Anonymous Attribute-based Credentials in Collaborative Indoor

Positioning Systems

ul Casanova-Marqu

1,2 a

, Pavel Pascacio

2,3 b

, Jan Hajny

1 c

and Joaqu

ın Torres-Sospedra

2,4 d

Department of Telecommunications, Brno University of Technology, Brno, Czech Republic

Institute of New Imaging Technologies, Universitat Jaume I, Castell

on, Spain

Electrical Engineering Unit, Tampere University, Tampere, Finland

UBIK Geospatial Solutions S.L., Castell

on, Spain

Keywords:

Attribute-based Credentials, Collaborative Indoor Positioning Systems, Privacy, Anonymity, Bluetooth Low

Energy, Wearables.

Abstract:

Collaborative Indoor Positioning Systems (CIPSs) have recently received considerable attention, mainly be-

cause they address some existing limitations of traditional Indoor Positioning Systems (IPSs). In CIPSs, Blue-

tooth Low Energy (BLE) can be used to exchange positioning data and provide information (the Received

Signal Strength Indicator (RSSI)) to establish the relative distance between the actors. The collaborative mod-

els 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. This paper deals with privacy-enhancing technologies and

their application in CIPS. In particular, we focus on cryptographic schemes which allow the veriﬁcation of

users without their identiﬁcation, so-called Anonymous Attribute-based Credential (ABC) schemes. As the

main contribution, we present a cryptographic scheme that allows security and privacy-friendly sharing of

location information sent through BLE advertising packets. In order to demonstrate the practicality of our

scheme, we also present the results from our implementation and benchmarks on different devices.

1 INTRODUCTION

Wi-Fi and BLE ﬁngerprinting are popular techniques

for providing indoor positioning. They rely on the

signal strength from multiple emitters (either Wi-Fi

Access Points (APs) or BLE beacons). However, the

Received Signal Strength (RSS) data used to estimate

the positions have a strong noise component, there-

fore having a general positioning accuracy at meter

level (i.e., a few meters) (Mautz, 2012). In addition,

the deployment complexity, as well as relying on an

infrastructure of beacons and servers, have become

a serious problem for large organizations all around

the world. Wearable-based CIPSs are increasingly be-

coming popular over the last few years as a solution

to address those problems in indoor positioning (Pas-

cacio et al., 2021). The CIPSs use the location of a

https://orcid.org/0000-0001-6653-867X

https://orcid.org/0000-0001-9206-7496

https://orcid.org/0000-0003-2831-1073

https://orcid.org/0000-0003-4338-4334

set of actors and their relative distances to 1) provide

localization to external users and/or 2) improve the

positioning accuracy of the current users that are col-

laborating (Khandker et al., 2019).

Wearable devices are equipped with built-in tech-

nologies such as Wi-Fi and Bluetooth, which are ca-

pable of estimating users’ positions and ubiquity in

indoor environments. In particular, CIPSs approaches

based on BLE are one of the most popular solutions.

Smartphones can send BLE advertisement packets

(e.g., iBeacon), which can be used later to compute

the relative distances between them with the gathered

RSS values. This alternative presents a straightfor-

ward deployment and a good trade-off between power

consumption and performance (Yang et al., 2020).

However, high inter-connectivity to unknown devices

and the lack of strong communication protocols in

terms of security and privacy represent a weakness.

Malicious actors could perform attacks such as spoof-

ing or eavesdropping in unsecured collaborative sys-

tems (Chebli et al., 2019). These issues encouraged

the development of more privacy-friendly solutions.

Casanova-Marqués, R., Pascacio, P., Hajny, J. and Torres-Sospedra, J.

Anonymous Attribute-based Credentials in Collaborative Indoor Positioning Systems.

DOI: 10.5220/0010582507910797

In Proceedings of the 18th International Conference on Security and Cryptography (SECRYPT 2021), pages 791-797

ISBN: 978-989-758-524-1

791

Existing CIPS are based on centralized schemes.

A server-based solution facilitates more complex at-

tacks such as user traceability or linkability of com-

munications. Thus, it is necessary to move to a decen-

tralized approach. Building a decentralized CIPS us-

ing BLE technology represents a major challenge due

to the reduced space available in BLE packets, which

depends on the Bluetooth version (Bluetooth SIG,

2019). This hinders the use of standard algorithms

such as RSA, DH or AES for information encryp-

tion and user authentication. Attribute-based authen-

tication schemes allow users to anonymously and se-

lectively prove possession of their personal attributes

such as age, citizenship, negative SARS-CoV-2 test

or location data. Elliptic curve cryptography and the

reduced size of its keys can tackle the aforementioned

problems and provide anonymity, unlinkability, un-

traceability and selective disclosure of attributes.

In light of the identiﬁed problems, in this work

we pursue the decentralization of collaborative indoor

positioning systems based on BLE beacons, while

providing, through elliptic curve cryptography and

anonymous attribute-based authentication schemes, a

secure and privacy-friendly system for exchanging lo-

cation data.

1.1 Related Work

In the literature there are several articles that address

anonymity, security, and privacy issues of IPSs from

different perspectives (e.g., Wireless Sensor Network

(WSN) or Internet of Things (IoT)). However, only

very few of them focus on collaborative positioning

and/or BLE beacon approaches. One of the most in-

teresting papers is proposed by (Kang et al., 2018),

which is based on the open-source Eddystone BLE

format proposed by (Hassidim et al., 2016). Eddy-

stone BLE format introduces a scheme to mitigate

tracking and security threats. The core of the scheme

is cloud-based Ephemeral Identiﬁers (EID), which en-

sures only authorized members to properly identify

the BLE standard beacons.

Up to now, far too little attention has been paid

to CIPSs. (Zidek et al., 2018) developed a scheme

named Bellrock which combines an ecosystem of

standard beacons and user-based beacons. Bell-

rock provides access control to standard beacons

and anonymity to user-based beacons by using three

techniques to generate pseudo-anonymous identiﬁers

(random, synchronized, and encrypted) that can be

unmasked by a server.

(Destiarti et al., 2017) proposed the Data Trans-

mission Scheme for Indoor Mobile Cooperative Lo-

calization System that works over Wi-Fi. The scheme

uses AES-128 to encrypt the message, MD5 to con-

struct the HMAC and RSA-2048 to encrypt both the

AES key and the HMAC. Their approach might not

be well suited as the system remains vulnerable to

man-in-the-middle attacks during the RSA key-pair

exchange, leading to a total loss of conﬁdentiality and

message integrity.

A key problem with the aforementioned propos-

als lies in server-dependency, turning the system into

a centralized environment. This method could suffer

from a plethora of pitfalls such as denial of service

(DoS), spooﬁng, etc.

Some preliminary work on ABCs was carried

out several decades ago by (Chaum, 1985), (Brands,

2000), and (Camenisch and Lysyanskaya, 2001). In

subsequent years, much more information on ABCs

has become available about practical applications

(Arfaoui et al., 2015), revocation protocols (Ca-

menisch et al., 2010) and the use of more efﬁcient

algebraic structures (Chase et al., 2014), (Barki et al.,

2016), and (Ringers et al., 2017). Due to the high

computational complexity required by these schemes,

practical implementation on constrained devices is

still a challenge, researchers need to ﬁnd a scheme

that satisﬁes their speciﬁc needs. As far as we know,

no studies have been published on ABC schemes ap-

plied to CIPSs.

1.2 Our Contribution

In order to ﬁll the gap found in the literature, we pro-

pose a new decentralized privacy-preserving user au-

thentication mechanism for CIPSs with ABC. Our

solution combines zero-knowledge proofs with a re-

vocation scheme based on lifetime and provides

anonymity, unlinkability and untraceability.

The main contributions are the following:

• To the best of our knowledge, we present the ﬁrst

decentralized security scheme for CIPSs based on

BLE advertising.

• We deﬁne a scheme that provide anonymous lo-

cation data sharing, decentralized authentication

and ofﬂine revocation.

• We provide a security analysis on the proposed

solution.

• We describe the implementation results.

The rest of the paper is organized as follows. Sec-

tion 2 gives a brief overview of the system model.

Section 3 thoroughly details the cryptographic design

of the scheme. Section 4 analyzes the security. Sec-

tion 5 presents the implementation results. Finally,

Section 6 reports some conclusions.

SECRYPT 2021 - 18th International Conference on Security and Cryptography

792

2 SYSTEM OVERVIEW

The system model is depicted in Figure 1 and consists

of the following entities:

• Issuer (I): is responsible for issuing the personal

attribute m

to a user using the Issue algorithm.

The m

attribute represents the user identiﬁer

obtained during registration in the system. The

Issue algorithm provides the user with crypto-

graphically signed credentials that are valid within

the system. Access credentials consist of the val-

ues σ,σ

,σ

• Revocation Authority (RA): removes invalid

users. Access control is achieved through the re-

vocation credential σ

, issued from the revoca-

tion attribute m

. Attempting to authenticate with

the revoked credentials will result in failure. To

reduce the complexity of the system, we will con-

sider from now on the revocation authority and the

issuer as the same entity (I).

• User (U): enrolls in the system to use the col-

laborative indoor positioning system. Every user

holds unique credentials generated by the issuer

and the revocation entity (σ,σ

,σ

), then,

the user can anonymously prove their possession

to the veriﬁer using the Show algorithm.

• Veriﬁer (V): veriﬁes the possession of the user at-

tribute using the Verify algorithm. If the attribute

ownership is veriﬁed successfully, the veriﬁer will

be able to use the location information to position

themself. Otherwise, the information is rejected.

Issuer Revocation Authority

Show

Verify

IssueIssue

User Verifier

Figure 1: System structure.

Initially, all devices registered in the system behave as

Users. When one of them needs to improve its posi-

tion, it temporarily assumes the role of Veriﬁer. This

role can be used to ensure that the information is pro-

vided by a valid user.

There are two modes of operation for the User and

Veriﬁer roles: the active mode and the passive mode.

User operation modes: 1) Active mode: responds

to veriﬁer requests and assists them with positioning

by actively participating in the system. 2) Passive

mode: remains unnoticed in the system and does not

respond to veriﬁer requests. In this mode, the user

does not interact with the system.

Veriﬁer operation modes: 1) Active mode: actively

requests location data from nearby users to improve

its position. 2) Passive mode: remains hidden in the

system and waits for other veriﬁers to request location

data. In this mode, positioning is not possible if there

are no active veriﬁers nearby.

Figure 2 depicts each of these four modes. User

A acts in active mode, user B acts in passive mode,

veriﬁer C acts in active mode, and veriﬁer D acts in

passive mode.

Verifier C

User B

Verifier D

User A

Verifier request

User response

Figure 2: System roles and operation modes.

These modes of operation are useful for improving

privacy in environments where there are few devices,

and it may be easy to determine their identity. In this

case, the user or veriﬁer can go into passive mode and

disappear from the system. We will not go into further

details because it is beyond the scope of this paper.

3 CRYPTOGRAPHIC DESIGN

This section describes the protocol for user authenti-

cation in a collaborative indoor positioning system.

3.1 Preliminaries

The symbol H denotes a secure hash function. We

write a

← A when a is sampled uniformly at random

from A. Let e denote a bilinear map G

× G

→ G

3.2 Protocol Speciﬁcation

First, the user must be registered in the system by

sending their personal information to the issuer. The

issuer generates the credentials and sends them to the

user. The user can now legitimately participate in the

system. This algorithm will be denoted as Issue.

Anonymous Attribute-based Credentials in Collaborative Indoor Positioning Systems

793

When the veriﬁer wants to improve their position,

a BLE advertisement packet with a timestamp will be

broadcast. The nearby users will compute the proof

of knowledge and transmit it via another BLE packet.

This algorithm will be denoted as Show.

Finally, the veriﬁer will ensure through the proof

of knowledge that the user has valid credentials and

will use the location information to position themself.

This algorithm will be denoted as Verify.

3.2.1 Issue

To register the user in the system and generate the

valid credentials, the user and the issuer perform the

algorithm shown in Figure 3.

User Issuer

q,G

,e,g

← G (1

), pk

) sk

= (x

), pk

= g

= ww/yyyy

−−−−−−−−−−−−−−−−−−−−→

σ = g

← σ

σ,σ

,σ

←−−−−−−−−−−−−−−−−−−−−

Store: σ,σ

,σ

Figure 3: Deﬁnition of the Issue algorithm.

• The user U performs the following steps:

1) Introduces personal information in the app.

2) Establishes a secure communication channel

with I and sends m

• The issuer I performs the following steps:

1) Generates the user’s private keys x

and x

2) Signs the attributes m

and m

with the is-

suer’s secret key as σ = g

; and

computes auxiliary values σ

,σ

, where

← σ

, σ

← σ

, and σ

← σ

3) Sends σ and auxiliary values σ

,σ

the user U.

3.2.2 Show and Verify

To broadcast the location information and the proof

of knowledge, and subsequently verify that it comes

from a legitimate user, the user and the veriﬁer will

run the algorithms shown in Figure 4.

• The veriﬁer V performs the following steps:

1) Generates a nonce based on the current time.

2) Creates a BLE advertising packet with the

nonce and broadcasts it.

User Veriﬁer

q,G

,e,g

← G (1

), pk

λ = (altitude, longitude,latitude,...) m

= ww/yyyy

timestamp

←−−−−−−−−−−−−−−−−−−

ρ,ρ

,ρ

← Z

∗

σ ← σ

← σ

← g

e ← H (t

σ,

,timestamp,λ)

← ρ

+ eρ

← ρ

− em

ψ ← (

σ,

)

π ← (e,s

)

ψ,π,λ

−−−−−−−−−−−−−−−−−→

← g

−e

−em

= H (t

σ,

,timestamp,λ)

) = e(

σ, pk

)

Figure 4: Deﬁnition of the Show and Verify algorithms.

• Nearby users u

perform the following steps after

they have received the timestamp:

1) Checks that the timestamp generated as a nonce

by the veriﬁer is less than 2 seconds from the

current time to avoid replay attacks.

2) Generates the randomizers ρ, ρ

,ρ

3) Computes

σ,

, where

σ ← σ

← σ

, and

← σ

4) Computes the attribute proof of knowledge t

as t

← g

5) Calculates the cryptographic hash e as

H (t

σ,

,timestamp,λ) and the

auxiliary proof of knowledge values s

where s

← ρ

+ eρ and s

← ρ

− em

6) Creates a BLE advertising packet with the lo-

cation information λ and the proof of knowl-

edge ψ ← (

σ,

),π ← (e,s

and transmits it by broadcast.

• Nearby veriﬁers v

execute the following steps af-

ter receiving ψ, π, and λ.

1) Recomputes the attribute proof of knowledge t

as t

← g

−e

−em

2) Calculates the cryptographic hash using t

as H (t

σ,

,timestamp,λ) and

checks if it matches the value of e.

3) Computes the bilinear pairing e(

) =

σ, pk

) to verify the credentials were emitted

by the issuer.

3.3 Bluetooth Integration

The proposed scheme transmits information using

BLE advertising packets. Bluetooth 4.2 and earlier

SECRYPT 2021 - 18th International Conference on Security and Cryptography

794

deﬁned a payload size of 31 bytes for a single BLE

advertisement packet. However, since it is impracti-

cal to pack both the positioning data and the proof of

knowledge into 31 bytes, we have considered version

5.0 of the BLE protocol. Bluetooth 5.0 introduced

a signiﬁcant improvement by increasing the capacity

of advertising packets. With Low-Energy Advertising

Extensions, the advertisement payload can hold up to

254 bytes (Bluetooth SIG, 2019).

To build the structure of the BLE advertising

packet, we have assumed the structure depicted in

Figure 5.

1 2

5 6

11 12

UUID

ψ ← (

σ,

)

120B

21 22

25 26

30 31

12B

π ← (e,s

)

92B

22B

Figure 5: Deﬁnition of the BLE advertising payload.

The UUID ﬁeld occupies 8 bytes, the proof of knowl-

edge (ψ, π) consists of 224 bytes, and there is a free

space of 22 bytes for the positioning data (λ). This

part is explained in detail in Section 5.

4 SECURITY ANALYSIS

The protocol is composed of the following parts:

the anonymous credentials scheme, the revocation

scheme and the positioning information. In this sec-

tion we provide a brief informal security analysis of

each of them.

4.1 Anonymous Credentials

Following the security model deﬁned by (Chase et al.,

2014) for anonymous credential schemes, it is re-

quired to meet the following properties:

• Correctness: honest users can produce an accept-

ing proof.

• Unforgeability: dishonest users cannot produce an

accepting proof.

• Anonymity: users cannot be identiﬁed based on

the values of attributes.

• Key-Parameter Consistency: every public key has

a distinct private key.

Based on the random Oracle model in (Camenisch

et al., 2019), the credential scheme used in our system

has proven to be secure under the n-SCDHI assump-

tion (deﬁned below).

Deﬁnition 1 (n-Strong Computational Difﬁe-Hell-

man Inversion Problem (SCDHI)). Let O

(·) on in-

put (m

,...,m

) ∈ Z

∗

add (m

,...,m

) to Q and out-

put g

1/(x

∑

i=1

)

. Let O

(·) on input h output h

Deﬁne the advantage of A as follows.

Adv

n−SCDHI

(A) = Pr

(G,g,q) ← GroupSetup(1

,...,x

)

← Z

∗

n+1

,(y,m

∗

,...,m

∗

) ←

(·),O

(·),...,O

(·)

(g) : y = g

∑

i=1

∗

∧

∗

,...,m

∗

/∈ Q

. (1)

SCDHI is (t,ε)-hard if no t-time adversary has ad-

vantage at least ε.

4.2 Revocation Scheme

Following the security model deﬁned by (Camenisch

et al., 2016), it is required to meet the following prop-

erties:

• Revocation Completeness: unrevoked users will

pass the revocation check.

• Revocation Soundness: revoked users will not

pass the revocation check.

• Revocation Privacy: users will not lose privacy by

proving an unrevoked revocation handle.

The revocation scheme is based on the epoch’s ex-

piration. In our solution, we set the epoch to one

week and deﬁne the revocation attribute as follows:

← ww/yyyy. When the credentials expire, the user

must renew them, allowing weekly access control.

We assume that all users in the system have the

date correctly set on their devices. When a user tem-

porarily switches to the veriﬁer role, they can produce

the revocation attribute without requiring the user to

disclose it.

4.3 Location Information

The positioning information is the critical component

of the system. We outline the resistance of the pro-

tocol against some of the most common attacks that

affect the integrity and conﬁdentiality.

Eavesdropping Attack: A fraudulent user could

monitor the communications and attempt to de-

anonymize a legitimate user because the information

is sent in plain text. Capturing communications is

possible since the information is freely transmitted.

However, removing anonymity and attempting to link

a speciﬁc user to positioning data is extremely difﬁ-

cult. Each time a user transmits the position, the cre-

dentials are randomized.

Anonymous Attribute-based Credentials in Collaborative Indoor Positioning Systems

795

Replay Attack: A malicious user could carry out

this attack because the information is not altered.

But after two seconds, the veriﬁer transmits a new

timestamp, invalidating the packet’s authentication.

During veriﬁcation with the new timestamp, e

H (...,timestamp, . ..) will produce a different value

and the veriﬁcation will fail.

Spooﬁng Attack: A dishonest user could attempt to

impersonate a legitimate user and try to provide fake

information to the system. This attack is impractical

because a user without valid credentials will be unable

to authenticate the data.

Tampering Attack: A malicious user could attempt

to modify an authenticated packet to alter the posi-

tioning of users. This attack is unfeasible because in

case λ is modiﬁed, e

= H (. . . , λ) will produce a dif-

ferent value and the veriﬁcation will fail.

5 IMPLEMENTATION RESULTS

In this section, we present the results of the imple-

mentation of our protocol on different smartphones.

Table 1 shows the hardware and software speciﬁca-

tions of the devices used.

Table 1: HW and SW speciﬁcations of the smartphones.

Device CPU OS Bluetooth RAM

Samsung Galaxy

S21+ 5G

Exynos 2100 Android 11 5.0 LE 8 GB

Samsung Galaxy

S20 FE

Exynos 990 Android 10 5.0 LE 6 GB

Samsung Galaxy

A52

SDM720G Android 11 5.0 LE 6 GB

Samsung Galaxy

A32

MTK D720 Dual

+ Hexa

Android 11 5.0 LE 4 GB

Samsung Galaxy

Exynos 8895 Android 9 5.0 LE 4 GB

We have used elliptic curve cryptography for the im-

plementation and design of the protocol. The im-

plementation is based on the mcl cryptographic li-

brary (Shigeo, 2018). It allows us to create portable

applications between different devices and operating

systems and supports the optimal Ate pairing over BN

curves. We use the 256-bit Barreto-Naehrig curve

(BN254) provided by the library and compiled it

using OpenSSL (The OpenSSL Project, 2003) and

GMP (Granlund, 2014) as dependencies. We ob-

served in (Hajny et al., 2018) that the GMP library

achieves the best performance results for bilinear pair-

ing operations. With this curve size, an uncompressed

point occupies 64 bytes and a compressed point 33

bytes. The size of a scalar integer is 32 bytes. To

reduce the size of the information, we have consid-

ered transmitting the points in their compressed form.

Furthermore, we used the OpenSSL library to com-

pute cryptographic hashes. In particular, the SHA-2

function with a digest size of 224 bits.

We have implemented the protocol in C to mea-

sure the performance on a RaspberryPi 4. The execu-

tion on this device is in the order of µs. The smart-

phone application has been developed in Java and

Kotlin. The Android NDK (Native Development Kit)

allows us to implement parts of our app in native code,

using languages such as C and C++. This allows us to

call the mcl library functions through the Java Native

Interface (JNI).

Finally, we benchmark the entire protocol without

including the communication overhead to evaluate the

performance and speed of the algorithms. The results

of the execution are shown in Figure 6. We can see the

execution of both protocols takes slightly more than

10 ms on the slowest device. During the development

we used the Energy Proﬁler of Android Studio and

did not detect an energy consumption impact.

Show Verify

1.21

2.88

1.78

3.66

2.87

6.96

3.11

7.26

3.57

7.64

Time t [ms]

S. Galaxy S21+ 5G S. Galaxy S20 FE

S. Galaxy A52 S. Galaxy A32

S. Galaxy S8

Figure 6: Speed comparison of the Show and Verify algo-

rithms.

6 CONCLUSION

Our solution allows sharing anonymous location in-

formation in CIPSs using BLE advertisement, provid-

ing real privacy in a fully decentralized system where

sources of data can be trusted. Notwithstanding the

relatively limited implementation on constrained de-

vices, this work offers valuable insights into the se-

curity and performance of implementing anonymous

ABC schemes on CIPSs. As far as we know, we

present the ﬁrst fully decentralized scheme running

over BLE. Our approach has proven to be efﬁcient on

SECRYPT 2021 - 18th International Conference on Security and Cryptography

796

mobile devices and completely feasible for real-time

environments. Next, we plan to improve the protocol

to reduce the size of the cryptographic part, improve

the revocation scheme, provide information encryp-

tion, and start deployment in real environments.

ACKNOWLEDGMENT

The authors gratefully acknowledge funding from

European Union’s Horizon 2020 Research and In-

novation programme under the Marie Sklodowska

Curie grant agreement No. 813278 (A-WEAR, http:

//www.a-wear.eu/) and from Ministry of the Inte-

rior of the Czech Republic under grant VJ01030002.

J. Torres-Sospedra acknowledges funding from IN-

SIGNIA project (ref. PTQ2018-009981, MICINN).

REFERENCES

Arfaoui, G., Lalande, J.-F., Traor

e, J., Desmoulins, N.,

Berthom

e, P., and Gharout, S. (2015). A practical set-

membership proof for privacy-preserving NFC mobile

ticketing. Proceedings on Privacy Enhancing Tech-

nologies, 2015(2):25–45.

Barki, A., Brunet, S., Desmoulins, N., and Traor

e, J.

(2016). Improved algebraic MACs and practical

keyed-veriﬁcation anonymous credentials. In Pro-

ceedings of the 2016 Selected Areas in Cryptography

- SAC 2016.

Bluetooth SIG (2019). Core speciﬁcation 5.2. Technical

Report 5.2, Bluetooth Special Interest Group.

Brands, S. A. (2000). Rethinking Public Key Infrastructures

and Digital Certiﬁcates: Building in Privacy. MIT

Press, Cambridge, MA, USA.

Camenisch, J., Drijvers, M., Dzurenda, P., and Hajny, J.

(2019). Fast keyed-veriﬁcation anonymous creden-

tials on standard smart cards. In IFIP International

Conference on ICT Systems Security and Privacy Pro-

tection, pages 286–298. Springer.

Camenisch, J., Drijvers, M., and Hajny, J. (2016). Scalable

revocation scheme for anonymous credentials based

on n-times unlinkable proofs. In Proceedings of the

2016 ACM on Workshop on Privacy in the Electronic

Society, WPES ’16, pages 123–133, New York, NY,

USA. ACM.

Camenisch, J., Kohlweiss, M., and Soriente, C. (2010).

Solving revocation with efﬁcient update of anony-

mous credentials. In Garay, J. A. and De Prisco,

R., editors, Security and Cryptography for Networks,

pages 454–471, Berlin, Heidelberg. Springer Berlin

Heidelberg.

Camenisch, J. and Lysyanskaya, A. (2001). An Efﬁcient

System for Non-transferable Anonymous Credentials

with Optional Anonymity Revocation, pages 93–118.

Springer Berlin Heidelberg, Berlin, Heidelberg.

Chase, M., Meiklejohn, S., and Zaverucha, G. (2014). Al-

gebraic MACs and keyed-veriﬁcation anonymous cre-

dentials. In Proceedings of the 2014 ACM SIGSAC

Conference on Computer and Communications Se-

curity, CCS ’14, pages 1205–1216, New York, NY,

USA. ACM.

Chaum, D. (1985). Security without identiﬁcation: Trans-

action systems to make big brother obsolete. Com-

mun. ACM, 28(10):1030–1044.

Chebli, M. S., Mohammad, H., and Al Amer, K. (2019).

An overview of wireless indoor positioning systems:

Techniques, security, and countermeasures. In Inter-

national Conference on Internet and Distributed Com-

puting Systems, pages 223–233. Springer.

Destiarti, A. R., Kristalina, P., and Sudarsono, A. (2017).

Secure data transmission scheme for indoor mobile

cooperative localization system. In 2017 International

Electronics Symposium on Engineering Technology

and Applications (IES-ETA), pages 50–56. IEEE.

Granlund, T. (2014). GNU MP: The GNU Multiple Preci-

sion Arithmetic Library. https://gmplib.org.

Hajny, J., Dzurenda, P., Ricci, S., Malina, L., and Vrba,

K. (2018). Performance analysis of pairing-based el-

liptic curve cryptography on constrained devices. In

2018 10th International Congress on Ultra Modern

Telecommunications and Control Systems and Work-

shops (ICUMT), pages 1–5. IEEE.

Hassidim, A., Matias, Y., Yung, M., and Ziv, A. (2016).

Ephemeral identiﬁers: Mitigating tracking & spoof-

ing threats to ble beacons. url: https://developers.

google. com/beacons/eddystone-eidpreprint. pdf (vis-

ited on 06/15/2016).

Kang, J., Seo, J., and Won, Y. (2018). Ephemeral id beacon-

based improved indoor positioning system. Symmetry,

10(11):622.

Khandker, S., Torres-Sospedra, J., and Ristaniemi, T.

(2019). Improving rf ﬁngerprinting methods by means

of d2d communication protocol. Electronics, 8(1):97.

Mautz, R. (2012). Indoor positioning technologies.

Geod

atisch-geophysikalische Arbeiten in der Schweiz.

Pascacio, P., Casteleyn, S., Torres-Sospedra, J., Lohan,

E. S., and Nurmi, J. (2021). Collaborative indoor

positioning systems: A systematic review. Sensors,

21(3):1002.

Ringers, S., Verheul, E. R., and Hoepman, J.-H. (2017).

An efﬁcient self-blindable attribute-based credential

scheme. IACR Cryptology ePrint Archive, 2017:115.

Shigeo, M. (2018). MCL library: A portable and fast

pairing-based cryptography library. https://github.

com/herumi/mcl.

The OpenSSL Project (2003). OpenSSL: The Open Source

toolkit for SSL/TLS. https://www.openssl.org.

Yang, J., Poellabauer, C., Mitra, P., and Neubecker, C.

(2020). Beyond beaconing: Emerging applications

and challenges of ble. Ad hoc networks, 97:102015.

Zidek, A., Tailor, S., and Harle, R. (2018). Bellrock:

Anonymous proximity beacons from personal de-

vices. In 2018 IEEE International Conference on Per-

vasive Computing and Communications (PerCom),

pages 1–10. IEEE.

Anonymous Attribute-based Credentials in Collaborative Indoor Positioning Systems

797