Multiresolution based power spectral density estimation

1. A method for estimating multi-resolution based power spectral density (PSD), the method comprising:

receiving, from a communication device, a set of wireless signals from a plurality of wireless devices;

sampling, at a base station control module, the received set of wireless signals;

computing, at the base station controller, a first PSD of a signal from the sampled set of wireless signals to determine a first resolution and a first error variance (EV) value associated with the first PSD;

computing, at the base station controller, a second PSD of the signal to determine a second resolution and a second EV value associated with the second PSD;

determining, at the base station controller, missing frequency variables of the second PSD using the first resolution and the second EV through an expectation maximization technique; and

facilitating, at the base station controller, wireless communications between the plurality of wireless devices based on the determined missing frequency variables.

2. The method of claim 1 , further comprising determining, at the base station controller, a sub-band energy level by computing an area under one of segments associated with the second PSD between noise level value and an adjacent noise level value.

3. The method of claim 2 , further comprising assigning, at the base station controller, the noise level value and the adjacent noise level value as approximate edge frequency values in response to detecting the noise level value and the adjacent noise level value bordering the area having a larger size in comparison to areas of other segments.

4. The method of claim 3 , further comprising detecting, at the base station controller, exact edge frequency values of the signal by locating a minimum value and a maximum value of a first derivative of the second PSD within the noise level value and the adjacent noise level value.

5. The method of claim 1 , further comprising computing, at the base station controller, a probability distribution of the missing frequency variables based on existing variables of the second PSD for the expectation maximization technique.

6. The method of claim 5 , further comprising executing, at the base station controller, a maximization of the probability distribution based on a mean variance and a co-variance of the second PSD.

7. The method of claim 6 , further comprising repeating, at the base station controller, the maximization of the probability distribution to determine a convergence.

8. The method of claim 7 , further comprising estimating, at the base station controller, missing frequency variables based on the convergence.

9. A controller integrated into a wireless base station for estimating multi-resolution based power spectral density (PSD), the controller comprising:

an expectation maximization module configured to:

receive, from a communication device, a set of wireless signals from a plurality of wireless devices;

sample the received set of wireless signals;

compute a first PSD of a signal from the sampled set of wireless signals to determine a first resolution and a first error variance (EV) value associated with the first PSD;

compute a second PSD of the signal to determine a second resolution and a second EV value associated with the second PSD;

determine missing frequency variables of the second PSD using the first resolution and the second EV through an expectation maximization technique;

determine an average value for the second PSD over consecutive non-overlapping segments of the signal; and

facilitate wireless communications between the plurality of wireless devices based on the determined missing frequency variables.

10. The controller according to claim 9 , wherein the expectation maximization module is further configured to:

determine a sub-band energy level by computing an area under one of the segments associated with the second PSD between noise level value and an adjacent noise level value.

11. The controller according to claim 10 , wherein the expectation maximization module is further configured to:

assign the noise level value and the adjacent noise level value as approximate edge frequency values in response to detecting the noise level value and the adjacent noise level value bordering the area having a larger size in comparison to areas of other segments.

12. The controller according to claim 11 , wherein the expectation maximization module is further configured to:

detect exact edge frequency values of the signal by locating a minimum value and a maximum value of a first derivative of the second PSD within the noise level value and the adjacent noise level value.

13. A computing device for estimating multi-resolution based power spectral density (PSD), the device comprising:

one or more communication devices configured to receive a set of wireless signals from a plurality of wireless devices; and

a controller coupled to the one or more communication devices, to manage an expectation management module, the controller configured to:

receive, from the one or more communication devices, the set of wireless signals;

sample the received set of wireless signals;

compute a first PSD of a signal from the sampled set of wireless signals to determine a first resolution and a first error variance (EV) value associated with the first PSD;

compute a second PSD of the signal to determine a second resolution and a second EV value associated with the second PSD;

determine missing frequency variables of the second PSD using the first resolution and the second EV through an expectation maximization technique;

determine an average value for the second PSD over consecutive non-overlapping segments of the signal;

compute an integer value for the average value;

assign the integer value to a noise level value in response to determining the integer value as a mode value of the second PSD based on a predetermined spectrum assumption of a cognitive radio associated with the signal; and

facilitate wireless communications between the plurality of wireless devices based on the determined missing frequency variables.

14. The device according to claim 13 , wherein the controller is further configured to:

determine a sub-band energy level by computing an area under one of the segments associated with the second PSD between the noise level value and an adjacent noise level value.

15. The device according to claim 14 , wherein the controller is further configured to:

assign the noise level value and the adjacent noise level value as approximate edge frequency values in response to detecting the noise level value and the adjacent noise level value bordering the area having a larger size in comparison to areas of other segments.

16. The device according to claim 15 , wherein the controller is further configured to:

detect exact edge frequency values of the signal by locating a minimum value and a maximum value of a first derivative of the second PSD within the noise level value and the adjacent noise level value.

17. The device according to claim 16 , wherein the controller is further configured to:

compute a probability distribution of the missing frequency variables based on existing variables of the second PSD for the expectation maximization technique.

18. The device according to claim 17 , wherein the controller is further configured to:

execute a maximization of the probability distribution based on a mean variance and a co-variance of the second PSD.

19. The device according to claim 18 , wherein the controller is further configured to:

repeat the maximization of the probability distribution to determine a convergence.

20. The device according to claim 19 , wherein the controller is further configured to:

estimate the missing frequency variables based on the convergence.