Page 1 of 7
Journal for Studies in Management and Planning
Available at
http://edupediapublications.org/journals/index.php/JSMaP/
ISSN: 2395-0463
Volume 03 Issue 12
November 2017
Available online: http://edupediapublications.org/journals/index.php/JSMaP/ P a g e | 326
Designing Of Robust Algorithm for Convolutive Blind
Source Separartion
1.BITLA SRUTHI,2.M DEVADAS
1.Pg Scholar, Department of ECE, Vaagdevi College of Engineering, Bollikunta Warangal,
Telangana
2.Assosciate Professor, Department of ECE, Vaagdevi College of Engineering, Bollikunta
Warangal, Telangana
ABSTRACT
Brief presents an efficient very-large-scale
integration architecture design for
convolutive blind source separation (CBSS).
The CBSS separation network derived from
the information maximization (Infomax)
approach is adopted. The proposed CBSS
chip design consists mainly of Infomax
filtering modules and scaling factor
computation modules. In an Infomax
filtering module, input samples are filtered
by an Infomax filter with the weights
updated by Infomax-driven stochastic
learning rules. As for the scaling factor
computation module, all operations
including logistic sigmoid are integrated and
implemented by the circuit design based on
a piecewise-linear approximation scheme.
The proposed prototype chipis implemented
via a semicustom design using90-nmCMOS
technology on a die size of approximately
0.54×0.54 mm2.
I. INTRODUCTION
Blind source separation (BSS) attempts to
separate sources from mixed signals when
most of the information for sources and
mixing process is unknown. Such
restrictions make BSS a challenging task for
researchers. BSS has become a very
important research topic in a lot of fields.
Notable examples include audio signal
processing, biomedical signal processing,
communication systems, and image
processing . Without a filtering effect,
instantaneous mixing is considered a simple
version of the mixing process of the source
Page 2 of 7
Journal for Studies in Management and Planning
Available at
http://edupediapublications.org/journals/index.php/JSMaP/
ISSN: 2395-0463
Volume 03 Issue 12
November 2017
Available online: http://edupediapublications.org/journals/index.php/JSMaP/ P a g e | 327
signals. However, for audio sources passing
through an environmental filtering before
arriving at the microphones, a convolutive
mixing process occurs, and convolutive BSS
(CBSS) is used to recover the original audio
sources. Independent component analysis
(ICA) is the conventional means of solving
the BSS or CBSS problem. However, this
method is often highly computationally
intensive and introduces time-consuming
processes for software implementation.
More than a faster solution than software
implementation, hardware solution achieves
optimal parallelism. Providing hardware
solutions for ICA-based BSS has drawn
considerable attention recently. Cohen and
Andreou explored the feasibility of
combining above-andsubthreshold CMOS
circuit techniques for implementing an
analog BSS chip that integrates an analog
I/O interface, weight coefficients, and
adaptation blocks. This chip incorporates the
use of the Herault–Jutten ICA algorithm.
Cho and Lee implemented a fully analog
CMOS chip based on information
maximization (Infomax) ICA, as developed
by Bell and Sejnowski. The chip
incorporated a modular architecture to
extend its use as a multichip. Apart from
these analog BSS chips, various
fieldprogrammable gate array(FPGA)
implementations with digital architectures
have been developed. Li and Lin realized the
Infomax BSS algorithm based on system- level FPGA design, by using Quartus II,
DSP builder, and Simulink. Du and Qi
presented an FPGA implementation for the
parallel ICA (pICA) algorithm, which
focuses on reducing dimensionality in
hyperspectral image analysis. The pICA
algorithm consists of three temporally
independent functional modules that are
synthesized individually with some
reconfigurable components developed for
reuse. Based on Infomax BSS, Ounas et al.
introduced a low-cost digital architecture
implemented on FPGA. This design used
merely one neuron to support sequential
operations of the neurons in neural network.
In 2008, Shyu et al.designed a pipelined
architecture for FPGA implementation based
on FastICA for separating mixtures of
biomedical signals, including
electroencephalogram (EEG), magneto
encephalography (MEG), and
electrocardiogram (ECG). In this design,
floating-point arithmetic units were used to
Page 3 of 7
Journal for Studies in Management and Planning
Available at
http://edupediapublications.org/journals/index.php/JSMaP/
ISSN: 2395-0463
Volume 03 Issue 12
November 2017
Available online: http://edupediapublications.org/journals/index.php/JSMaP/ P a g e | 328
increase the precision of the numbers and
ensure the FastICA performance.
II. LITERATURE SURVEY
Separating brain imaging signals by
maximizing their autocorrelations is an
important component of blind source
separation (BSS). Canonical correlation
analysis (CCA), one of leading BSS
techniques, has been widely used for
analyzing optical imaging (OI) and
functional magnetic resonance imaging
(fMRI) data. However, because of the need
to reduce dimensionality and ignore spatial
autocorrelation, CCA is problematic for
separating temporal signal sources. To solve
the problems of CCA, "straightforward
image projection" (SIP) has been
incorporated into temporal BSS. This novel
method, termed low-dimensional canonical
correlation analysis (LD-CCA), relies on the
spatial and temporal autocorrelations of all
genuine signals of interest. Incorporating
both spatial and temporal information, here
we introduce a "generalized timecourse"
technique in which data are artificially
reorganized prior to separation. The quantity
of spatial plus temporal autocorrelations can
then be defined. By maximizing temporal
and spatial autocorrelations in combination,
LD-CCA is able to obtain expected "real"
signal sources. Generalized timecourses are
low-dimensional, eliminating the need for
dimension reduction. This removes the risk
of discarding useful information. The new
method is compared with temporal CCA and
temporal independent component analysis
(tICA). Comparison of simulated data
showed that LD-CCA was more effective
for recovering signal sources. Comparisons
using real intrinsic OI and fMRI data also
supported the validity of LD-CCA. Online
blind source separation (BSS) is proposed to
overcome the high computational cost
problem, which limits the practical
applications of traditional batch BSS
algorithms
III PROPOSED VLSI BLIND SOURCE
SEPARATOR
The proposed CBSS system is shown in the
FIG.1. 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
