System and methods for estimation and improvement of user, service and network QOE metrics
Methods for estimating Subscriber quality of experience (QOE) for mobile users accessing networks for different services from observed data in control and user planes in mobile wireless networks and then summarizing inferences per user, per service, per sector, group of sectors and other aggregate points, and exporting this information for reducing user churn, network planning and network tuning, application adaptation to improve QOE are disclosed. Methods for improving subscriber QOE metrics for certain applications, services and web-sites for improved monetization methods are also presented. The methods facilitate quantifying network goodness from the user application point of view, and exporting triggers to other network elements, such as SON Server, OSS and PCRF, when QOE anomalies are detected. Additionally, this exported information could also trigger content adaptation, delivery optimizations and other actions. Finally, client, server and intermediary enhancements to improve QOE of certain applications in delay/capacity varying networks are presented.
1. A method of monitoring and improving the quality of experience for one or more users in a mobile cellular network, comprising:
monitoring network behavior on said mobile cellular network;
determining a quality of experience metric for a plurality of activities based on said monitored network behavior, said activities selected from the group consisting of SMS messaging, web browsing, multimedia downloads, multimedia uploads, advertisement delivery, voice calls, and paging;
consolidating, for a particular user, said plurality of said determined quality of experience metrics, to generate a consolidated QoE metric for said user;
exporting said consolidated QoE metric to a component in said mobile cellular network, wherein said components takes an action based on said consolidated QoE metric to improve a quality of experience for said user.
2. The method of claim 1 , wherein said activity comprises web browsing, and said quality of experience metric is determined by:
identifying a start of a browsing session based on establishment of said user's TCP connection to a server;
identifying an end of said browsing session when a significant idle time where there is no uplink or downlink traffic to said TCP connection;
determining a duration of said browsing session;
determining a number of bytes delivered during said browsing session; and
determining a quality of experience for web browsing based on said number of bytes or objects delivered and said duration of browsing session.
3. The method of claim 1 , wherein said activity comprises web browsing, and said quality of experience metric is determined by:
estimating page down load time by identifying a time period between a top level domain request to a subsequent top level domain request;
estimating an amount of data downloaded during said time period; and
determining a quality of experience for web browsing based on said amount of downloaded data and said time period.
4. The method of claim 1 , wherein said activity comprises web browsing, and said quality of experience metric is determined by:
estimating a non-idle time user throughput; and
determining a quality of experience for web browsing based on said non-idle time user throughput.
5. The method of claim 4 , wherein said non-idle time user throughput is estimated by counting a number of bytes acknowledged per TCP connection over a fixed time interval.
6. The method of claim 1 , wherein said activity comprises SMS messaging, and said quality of experience metric is determined by:
monitoring a plurality of parameters, said parameters selected from the group consisting of a round trip interval from a service request to a SMS-submit; a number of service request retries; a number of paging retries; a time between a Type1-paging message and an associated paging response; round trip time between sending a Type1-paging message and receiving a CP-ACK; and estimated delivery completion based on a time between sending a message to a SMS server and receiving a delivery notification completion; and
combining said parameters to estimate a quality of experience for SMS service.
7. The method of claim 1 , wherein said activity comprises multimedia downloads, and said quality of experience metric is determined by:
estimating, based on a time between a client request for content, and a corresponding acknowledgement from a content server, a number of times a keep-ahead buffer in said client is empty; and
determining a quality of experience for multimedia download based on said number.
8. The method of claim 1 , further comprising combining consolidated QoE metrics for a plurality of users located in a common geographic region prior to taking said action.
9. The method of claim 8 , wherein said action comprises using a microcell or femtocell in said common geographic region if said combined consolidated QoE metric is below a predetermined threshold.
10. The method of claim 8 , wherein said action comprises moving said plurality of users to an alternative network.
11. The method of claim 1 , further comprising combining consolidated QoE metrics for a plurality of users accessing a common website prior to taking said action.
12. The method of claim 11 , wherein said action comprises introducing caching or site acceleration improvements.
13. The method of claim 1 , further comprising combining consolidated QoE metrics for a plurality of users using a common device class prior to taking said action.
14. The method of claim 13 , wherein said action comprises diagnosing a potential compatibility issue with said device class.