link below to submit a proposal for research responding to this RFP. After a preliminary review, we may ask you to revise and resubmit your proposal.

Create/submit a proposal for this RFP

RFPs may be withdrawn as research proposals are funded, or interest in the specific topic is satisfied. Researchers should plan to submit their proposals as soon as possible. Submissions-to-date are reviewed at the beginning of each quarter (the first business day of: January, April, July, October).

Questions? Contact: research@cisco.com