Predictive spectral allocation in mobile networks

1. A method to analyze sub-carrier allocation data to categorize clusters for predictive spectral allocation in a wireless network, the method comprising:

receiving information on device location, time slots, at least one sub-carrier allotted for each time slot, and quality indicators from a plurality of mobile devices;

grouping the time slots to form analysis intervals;

dynamically adjusting a length of the analysis intervals based on an expected population change within a geographic area that encompasses the plurality of mobile devices;

identifying a time-frequency vector for each analysis interval, wherein the time-frequency vector associates the analysis interval with the device location;

computing a goodness indicator for each time-frequency vector;

identifying clusters of time-frequency vectors; and

categorizing the clusters of time-frequency vectors into two or more categories.

2. The method according to claim 1 , wherein categorizing the clusters into two or more categories comprises labeling the clusters as good clusters or bad clusters.

3. The method according to claim 1 , further comprising:

employing a machine learning technique to identify the clusters of time-frequency vectors with substantially similar values of goodness indicators.

4. The method according to claim 1 , wherein computing the goodness indicator comprises:

computing the goodness indicator based on one or more of: channel quality indicators from the plurality of the mobile devices, a received signal strength indicator (RSSI), a bit error rate (BER), a number of packet retransmissions, a signal to noise ratio (S/N), a number of call failures, a user feedback, a user requested power-down at a same bit-rate, and a user requested power-up.

5. The method according to claim 1 , wherein the device location is based on a plurality of units of a geographic grid, and the method further comprises:

dynamically adjusting a size of each unit of the geographic grid based on one or more of whether the device location is in a densely populated area, whether the device location is in a sparsely populated area, a communication technology of the wireless network, and a selected frequency of communication.

6. The method according to claim 5 , further comprising:

further adjusting the size of each unit of geographic grid based on one or more of a time of day, a day of week, a day of month, a season, and an expected population change within a geographic area.

7. The method according to claim 5 , wherein receiving the information from the plurality of mobile devices comprises receiving the information over a wireless communication technology that utilizes a multiple-access technology.

8. An analysis server to analyze sub-carrier allocation data to categorize clusters for predictive spectral allocation in a wireless network, the analysis server comprising:

a memory configured to store instructions; and

a processor coupled to the memory, the processor configured to execute a communication application in conjunction with the stored instructions, the communication application comprising:

an analysis module configured to:

evaluate information on device location, time slots, at least one sub-carrier allotted for each of the time slots, and quality indicators received from a plurality of mobile devices;

group the time slots by location to form analysis intervals;

dynamically adjust a length of the analysis intervals based on an expected population change within a geographic area that encompasses the plurality of mobile devices;

identify a time-frequency vector for each analysis interval, wherein the time-frequency vector associates the analysis interval with the device location; and

compute a goodness indicator for each time-frequency vector; and

a support vector machine (SVM) module configured to:

identify clusters of time-frequency vectors; and

categorize clusters of time-frequency vectors into two or more categories.

9. The analysis server according to claim 8 , wherein the analysis module is further configured to:

dynamically adjust a size of each unit of a geographic grid based on one or more of whether the device location is in a densely populated area, whether the device location is in a sparsely populated area, a communication technology of the wireless network, and a selected frequency of communication.

10. The analysis server according to claim 9 , wherein the analysis module is further configured to:

further adjust the size of each unit of the geographic grid based on one or more of a time of day, a day of week, a day of month, a season, and an expected population change within a geographic area.

11. The analysis server according to claim 8 , wherein a communication technology of the wireless network is one of frequency division multiple access (FDMA), orthogonal frequency division multiple access (OFDMA), or a narrow band allocation technology.

12. The analysis server according to claim 8 , wherein the analysis module is further configured to:

determine the time-frequency vector for one of a single sub-carrier frequency or a group of sub-carrier frequencies.

13. The analysis server according to claim 8 , wherein the analysis module is further configured to:

compute the goodness indicator based on a type of communication associated with a user.

14. The analysis server according to claim 8 , wherein the analysis server is part of a base station.

15. The analysis server according to claim 8 , wherein the analysis module is configured to receive the information from the plurality of mobile devices via one or more base stations.

16. The analysis server according to claim 8 , wherein the wireless network is one of an Evolved Universal Mobile Telecommunications System Terrestrial Radio Access Network (eUTRAN), a long term evolution (LTE) network, an LTE-Advanced network, a high speed packet access (HSPA) network, or an HSPA-Advanced network.

17. A computer readable memory device with instructions stored thereon to analyze sub-carrier allocation data to categorize clusters for predictive spectral allocation in a wireless network, the instructions being executable by a computing device to perform or cause to be performed:

identifying information on device location, time slots, at least one sub-carrier allotted for each time slot, and quality indicators from a plurality of mobile devices;

grouping the time slots to form analysis intervals;

dynamically adjusting a length of the analysis intervals based on an expected population change within a geographic area that encompasses the plurality of mobile devices;

identifying a time-frequency vector for each analysis interval, wherein the time-frequency vector associates the analysis interval with the device location;

computing a goodness indicator for each time-frequency vector;

identifying clusters of time-frequency vectors based on the computed goodness indicator for each time-frequency vector; and

categorizing the clusters of time-frequency vectors into two or more categories.

18. The computer readable memory device of claim 17 , wherein the instructions are further executable by the computing device to perform or cause to be performed:

employing a machine learning technique to identify the clusters of time-frequency vectors with substantially similar values of goodness indicators, wherein the machine learning technique includes one or more of a neural network, a support vector machine (SVM), and a Bayes Classifier.

19. The computer readable memory device of claim 17 , wherein

dynamically adjusting the length of the analysis intervals is further based on one or more of a time of day, a day of week, a day of month, a season, and a location of a mobile device.

20. The computer readable memory device of claim 17 , wherein the instructions are further executable by the computing device to perform or cause to be performed:

determining the time-frequency vector for one of a single sub-carrier frequency or a group of sub-carrier frequencies.

21. The computer readable memory device of claim 17 , wherein the instructions are further executable by the computing device to perform or cause to be performed:

computing the goodness indicator based on a type of communication associated with a user.