scieee Science in your language
[en] (orig)

Anonymous Attribute-Based Credentials in Collaborative Indoor Positioning Systems

Abstract

Collaborative Indoor Positioning Systems have recently received considerable attention, mainly because they address some of the existing limitations of traditional Indoor Positioning System. In Collaborative Indoor Positioning Systems, Bluetooth Low Energy can be used to exchange positioning data and provide information (the Received Signal Strength Indicator) to establish the relative distance between the actors. The collaborative models exploit the position of actors and the relative position among them to allow positioning to external actors or improve the accuracy of the existing actors. However, the traditional protocols (e.g. iBeacon) are not yet ready for providing sufficient privacy protection. Therefore, this paper deals with privacy-enhancing technologies and their application in Collaborative Indoor Positioning System. In particular, we focus on cryptographic schemes which allow the verification of users without their identification, so-called Anonymous Attribute-Based Credentials schemes. As the main contribution, we present a cryptographic scheme that allows security and privacy-friendly sharing of location information sent through Bluetooth Low Energy advertising packets. In order to demonstrate the practicality of our scheme, we also present the results from our implementation and benchmarks on different devices.

Read accessible full text

Anonymous Attribute-Based Credentials in Collaborative Indoor Positioning Systems

Author: Casanova-Marqués, Raúl; Pascacio, Pavel; Hajný, Jan; Torres-Sospedra, Joaquín
Publisher: SciTePress
Year: 2021
DOI: 10.5220/0010582507910797
Source: https://dspace.vut.cz/bitstreams/635684ef-3a79-4a37-89cc-82af075ab086/download
Anonymous A ibu e-based C eden ials in Collabo a i e Indoo
Posi ioning Sys ems
Ra´
ul Casano a-Ma qu´
es1,2 a, Pa el Pascacio2,3 b, Jan Hajny1 c and Joaqu´
ın To es-Sosped a2,4 d
1Depa men o Telecommunica ions, B no Uni e si y o Technology, B no, Czech Republic
2Ins i u e o New Imaging Technologies, Uni e si a Jaume I, Cas ell´
on, Spain
3Elec ical Enginee ing Uni , Tampe e Uni e si y, Tampe e, Finland
4UBIK Geospa ial Solu ions S.L., Cas ell´
on, Spain
Keywo ds: A ibu e-based C eden ials, Collabo a i e Indoo Posi ioning Sys ems, P i acy, Anonymi y, Blue oo h Low
Ene gy, Wea ables.
Abs ac : Collabo a i e Indoo Posi ioning Sys ems (CIPSs) ha e ecen ly ecei ed conside able a en ion, mainly be-
cause hey add ess some exis ing limi a ions o adi ional Indoo Posi ioning Sys ems (IPSs). In CIPSs, Blue-
oo h Low Ene gy (BLE) can be used o exchange posi ioning da a and p o ide in o ma ion ( he Recei ed
Signal S eng h Indica o (RSSI)) o es ablish he ela i e dis ance be ween he ac o s. The collabo a i e mod-
els exploi he posi ion o ac o s and he ela i e posi ion among hem o allow posi ioning o ex e nal ac o s
o imp o e he accu acy o he exis ing ac o s. Howe e , he adi ional p o ocols (e.g., iBeacon) a e no ye
eady o p o iding su icien p i acy p o ec ion. This pape deals wi h p i acy-enhancing echnologies and
hei applica ion in CIPS. In pa icula , we ocus on c yp og aphic schemes which allow he e i ica ion o
use s wi hou hei iden i ica ion, so-called Anonymous A ibu e-based C eden ial (ABC) schemes. As he
main con ibu ion, we p esen a c yp og aphic scheme ha allows secu i y and p i acy- iendly sha ing o
loca ion in o ma ion sen h ough BLE ad e ising packe s. In o de o demons a e he p ac icali y o ou
scheme, we also p esen he esul s om ou implemen a ion and benchma ks on di e en de ices.
1 INTRODUCTION
Wi-Fi and BLE inge p in ing a e popula echniques
o p o iding indoo posi ioning. They ely on he
signal s eng h om mul iple emi e s (ei he Wi-Fi
Access Poin s (APs) o BLE beacons). Howe e , he
Recei ed Signal S eng h (RSS) da a used o es ima e
he posi ions ha e a s ong noise componen , he e-
o e ha ing a gene al posi ioning accu acy a me e
le el (i.e., a ew me e s) (Mau z, 2012). In addi ion,
he deploymen complexi y, as well as elying on an
in as uc u e o beacons and se e s, ha e become
a se ious p oblem o la ge o ganiza ions all a ound
he wo ld. Wea able-based CIPSs a e inc easingly be-
coming popula o e he las ew yea s as a solu ion
o add ess hose p oblems in indoo posi ioning (Pas-
cacio e al., 2021). The CIPSs use he loca ion o a
ah ps://o cid.o g/0000-0001-6653-867X
bh ps://o cid.o g/0000-0001-9206-7496
ch ps://o cid.o g/0000-0003-2831-1073
dh ps://o cid.o g/0000-0003-4338-4334
se o ac o s and hei ela i e dis ances o 1) p o ide
localiza ion o ex e nal use s and/o 2) imp o e he
posi ioning accu acy o he cu en use s ha a e col-
labo a ing (Khandke e al., 2019).
Wea able de ices a e equipped wi h buil -in ech-
nologies such as Wi-Fi and Blue oo h, which a e ca-
pable o es ima ing use s’ posi ions and ubiqui y in
indoo en i onmen s. In pa icula , CIPSs app oaches
based on BLE a e one o he mos popula solu ions.
Sma phones can send BLE ad e isemen packe s
(e.g., iBeacon), which can be used la e o compu e
he ela i e dis ances be ween hem wi h he ga he ed
RSS alues. This al e na i e p esen s a s aigh o -
wa d deploymen and a good ade-o be ween powe
consump ion and pe o mance (Yang e al., 2020).
Howe e , high in e -connec i i y o unknown de ices
and he lack o s ong communica ion p o ocols in
e ms o secu i y and p i acy ep esen a weakness.
Malicious ac o s could pe o m a acks such as spoo -
ing o ea esd opping in unsecu ed collabo a i e sys-
ems (Chebli e al., 2019). These issues encou aged
he de elopmen o mo e p i acy- iendly solu ions.
Casano a-Ma qués, R., Pascacio, P., Hajny, J. and To es-Sosped a, J.
Anonymous A ibu e-based C eden ials in Collabo a i e Indoo Posi ioning Sys ems.
DOI: 10.5220/0010582507910797
In P oceedings o he 18 h In e na ional Con e ence on Secu i y and C yp og aphy (SECRYPT 2021), pages 791-797
ISBN: 978-989-758-524-1
Copy igh c
2021 by SCITEPRESS – Science and Technology Publica ions, Lda. All igh s ese ed
791
Exis ing CIPS a e based on cen alized schemes.
A se e -based solu ion acili a es mo e complex a -
acks such as use aceabili y o linkabili y o com-
munica ions. Thus, i is necessa y o mo e o a decen-
alized app oach. Building a decen alized CIPS us-
ing BLE echnology ep esen s a majo challenge due
o he educed space a ailable in BLE packe s, which
depends on he Blue oo h e sion (Blue oo h SIG,
2019). This hinde s he use o s anda d algo i hms
such as RSA, DH o AES o in o ma ion enc yp-
ion and use au hen ica ion. A ibu e-based au hen-
ica ion schemes allow use s o anonymously and se-
lec i ely p o e possession o hei pe sonal a ibu es
such as age, ci izenship, nega i e SARS-CoV-2 es
o loca ion da a. Ellip ic cu e c yp og aphy and he
educed size o i s keys can ackle he a o emen ioned
p oblems and p o ide anonymi y, unlinkabili y, un-
aceabili y and selec i e disclosu e o a ibu es.
In ligh o he iden i ied p oblems, in his wo k
we pu sue he decen aliza ion o collabo a i e indoo
posi ioning sys ems based on BLE beacons, while
p o iding, h ough ellip ic cu e c yp og aphy and
anonymous a ibu e-based au hen ica ion schemes, a
secu e and p i acy- iendly sys em o exchanging lo-
ca ion da a.
1.1 Rela ed Wo k
In he li e a u e he e a e se e al a icles ha add ess
anonymi y, secu i y, and p i acy issues o IPSs om
di e en pe spec i es (e.g., Wi eless Senso Ne wo k
(WSN) o In e ne o Things (IoT)). Howe e , only
e y ew o hem ocus on collabo a i e posi ioning
and/o BLE beacon app oaches. One o he mos in-
e es ing pape s is p oposed by (Kang e al., 2018),
which is based on he open-sou ce Eddys one BLE
o ma p oposed by (Hassidim e al., 2016). Eddy-
s one BLE o ma in oduces a scheme o mi iga e
acking and secu i y h ea s. The co e o he scheme
is cloud-based Epheme al Iden i ie s (EID), which en-
su es only au ho ized membe s o p ope ly iden i y
he BLE s anda d beacons.
Up o now, a oo li le a en ion has been paid
o CIPSs. (Zidek e al., 2018) de eloped a scheme
named Bell ock which combines an ecosys em o
s anda d beacons and use -based beacons. Bell-
ock p o ides access con ol o s anda d beacons
and anonymi y o use -based beacons by using h ee
echniques o gene a e pseudo-anonymous iden i ie s
( andom, synch onized, and enc yp ed) ha can be
unmasked by a se e .
(Des ia i e al., 2017) p oposed he Da a T ans-
mission Scheme o Indoo Mobile Coope a i e Lo-
caliza ion Sys em ha wo ks o e Wi-Fi. The scheme
uses AES-128 o enc yp he message, MD5 o con-
s uc he HMAC and RSA-2048 o enc yp bo h he
AES key and he HMAC. Thei app oach migh no
be well sui ed as he sys em emains ulne able o
man-in- he-middle a acks du ing he RSA key-pai
exchange, leading o a o al loss o con iden iali y and
message in eg i y.
A key p oblem wi h he a o emen ioned p opos-
als lies in se e -dependency, u ning he sys em in o
a cen alized en i onmen . This me hod could su e
om a ple ho a o pi alls such as denial o se ice
(DoS), spoo ing, e c.
Some p elimina y wo k on ABCs was ca ied
ou se e al decades ago by (Chaum, 1985), (B ands,
2000), and (Camenisch and Lysyanskaya, 2001). In
subsequen yea s, much mo e in o ma ion on ABCs
has become a ailable abou p ac ical applica ions
(A aoui e al., 2015), e oca ion p o ocols (Ca-
menisch e al., 2010) and he use o mo e e icien
algeb aic s uc u es (Chase e al., 2014), (Ba ki e al.,
2016), and (Ringe s e al., 2017). Due o he high
compu a ional complexi y equi ed by hese schemes,
p ac ical implemen a ion on cons ained de ices is
s ill a challenge, esea che s need o ind a scheme
ha sa is ies hei speci ic needs. As a as we know,
no s udies ha e been published on ABC schemes ap-
plied o CIPSs.
1.2 Ou Con ibu ion
In o de o ill he gap ound in he li e a u e, we p o-
pose a new decen alized p i acy-p ese ing use au-
hen ica ion mechanism o CIPSs wi h ABC. Ou
solu ion combines ze o-knowledge p oo s wi h a e-
oca ion scheme based on li e ime and p o ides
anonymi y, unlinkabili y and un aceabili y.
The main con ibu ions a e he ollowing:
• To he bes o ou knowledge, we p esen he i s
decen alized secu i y scheme o CIPSs based on
BLE ad e ising.
• We de ine a scheme ha p o ide anonymous lo-
ca ion da a sha ing, decen alized au hen ica ion
and o line e oca ion.
• We p o ide a secu i y analysis on he p oposed
solu ion.
• We desc ibe he implemen a ion esul s.
The es o he pape is o ganized as ollows. Sec-
ion 2 gi es a b ie o e iew o he sys em model.
Sec ion 3 ho oughly de ails he c yp og aphic design
o he scheme. Sec ion 4 analyzes he secu i y. Sec-
ion 5 p esen s he implemen a ion esul s. Finally,
Sec ion 6 epo s some conclusions.
SECRYPT 2021 - 18 h In e na ional Con e ence on Secu i y and C yp og aphy
792
2 SYSTEM OVERVIEW
The sys em model is depic ed in Figu e 1 and consis s
o he ollowing en i ies:
•Issue (I): is esponsible o issuing he pe sonal
a ibu e mID o a use using he Issue algo i hm.
The mID a ibu e ep esen s he use iden i ie
ob ained du ing egis a ion in he sys em. The
Issue algo i hm p o ides he use wi h c yp o-
g aphically signed c eden ials ha a e alid wi hin
he sys em. Access c eden ials consis o he al-
ues σ,σx0,σx ,σxID .
•Re oca ion Au ho i y (RA): emo es in alid
use s. Access con ol is achie ed h ough he e-
oca ion c eden ial σx , issued om he e oca-
ion a ibu e m . A emp ing o au hen ica e wi h
he e oked c eden ials will esul in ailu e. To
educe he complexi y o he sys em, we will con-
side om now on he e oca ion au ho i y and he
issue as he same en i y (I).
•Use (U): en olls in he sys em o use he col-
labo a i e indoo posi ioning sys em. E e y use
holds unique c eden ials gene a ed by he issue
and he e oca ion en i y (σ,σx0,σx ,σxID ), hen,
he use can anonymously p o e hei possession
o he e i ie using he Show algo i hm.
•Ve i ie (V): e i ies he possession o he use a -
ibu e using he Ve i y algo i hm. I he a ibu e
owne ship is e i ied success ully, he e i ie will
be able o use he loca ion in o ma ion o posi ion
hemsel . O he wise, he in o ma ion is ejec ed.
Issue Re oca ion Au ho i y
Show
Ve i y
IssueIssue
Use Ve i ie
Figu e 1: Sys em s uc u e.
Ini ially, all de ices egis e ed in he sys em beha e as
Use s. When one o hem needs o imp o e i s posi-
ion, i empo a ily assumes he ole o Ve i ie . This
ole can be used o ensu e ha he in o ma ion is p o-
ided by a alid use .
The e a e wo modes o ope a ion o he Use and
Ve i ie oles: he ac i e mode and he passi e mode.
Use ope a ion modes: 1) Ac i e mode: esponds
o e i ie eques s and assis s hem wi h posi ioning
by ac i ely pa icipa ing in he sys em. 2) Passi e
mode: emains unno iced in he sys em and does no
espond o e i ie eques s. In his mode, he use
does no in e ac wi h he sys em.
Ve i ie ope a ion modes: 1) Ac i e mode: ac i ely
eques s loca ion da a om nea by use s o imp o e
i s posi ion. 2) Passi e mode: emains hidden in he
sys em and wai s o o he e i ie s o eques loca ion
da a. In his mode, posi ioning is no possible i he e
a e no ac i e e i ie s nea by.
Figu e 2 depic s each o hese ou modes. Use
A ac s in ac i e mode, use B ac s in passi e mode,
e i ie C ac s in ac i e mode, and e i ie D ac s in
passi e mode.
º
Ve i ie C
Use B
Ve i ie D
Use A
Ve i ie eques
Use esponse
Figu e 2: Sys em oles and ope a ion modes.
These modes o ope a ion a e use ul o imp o ing
p i acy in en i onmen s whe e he e a e ew de ices,
and i may be easy o de e mine hei iden i y. In his
case, he use o e i ie can go in o passi e mode and
disappea om he sys em. We will no go in o u he
de ails because i is beyond he scope o his pape .
3 CRYPTOGRAPHIC DESIGN
This sec ion desc ibes he p o ocol o use au hen i-
ca ion in a collabo a i e indoo posi ioning sys em.
3.1 P elimina ies
The symbol Hdeno es a secu e hash unc ion. We
w i e a$
←Awhen ais sampled uni o mly a andom
om A. Le edeno e a bilinea map G1×G2→GT.
3.2 P o ocol Speci ica ion
Fi s , he use mus be egis e ed in he sys em by
sending hei pe sonal in o ma ion o he issue . The
issue gene a es he c eden ials and sends hem o he
use . The use can now legi ima ely pa icipa e in he
sys em. This algo i hm will be deno ed as Issue.
Anonymous A ibu e-based C eden ials in Collabo a i e Indoo Posi ioning Sys ems
793
When he e i ie wan s o imp o e hei posi ion,
a BLE ad e isemen packe wi h a imes amp will be
b oadcas . The nea by use s will compu e he p oo
o knowledge and ansmi i ia ano he BLE packe .
This algo i hm will be deno ed as Show.
Finally, he e i ie will ensu e h ough he p oo
o knowledge ha he use has alid c eden ials and
will use he loca ion in o ma ion o posi ion hemsel .
This algo i hm will be deno ed as Ve i y.
3.2.1 Issue
To egis e he use in he sys em and gene a e he
alid c eden ials, he use and he issue pe o m he
algo i hm shown in Figu e 3.
Use Issue
q,G1,G2,GT,e,g1,g2←G(1τ),pkI
(mID)skI= (x0,x ,xID),pkI=gx0
2
m =ww/yyyy
mID
−−−−−−−−−−−−−−−−−−−−→
σ=g
1
x0+m x +mIDxID
1σx0←σx0
σx ←σx
σxID ←σxID
σ,σx0,σx ,σxID
←−−−−−−−−−−−−−−−−−−−−
S o e: σ,σx0,σx ,σxID
Figu e 3: De ini ion o he Issue algo i hm.
• The use U pe o ms he ollowing s eps:
1) In oduces pe sonal in o ma ion in he app.
2) Es ablishes a secu e communica ion channel
wi h I and sends mID.
• The issue I pe o ms he ollowing s eps:
1) Gene a es he use ’s p i a e keys x and xID.
2) Signs he a ibu es m and mID wi h he is-
sue ’s sec e key as σ=g
1
x0+m x +mIDxID
1; and
compu es auxilia y alues σx0,σx ,σxID , whe e
σx0←σx0,σx ←σx , and σxID ←σxID .
3) Sends σand auxilia y alues σx0,σx ,σxID o
he use U.
3.2.2 Show and Ve i y
To b oadcas he loca ion in o ma ion and he p oo
o knowledge, and subsequen ly e i y ha i comes
om a legi ima e use , he use and he e i ie will
un he algo i hms shown in Figu e 4.
• The e i ie V pe o ms he ollowing s eps:
1) Gene a es a nonce based on he cu en ime.
2) C ea es a BLE ad e ising packe wi h he
nonce and b oadcas s i .
Use Ve i ie
q,G1,G2,GT,e,g1,g2←G(1τ),pkI
λ= (al i ude,longi ude,la i ude,...)m =ww/yyyy
imes amp
←−−−−−−−−−−−−−−−−−−
ρ,ρκ,ρ ,ρID
$
←Z∗
q
ˆ
σ←σρ,ˆ
σx0←σρ
x0
ˆ
σx ←σρ
x ,ˆ
σxID ←σρ
xID
κ←gρκ
1σρIDρ
xID
e←H( κ,ˆ
σ,ˆ
σx0,ˆ
σx ,ˆ
σxID , imes amp,λ)
sκ←ρκ+eρ
sID ←ρID −emID
ψ←(ˆ
σ,ˆ
σx0,ˆ
σx ,ˆ
σxID )
π←(e,sκ,sID)ψ,π,λ
−−−−−−−−−−−−−−−−−→
0
κ←gsκ
1ˆ
σ−e
x0ˆ
σ−em
x ˆ
σsID
xID
e?
=H( 0
κ,ˆ
σ,ˆ
σx0,ˆ
σx ,ˆ
σxID , imes amp,λ)
e(ˆ
σx0,g2) = e(ˆ
σ,pkI)
Figu e 4: De ini ion o he Show and Ve i y algo i hms.
• Nea by use s uipe o m he ollowing s eps a e
hey ha e ecei ed he imes amp:
1) Checks ha he imes amp gene a ed as a nonce
by he e i ie is less han 2 seconds om he
cu en ime o a oid eplay a acks.
2) Gene a es he andomize s ρ,ρκ,ρ ,ρID.
3) Compu es ˆ
σ,ˆ
σx0,ˆ
σx ,ˆ
σxID , whe e ˆ
σ←σρ,
ˆ
σx0←σρ
x0,ˆ
σx ←σρ
x , and ˆ
σxID ←σρ
xID .
4) Compu es he a ibu e p oo o knowledge κ
as κ←gρκ
1σρIDρ
xID .
5) Calcula es he c yp og aphic hash eas
H( κ,ˆ
σ,ˆ
σx0,ˆ
σx ,ˆ
σxID , imes amp,λ)and he
auxilia y p oo o knowledge alues sκ,sID,
whe e sκ←ρκ+eρand sID ←ρID −emID.
6) C ea es a BLE ad e ising packe wi h he lo-
ca ion in o ma ion λand he p oo o knowl-
edge ψ←(ˆ
σ,ˆ
σx0,ˆ
σx ,ˆ
σxID ),π←(e,sκ,sID),
and ansmi s i by b oadcas .
• Nea by e i ie s iexecu e he ollowing s eps a -
e ecei ing ψ,π, and λ.
1) Recompu es he a ibu e p oo o knowledge 0
κ
as 0
κ←gsκ
1ˆ
σ−e
x0ˆ
σ−em
x ˆ
σsID
xID .
2) Calcula es he c yp og aphic hash using 0
κ
as H( 0
κ,ˆ
σ,ˆ
σx0,ˆ
σx ,ˆ
σxID , imes amp,λ)and
checks i i ma ches he alue o e.
3) Compu es he bilinea pai ing e(ˆ
σx0,g2) =
e(ˆ
σ,pkI) o e i y he c eden ials we e emi ed
by he issue .
3.3 Blue oo h In eg a ion
The p oposed scheme ansmi s in o ma ion using
BLE ad e ising packe s. Blue oo h 4.2 and ea lie
SECRYPT 2021 - 18 h In e na ional Con e ence on Secu i y and C yp og aphy
794
de ined a payload size o 31 by es o a single BLE
ad e isemen packe . Howe e , since i is imp ac i-
cal o pack bo h he posi ioning da a and he p oo o
knowledge in o 31 by es, we ha e conside ed e sion
5.0 o he BLE p o ocol. Blue oo h 5.0 in oduced
a signi ican imp o emen by inc easing he capaci y
o ad e ising packe s. Wi h Low-Ene gy Ad e ising
Ex ensions, he ad e isemen payload can hold up o
254 by es (Blue oo h SIG, 2019).
To build he s uc u e o he BLE ad e ising
packe , we ha e assumed he s uc u e depic ed in
Figu e 5.
01 2 345 6 78910 11 12 13 14 15
UUID
ψ←(ˆ
σ,ˆ
σx0,ˆ
σx ,ˆ
σxID )
120B
16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
ψ
12B
π←(e,sκ,sID)
92B
λ
22B
Figu e 5: De ini ion o he BLE ad e ising payload.
The UUID ield occupies 8 by es, he p oo o knowl-
edge (ψ,π) consis s o 224 by es, and he e is a ee
space o 22 by es o he posi ioning da a (λ). This
pa is explained in de ail in Sec ion 5.
4 SECURITY ANALYSIS
The p o ocol is composed o he ollowing pa s:
he anonymous c eden ials scheme, he e oca ion
scheme and he posi ioning in o ma ion. In his sec-
ion we p o ide a b ie in o mal secu i y analysis o
each o hem.
4.1 Anonymous C eden ials
Following he secu i y model de ined by (Chase e al.,
2014) o anonymous c eden ial schemes, i is e-
qui ed o mee he ollowing p ope ies:
•Co ec ness: hones use s can p oduce an accep -
ing p oo .
•Un o geabili y: dishones use s canno p oduce an
accep ing p oo .
•Anonymi y: use s canno be iden i ied based on
he alues o a ibu es.
•Key-Pa ame e Consis ency: e e y public key has
a dis inc p i a e key.
Based on he andom O acle model in (Camenisch
e al., 2019), he c eden ial scheme used in ou sys em
has p o en o be secu e unde he n-SCDHI assump-
ion (de ined below).
De ini ion 1 (n-S ong Compu a ional Di ie-Hell-
man In e sion P oblem (SCDHI)).Le Obb(·)on in-
pu (m1,...,mn)∈Z∗
q
nadd (m1,...,mn) o Q and ou -
pu g1/(x0+∑n
i=1ximi). Le Odhi(·)on inpu h ou pu hxi.
De ine he ad an age o Aas ollows.
Ad n−SCDHI(A) = P h(G,g,q)←G oupSe up(1κ),
(x0,...,xn)$
←Z∗
q
n+1,(y,m∗
1,...,m∗
n)←
AObb(·),Odh0(·),...,Odhn(·)(g):y=g
1
x0+∑n
i=1xim∗
i∧
(m∗
1,...,m∗
n/∈Qi.(1)
SCDHI is ( ,ε)-ha d i no - ime ad e sa y has ad-
an age a leas ε.
4.2 Re oca ion Scheme
Following he secu i y model de ined by (Camenisch
e al., 2016), i is equi ed o mee he ollowing p op-
e ies:
•Re oca ion Comple eness: un e oked use s will
pass he e oca ion check.
•Re oca ion Soundness: e oked use s will no
pass he e oca ion check.
•Re oca ion P i acy: use s will no lose p i acy by
p o ing an un e oked e oca ion handle.
The e oca ion scheme is based on he epoch’s ex-
pi a ion. In ou solu ion, we se he epoch o one
week and de ine he e oca ion a ibu e as ollows:
m ←ww/yyyy. When he c eden ials expi e, he use
mus enew hem, allowing weekly access con ol.
We assume ha all use s in he sys em ha e he
da e co ec ly se on hei de ices. When a use em-
po a ily swi ches o he e i ie ole, hey can p oduce
he e oca ion a ibu e wi hou equi ing he use o
disclose i .
4.3 Loca ion In o ma ion
The posi ioning in o ma ion is he c i ical componen
o he sys em. We ou line he esis ance o he p o-
ocol agains some o he mos common a acks ha
a ec he in eg i y and con iden iali y.
Ea esd opping A ack: A audulen use could
moni o he communica ions and a emp o de-
anonymize a legi ima e use because he in o ma ion
is sen in plain ex . Cap u ing communica ions is
possible since he in o ma ion is eely ansmi ed.
Howe e , emo ing anonymi y and a emp ing o link
a speci ic use o posi ioning da a is ex emely di i-
cul . Each ime a use ansmi s he posi ion, he c e-
den ials a e andomized.
Anonymous A ibu e-based C eden ials in Collabo a i e Indoo Posi ioning Sys ems
795

Replay A ack: A malicious use could ca y ou
his a ack because he in o ma ion is no al e ed.
Bu a e wo seconds, he e i ie ansmi s a new
imes amp, in alida ing he packe ’s au hen ica ion.
Du ing e i ica ion wi h he new imes amp, e?
=
H(..., imes amp,...)will p oduce a di e en alue
and he e i ica ion will ail.
Spoo ing A ack: A dishones use could a emp o
impe sona e a legi ima e use and y o p o ide ake
in o ma ion o he sys em. This a ack is imp ac ical
because a use wi hou alid c eden ials will be unable
o au hen ica e he da a.
Tampe ing A ack: A malicious use could a emp
o modi y an au hen ica ed packe o al e he posi-
ioning o use s. This a ack is un easible because in
case λis modi ied, e?
=H(...,λ)will p oduce a di -
e en alue and he e i ica ion will ail.
5 IMPLEMENTATION RESULTS
In his sec ion, we p esen he esul s o he imple-
men a ion o ou p o ocol on di e en sma phones.
Table 1 shows he ha dwa e and so wa e speci ica-
ions o he de ices used.
Table 1: HW and SW speci ica ions o he sma phones.
De ice CPU OS Blue oo h RAM
Samsung Galaxy
S21+ 5G Exynos 2100 And oid 11 5.0 LE 8 GB
Samsung Galaxy
S20 FE Exynos 990 And oid 10 5.0 LE 6 GB
Samsung Galaxy
A52 SDM720G And oid 11 5.0 LE 6 GB
Samsung Galaxy
A32
MTK D720 Dual
+ Hexa And oid 11 5.0 LE 4 GB
Samsung Galaxy
S8 Exynos 8895 And oid 9 5.0 LE 4 GB
We ha e used ellip ic cu e c yp og aphy o he im-
plemen a ion and design o he p o ocol. The im-
plemen a ion is based on he mcl c yp og aphic li-
b a y (Shigeo, 2018). I allows us o c ea e po able
applica ions be ween di e en de ices and ope a ing
sys ems and suppo s he op imal A e pai ing o e BN
cu es. We use he 256-bi Ba e o-Naeh ig cu e
(BN254) p o ided by he lib a y and compiled i
using OpenSSL (The OpenSSL P ojec , 2003) and
GMP (G anlund, 2014) as dependencies. We ob-
se ed in (Hajny e al., 2018) ha he GMP lib a y
achie es he bes pe o mance esul s o bilinea pai -
ing ope a ions. Wi h his cu e size, an uncomp essed
poin occupies 64 by es and a comp essed poin 33
by es. The size o a scala in ege is 32 by es. To
educe he size o he in o ma ion, we ha e consid-
e ed ansmi ing he poin s in hei comp essed o m.
Fu he mo e, we used he OpenSSL lib a y o com-
pu e c yp og aphic hashes. In pa icula , he SHA-2
unc ion wi h a diges size o 224 bi s.
We ha e implemen ed he p o ocol in C o mea-
su e he pe o mance on a Raspbe yPi 4. The execu-
ion on his de ice is in he o de o µs. The sma -
phone applica ion has been de eloped in Ja a and
Ko lin. The And oid NDK (Na i e De elopmen Ki )
allows us o implemen pa s o ou app in na i e code,
using languages such as C and C++. This allows us o
call he mcl lib a y unc ions h ough he Ja a Na i e
In e ace (JNI).
Finally, we benchma k he en i e p o ocol wi hou
including he communica ion o e head o e alua e he
pe o mance and speed o he algo i hms. The esul s
o he execu ion a e shown in Figu e 6. We can see he
execu ion o bo h p o ocols akes sligh ly mo e han
10 ms on he slowes de ice. Du ing he de elopmen
we used he Ene gy P o ile o And oid S udio and
did no de ec an ene gy consump ion impac .
Show Ve i y
0
2
4
6
8
1.21
2.88
1.78
3.66
2.87
6.96
3.11
7.26
3.57
7.64
Time [ms]
S. Galaxy S21+ 5G S. Galaxy S20 FE
S. Galaxy A52 S. Galaxy A32
S. Galaxy S8
Figu e 6: Speed compa ison o he Show and Ve i y algo-
i hms.
6 CONCLUSION
Ou solu ion allows sha ing anonymous loca ion in-
o ma ion in CIPSs using BLE ad e isemen , p o id-
ing eal p i acy in a ully decen alized sys em whe e
sou ces o da a can be us ed. No wi hs anding he
ela i ely limi ed implemen a ion on cons ained de-
ices, his wo k o e s aluable insigh s in o he se-
cu i y and pe o mance o implemen ing anonymous
ABC schemes on CIPSs. As a as we know, we
p esen he i s ully decen alized scheme unning
o e BLE. Ou app oach has p o en o be e icien on
SECRYPT 2021 - 18 h In e na ional Con e ence on Secu i y and C yp og aphy
796
mobile de ices and comple ely easible o eal- ime
en i onmen s. Nex , we plan o imp o e he p o ocol
o educe he size o he c yp og aphic pa , imp o e
he e oca ion scheme, p o ide in o ma ion enc yp-
ion, and s a deploymen in eal en i onmen s.
ACKNOWLEDGMENT
The au ho s g a e ully acknowledge unding om
Eu opean Union’s Ho izon 2020 Resea ch and In-
no a ion p og amme unde he Ma ie Sklodowska
Cu ie g an ag eemen No. 813278 (A-WEAR, h p:
//www.a-wea .eu/) and om Minis y o he In e-
io o he Czech Republic unde g an VJ01030002.
J. To es-Sosped a acknowledges unding om IN-
SIGNIA p ojec ( e . PTQ2018-009981, MICINN).
REFERENCES
A aoui, G., Lalande, J.-F., T ao ´
e, J., Desmoulins, N.,
Be hom´
e, P., and Gha ou , S. (2015). A p ac ical se -
membe ship p oo o p i acy-p ese ing NFC mobile
icke ing. P oceedings on P i acy Enhancing Tech-
nologies, 2015(2):25–45.
Ba ki, A., B une , S., Desmoulins, N., and T ao ´
e, J.
(2016). Imp o ed algeb aic MACs and p ac ical
keyed- e i ica ion anonymous c eden ials. In P o-
ceedings o he 2016 Selec ed A eas in C yp og aphy
- SAC 2016.
Blue oo h SIG (2019). Co e speci ica ion 5.2. Technical
Repo 5.2, Blue oo h Special In e es G oup.
B ands, S. A. (2000). Re hinking Public Key In as uc u es
and Digi al Ce i ica es: Building in P i acy. MIT
P ess, Camb idge, MA, USA.
Camenisch, J., D ij e s, M., Dzu enda, P., and Hajny, J.
(2019). Fas keyed- e i ica ion anonymous c eden-
ials on s anda d sma ca ds. In IFIP In e na ional
Con e ence on ICT Sys ems Secu i y and P i acy P o-
ec ion, pages 286–298. Sp inge .
Camenisch, J., D ij e s, M., and Hajny, J. (2016). Scalable
e oca ion scheme o anonymous c eden ials based
on n- imes unlinkable p oo s. In P oceedings o he
2016 ACM on Wo kshop on P i acy in he Elec onic
Socie y, WPES ’16, pages 123–133, New Yo k, NY,
USA. ACM.
Camenisch, J., Kohlweiss, M., and So ien e, C. (2010).
Sol ing e oca ion wi h e icien upda e o anony-
mous c eden ials. In Ga ay, J. A. and De P isco,
R., edi o s, Secu i y and C yp og aphy o Ne wo ks,
pages 454–471, Be lin, Heidelbe g. Sp inge Be lin
Heidelbe g.
Camenisch, J. and Lysyanskaya, A. (2001). An E icien
Sys em o Non- ans e able Anonymous C eden ials
wi h Op ional Anonymi y Re oca ion, pages 93–118.
Sp inge Be lin Heidelbe g, Be lin, Heidelbe g.
Chase, M., Meiklejohn, S., and Za e ucha, G. (2014). Al-
geb aic MACs and keyed- e i ica ion anonymous c e-
den ials. In P oceedings o he 2014 ACM SIGSAC
Con e ence on Compu e and Communica ions Se-
cu i y, CCS ’14, pages 1205–1216, New Yo k, NY,
USA. ACM.
Chaum, D. (1985). Secu i y wi hou iden i ica ion: T ans-
ac ion sys ems o make big b o he obsole e. Com-
mun. ACM, 28(10):1030–1044.
Chebli, M. S., Mohammad, H., and Al Ame , K. (2019).
An o e iew o wi eless indoo posi ioning sys ems:
Techniques, secu i y, and coun e measu es. In In e -
na ional Con e ence on In e ne and Dis ibu ed Com-
pu ing Sys ems, pages 223–233. Sp inge .
Des ia i, A. R., K is alina, P., and Suda sono, A. (2017).
Secu e da a ansmission scheme o indoo mobile
coope a i e localiza ion sys em. In 2017 In e na ional
Elec onics Symposium on Enginee ing Technology
and Applica ions (IES-ETA), pages 50–56. IEEE.
G anlund, T. (2014). GNU MP: The GNU Mul iple P eci-
sion A i hme ic Lib a y. h ps://gmplib.o g.
Hajny, J., Dzu enda, P., Ricci, S., Malina, L., and V ba,
K. (2018). Pe o mance analysis o pai ing-based el-
lip ic cu e c yp og aphy on cons ained de ices. In
2018 10 h In e na ional Cong ess on Ul a Mode n
Telecommunica ions and Con ol Sys ems and Wo k-
shops (ICUMT), pages 1–5. IEEE.
Hassidim, A., Ma ias, Y., Yung, M., and Zi , A. (2016).
Epheme al iden i ie s: Mi iga ing acking & spoo -
ing h ea s o ble beacons. u l: h ps://de elope s.
google. com/beacons/eddys one-eidp ep in . pd ( is-
i ed on 06/15/2016).
Kang, J., Seo, J., and Won, Y. (2018). Epheme al id beacon-
based imp o ed indoo posi ioning sys em. Symme y,
10(11):622.
Khandke , S., To es-Sosped a, J., and Ris aniemi, T.
(2019). Imp o ing inge p in ing me hods by means
o d2d communica ion p o ocol. Elec onics, 8(1):97.
Mau z, R. (2012). Indoo posi ioning echnologies.
Geod¨
a isch-geophysikalische A bei en in de Schweiz.
Pascacio, P., Cas eleyn, S., To es-Sosped a, J., Lohan,
E. S., and Nu mi, J. (2021). Collabo a i e indoo
posi ioning sys ems: A sys ema ic e iew. Senso s,
21(3):1002.
Ringe s, S., Ve heul, E. R., and Hoepman, J.-H. (2017).
An e icien sel -blindable a ibu e-based c eden ial
scheme. IACR C yp ology eP in A chi e, 2017:115.
Shigeo, M. (2018). MCL lib a y: A po able and as
pai ing-based c yp og aphy lib a y. h ps://gi hub.
com/he umi/mcl.
The OpenSSL P ojec (2003). OpenSSL: The Open Sou ce
oolki o SSL/TLS. h ps://www.openssl.o g.
Yang, J., Poellabaue , C., Mi a, P., and Neubecke , C.
(2020). Beyond beaconing: Eme ging applica ions
and challenges o ble. Ad hoc ne wo ks, 97:102015.
Zidek, A., Tailo , S., and Ha le, R. (2018). Bell ock:
Anonymous p oximi y beacons om pe sonal de-
ices. In 2018 IEEE In e na ional Con e ence on Pe -
asi e Compu ing and Communica ions (Pe Com),
pages 1–10. IEEE.
Anonymous A ibu e-based C eden ials in Collabo a i e Indoo Posi ioning Sys ems
797