A Spectrum Sensing Method Based on Empirical Mode Decomposition and K-Means Clustering Algorithm

Table 1

The algorithm of KEMDSS.

---

KEMDSS algorithm

---

Input: Spectrum signal

(1) Denoising with EMD algorithm to obtain reconstructed signal

(2) Construct the perceptual matrix and calculate the corresponding covariance matrix

(3) Constructed eigenvector

(4) Train the classifier to get the corresponding cluster center

(5) Import test set and get test result according to Eq. (17)

(6) Calculate and according to Eq. (3) and Eq. (4)

Output: and

---