International Research Journal of Engineering and Technology (IRJET) Volume: 04 Issue: 05 | May -2017
www.irjet.net
e-ISSN: 2395 -0056 p-ISSN: 2395-0072
VLSI DESIGN FOR CONVOLUTIVE BLIND SOURCE SEPARATION Ramya K1, Rohini A B2, Apoorva3, Lavanya M R4, Vishwanath A N5 1Asst.
professor, Dept. Of ECE, BGS Institution of Technology, Karnataka, India Students, Dept. Of ECE, BGS Institute of Technology, Karnataka, India -----------------------------------------------------------------------***-----------------------------------------------------------------------ABSTRACT- This will presents an efficient Very Large Scale Integration (VLSI) design for Convolutive Blind Source Separation (CBSS). Information maximization (Infomax) approach is adopted for CBSS network. CBSS chip design mainly includes Infomax filtering modules and scaling factor computation modules. In an Infomax filtering module, filtering of input samples are done by Infomax filter with the weights updated by Infomax driven stochastic learning rules. And for scaling factor computation module all operations are implemented by the circuit design based on a piecewise-linear approximation scheme. An efficient and high performance and less delayed blind source separation technique is described . 2,3,4,5 UG
Key Words- Convolutive Blind Source Separation (CBSS), Infomax, Scaling factor. 1. INTRODUCTION
Blind source separation is a kind of a filtering process used to separate different sources from the mixed signals in which most of the information about sources and mixed signals is not known. This restriction makes the blind source separation a challenging task. Blind source separation becomes a very important research topics in a lot of fields such as audio signal processing, biomedical signal processing, communication systems and image processing. Simple version of mixing process is one in which without filtering effect instantaneous mixing occurs. Convolutive mixing process should be done for the audio source passing through a filtering environment before arriving at the microphones and in order to recover the original audio source convoluted blind source separation should be done. One of the conventional methods is Independent component analysis (ICA) which is used to solve the CBSS problem. Major drawback of software implementation using this technique is often highly computational intensive and more time consuming process. Providing hardware solutions for ICA-based blind source separation has drawn considerable attention because of the hardware solution achieves optimal parallelism. An analog BSS chip can be designed using above-and–sub threshold CMOS circuit techniques which integrates an i/o interface of analog, weight coefficients and adoption blocks. 2. PROPOSED VLSI BLIND SOURCE SEPARATOR The proposed CBSS system is shown in the FIG.2. The CBSS chip mainly consists of two functional cores: Infomax filtering module and scaling factor computation module. Additionally, the Infomax filtering outputs are added with the help of two small carry-save adders (CSAs). The current prototype chip is used for two sources and two sensors by utilizing four Infomax filtering modules along with two scaling factor computation modules.
Š 2017, IRJET
|
Impact Factor value: 5.181
|
ISO 9001:2008 Certified Journal
| Page 1535