Multi-frequency signal classification by multilayer neural networks and linear filter methods
スポンサーリンク
概要
- 論文の詳細を見る
This paper compares signal classification performance of multilayer neural networks (MLNNs) and linear filters (LFs). The MLNNs are useful for arbitrary waveform signal classification. On the other hand, LFs are useful for the signals, which are specified with frequency components. In this paper, both methods are compared based on frequency selective performance. The signals to be classified contain several frequency components. Furthermore, effects of the number of the signal samples are investigated. In this case, the frequency information may be lost to some extent. This makes the classification problems difficult. From practical viewpoint, computational complexity is also limited to the same level in both methods. IIR and FIR filters are compared. FIR filters with a direct form can save computations, which is independent of the filter order. IIR filters, on the other hand, cannot provide good signal classification due to their phase distortion, and require a large amount of computations due to their recursive structure. When the number of the input samples is strictly limited, the signal vectors are widely distributed in the multi-dimensional signal space. In this case, signal classification by the LF method cannot provide a good performance. Because, they are designed to extract the frequency components. On the other hand, the MLNN method can form class regions in the signal vector space with high degree of freedom. When the number of the signal samples is not so limited, both the MLNN and LF methods can provide the same high classification rates. In this case, since the signal vectors are distributed in the specific region, the MLNN method has some convergence problem, that is local minimum problem. The initial weights should be carefully determined around the optimum solution. Another point is robustness for noisy signal. The LFs can suppress wide-band noise by using very high-Q filters. However, the MLNN method can be also robust. Rather, it is a little superior to the LF method when the computational load is limited.
- 1997-05-25
著者
-
Hara Kazuyuki
Toyama Polytechnic College:(present Addres)graduate School Of Natural Science And Technology Kanazaw
-
Nakayama Kenji
Department Of Cardiothoracic Surgery Niigata Prefectural Shibata Hospital
関連論文
- Clinicopathological study of pleomorphic xanthoastrocytoma : Correlation between histological features and prognosis
- Contralateral Development of Acute Subdural Hematoma Following Surgery for Chronic Subdural Hematoma : Case Report
- Change of Patient Position Using a Transportation Board During Lumboperitoneal Shunting : Technical Note
- Modified Draping to Avoid Fluid Leakage in Cranial Surgery : Technical Note
- Video-Assisted Thoracic Surgery for Thorascopic Resection of Giant Bulla
- A-2-13 Comparison of Back-Propagation, Hybrid Learning and Support Vector Machine based on Generalization
- Video-assisted transaortic left ventricular thrombectomy and coronary artery bypass grafting
- Nonlinear resonance effects in a linear Paul trap
- Anchoring of Methylmethacrylate Filler to the Calvarium : Technical Note
- Inclined Foot Switches for Surgical Microscopes : A Comfortable Design for Seated Surgeons : Technical Note
- The tumor of the third ventricle
- Multi-frequency signal classification by multilayer neural networks and linear filter methods
- Depth-Adjustable Fixation of External Ventricular Drains to Counteract Obstruction in Tight Ventricles : Technical Note
- Determination of Transverse Distributions of Ion Plasmas Confined in a Linear Paul Trap by Imaging Diagnostics
- A hybrid nonlinear predictor: Analysis of learning process and predictability for noisy time series
- Hanging Foot Switch for Bipolar Forceps: A Device for Surgeons Operating in the Standing Position:—Technical Note—
- Decompressive hemicraniectomy for acute subdural hematoma.
- RADIOLOGICAL FEATURES OF LIPOMA OF THE CORPUS CALLOSUM
- CT FINDINGS IN AN ALOBAR HOLOPROSENCEPHALY ASSOCIATED WITH DANDY-WALKER'S CYST
- Hanging Foot Switch for Bipolar Forceps : A Device for Surgeons Operating in the Standing Position : Technical Note