SIP-based VoIP traffic behavior profiling method
View Patent ↗With the widespread adoption of SIP-based VoIP, understanding the characteristics of SIP traffic behavior is critical to problem diagnosis and security protection of VoIP services. A general methodology is provided for profiling SIP-based VoIP traffic behavior at several levels: SIP server host, server entity (e.g., registrar and call proxy) and individual user levels. Using SIP traffic traces captured in a production VoIP network, the characteristics of SIP-based VoIP traffic behavior in an operational environment is illustrated and the effectiveness of the general profiling methodology is demonstrated. In particular, the profiling methodology identifies anomalies due to performance problems and/or implementation flaws through a case study. The efficacy of the methodology in detecting potential VoIP attacks is also demonstrated through a test-bed experimentation.
1. A method for profiling session initiation protocol (SIPS network traffic, comprising:
tallying a plurality of SIP messages associated with a SIP registrar according to a plurality of distinct user resource indicators (URIs) to produce a plurality of message tally counts;
determining a measure of randomness of user activity of the SIP network traffic based on the plurality of message tally counts according to a user activity diversity (UAD) metric; and
identifying an anomaly of the SIP network traffic based on the measure of randomness.
2. The method of claim 1 wherein identifying the anomaly based on the measure of randomness comprises:
determining a rate of change by comparing the measure of randomness to a previously determined measure of randomness within a pre-determined time period;
determining a deviation from norm by comparing the rate of change to a moving average of the rate of change;
comparing the deviation from norm to a previously determined deviation from norm to generate a comparison result, wherein the previously determined deviation from norm is determined in a learning period, and
identifying the anomaly when the comparison result matches a pre-determined criterion.
3. The method of claim 1 wherein identifying the anomaly comprises:
calculating a first parameter representing a rate of change based on the measure of randomness,
calculating a second parameter representing an instantaneous deviation from a moving average of the first parameter;
incrementing an alert level and locking the first parameter if the second parameter is greater than a first threshold, wherein the first threshold is based on a maximum of the second parameter in a learning period prior to identifying the anomaly;
decrementing the alert level if the second parameter is lower than a second threshold;
unlocking the first parameter if the alert level is lower than a third threshold; and
identifying the anomaly when the alert level is greater than a fourth threshold.
4. The method of claim 3 further comprising:
adjusting the third threshold or the fourth threshold if the identified anomaly is a false anomaly.
5. The method of claim 1 further comprising:
generating a reference based on another plurality of message tally counts tallied when no anomaly is identified;
comparing the plurality of message tally counts to the reference to generate a comparison result when the anomaly is identified; and
identifying a cause of the anomaly based on the comparison result.
6. The method of claim 5 ,
wherein the reference comprises a histogram of the plurality of message tally counts when no anomaly is identified,
wherein a current histogram is a histogram of the plurality of message tally counts when the anomaly is identified, and
wherein the comparison result is a plurality of elements contributing to a relative entropy of the current histogram with respect to the reference.
7. A method for profiling SIP network traffic comprising:
tallying a plurality of SIP call request messages associated with a SIP call proxy according to a plurality of distinct user resource indicators (URIs) to produce a plurality of message tally counts;
determining a measure of randomness of user activity of the SIP network traffic based on the plurality of message tally counts; and
identifying an anomaly of the SIP network traffic based on the measure of randomness, comprising:
determining a rate of change by comparing the measure of randomness to a previously determined measure of randomness within a pre-determined time period;
determining a deviation from norm by comparing the rate of change to a moving average of the rate of change;
comparing the deviation from norm to a previously determined deviation from norm to generate a comparison result, wherein the previously determined deviation from norm is determined in a learning period, and
identifying the anomaly when the comparison result matches a pre-determined criterion.
8. The method of claim 7 wherein identifying the anomaly comprises:
calculating a first parameter representing a rate of change based on the measure of randomness,
calculating a second parameter representing a instantaneous deviation from a moving average of the first parameter;
incrementing an alert level and locking the first parameter if the second parameter is greater than a first threshold, wherein the first threshold is based on a maximum of the second parameter in a learning period prior to identifying the anomaly;
decrementing the alert level if the second parameter is lower than a second threshold;
unlocking the first parameter if the alert level is lower than a third threshold; and
identifying the anomaly when the alert level is greater than a fourth threshold.
9. The method of claim 8 further comprising:
adjusting the third threshold or the fourth threshold if the identified anomaly is a false anomaly.
10. The method of claim 7 further comprising:
generating a reference based on another plurality of message tally counts tallied when no anomaly is identified;
comparing the plurality of message tally counts to the reference to generate a comparison result when the anomaly is identified; and
identifying a cause of the anomaly based on the comparison result.
11. The method of claim 7 ,
wherein the plurality of distinct URIs correspond to source IP addresses of the plurality of call request messages, and
wherein determining the measure of randomness of user activity of the SIP network traffic based on the plurality of message tally counts is determining a measure of randomness of caller activity of the SIP network traffic.
12. The method of claim 11 , wherein determining the measure of randomness of the caller activity of the SIP network traffic is according to a user activity diversity metric.
13. The method of claim 7 ,
wherein the plurality of distinct URIs correspond to destination IP addresses of the plurality of call request messages, and
wherein determining the measure of randomness of user activity of the SIP network traffic based on the plurality of message tally counts is determining a measure of randomness of callee activity of the SIP network traffic.
14. The method of claim 13 , wherein determining the measure of randomness of the callee activity of the SIP network traffic is according to a user activity diversity metric.
15. The method of claim 7 ,
wherein the plurality of distinct URIs comprise a plurality of distinct URI pairs, wherein each distinct URI pair corresponds to a source IP address and a destination IP address of at least one of the plurality of call request messages, and
wherein determining the measure of randomness of the SIP network traffic based on the plurality of message tally counts is determining a measure of randomness of caller-callee pair activity of the SIP network traffic.
16. The method of claim 15 , wherein determining the measure of randomness of the caller-callee pair activity of the SIP network traffic is according to a user activity diversity metric.
17. The method of claim 7 ,
wherein the plurality of distinct URIs correspond to source IP addresses of a portion of the plurality of call request messages associated with a callee, and
wherein determining the measure of randomness of the SIP network traffic based on the plurality of message tally counts is determining a measure of randomness of caller activity of the SIP network traffic associated with the callee.
18. The method of claim 17 , wherein determining the measure of randomness of the caller activity of the SIP network traffic associated with the callee is according to a user activity diversity metric.
19. The method of claim 7 ,
wherein the plurality of distinct URIs correspond to destination IP addresses of a portion of the plurality of call request messages associated with a caller, and
wherein determining the measure of randomness of the SIP network traffic based on the plurality of message tally counts is determining a measure of randomness of callee activity of the SIP network traffic associated with the caller.
20. The method of claim 19 , wherein determining the measure of randomness of the callee activity of the SIP network traffic associated with a caller is according to a user activity diversity metric.