Senso s 2014, 14, 10479-10496; doi:10.3390/s140610479
senso s
ISSN 1424-8220
www.mdpi.com/jou nal/senso s
A icle
Sui abili y E alua ion o Mul ipoin Simul aneous CO2
Sampling Wi eless Senso s o Li es ock Buildings
Sal ado Cal e 2, José Ca los Campelo 1, Fe nando Es ellés 2, Angel Pe les 1,*,
Rica do Me cado 1 and Juan José Se ano 1
1 Ins i u e o he Applica ions o Ad anced In o ma ion and Communica ions Technologies (ITACA),
Uni e si a Poli ècnica de Valencia, Camino de Ve a, 46022 Valencia, Spain;
E-Mails: jcampelo@i aca.up .es (J.C.C.); me cado@i aca.up .es (R.M.); jse ano@i aca.up .es (J.J.S.)
2 Ins i u e o Animal Science and Technology, Uni e si a Poli ècnica de València, Camino de Ve a,
46022 Valencia, Spain; E-Mails: [email p o ec ed] (S.C.); [email p o ec ed] (F.E.)
* Au ho o whom co espondence should be add essed; E-Mail: [email p o ec ed];
Tel.: +34-963-877-007; Fax: +34-963-877-579.
Recei ed: 11 Ap il 2014; in e ised o m: 30 May 2014 / Accep ed: 3 June 2014 /
Published: 13 June 2014
Abs ac : The en i onmen in li es ock buildings mus be con olled o ensu e he heal h
and wel a e o bo h wo ke s and animals, as well as o es ic he emission o pollu an s o
he a mosphe e. Among he pollu an s gene a ed inside hese p emises, ca bon dioxide
(CO2) is o g ea in e es in e ms o animal wel a e and en ila ion con ol. The use o
inexpensi e senso s means ha comple e sys ems can be designed wi h a numbe o
senso s loca ed a ound he building. This pape desc ibes a s udy o he sui abili y o
mul ipoin simul aneous CO2 senso s ope a ing in a wi eless senso ne wo k, which was
ound o ope a e sa is ac o ily unde labo a o y condi ions and was ound o be he bes
al e na i e o hese applica ions. The senso s showed a highly linea esponse o CO2
concen a ions, anging om 500 o 5000 ppm. Howe e , indi idual senso esponse was
ound o di e , which made i necessa y o calib a e each one sepa a ely. Senso p ecision
anged be ween 80 and 110 ppm CO2, and senso esponse o egis e a 95% change in
concen a ion was es ima ed a a ound 5 min. These ea u es mean his ype o senso
ne wo k can be used o moni o animal wel a e and also o en i onmen al con ol in
poo ly en ila ed li es ock p emises. Acco ding o he es s conduc ed in his s udy, a
empo al d i may occu and he e o e a egula calib a ion o senso s would be needed.
OPEN ACCESS
Senso s 2014, 14 10480
Keywo ds: ca bon dioxide; li es ock; en ila ion; wi eless senso ne wo k
1. In oduc ion
In ensi e li es ock ea ing is associa ed wi h he emission o la ge amoun s o a mosphe ic
pollu an s, which impai animal heal h and wel a e inside he buildings and ha e a se ious
en i onmen al impac . The accumula ion o ammonia (NH3) inside li es ock houses a ec s animal
heal h and wel a e [1,2], and he emission o his gas is associa ed wi h he acidi ica ion o soils and
he eu ophica ion o ecosys ems [3]. G eenhouse gases such as me hane (CH4) and ni ous oxide
(N2O), which a e gene a ed bo h by he animals hemsel es and hei manu e, con ibu e o clima e
change [4]. Due o he accumula ion o hese e ec s, many egula ions ha e been issued in ecen
yea s (e.g., he Kyo o P o ocol, Eu opean Ceilings Di ec i e), which di ec ly o indi ec ly a ec he
design and ope a ion o li es ock buildings.
Ca bon dioxide (CO2) is p oduced in li es ock a ms mainly as a esul o animal espi a ion and
he ae obic decomposi ion o hei manu e. Al hough CO2 is a g eenhouse gas, animal- ela ed CO2
emissions a e no conside ed a ne addi ion o he a mosphe e, since his gas pa icipa es in a
sho - e m closed cycle [5]. Howe e , CO2 has been widely used as a ace gas o quan i y en ila ion
in li es ock buildings [6] and exposu e o high concen a ions o his gas has also been associa ed wi h
impai ed heal h o bo h humans and animals [7].
Ven ila ion o li es ock housing p o ides esh ai o he animals and emo es any po en ial
ai bo ne pollu an s eleased inside he building. Measu ing en ila ion a es in li es ock buildings is
he e o e essen ial o p ope building design and ope a ional con ol. Fu he mo e, en ila ion should
be quan i ied and he ai should be moni o ed wi h a iew o de e mining gas concen a ions in o de o
es ima e he emission o o he a mosphe ic pollu an s by means o mass balances. Cu en ly, mos
pollu an s can be measu ed wi h accep able p ecision using a wide a ie y o measu emen echniques [8],
bu quan i ying ai low a es may be mo e complica ed [9].
Measu ing en ila ion a es in comme cial li es ock buildings is complex and ime-consuming, and
is pa icula ly challenging in na u ally en ila ed li es ock buildings, whe e changing ou doo
condi ions de e mine i egula and con inuously a ying en ila ion pa e ns. Na u ally en ila ed
buildings a e commonly used o house ce ain animal species such as dai y ca le and a e widely used
o o he species unde mild o wa m condi ions. They a e also seen as a good solu ion in de eloping
coun ies and a eas wi h es ic ed access o elec ici y. Howe e , he measu emen and con ol o a
en ila ion sys em emains an elusi e a ge o a me s and esea che s [10]. As discussed by [9], CO2
can be used as a ace o de e mine a m en ila ion and pollu an emissions, bu mo e de elopmen is
needed ega ding CO2 p oduc ion inside he a m and also in e ms o CO2 measu emen echnology.
Fa me s a e inc easingly equi ed o imp o e he con ol o en ila ion in hei buildings. Wel a e
egula ions also include a en ila ion sys em which is able o main ain indoo gas concen a ions below
ce ain h esholds (e.g., 20 ppm NH3 and 3000 ppm CO2 o b oile s, acco ding o Eu opean Council
Di ec i e 2007/43/EC [11]. I is also now widely accep ed ha main aining app op ia e ai quali y in
Senso s 2014, 14 10481
animal housing is e ec i e, no only o ensu e accep able wel a e s a us, bu i also has a posi i e
in luence on animal pe o mance.
Measu ing CO2 concen a ions in li es ock buildings is he e o e o g ea in e es o a me s.
Fi s ly, his in o ma ion could be included in he a m clima e con ol sys em o con ol en ila ion as
equi ed. Sly, he accu a e de e mina ion o en ila ion a es by means o CO2 balances would allow
al e na i e housing sys ems o be cha ac e ized in e ms o pollu an emissions. In addi ion, such a
sys em would p o ide in o ma ion on animal wel a e ela ed o exposu e o gaseous pollu an s.
CO2 balance has been p oposed as al e na i e me hod o es ima ing en ila ion a es. The CO2
na u ally emi ed in li es ock houses ul ills he p ope ies o a ace gas es ablished by [12]: low and
s able backg ound concen a ions, low isk in case o exposu e, easy o handle and measu e. The CO2
balance me hodology has been widely desc ibed [6,13] and applied o a numbe o mechanically
en ila ed li es ock houses. To apply his me hod wo inpu s mus be cha ac e ized accu a ely. The
i s equi es ha he CO2 p oduced by he animals and hei manu e be well cha ac e ized; his
p oduc ion can be es ima ed acco ding o [6] and is p opo ional o he me abolic ac i i y o he
animals, which depends on he ype o animal, i s weigh and i s p oduc i i y. The s inpu is an
accu a e measu e o he di e ence in CO2 concen a ion be ween he exhaus and incoming ai . This
equi es a high ime and space esolu ion in concen a ion measu emen s.
The ai inside li es ock houses is cha ac e ized by ela i ely low CO2 concen a ions, which a e
mos ly unde 5000 ppm [14–17]. In some cases (na u ally en ila ed, open buildings) indoo CO2
concen a ions a e lowe han 500 ppm, which in ol es a di e ence o less han 200 ppm wi h ou doo
ai . In mos mechanically en ila ed buildings he incoming ai mixes pe ec ly wi h he ai al eady
p esen and CO2 concen a ion g adien s a e s able and well cha ac e ized. Howe e , in na u ally
en ila ed buildings (pa icula ly hose o open s uc u e), indoo and ou doo concen a ions a e
simila , pe ec mixing is no achie ed and pe manen changes in CO2 g adien s a e common as a
consequence o changes in ai low pa e ns. CO2 concen a ion measu emen s in li es ock buildings
mus mee he ollowing equi emen s:
 They should be su icien ly p ecise o he pu pose o he s udy. Fo assessing animal wel a e, a
p ecision o 100 ppm may be enough, conside ing ha a h eshold o 3000 o 5000 ppm is
no mally es ablished. Fo en ila ion, he p ecision should be enough o p ope ly cha ac e ize
he di e ence be ween he indoo and ou doo en i onmen s. In closed, na u ally en ila ed
buildings, a di e ence o 150 ppm should be de ec ed [13], bu mo e open s uc u es will
equi e much mo e p ecise measu emen [9].
 Simul aneous mul i-poin measu emen s o p o ide eliable in o ma ion on he spa ial
dis ibu ion o gas concen a ions.
 Adequa e ime esolu ion o de ec changes de e mined by a ying ou doo condi ions.
F om he li es ock p oduce ’s poin o iew, i would be ad isable o ins all a se o CO2 senso s a
di e en poin s in a building. This would allow an in-dep h s udy o CO2 concen a ion a hese poin s
and would moni o he ai low e ec . O cou se, o make his a iable app oach, he measu emen
sys em mus ha e a compe i i e p ice. Ou p oposal includes a wi eless senso ne wo k consis ing o a
se o CO2 senso s placed h oughou he building. In o de o selec he mos app op ia e CO2 senso
o his wo k, an in-dep h analysis o comme cial senso s mus be ca ied ou . The senso will hen be
Senso s 2014, 14 10482
connec ed o he bes cu en echnology o s o e and communica e he esul s: a wi eless senso
ne wo k (WSN) a chi ec u e.
The huge ad ances in echnologies such as e y la ge scale in eg a ed mic oelec omechanical
sys ems and wi eless communica ions ha e con ibu ed o he widesp ead use o dis ibu ed senso
sys ems [18]. The minia u iza ion o compu ing and sensing echnologies has enabled he de elopmen
o iny, low-powe ed, inexpensi e senso s, ac ua o s, and con olle s. Embedded compu ing sys ems
(i.e., sys ems ha ypically in e ac closely wi h he physical wo ld and a e designed o pe o m only a
limi ed numbe o dedica ed unc ions) con inue o ind applica ions in an inc easing numbe o a eas.
While de ense and ae ospace sys ems s ill domina e he ma ke , he e is an inc easing ocus on sys ems
ha moni o and p o ec ci il in as uc u es such as b idges and unnels, he na ional powe g id and
pipelines. Ne wo ks o hund eds o senso nodes a e al eady being used o moni o la ge geog aphic
a eas o modeling and o ecas ing en i onmen al pollu ion and looding, using ib a ion senso s o
collec s uc u al heal h in o ma ion on b idges, and con olling usage o wa e , e ilize s, and
pes icides o imp o e c op heal h and quan i y.
In his pape , we deal wi h wo pa icula objec i es. Fi s , he selec ion o an app op ia e CO2
senso o be used in wi eless senso ne wo ks, and sly, o e alua e his senso unde labo a o y
condi ions in o de o assess i s po en ial applica ion o p ac ical aspec s o li es ock a ming, such as
en ila ion con ol, wel a e assessmen o en i onmen al applica ions. The pape is s uc u ed as
ollows: Sec ion 2 desc ibes he selec ion p ocess o he CO2 senso and he wi eless CO2 senso
ne wo k is p esen ed in Sec ion 3, including he expe imen al se -up o e alua ing he senso
pe o mance. Sec ion 4 includes he esul s and discussion and, inally, ou conclusions a e p esen ed
in Sec ion 5.
2. Choice o CO2 Senso
The ype o ins alla ion in which he senso s will be deployed equi es he measu emen o
concen a ions o CO2 in ai o a ange be ween 0 and 5000 ppm wi h an accu acy o ±100 ppm. The
choice o senso is c i ical, since ou expe imen al se -up equi es a wi eless ne wo k o CO2 senso s.
This imposes he ollowing es ic ions:
 Ene gy-cons ained wi eless senso nodes. Bo h he CO2 senso s and wi eless nodes mus ha e low
ene gy equi emen s. The se -up can a y om one eco ding pe min e e y week o one e e y
10 min o i e mon hs, so ha he ba e y sys em mus be sized o las o a whole campaign.
 Small senso size o embedding in he physical node.
 No ecalib a ion while he expe imen is unning. E o speci ica ions mus be limi ed and
known o he ull expe imen , which can mean as long as i e mon hs o known measu emen
beha io . Measu emen accu acy mus be as high as possible wi hou he need o ecalib a ion
du ing he en i e expe imen [19].
 Long senso li e. Some CO2 senso s ha e a limi ed li e due o chemical con amina ion. This
deg ada ion in luences measu emen , implying ex a main enance.
 Low-cos . To ob ain simul aneous mul i-poin measu emen s a conside able numbe o nodes
a e needed, so senso p ice is impo an .
 Accu a e senso s. Known accu acy is necessa y o de ec di e ences in concen a ion.
Senso s 2014, 14 10483
Taking he abo e es ic ions in o conside a ion, he choice o he igh senso s was based on an
analysis o di e en CO2 measu emen echniques. Technologies o measu ing CO2 in he ai can be
g ouped in o wo ca ego ies: elec ochemical and spec al abso p ion.
Comme cially a ailable elec ochemical senso s we e uled ou , as hey ha e a limi ed li e and high
ene gy equi emen s due o he need o p ehea ing and s abilizing be o e use.
Spec al abso p ion-based non-dispe si e in a ed (NDIR) senso s seemed he mos sui able o he
scena ios o he s udy. NDIR senso s a e widely employed in a mosphe ic CO2 concen a ion
measu emen s since hey a e conside ed s able, ha e excellen du abili y and a e obus agains
in e e ence om o he ai componen s.
Low-cos NDIR [19] senso s a e now widely a ailable, howe e hei speci ica ions and long- e m
cha ac e is ics a e in gene al ague o ha e no been au hen ica ed.
Bea ing in mind he equi emen s o ou s udy, he ma ke o e was analyzed and he speci ica ions
compa ed. Table 1 summa izes a selec ion o models ha appa en ly sui ed ou equi emen s.
Table 1. Lis o non-dispe si e in a ed (NDIR) CO2 senso s conside ed.
Manu ac u e
Hanwei
SST
Sensing
Alphasens
e
Dynamen
E2V
Senseai
Model
MH-Z12
CO2S-A
IRC-A1
TDS0037
IR11BD
K30
Measu emen ange
(ppm)
0–5,000
0–5,000
0–5,000
a iable
a iable
0–10,000
Accu acy
±50 ppm
±50ppm
±3%
measu e
1%
2%
±100 ppm
±30ppm
±5%
measu e
Response ime (s)
<30 + 180
p ehea
30 o 180
<40
< 30 + 60
p ehea
< 20 + 30 o
1800 p ehea
20
Ope a ing ol age (V)
4–6
3.25–5
2–5
3–5
3–15
4.5–14
Size (mm) (Heigh ×
heigh × dep h /Ø
Diame e × heigh )
60 × 42 × 15
Ø 60 × 20
Ø 20 × 17
Ø 20 × 17
Ø 14 × 19
51 × 51 × 41
App oxima e p ice
(Eu os)
120
80
75
100
150
70
Since ene gy consump ion is an impo an es ic ion, we es ima ed he o al amoun o ene gy
equi ed o di e en con igu a ions. I should be no ed ha a measu emen equi es ini ial p ehea and
s able senso eadings, which mus be allowed o in he calcula ions.
Table 2 summa izes he esul s o di e en eading pe iods wi h a 1.5 ol 3000 mAh ba e y. Some
modules need a DC-DC con e e . The example in he able is o an ou pu ol age o 5 V and e iciency
o 80%. I should be emembe ed ha in gene al lowe ol ages p o ide be e ene gy e iciency.

Senso s 2014, 14 10484
Table 2. Es ima ed li e in days o 1.5 V, 3000 mAh ba e y, DC-DC 80% e iciency and 5 V senso s.
Measu e
in e al (min)
Model
Hanwei
MH-Z12
SST Sensing
CO2S-A
Alphasense
IRC-A1
Dynamen
TDS0037
E2V IR11BD
Senseai K30
Con .
1.2
3.6
1.2
0.9
2
1.8
5
1.7
7.2
5.1
3.0
15.0
6.0
10
3.4
14.4
10.3
6.0
30.0
12.0
15
5.1
21.6
15.4
9.0
45.0
18.0
20
6.9
28.8
20.6
12.0
60.0
24.0
25
8.6
36.0
25.7
15.0
75.0
30.0
30
10.3
43.2
30.9
18.0
90.0
36.0
120
41.1
172.8
123.4
72.0
360.0
144.0
360
123.4
518.4
370.3
216.0
1080.0
432.0
F om he ba e y li e poin o iew, he E2V IR11BD senso seems he bes op ion, bu has se ious
d awbacks, such as he lack o an ins umen a ion ampli ie ( equi ing i s implemen a ion in he node),
so his pa o he ene gy equi emen s is no e alua ed. Also, he lack o double channel ou pu means
i needs con inuous ecalib a ions. Based on conside a ions o ene gy, speci ica ions and inal senso
p ice, he CO2S-A model (SST Sensing L d., Coa b idge, UK) was chosen o he wi eless node
sys em. A iew o his senso is gi en in Figu e 1.
Figu e 1. View o he SST Sensing CO2S-A senso .
3. Wi eless CO2 Senso s Ne wo k
A wi eless senso ne wo k (WSN) consis s o spa ially dis ibu ed au onomous senso s o moni o
physical o en i onmen al condi ions, such as empe a u e, sound, p essu e, CO2, e c. and o pass he
da a acqui ed o a cen al pe sonal compu e (PC). As cu en ne wo ks a e bi-di ec ional, his means
senso ac i i y can be con olled. Wi eless senso ne wo ks a e now used in many indus ial and
consume applica ions, such as indus ial p ocess moni o ing, machine heal h moni o ing, and so on.
The WSN is buil o ―nodes‖—which can a y in numbe om a ew o se e al hund ed o e en
housands, in which each node is connec ed o one o se e al senso s. Each senso ne wo k node is
ypically di ided in o se e al pa s: a adio anscei e wi h an in e nal an enna o connec ion o an
ex e nal an enna, a mic ocon olle , an elec onic ci cui o in e acing wi h he senso s and an ene gy
sou ce, usually a ba e y o an embedded o m o ene gy ha es ing. A senso node can a y in size
Senso s 2014, 14 10485
om ha o a shoebox down o he size o a g ain o dus , al hough unc ioning ―mo es‖ o genuine
mic oscopic dimensions ha e ye o be c ea ed. The cos o senso nodes is simila ly a iable, anging
om a ew o hund eds o eu os, depending on he complexi y o he indi idual senso nodes. Size and
cos cons ain s on senso nodes esul in co esponding cons ain s on esou ces such as ene gy,
memo y, compu a ional speed and communica ions bandwid h. The opology o he WSNs can a y
om a simple s a ne wo k o an ad anced mul i-hop wi eless mesh ne wo k. The p opaga ion
echnique be ween he hops o he ne wo k can be ou ing o looding [18,20].
Figu e 2 shows a ypical scheme o a wi eless senso ne wo k. A se o nodes (S) ansmi s by a
wi eless p o ocol he in o ma ion ob ained h ough he ne wo k o he sink node. As can be seen in he
Figu e, some senso nodes can ac as epea e s, which allows low powe ansmi modes o be used o
ex end ba e y li e. The sink node is connec ed by Uni e sal Se ial Bus (USB) o E he ne o a
pe sonal compu e o s o e he in o ma ion ecei ed. Dis ance be ween sink nodes and nodes could be
(using app op ia e an ennas and wi eless echnology) hund eds o me e s.
Figu e 2. Wi eless senso ne wo k (WSN) scheme.
3.1. Sys em A chi ec u e
To demons a e he easibili y o using a wi eless senso ne wo k o sense CO2 le els, 12 wi eless
nodes we e implemen ed and ins alled in a con olled chambe (Figu e 3).
Figu e 3. Sys em deploymen in a chambe .
Senso s 2014, 14 10486
The chambe is equipped wi h a se o ans ha can ep oduce ai cu en s. As can be seen in Figu e 3,
he nodes we e ins alled in he o m o a ma ix o measu e he CO2 a di e en heigh s.
3.2. Node Cha ac e is ics
Figu e 4 shows wo iews o he wi eless CO2 node, which mus be ins alled in a e ical
a angemen o ensu e ha he CO2 senso is below he node box. I allows ai low h ough he senso
and is p o ec ed om dus .
Figu e 4. CO2 Node pho og aphs: (a) Ex e nal iew; and (b) In e nals.
(a)
(b)
Figu e 4a also shows h ee Ligh Emission Diodes (LED), wo pushbu ons and he powe on/o
bu on. LEDs and pushbu ons a e used basically o p og am he ime be ween measu es (sample and
ansmi equency o he sink node). This ime can be con igu ed by p essing one o he bu ons and
selec ing one o he ollowing alues: 15, 30, 60, 120, 300, 600, 1200 and 3600 s.
Figu e 4b shows he node p in ed ci cui boa d and he ba e y pack ha ensu es a measu emen
campaign o se e al weeks. Ba e y le els a e moni o ed and sen o he sink node o wa n use s i a
ba e y eplacemen is necessa y.
The node is based on he CC1110F32 mic ocon olle (Texas Ins umen s, Dallas, TX, USA) [21].
The CC1110F32, is a ue low-powe sub-1 GHz sys em-on-chip designed o low powe wi eless
applica ions. I combines he excellen pe o mance o he s a e-o - he-a RF anscei e CC1101 wi h
an indus y-s anda d enhanced 8051 MCU, up o 32 KB o in-sys em p og ammable lash memo y and
up o 4 KB o Random Access Memo y (RAM), and many o he powe ul ea u es. The adio
equency ange can be chosen om: 300–348 MHz, 391–464 MHz and 782–928 MHz. The Eu opean
Indus ial, Scien i ic and Medical (ISM) band 868 MHz equency and 500 kbps we e chosen o he
p esen implemen a ion. This ISM band allows longe anges han ypical sys ems using 2.4 GHz (up
o almos 3 imes highe o he same ansmi powe and sensi i i y ecep ion). Addi ionally, i s low
powe demands (16 mA o ansmission a 10 mW and 18 mA o ecep ion) make i sui able o
Senso s 2014, 14 10487
ba e y-powe ed sys ems. In ou es s, anges up o 290 m wi h an omnidi ec ional an enna we e
achie ed, which was in excess o he equi emen s in li es ock buildings.
The Sensi ion SHT11 [22] digi al humidi y and empe a u e senso was included o analyze he
in luence o empe a u e and humidi y on CO2 measu ing. This is he all- ound e sion o he e low
solde able humidi y senso se ies ha combines decen accu acy wi h a compe i i e p ice and is ully
calib a ed. The in o ma ion ansmi ed o he sink node was as ollows:
 CO2 concen a ion
 Ambien empe a u e
 Rela i e humidi y
 Ba e y le el
Fo he i s h ee i ems, he node so wa e a e aged he las en measu es be o e sending he esul
o he sink node.
3.3. Sink Node
The sink node (see Figu e 5) ecei es all he in o ma ion ansmi ed by he se o nodes and
communica es his da a by means o a USB connec ion o a pe sonal compu e ha s o es all he
in o ma ion ecei ed in a Comma Sepa a ed Values (CSV) o ma ile ha can be p ocessed by many
s anda d PC applica ions (da abase o sp eadshee so wa e).
The sink node ha dwa e is he same as ha o he o he nodes, wi hou he CO2 and empe a u e and
humidi y senso s. As i is connec ed by a USB po o he PC i does no need ba e ies. Al hough a
USB connec ion was chosen o his sys em, o he al e na i es such as E he ne connec ion o Gene al
Packe Radio Se ices (GPRS) can be implemen ed o send in o ma ion o he PC.
Figu e 5. Views o he sink node.
Figu e 6 shows he use - iendly applica ion used o s o e he acqui ed in o ma ion in a CSV da abase.
Senso s 2014, 14 10494
a iables, i mus be conside ed ha senso esponse is a ec ed by gas di usion h ough a memb ane
o he senso chambe , so ha he ac o s a ec ing memb ane pe meabili y (e.g., dus accumula ion)
may slow down he senso esponse. Acco ding o [26,27], in hese senso s he e is a po en ial c oss
in e e ence wi h dus due o ligh sca e ing and he e o e senso s a e placed in closed o di usion
cells p o ec ed by dus il e s. Fo his eason, dus accumula ion in hese il e s may educe di usion
o he measu emen cell, hus a ec ing esponse ime o hese NDIR senso s
5. Conclusions
This pape desc ibes a s udy on he use o a low cos CO2 senso implemen ed in a wi eless senso
ne wo k sys em ha showed an excellen h oughpu du ing measu emen campaigns. This sys em
combines mode a e cos wi h low ene gy equi emen s and can be powe ed by s anda d ba e ies,
which simpli ies i s ins alla ion. A se o nodes can be placed a di e en poin s h oughou a li es ock
building o acqui e eadings. The nodes a e also designed o make i possible o include o he CO2
senso s o addi ional senso s o de ec he p esence o o he gases (i.e., me hane).
The CO2 senso selec ed may be app op ia e o animal wel a e and en i onmen al applica ions in
li es ock p oduc ion, pa icula ly in animal houses which a e ela i ely open o he a mosphe e. The
senso s we e seen o espond linea ly o changes in CO2 concen a ions. Howe e , he di e ences
be ween indi idual senso s and a po en ial ime d i make i ad isable o calib a e be o e a
measu emen campaign. The senso ime esponse (abou 5 min) makes i app op ia e o mos
applica ions, excep hose equi ing a as esponse. Fo his eason hese senso s may no be
app op ia e o moni o sligh , sudden changes occu ing in open, na u ally en ila ed buildings.
Acknowledgmen s
This p ojec was suppo ed by he Vice ec o ado de In es igación o he UPV (P og ama de Apoyo
a la In es igación y Desa ollo, PAID-06-11 P og am, P ojec No. 2843) and he Spanish Go e nmen
unde P ojec s CTM2011-29691-C02-01 and TIN2011-28435-C03-0. The ansla ion o his pape was
unded by he Uni e si a Poli ècnica de València, Valencia, Spain.
Au ho Con ibu ions
Sal ado Cal e coo dina ed he calib a ion es s and w o e he co esponding pa o he
manusc ip . José Ca los Campelo and Angel Pe les selec ed he se o CO2 senso s and de eloped he
i mwa e o he wi eless senso ne wo k and w i e he ela ed pa s o his pape . Rica do Me cado
designed he ha dwa e subsys em. Fe nando Es ellés pe o med senso calib a ions. Juan José Se ano
designed he wi eless senso ne wo k.
Con lic s o In e es
The au ho s decla e no con lic o in e es .

Senso s 2014, 14 10495
Re e ences
1. Wa hes, C.M.; Jones, J.B.; K is ensen, H.H.; Jones, E.K.M.; Webs e , A.J.F. A e sion o pigs
and domes ic owl o a mosphe ic ammonia. T ans. ASAE 2002, 45, 1605–1610.
2. Homidan, A.A.; Robe son, J.F.; Pe chey, A.M. Re iew o he e ec o ammonia and dus
concen a ions on b oile pe o mance. Wo lds Poul . Sci. J. 2003, 59, 340–349.
3. K upa, S.V. E ec s o a mosphe ic ammonia (NH3) on e es ial ege a ion: A e iew.
En i on. Pollu . 2003, 124, 179–221.
4. IPCC Fou h Assessmen Repo o he In e go e nmen al Panel on Clima e Change;
In e go e nmen al Panel on Clima e Change (IPCC): Gene a, Swi ze land, 2007; pp. 1–74.
5. 2006 IPCC Guidelines o G eenhouse Gas In en o ies, P epa ed by he Na ional G eenhouse
Gas In en o ies P og amme; In e go e nmen al Panel on Clima e Change (IPCC): Gene a,
Swi ze land, 2006.
6. 4 h Repo o Wo king G oup on Clima iza ion o Animal Houses: Hea and Mois u e P oduc ion
a Animal and House Le els; Resea ch Cen e Bygholm, Danish Ins i u e o Ag icul u al Sciences:
Ho sens, Denma k, 2002.
7. Kaye, J.; Buchanan, J.; Kend ick, A.; Johnson, P.; Low y, C.; Bailey, J.; Nu , D.; Ligh man, D.
Acu e ca bon dioxide exposu e in heal hy adul s: E alua ion o a no el means o in es iga ing he
s ess esponse. J. Neu oendoc inol. 2004, 16, 256–264.
8. Ni, J.Q.; Hebe , A.J. Sampling and measu emen o ammonia a animal acili ies. Ad . Ag on.
2008, 98, 201–269.
9. Ogink, N.W.M.; Mosque a, J.; Cal e , S.; Zhang, G. Me hods o measu ing gas emissions om
na u ally en ila ed li es ock buildings: De elopmen s o e he las decade and pe spec i es o
imp o emen . Biosys . Eng. 2013, 116, 297–308.
10. Cal e , S.; Ga es, R.S.; Zhang, G.; Es ellés, F.; Ogink, N.W.M.; Pede sen, S.; Be ckmans, D.
Measu ing gas emissions om li es ock buildings: A e iew on unce ain y analysis and e o
sou ces. Biosys . Eng. 2013, 116, 221–231.
11. Council di ec i e 2007/43/EC laying down minimum ules o he p o ec ion o chickens kep o
mea p oduc ion. O . J. Eu . Union 2007, L182, 19–28.
12. Phillips, V.R.; Holden, M.R.; Snea h, R.W.; Sho , J.L.; Whi e, R.P.; Ha ung, J.; Seedo , J.;
Sch ode , M.; Linke , K.H.; Pede sen, S.; e al. The de elopmen o obus me hods o measu ing
concen a ions and emission a es o gaseous and pa icula e ai pollu an s in li es ock buildings.
J. Ag ic. Eng. Res. 1998, 70, 11–24.
13. Van Ouwe ke k, E.N.J.; Pede sen, S. Applica ion o he ca bon dioxide mass balance me hod o
e alua e en ila ion a es in li es ock buildings. In P oceedings o he XII Wo ld Cong ess on
Ag icul u al Enginee ing, Milan, I aly, 29 Augus –1 Sep embe 1994; pp. 516–529.
14. Cal e , S.; Camb a-Lopez, M.; Es elles, F.; To es, A.G. Cha ac e iza ion o gas emissions om a
Medi e anean b oile a m. Poul . Sci. 2011, 90, 534–542.
15. Cal e , S.; Camb a-Lopez, M.; Es elles, F.; To es, A.G. Cha ac e iza ion o he indoo
en i onmen and gas emissions in abbi a ms. Wo ld Rabbi Sci. 2011, 19, 49–61.
Senso s 2014, 14 10496
16. Wa hes, C.M.; Holden, M.R.; Snea h, R.W.; Whi e, R.P.; Phillips, V.R. Concen a ions and
emission a es o ae ial ammonia, ni ous oxide, me hane, ca bon dioxide, dus and endo oxin in
UK b oile and laye houses. B . Poul . Sci. 1997, 38, 14–28.
17. Pede sen, S.; Takai, H.; Johnsen, J.O.; Me z, J.H.M.; G oo Koe kamp, P.W.G.; Uenk, G.H.;
Phillips, V.R.; Holden, M.R.; Snea h, R.W.; Sho , J.L.; e al. A compa ison o h ee balance
me hods o calcula ing en ila ion a es in li es ock buildings. J. Ag ic. Eng. Res. 1998, 70, 25–37.
18. Da gie, W.; Poellabaue , C. Fundamen als o Wi eless Senso Ne wo ks: Theo y and P ac ice;
John Wiley and Sons: New Yo k, NY, USA, 2010.
19. Yasuda, T.; Yonemu a, S.; Tani, A. Compa ison o he cha ac e is ics o small comme cial NDIR
CO2 senso models and de elopmen o a po able CO2 measu emen de ice. Senso s 2012, 12,
3641–3655.
20. Soh aby, K.; Minoli, D.; Zna i, T. Wi eless Senso Ne wo ks: Technology, P o ocols, and
Applica ions; John Wiley and Sons: New Yo k, NY, USA, 2007.
21. Texas Ins umen s. CC1110F32. Low-Powe SoC (Sys em-on-Chip) wi h MCU, Memo y, Sub-1
GHz RF T anscei e , and USB Con olle . A ailable online: www. i.com/p oduc /cc1110 32
(accessed on 9 June 2014).
22. Sensi ion. SHT11: Digi al Humidi y Senso (RH&T). A ailable online: h p://www.sensi ion.com/
en/p oduc s/humidi y- empe a u e/humidi y-senso -sh 11/ (accessed on 9 June 2014).
23. S a is ics—Vocabula y and symbols—Pa 2: Applied S a is ics; ISO 3534-2:2006; In e na ional
O ganiza ion o S anda diza ion (ISO): Gene a, Swi ze land, 2006.
24. Bje g, B.; Zhang, G.; Madsen, J.; Rom, H.B. Me hane emission om na u ally en ila ed li es ock
buildings can be de e mined om gas concen a ion measu emen s. En i on. Moni . Assess. 2012,
184, 5989–6000.
25. Kaasik, A.; Maasikme s, M. Concen a ions o ai bo ne pa icula e ma e , ammonia and ca bon
dioxide in la ge scale uninsula ed loose housing cowsheds in Es onia. Biosys . Eng. 2013, 114,
223–231.
26. Hodgkinsona, J.; Smi h, R; On Ho, W.; Sa ell, J.R.; Ta ama, R.P. Non-dispe si e in a- ed
(NDIR) measu emen o ca bon dioxide a 4.2 µm in a compac and op ically e icien senso .
Sens. Ac ua o s B 2013, 186, 580–588.
27. Zosel, J.; Oelßne , W.; Decke , M.; Ge lach, G.; Gu h, U. The measu emen o dissol ed
and gaseous ca bon dioxide concen a ion. Meas. Sci. Technol. 2011, 22, doi:10.1088/
0957-0233/22/7/072001.
© 2014 by he au ho s; licensee MDPI, Basel, Swi ze land. This a icle is an open access a icle
dis ibu ed unde he e ms and condi ions o he C ea i e Commons A ibu ion license
(h p://c ea i ecommons.o g/licenses/by/3.0/).