Introduction:
.
.
.
.
.
Olfactory Physiology
Organic Chemistry
Signal Processing
Pattern Recognition Computational Learning
Electronic
Nose
Chemical Sensors /
Analytical Chemistry
Electronic Nose
Systems
C IM C o a t e d Sensor
Se nso r Pr o c e ssor
Si g na l Pr o c e ss o r
Pat t e r n R e c o g ni t i o n
Know l e dg e
B a se
Odoran t M o l e c ul e s
Ide nt i f i c a t i o n
R a w el ec tri ca l s i g na l
P ro ce s s ed El ec tri ca l s i g na l
D en o i s ed E l ec tri ca l s i g na l
T ra i ni ng
&
T es ti ng
C he m i ca l I nf o rm a ti o n
Fe a t ur e E x t r a c t o r
R el ev a nt f ea ture s
El e c tr o n i c N o s e S y s te m
M a n uf a c t u r e r
S e n s o r
Te c h n o l o gy
# o f
Se n s o r s
P atte r n
R e c o gn i ti o n
A l go r i th m s
P r i c e
($)
C ou n tr y o f
Or i gi n
WM A A i r s e n se
A n a l y se n t e c hn i k G m bH
MOS 10 A N N,D C,P C A
20,000 -
43,000
G e r m a ny
F ox x 000
A l p h a MOS
Q C M,S A W,
C P,MOS
6 - 24 A N N,D F A,P C A
20,000 -
100,000
F r a n c e
A r om aS c an
O s m e T e c h I n c,
CP 48 A N N,F L
20,000 -
75,000
U,K,
B H 114
B l oodhou nd S e n s o r s I n c,
CP 14
A N N,C A,D A,
P C A
U,K,
C yr an ose 320
C y r a no S c i e n c e s Lt d,
CP 32 P C A 5,000 U S A
E n ose 5000
Ma r c o n i L t d,
Q C M,M O S,
C P,S A W
8 - 28 AN N,D A,P C A? U,K,
Z n ose 4200
Ele c t r on i c S e n s o r Te c h,
S A W,G C 6 - 15 SPR
19,500 -
25,000
U S A
Q M B 6 – H S 40X L
H K R S e n s o r sy st e m G m bH
Q C M 6 A N N,P C A? G e r m a ny
M O S E S II
L e nn a r t z Ele c t r on i k G m bH
Q C M,M O S 16 A N N,P C A? G e r m a ny
N S T 3210
N o r d i c S e n s o r Te c hno l o gi e s
MOS,F ET,
Q C M
22 A N N,P C A
40,000 -
60,000
S w e de n
O l i goS e n se
O l i go S e n s e
CP N D / P R N D / P R? B e l g i um
S A M
D a i m l e r R S T R o s t oc k
Q C M,S A W,
MOS
6 - 10 A N N,P C A 50,000 G e r m a ny
S M ar t N os e 300
S Ma r t N o se
MS N / A D A,P C A? S w i t z e r l a nd
V O C m e t e r
Mo Te c h S e n s o r i k,G m bH
Q C M,M O S 8 A N N,P C A? G e r m a ny
F r e sh S e n se
Ele m e n t Ltd,
MOS N D / P R N D / P R? I c e l a nd
4440B
H P – A g i l e n t T e c hno l og i e s
MS N / A
V a r i o u s
C h e m o m e t r i c s
79,900 U S A
V ap or L ab
Mi c r o S e n so r I nc,
S A W 2 N D / P R 5,000 U S A
L i b r a N ose
T e c hno b i o c hi p
Q C M 8 N D / P R 5,000 I t a l y
Commercially
Available
Systems
Block Diagram of the
Experimental Setup
P ERSON AL
CO M P UTER
M ASS F LOW
CO NT RO LLER
NET W O RK
AN AL Y Z ER
S ENSOR CELL
6 Q C M s
VOC
B U BB LERS
Bi - d irec ti o n a l i n f o r m a ti o n f lo w
Un id irec ti o n a l in f o rm a ti o n f lo w
Ga s f lo w
3 - WAY
VA L V E
MU L T I P L E X E R
CA R RI ER G AS
(NI TR O G EN )
Experimental
Setup
Switching
Box
Mass Flow
Controller
Network
Analyzer
VOC in
bubbler
Nitrogen
VOC
PC
Sensor
Cell
EXPERIMENTAL
SETUP
Mass Flow
Controller
Network
Analyzer
Gas
Bubbler
Sensor
Cell
Mass Flow
Meter
Switching
Box
Post-It
notes
How Does Odor Signal
look Like?
Identification of VOCs
Preprocessing
Increasing Pattern
Separability
Neural Network
Training
Neural Network
Validation
VOC Identification
Raw Sensor Readings (6-D)
Filtering,Normalization,
De-trending,etc.
Fuzzy nose (FNOSE),
Feature range stretching,or
Nonlinear cluster transformation
Multilayer perceptron
LEARN++ (for incremental learning)
Classification
.
.
.
.
.