Entropy-based subspace separation for multiple frequency estimation

Ince, Erhan A.; Yu, Runyi; Hocanin, Aykut

doi:10.1016/j.dsp.2012.09.014

Cited by 9 publications

(9 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The SAS-Est was used to estimate order and frequencies for multiple sinusoids buried in noise. The results were compared with the ones from subspace-based estimators such as the A-MUSIC [6] and E-MUSIC [7]. The percentage for correct order estimation (PCOE) has been used in evaluating the performance for model order estimation; whereas for frequency estimation, we examine the mean square error (MSE) of each frequency in the case of random multiple complex tone models or the fundamental frequency in the case of harmonic multiple complex tone models.…”

Section: Simulation Resultsmentioning

confidence: 99%

“…2 depicts the PCOE assuming random complex tones and an SNR of 20 dB. The performance of the A-MUSIC [6], E-MUSIC [7], and SAS-Est are compared. We observe that for values of 3 and 4 all three estimators perform very close to each other.…”

Section: Simulation Resultsmentioning

confidence: 99%

“…Note that, this would not be the case should the objective function be selected differently. Consequently Therefore (5) or (6) Next, we compare objective function with the following two functions: one used in A-MUSIC [6]: (7) and the other used in the E-MUSIC [7]: (8) where for small 's. We first state for in (7) that (9) where is the Euler's constant.…”

Section: Subspace Alignment and Separation For Estimationmentioning

confidence: 99%

“…This algorithm was later improved/extended in the constraint MUSIC [3], time-frequency MUSIC [4], the harmonic MUSIC [5]. Recently, the angle-based MUSIC [6] (referred to as A-MUSIC herein) and the entropy-based E-MUSIC [7] have also employed subspace separation for order and frequency estimation. The property of subspace alignment has previously been used in the rotational invariance technique (ESPRIT) [8].…”

Section: Introductionmentioning

confidence: 98%

“…In order to enhance performance, we consider angles between subspaces in characterizing subspace alignment and separation. Using the angles rather than simple and direct measures such as distance gives more robust estimation [6], [7] since the angles used in computation are independent of the bases (of respective subspaces). Furthermore, to achieve an almost uniform distribution for the angles, we use the generalized Kullback-Leibler divergence (i.e., the -divergence) [9] as the objective function in our optimization problem.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Subspace Alignment and Separation for Multiple Frequency Estimation

Yu¹,

Ince²,

Hocanin³

2015

IEEE Signal Process. Lett.

Self Cite

View full text Add to dashboard Cite

In this letter, a new subspace based estimator that can effectively provide the order and frequencies of multiple sinusoids in noise is proposed. The estimator, referred to as SAS-Est (Subspace Aligning and Separating Estimator), simultaneously seeks to separate the steering vectors from the noise subspace and align them to the signal subspace. The angles between subspaces and the generalized Kullback-Leibler divergence are used in characterizing the alignment and separation. Minimizing the divergence leads to maximal subspace separation and best alignment, thus allowing improved performance. Simulations in additive white Gaussian noise show that the new estimator offers an improvement for both model order and frequency estimation. When compared with other methods, the improvement is more pronounced for high model orders and low signal-to-noise ratio values.

show abstract

Section: Simulation Resultsmentioning

confidence: 99%

Section: Simulation Resultsmentioning

confidence: 99%