Methods and apparatus for statistical mobile device identification

1. A method for mobile device identification, the method comprising:

receiving, at a server of a processing system including a processor, a request to establish a connection with the server from a source device;

receiving, by the processing system, identifying information associated with the source device, the identifying information comprising a timestamp generated by the source device and a device-type unique identifier;

determining, by the processing system, a statistically unique identifier of the source device according to the identifying information;

retrieving, by the processing system, a stored record of a previous connection associated with the statistically unique identifier of the source, wherein the stored record of the previous connection comprises a timestamp generated by a source of a request to establish the previous connection; and

determining, by the processing system, based on the identifying information, the stored record of the previous connection and a current timestamp of the server, a likelihood that the previous connection associated with the statistically unique identifier was made between the source device and the server,

wherein determining the likelihood comprises determining whether a ratio of a first difference between the timestamp generated by the source device and the timestamp generated by the source of the request to establish the previous connection, and a second difference between the current timestamp of the server and a timestamp generated by the server in response to the request to establish the previous connection, exceeds a threshold, and

wherein the threshold comprises a variable value subject to exponential decay.

2. The method of claim 1 , wherein the source device is a mobile device; and wherein the device-type unique identifier comprises one of a pixel-ratio, a screen height and width, locale information, a time zone offset, or any combination thereof.

3. The method of claim 1 , wherein the identifying information is provided in one or more TCP header fields.

4. The method of claim 1 , wherein the identifying information is provided in one or more TCP SYN packets.

5. The method of claim 1 , wherein the identifying information further comprises at least one of an IP address, a port number, and a plurality of HTTP headers, and wherein the device-type unique identifier is based on an HTTP header of the plurality of HTTP headers.

6. The method of claim 5 , further comprising:

generating, by the processing system, a device identifier to obtain a generated device identifier based on at least one of the HTTP headers and at least the device-type unique identifier; and

storing, by the processing system, the identifying information and the current timestamp in association with the generated device identifier.

7. The method of claim 6 , further comprising storing, by the processing system, records of previously established connections, each record comprising a device identifier, a timestamp generated by a source of a particular request to establish a connection, resulting in a requested connection, and a timestamp generated by the server in response to receiving the particular request.

8. The method of claim 7 , wherein determining the likelihood further comprises, for each record having a device identifier equal to the generated device identifier, determining, by the processing system, whether an alignment between the timestamps associated with the record and timestamps associated with the requested connection exceeds a threshold value, wherein the timestamps associated with the requested connection comprise the timestamp generated by the source device and the current timestamp of the server.

9. The method of claim 1 , wherein the server further comprises an advertising server; and

wherein the request to establish a connection further comprises a request for an advertisement to be served; and

wherein the likelihood that a previous connection was made further comprises a likelihood that a previous request for an advertisement to be served was received from the source device.

10. A system for mobile device identification, the system comprising:

a processing system including a processor; and

a memory that stores executable instructions that, when executed by the processing system, facilitate performance of operations, the operations comprising:

receiving, at a server, a request to establish a connection with the server from a source device;

receiving identifying information associated with the source device, the identifying information comprising a timestamp generated by the source device and a device-type unique identifier;

determining a statistically unique identifier of the source device according to the identifying information;

obtaining a stored record of a previous connection associated with the statistically unique identifier of the source, wherein the stored record of the previous connection comprises a timestamp generated by a source of a request to establish the previous connection; and

determining, based on the identifying information, the stored record of the previous connection and a current timestamp of the server, a likelihood that the previous connection associated with the statistically unique identifier was made between the source device and the server,

wherein determining the likelihood comprises determining whether a ratio of a first difference between the timestamp generated by the source device and the timestamp generated by the source of the request to establish the previous connection and a second difference between the current timestamp of the server and a timestamp generated by the server in response to the request to establish the previous connection exceeds a threshold, and

wherein the threshold comprises a variable value subject to exponential decay.

11. The system of claim 10 , wherein the source device is a mobile device.

12. The system of claim 10 , wherein the identifying information is provided in one or more TCP header fields.

13. The system of claim 10 , wherein the identifying information is provided in one or more TCP SYN packets.

14. The system of claim 10 , wherein the identifying information further comprises at least one of an IP address, a port number, and a plurality of HTTP headers, and wherein the device-type unique identifier is based on an HTTP header of the plurality of HTTP headers.

15. The system of claim 14 , wherein the operations further comprise:

generating a device identifier to obtain a generated device identifier based on at least one of the HTTP headers and at least the device-type unique identifier; and

storing the identifying information and the current timestamp in association with the generated device identifier.

16. The system of claim 15 , wherein the operations further comprise storing records of previously established connections, each record comprising a device identifier, a timestamp generated by a source of a particular request to establish a connection, resulting in a requested connection, and a timestamp generated by the server in response to receiving the particular request.

17. The system of claim 16 , wherein determining the likelihood further comprises, for each record having a device identifier equal to the generated device identifier, determining whether an alignment between the timestamps associated with the record and timestamps associated with the requested connection exceeds a threshold value, wherein the timestamps associated with the requested connection further comprise the timestamp generated by the source device and the current timestamp of the server.

18. The system of claim 10 , wherein the server further comprises an advertising server; and

wherein the request to establish a connection further comprises a request for an advertisement to be served; and wherein the likelihood that a previous connection was made comprises a likelihood that a previous request for an advertisement to be served was received from the source device.

19. The method of claim 1 , further comprising:

storing, by the processing system, a set of records of previously established connections to the server; and

generating, by the processing system, a subset of the set of records by determining, based on the identifying information and a current timestamp of the server, which records in the set of records meet a threshold likelihood that a previous connection was made between the source device and the server.

20. The method of claim 1 , further comprising normalizing, by the processing system, the timestamp generated by the source device to a constant time base matching a timestamp resolution of the server by multiplying or dividing by an increment frequency conversion value.

21. The method of claim 6 , wherein the generated device identifier further represents a category of devices rather than an individual device.