A Lightweight Access Control Scheme with Attribute Policy for

Blockchain-Enabled Internet of Things

Syed Sajid Ullah

, Vladimir Oleshchuk

and Harsha S. Gardiyawasam Pussewalage

Department of Information and Communication Technology, University of Agder, Grimstad N-4898, Norway

Keywords:

Blockchain, Internet of Things (IoT), Access Control, blockchain-enabled Internet of Things, Digital

Signature, Hyper Elliptic Curve Cryptography.

Abstract:

The Internet of Things (IoT) revolution has increased the number of connected devices, leading to new security

challenges. One of these challenges is the management of access control for these devices. Traditional access

control methods may not be able to address these challenges effectively. This paper proposes attribute-based

access control (ABAC) for the blockchain-enabled Internet of Things (BE-IoT). ABAC allows access to be

granted or denied based on the attributes of the user or device requesting access rather than relying on a central

authority to manage access control information. This allows for more ﬂexible and dynamic access control poli-

cies that can adapt to changing situations and minimize the risk of unauthorized access. Blockchain technology

provides a secure and transparent way to manage access control information for IoT devices. Blockchain is a

decentralized and distributed ledger allowing secure and tamper-proof information storage. By storing access

control information on the blockchain, it can be shared across a network of devices transparently and securely.

The authors conduct a security analysis to evaluate and compare the proposed scheme to existing schemes.

The analysis results indicate that the proposed scheme has the advantage of using minimal computation time

and communication overhead compared to previous solutions. The authors suggest that the ABAC scheme

using blockchain combined with a lightweight Hyperelliptic Curve Cryptosystem (HCC) is well suited for

secure deployment in IoT.

1 INTRODUCTION

The Internet of Things (IoT) refers to the integration

of diverse physical entities equipped with sensors,

embedded software, and network interfaces, such

as household appliances, automobiles, and industrial

machinery (Kumar et al., 2019), (Sinha et al., 2022).

These devices collect and exchange data, thereby ex-

panding their capabilities. The remote interaction and

control of these devices have the potential to radi-

cally alter lifestyle and occupational norms, with sub-

stantial implications for business dynamics. How-

ever, IoT also brings new security challenges to the

table; as the number of connected devices increases,

so does the number of potential vulnerabilities (Omo-

lara et al., 2022), (Swessi and Idoudi, 2022). Protect-

ing IoT devices from unauthorized access is crucial

for preventing data leakage and protecting valuable

and sensitive information stored on these IoT devices.

https://orcid.org/0000-0002-5406-0389

https://orcid.org/0000-0002-7905-1016

https://orcid.org/0000-0002-3623-3362

Access control is a vital aspect of cybersecurity that

ensures that only individuals with the proper autho-

rization can access computer systems, devices, and

the sensitive information stored within them. By im-

plementing effective access control measures, orga-

nizations can safeguard their critical assets and pre-

vent potential security breaches (Sandhu and Sama-

rati, 1994). When it comes to providing a secure and

efﬁcient mechanism for the protection of information

in the large-scale and complex environment of the

IoT, traditional access control models, such as discre-

tionary access control (DAC), mandatory access con-

trol (MAC), and role-based access control (RBAC),

have limitations (Hu et al., 2015). These models were

designed for use in closed systems and did not have

the capability to deal with the dynamic interactions

that can arise between a large number of connected

devices in an IoT environment.

The attribute-based access control (ABAC) (Wang

et al., 2004), allows access to be granted or de-

nied depending on the attributes of the user or de-

vice requesting access. A user’s identity, location,

528

Ullah, S., Oleshchuk, V. and Pussewalage, H.

A Lightweight Access Control Scheme with Attribute Policy for Blockchain-Enabled Internet of Things.

DOI: 10.5220/0012138900003555

In Proceedings of the 20th International Conference on Security and Cryptography (SECRYPT 2023), pages 528-539

ISBN: 978-989-758-666-8; ISSN: 2184-7711

 2023 by SCITEPRESS – Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)

role, or even the device can be considered among

these attributes. Traditional access control methods

have been replaced by ABAC, which offers a ﬁner-

grained level of control over who can access various

resources. In addition, ABAC makes it possible to

have access control policies that are more adaptable

and dynamic, which means they can accommodate

shifting conditions and reduce the likelihood of unau-

thorized access.

Access control information can be stored in

a ledger that is decentralized and distributed if

blockchain technology is used, providing an increased

level of security as well as transparency. In addition,

because blockchain is a decentralized and distributed

ledger, it enables all of the parties in the network to

have a copy of the same data. This makes it extremely

difﬁcult to alter the data, further increasing its secu-

rity. The combination of blockchain technology, IoT,

and ABAC offers a more efﬁcient and secure access

control approach to handle a large number of IoT de-

vices. In a Blockchain-Enabled ABAC (BE-ABAC)

for IoT, each device has a unique identiﬁer stored

on the blockchain. Access requests are sent to the

blockchain network for evaluation. Furthermore, the

blockchain network would verify the attributes of the

user or device against the device’s access control pol-

icy, which is also recorded on the blockchain. Imple-

menting blockchain in this setting offers several ad-

vantages over more conventional approaches to con-

trolling access. Because it is decentralized and dis-

tributed, the blockchain does not have a single point

that can be attacked or compromised. In addition, the

utilization of smart contracts enables the automated

execution of access control policies, which can help

to contribute to an increase in efﬁciency while simul-

taneously decreasing the likelihood of a mistake being

made by a human.

To guarantee the security of the attributes uti-

lized in the access control determination, ABAC sys-

tems often use various cryptographic methods, such

as RSA, Bilinear pairing (BP), Difﬁe-Hellman (DH),

Elliptic Curve Cryptosystem (ECC), and HCC (Hus-

sain et al., 2021). It has been demonstrated that HCC

provides superior performance in terms of key length

and efﬁciency compared to ECC, BP, DH, and RSA

(Hussain et al., 2020). HCC also has a shorter key

length. In addition, HCC uses an 80-bit key, which

offers a high level of protection and has a size ap-

propriate for the resource-constrained devices typi-

cal for IoT environments. In general, the utiliza-

tion of HCC in ABAC systems for IoT, in combi-

nation with Blockchain technology, provides a bal-

ance of security and efﬁciency, making it an appropri-

ate choice to secure communication in IoT systems.

Also, blockchain allows for creating a decentralized

and tamper-proof record of access control decisions.

This can be used to ensure the integrity of the at-

tributes used in the access control decision and the

authenticity of the devices accessing the resources.

Inspired by the above considerations, we present a

lightweight ABAC scheme for BE-IoT. The proposed

scheme is considered lightweight because it is built on

the fundamental idea of HCC. The main contributions

of the paper are listed below.

• We designed an ABAC system for blockchain-

enabled Internet of Things (BE-IoT) that is both

secure and lightweight.

• We proved that the proposed scheme is secure

based on the hardness of the Hyperelliptic Curve

Discrete Logarithm Problem in the random oracle

model.

• We compare the efﬁciency of the proposed

scheme with previous schemes, and the results in-

dicate that the proposed scheme has superior ef-

ﬁciency in terms of computational and communi-

cation costs.

1.1 Paper Organization

The paper is organized into nine sections: Intro-

duction and Motivation (Section 1), Literature Re-

view (Section 2), Threat Model (Section 3), Network

Model for Access Control (Section 4), Proposed Ac-

cess Control Mechanism (Section 5), Deployment and

Workﬂow (Section 6), Formal Security Analysis (Sec-

tion 7), Performance Analysis (Section 8), and Con-

clusion (Section 9).

2 RELATED WORK

(Alansari et al., 2017), presented an innovative iden-

tity and access management solution in cloud feder-

ations. The proposed solution utilizes ABAC poli-

cies and blockchain technology with trusted hardware

from Intel to protect users’ identities while ensuring

the policy review process is honest and transparent.

Due to the intensive Bilinear Pairing algorithm, the

proposed method consumes a great deal of compu-

tational and communication resources, and the pro-

posed system requires formal security proof. (Wang

et al., 2018), examine a data storage and sharing

scheme in decentralized storage systems and presents

a framework that integrates the interplanetary ﬁle sys-

tem, the Ethereum blockchain, and ABE technology.

This framework allows data owners to encrypt shared

data and control who can access it through access

A Lightweight Access Control Scheme with Attribute Policy for Blockchain-Enabled Internet of Things

529

policies while implementing a keyword search func-

tion on the ciphertext of the data using smart contracts

on the Ethereum blockchain. The proposed method

has a few limitations: it employs a computationally

and communicatively intensive Bilinear Pairing algo-

rithm. Formal security proof is required to ensure

its security and reveal private key attacks. Besides,

the proposed solution improves data storage and shar-

ing in decentralized storage systems; it may require

additional security guarantees for resource use to be

practical. (Alniamy and Taylor, 2020) propose an ar-

chitecture model to offer greater ﬁne-grained access

control over data stored in the cloud. The model in-

tegrates the Hyperledger blockchain technology and

the ABE scheme. This allows data owners to encrypt

shared data and control access through access control

policies connected with attributes. However, the au-

thors fail to provide a formal proof.

(Yang et al., 2021), propose a non-interactive ac-

cess control strategy for the IoT based on blockchain

technology and private set instruction (PSI). The pro-

posed strategy allows data holders to store their infor-

mation on a cloud server and users to access this data

by adding attributes as a transaction to the blockchain.

A smart contract then executes the PSI protocol to

determine whether the characteristics set satisﬁes the

threshold structure, and if it does, the user is granted

access to the data. However, the proposed method

employs a computationally intensive Bilinear Pairing

algorithm, which is not suitable for a resource-limited

environment. (Liu et al., 2021), offer a revocable

ABAC system for blockchain applications. The de-

sign approach allows for attribute-level access con-

trol and user revocation. The proposed method em-

ploys a resource-intensive Bilinear Pairing algorithm,

which necessitates extensive computational and com-

munication resources. (Rouhani et al., 2021), ABAC

on blockchain will allow reliable auditing of access

attempts. The system’s audibility and openness will

beneﬁt both parties. With Hyperledger Fabric, the

authors claim high efﬁciency with low computational

overhead. However, due to the intensive Bilinear Pair-

ing algorithm, the proposed method consumes a great

deal of computational and communication resources.

In 2021, (Zaidi et al., 2021) propose a blockchain-

based ABAC approach for the IoT. IoT devices can

control the user’s environment and collect personal

data. Smart contracts automate data access, while

Proof of Authority improves system performance and

reduces gas usage. Data are encrypted and can only

be decrypted within a valid access time when using

blockchains. Due to the intensive Bilinear Pairing al-

gorithm, the proposed method consumes a great deal

of computational and communication resources. (Lu

et al., 2021), propose using ABE with blockchain

technologies to regulate IoT data access. Data encryp-

tion and ABE algorithms create ﬁne-grained access

control while ensuring IoT security. Aside from data

hash values, the blockchain currently stores cipher-

text location, access control policy, and other critical

information in hash values. However, due to the inten-

sive Bilinear Pairing algorithm, the proposed method

consumes a great deal of computational and commu-

nication resources. There also needs to be explicit se-

curity proof for the proposed approach. (Arasi et al.,

2022)propose a new data-sharing system combining

blockchain technology and ABAC management. The

author creates a trustworthy blockchain-based scheme

for safe data sharing with integrity audits that main-

tain data integrity. Due to the intensive Bilinear

Pairing algorithm, the proposed method consumes a

great deal of computational and communication re-

sources, and the proposed system requires formal se-

curity proof. A blockchain-enabled ABAC that main-

tains conﬁdentiality was proposed by (Zhang et al.,

2022) for use in intelligent healthcare systems. The

proposed method prevents failure at a single point

and lowers the expenses of online operations thanks

to online-ofﬂine encryption.

(Zhu et al., 2018b), focus on creating a secure

resource-sharing platform that leverages the advan-

tages of a decentralized blockchain environment, ﬂex-

ible and diverse permission management, and a ver-

iﬁable and transparent access mechanism. They

propose a transaction-based access control (TBAC)

platform that combines the blockchain system and

the industry-standard ABAC model The authors pro-

pose a TBAC-related cryptosystem (CryptoTBAC)

for secure attribute exchanging and dynamic policy

decision-making. (Zhu et al., 2018a), presented a dig-

ital asset management platform, DAM-Chain, which

utilizes TBAC and combines the beneﬁts of the distri-

bution ABAC paradigm and blockchain technology.

The ABAC in the platform offers ﬂexible and de-

centralized authorization mechanisms for digital asset

management on the blockchain. Blockchain transac-

tions serve as a traceable and veriﬁable method for

access requests. (Dukkipati et al., 2018), proposed an

ABAC system for the IoT that enables users to access

and manage their data. The authors used a blockchain

model to create access control measures, but there

were privacy concerns and a lack of explicit security

evidence. (Ding et al., 2019), introduced a new ABAC

approach for IoT systems that leverages blockchain

technology to prevent single points of failure and data

tampering. Streamlining access control methods and

technologies has also enhanced low-power processing

efﬁciency for IoT hardware.

SECRYPT 2023 - 20th International Conference on Security and Cryptography

530

(Jiang, 2021), introduce smart contract-based ac-

cess control, which eliminates the need for a central

trusted server and instead uses the blockchain to com-

plete access authorization. The proposed technique

addresses the standard access control strategy’s cen-

tralization issue. The scheme contains a security ﬂaw

and uses heavy computation and communication re-

sources. (Ghorbel et al., 2021), reveal blockchain-

based ABAC and ﬁne-grained access control, which

maintains user privacy and accountability. The au-

thors constructed a permission blockchain prototype

and ran numerous tests to establish the solution’s scal-

ability, which they presented at the conference. Un-

fortunately, Due to the intensive Bilinear Pairing al-

gorithm, the proposed method consumes a great deal

of computational and communication resources, and

the proposed system requires formal security proof.

After conducting a comprehensive study, it has

been found that the recommended methods, partic-

ularly those using bilinear pairing, have signiﬁcant

computational and communication resource utiliza-

tion drawbacks. The underlying cryptographic al-

gorithms used in these methods are computationally

intensive and result in substantial overheads, mak-

ing them impractical for many applications. More-

over, despite their popularity, these methods have not

been rigorously tested and analyzed for security, and

proper formal security proof is lacking. This raises

concerns regarding the security of these methods, as

they rely heavily on theoretical assumptions and lack

empirical evidence to support their claims. As a re-

sult, it is essential to conduct more extensive secu-

rity analysis and testing before these methods can be

widely adopted and relied upon. Despite these chal-

lenges, developing a secure and efﬁcient scheme is es-

sential for ensuring the privacy and security of digital

communications and transactions. As such, contin-

ued research and development in this ﬁeld are critical

for advancing the state of the art and addressing the

current limitations of existing methods.

3 NETWORK MODEL FOR

ACCESS CONTROL IN

BLOCKCHAIN-ENABLED IOT

We used a lightweight ID-based signature and at-

tributes access policy to provide and ﬁll the ac-

cess control requirements for IoT using blockchain.

Figure 1 presents a network model for secure and

well-managed access control in blockchain-enabled

IoT. This modal consists of IoT Devices, Users,

Blockchain, and Attributes Authority participants.

Figure 1: Proposed network modal for access control in

blockchain-enabled IoT.

IoT Device: The IoT is a network of physical de-

vices equipped with sensors, software, and network

connectivity that allow them to gather and exchange

data. In the context of a blockchain-enabled IoT (BE-

IoT) infrastructure, an IoT device refers to any phys-

ical device that is used in IoT applications such as

smart healthcare, smart homes, banking, and smart

contract trading. These devices can include devices

like smart thermostats, cameras, door locks, and many

other types of connected devices. These devices pro-

vide access and data to users using blockchain tech-

nology.

Users: Users refer to end-users or devices that want to

access and retrieve data from IoT devices. The users

can be individuals or organizations looking to access

data from IoT devices for various reasons like mon-

itoring, control, and decision-making. These users

need to be authenticated and authorized to access the

IoT data.

Blockchain: In a BE-IoT infrastructure, the data is

stored in a combination of blocks containing IoT de-

vice requests. Blockchain is a decentralized tech-

nology that uses a digital ledger to record and se-

cure transactions between multiple parties. It employs

cryptography to ensure the authenticity, integrity, and

immutability of the data stored on the ledger. The

data are organized in blocks, each containing a set of

transactions and a cryptographic hash of the previous

block. This structure creates a blockchain that forms

the blockchain, making it tamper-resistant.

Attributes Authority (AA): An essential aspect of

a BE-IoT infrastructure is the Attributes Authority

(AA), which is a trustworthy authority that establishes

secure access. The AA assigns access attributes and

provides participants with a key pair. These attributes

can be things like user identiﬁcation, location, time,

A Lightweight Access Control Scheme with Attribute Policy for Blockchain-Enabled Internet of Things

531

and other information that can be used to make access

decisions. The key pair is used to sign and verify data,

ensuring only authorized parties can access it. The AA

is a critical element in a BE-IoT system, playing a cru-

cial role in maintaining data security and protection.

It veriﬁes the attributes and identity of users to ensure

that only authorized entities are granted access to IoT

device data, preventing any unauthorized access and

safeguarding the integrity of the data.

The algorithm sections explain in detail the activ-

ities of all participants in the proposed access control

scheme.

4 THREAT MODEL

In a BE-IoT infrastructure, the main issue that could

compromise the system is illegal access to IoT data.

Threats in a BE-IoT include actions that compro-

mise data access, integrity, availability, and veriﬁca-

tion of data owner. Individuals with malicious in-

tent and technical capabilities can exploit these vul-

nerabilities. Researchers have developed various ac-

cess control cryptographic schemes to address secu-

rity and privacy concerns in the open BE-IoT envi-

ronment. These schemes aim to ensure secure data

exchange between peers and prevent adversaries from

accessing, retrieving, or tampering with the informa-

tion. The typical threat model for designing these se-

curity solutions is the (Dolev and Yao, 1983), which

assumes an insecure public channel and powerful ad-

versaries who can access and receive data through the

BE-IoT network. However, this model does not ac-

count for the possibility of an attacker guessing ran-

dom numbers or obtaining access without a valid key.

In this work, our proposed access control scheme for

BE-IoT is based on the DY threat model and assumes

that only the AA is trusted and secure. The scheme

leverages digital signatures to verify devices and data

sources and grant secure access to end users. The dig-

ital signatures-based access control allows for secure

veriﬁcation of devices and data sources, granting se-

cure access to end-users. The proposed scheme aims

to provide enhanced security compared to previous

solutions while still operating within the constraints

of the DY threat model.

To enhance security in the BE-IoT environment

with minimal cost, an access control scheme has been

proposed. The objective of this scheme is to provide

secure access to the users while operating in the BE-

IoT where adversaries can access, retrieve, and ex-

change block data. The scheme is designed to be Ex-

istentially Unforgeable and Secure against Attacks on

Access Control (UAACA). In this scheme, a game

is played between the adversary (AV ) and the chal-

lenger (CA). The AV inputs the public parameter P,

which has a polynomial running time and tries to win

a game with a probability of non-negligibility. The

CA, on the other hand, must respond to the queries

made by the AV in order to play the game. The aim of

the game is to ensure that the access control scheme

remains secure and prevents unauthorized access to

the data and devices in the BE-IoT environment.

• For any participants identity ID

, the AV issues an

extraction query Q

, and the CA executes a key

generation query using ID

as input, obtaining a

private key PV K

, which is then sent back to the

AV as a response.

• When the AV receives a Signature Generation

Query Q

signature

, they provide PV K

as input. The

CA then runs the Signature Generation Algorithm

using PV K

and provides the resulting signature

Sd to the AV .

• The AV inputs ID

′

and user/object attributes or

data Π

′

. It generates Z

′

as the signature. The

extraction query Q

and the signature generation

query Q

signature

have not been executed previously

for ID

′

and Π

′

5 PROPOSED ACCESS CONTROL

ALGORITHM FOR

BLOCKCHAIN-ENABLED IOT

The following is the detail of the proposed algorithm

and its working steps.

Access Policy and Condition: Let S be a set of at-

tributes, U be a set of users with identities (ID

), and

Π be an object or data in our BE-IoT. Every user and

object must sign attributes or data (Sd). The access

will grant according to the following conditions and

rules R.

• Access grants if Sd veriﬁed (S |= R).

• Access denied if Sd veriﬁcation ﬁeld (S ⊭ R).

Setup for Registration: The Attribute Authority (AA)

executes this phase. The AA takes the parameter of se-

curity (ps), selects ﬁnite ﬁeld (FF) and divisor (Dv)

of HCC of order n.

• Choose the master secret key (Ms), such that

Ms ∈{1,2,3,. . . ,n-1}.

• calculate the master public key (Mb), Mb = Ms ·

Dv.

• choose hash functions H

, H

of SHA-256 nature.

SECRYPT 2023 - 20th International Conference on Security and Cryptography

532

The AA then publishes P = (n,H

,Dv,Mb) as

public parameters set in the blockchain-enabled IoT

network and keeps (Ms) secret with itself.

Key Generation: In this step, the IoT devices, and

users send the identities (attributes) (ID

) to AA. The

AA calculates keys for all participants by computing

follow steps.

• Calculate private keys for participants (PV K

• At the start, the AA pick a number randomly Rn ∈

1,2,3, ... ,n − 1.

• Compute h

= H

(ID

,Mb).

• Set the PV K

= Rn + h

· Ms mod n where PV K

of every participants will be different.

• In the same, the AA calculate public keys (PBK

)

for participants as PBK

= Rn · Dv where PBK

of every participants will be different from each

other. Then AA set a factor FA = h

· Mb.

After calculations, the AA sends (PV K

,PBK

) to the

participants via a secure channel.

Access: In this step, the IoT devices, users, or block

owner requests to access any data or block addition in

IoT networks.

• Grants access if the attribute matches S |= R.

• Denied access if the attribute does not match S ⊭

The following steps of signature generation and veri-

ﬁcation will be used to ﬁnd the validity of IoT devices,

and users, in our BE-IoT.

Signature Generation: The signature will be applied

to validate the user for access, or data sources authen-

tication. This algorithm will be run by IoT devices,

a user when requested to access IoT data, IoT data

owner for delivering data. It takes an attribute, object,

or IoT data as input represented by Π, a nonce (Fn),

, PV K

, PBK

, and executes the following steps.

• The participant picks a random number Rs ∈

{1,2,3, ... ,n − 1} and calculates Pn = Rs · Dv.

• The participant computes a message digest h

(Π,Pn,ID

,Fn) where ID

and h

of every

IoT device, users, and IoT Data Owner will be dif-

ferent from each other. Here We use an additional

fresh nonce (Fn) to achieve the goal of revoca-

tion.

• The participant generate Z = Rs + h

· PV K

mod

The participant delivers the signed attribute or data

Sd = ID

,Z,PV K

to the recipient.

Note: The user will execute this algorithm to re-

quest access, and the sender, after access is granted,

signs the data.

Signature Veriﬁcation: To ﬁnd the validation of the

user for access, veriﬁcation of block addition, IoT de-

vices, and data source veriﬁcation, the applied signa-

ture (Z) will be veriﬁed as.

• Compute the digest of a message h

= H

(Π,

Pn,ID

,Fn).

• Compute Z · Dv=Pn + h

(PBK

+ FA)

If Sd is veriﬁed it holds S |= R then access will be

granted. Otherwise, denied access according to S ⊭ R.

Note: The sender will execute this algorithm to

verify the requested users for access grant, IoT Data

Owner for allowing block addition, and receivers to

verify the data source.

5.1 Consistency of the Proposed Scheme

The following calculations proved the correctness of

the designed attribute for access.

Z = Rs + h

· PV K

To prove this, we multiply Dv by both sides;

Z · Dv = (Rs + h

· PV K

)Dv

Multiply Dv by both elements of the brackets;

Z · Dv = Rs · Dv + (h

· PV K

)Dv

As we know Rs · Dv = Pn, so put the value of Rs;

Z · Dv = Pn + h

(PV K

)Dv

Similarly PV K

= PBK

+ h

· Mb, put the value;

Z · Dv = Pn + h

(PBK

+ h

· Mb)

Also h

· Mb = FA, so put the value of h

· Mb;

Z · Dv = Pn + h

(PBK

+ FA)

So, the above equations securely calculate the

value of Z.

6 DEPLOYMENT AND

WORKFLOW OF THE

PROPOSED SCHEME

As depicted in Figure 2, our access control scheme is

intended to be used in a BE-IoT environment. This

scheme aims to allow multiple users to securely ac-

cess IoT devices while ensuring that only valid users

are granted access as per the set Blockchain policies.

To achieve this goal with minimal resource consump-

tion, the scheme is designed to operate in the follow-

ing consecutive steps:

Access Policy and Condition : Let S be a set of at-

tributes, U set of users with identities (ID

), and Π

be objects or data in our BE-IoT. Every user and ob-

ject must sign attributes or data (Sd). Access grants if

Sd veriﬁed (S |= R). Access denied if Sd veriﬁcation

ﬁeld (S ⊭ R).

A Lightweight Access Control Scheme with Attribute Policy for Blockchain-Enabled Internet of Things

533

Figure 2: Deployment and workﬂow of the proposed scheme.

Setup for Registration and Connectivity: To connect

to the network. First, the Attribute Authority (AA)

take the parameter of security (ps), selects ﬁnite ﬁeld

FF and divisor Dv of order n, chooses master se-

cret key (Ms), where Ms ∈ {1,2,3,...,n − 1}, cal-

culate the master public key (Mb), Mb = Ms· Dv, and

chose hash functions H

, H

. The AA then publishes

P = (n, H

,Dv,Mb) as public parameters set in the

IoT network publicly and keeps (Ms) secret with it-

self. To connect to the network, the IoT devices and

users send the identities (attributes) (ID

) to AA. The

AA calculates private keys for participants (PV K

At the start, the AA pick a number randomly Rn ∈

{1,2,3, ... ,n − 1}, compute h

= H

(ID

,Mb).Then

set PV K

= Rn + h

· Ms mod n where PV K

of every

participant will differ. In the same, the AA calculates

public keys (PBK

) for participants as PBK

= Rn·Dv

where PBK

of every participant will be different

from each other. Then AA set a factor FA = h

· Mb.

After calculations, the AA sends (PV K

,PBK

) to the

participants via a secure channel.

Request for Access: Let the user request to access data

(Π). The user will apply a digital signature (Z); for

that, the user takes the requested data name (Π), a

fresh nonce (Fn), ID

, PV K

, PBK

. To generate Z,

the user picks a random number Rs ∈ {1, 2,3,. .., n −

1} , calculate Pn = Rs · Dv, compute data name digest

= H

(Π,Pn, ID

, Fn). The user generates Z = Rs

+ h

· PV K

mod n and sends an access request. The

requested data must include attributes like the thing’s

name, identity, etc.

User Veriﬁcation to Grant or Denied Access: To val-

idate the user’s access, the attributes must be present

in S according to R, and the applied signature Z

will be veriﬁed by computing the digest data name

= H

(Π,Pn,ID

,Fn) and computing Z ·Dv = Pn+

(PBK

+ FA). If Z is veriﬁed according to R, then

the conditions S |= R are satisﬁed, and access will be

granted. Otherwise, access will be denied according

to the state S ⊭ R. When access is granted, the IoT

devices will apply a digital signature on the requested

data Sd = ID

,Z,PV K

using the same signature gen-

eration process as explained in the request for access

step. The user can verify the signature to validate the

data sources of the IoT device.

Block Veriﬁcation for Addition: let the owner want

to add a new block. IoT Data Owner will contain sig-

nature Z, and the block will be hash h and informa-

SECRYPT 2023 - 20th International Conference on Security and Cryptography

534

tion of the last block. Here, the same digital signature

process will be applied to generate the signature, and

every user and IoT Data Owner will verify the sig-

nature using our veriﬁcation process. If the IoT Data

Owner’s signature is conﬁrmed, it will be added to the

chain. Otherwise, it will remove.

7 FORMAL ANALYSIS OF

SECURITY

In this section of the paper, we will demonstrate

the security of our access control scheme under the

assumption of the hardness of the HCDLP prob-

lem using the Random Oracle Model (ROM). If

an attacker/adversary AV has non-negligible advan-

tage NNA and can make at most MosT queries for

hash Q

hash

, extraction Q

, and signature genera-

tion Q

signature

, then the security of the scheme is

guaranteed against malicious access, unveriﬁed data

sources, and unveriﬁed block additions. The CA

with an execution time of



23(Q

hash

)

NNA



and a prob-

ability of NNA

′

⩾

can solve the HCDLP prob-

lem if



NNA ⩾

10(Q

hash

+1)(Q

hash

signature

)



and

MosT

′

⩽



23(Q

hash

)

NNA



(Ullah et al., 2020).

PROOF: Let a be such that the CA is challenged with

an HCDLP. Also, the task of the CA is to calculate Z

from W = Z · Dv, where Z ∈ {1, 2,3,. .., n − 1}. The

CA sets P = (n,H

,Dv,Mb) as a set of parameters

for the public. The response to the AV queries is given

below.

1. Queries on H

: The CA in start maintains an

empty list H

containing the tuples (ID

,PBK

If the tuples (ID

,PBK

) exist in H

, the CA

returns h

; otherwise, the CA chooses a random

∈ {1, 2,3,. .., n −1} and stores (ID

,PBK

)

in H

and returns h

2. Queries on H

: The CA initially maintains an

empty H

containing the tuples (Π, Pn,ID

If the tuples (Π,Pn,ID

) exist in H

, the CA

returns h

; otherwise, the CA selects a num-

ber at random h

∈ {1,2, 3,.. .,n − 1}, stores

(Π,Pn,ID

) in H

, and returns h

3. Query on Extraction Q

: When CA submits

a request with ID

, then CA randomly selects

x,y ∈ {1, 2,3,. .., n − 1}, sets PBK

= x · Mb +

y · Dv, PV K

= x, and h

= (ID

,PBK

) = −i

mod n. The tuple (PV K

,PBK

) satisﬁes the

equation PV K

· Dv = Pub

+ FA in the key gen-

eration algorithm, where FA = h

· Mb and h

(ID

,PBK

). The CA generates (PV K

,PBK

)

as the public-private key pair of ID

and copies

the tuple (PV K

,PBK

(ID

,PBK

),ID

) in

the H

4. Queries on Signature Generation Q

signature

: Upon

request with ID

from the AV, CA performs the

following steps.

• The Q

signature

check determines whether it

has been queried for ID

. If it has been

queried for Q

hash

or Q

signature

, it retrieves

(ID

,PV K

,PBK

(ID

,FA),ID

) from the

list H

. Then, the Signature Generation pro-

cess is performed by the CA, which produces Z

and inserts it into H

(Π,Pn,ID

) within H

• If it is not asked to perform the Q Queries on

Signature Generation (Q

signature

), then the CA

processes the speciﬁed oracle and retrieves the

matching secret key to generate a signature for

access.

• If it is queried, then the CA selects two random

numbers x and y from the set 1, 2,3,. ..,n − 1.

It retrieves h

= H

(ID

,PBK

) from H

and

calculates Pn = x · Dv − y · PBK

− y · h

· Mb.

Then, it sets x = Z and h

= y, and inserts

(Π,Pn,ID

,y) into H

. If (Π,Pn,ID

,y) al-

ready exists in H

, the CA response is false,

and it exits. The failure probability is not more

than 1/n

′

as y is selected randomly.

When AV can produce the same signa-

ture for access as described in the Signature

Generation procedure by using the function



NNA⩾10(Q

hash

+1)·(Q

hash

signature

)



. It is im-

portant to note that if CA has not undergone the

Signature Generation oracle for access, two valid

signatures can be created: (Π,PBK

,Pn,h

,Z) and

(Π,PBK

′

,Pn

′

Thus, veriﬁcation processes are necessary to grant

access or verify data

Firstly,

Pn=g(Dv), PBK

= x · Mb + y · Dv, Mb = Ms · Dv

Z(Dv) = g · Dv + h

· x · sMs · Dv + h

· h

· Ms · Dv

Secondly,

′

(Dv) = g · Dv + h

′

· x · Ms · Dv + g · Dv = h

′

· y ·

Dv + h

′

· h

· Ms · Dv

′

·Z · Dv = Z

′

·g ·Dv+Z

′

·h

·x ·Ms ·Dv +Z

′

·h

y · Dv + Z

′

· h

· Ms · Dv

Z · Z

′

· Dv = Z · g · Dv + Z · h

′

· x · Ms · Dv + Z · h

′

y · Dv + Z · h

′

· h

· Ms · Dv

So, if we perform the subtraction, then we have

(Z · −Z

′

·g+Z ·h

′

·y−Z

′

·h

·y)·(Dv = Z · h

′

·x−Z

′

· x + Z ···h

′

· h

) · Ms · Dv.

Then we have k = (Z · g − Z

′

· g + Z · h

′

· x − Z

′

· x) mod n and (ζ) mod n = (Z · h

′

· x − Z · h

A Lightweight Access Control Scheme with Attribute Policy for Blockchain-Enabled Internet of Things

535

x + Z · h

′

· h

)

−1

mod n, which is the solution of the

HCDLP for CA with the time MosT

′

⩽



23(Q

hash

)

NNA



and probability NNA

′

⩾

(Ullah et al., 2020).

8 EFFICIENCY ANALYSIS

The goal is to determine how well the scheme per-

forms in terms of its resource requirements. For ef-

ﬁciency analysis, we selected four schemes ((Wang

et al., 2018), (Liu et al., 2021), (Zhang et al., 2022),

and (Zhu et al., 2018a)) based on their formal proofs

and clear cryptographic operations. This analysis is

essential because it helps to assess the feasibility of

using a particular scheme in a given application and

can also be used to compare different schemes and

choose the most appropriate one. Some factors that

are commonly considered in efﬁciency analysis in-

clude the time and number of bits required for secure

access, as well as the size of the keys and ciphertext.

8.1 Computation Time

In the ABAC system, the attributes of both users and

a resource are embedded in a digital signature, which

is then checked to ensure that the attributes are correct

and have not been altered in any way. Verifying a dig-

ital signature is a computationally expensive process.

The more attributes used in a policy, the more signa-

tures must be veriﬁed, which drives up the computa-

tion cost. The computation cost is also affected by

the size of the digital signature and the complexity of

the algorithm being used. The computation cost will

be increased, for instance, if a larger key size is used

for the algorithm that generates digital signatures or a

more complex algorithm, such as RSA, is used. Using

a more efﬁcient digital signature algorithm, such as

the ECC and HCC, which is more efﬁcient than RSA,

is one approach to reducing the computation cost in

an ABAC system. Similarly, reducing the number of

characteristics used in the policies by only utilizing

the attributes required to determine who has access to

the resources.

The cost of pairing-based point multiplication

), bilinear pairing (B

), exponentiation (E

and Hyperelliptic Curve Divisor Multiplication (H

)

are essential parameters in evaluating the perfor-

mance and efﬁciency of pairing-based cryptography

methods. The authors in (Khan et al., 2020) and (Ul-

lah et al., 2021), report that the cost of one pairing-

based point multiplication is 4.31 milliseconds, the

cost of a single bilinear pairing is approximately

14.90 milliseconds, the cost of single exponentiation

is approximately 1.25 milliseconds, and the cost of a

single HCDM is about 0.48 milliseconds. These val-

ues are used to measure the performance of the cryp-

tographic algorithm and can be used to compare dif-

ferent methods.

In Table 1, the performance of the designed

scheme is compared to previously related access con-

trol schemes, such as (Wang et al., 2018), (Liu et al.,

2021), (Zhang et al., 2022), and (Zhu et al., 2018a)

in terms of the computation time required in millisec-

onds. Table 1 suggests that our scheme is more ef-

ﬁcient than the previously related schemes regarding

the time required for cryptographic operations. The

designed scheme is evaluated for its performance by

comparing the cost of mathematical operations used

with the previously related schemes. The perfor-

mance is measured in terms of time required for the

cryptographic operations, and based on the compari-

son results, the designed scheme is more efﬁcient than

the previous ones as shown in Figure 3.

Figure 3: Computation time analysis.

Concluding Remarks: We evaluated the performance

of various ABAC schemes for IoT using blockchain

technology in terms of computation time. The ta-

ble above summarizes the results of the evaluation,

which show that the computation time for the existing

schemes evaluated ranges from 65.69 milliseconds to

75.75 milliseconds. In contrast, the computation time

for our proposed scheme is signiﬁcantly lower at 1.44

milliseconds. This indicates that the proposed scheme

is more efﬁcient in computation time as it takes less

time to perform cryptographic operations. This makes

it a more suitable option for real-world IoT applica-

tions requiring low computation time and real-time

processing, such as applications with many users or

real-time data processing.

8.2 Communication Overhead

In the context of ABAC, the term ”communication

overhead” refers to the amount of data that needs to

be exchanged between the client, the attribute author-

ity, and the access control decision point to arrive at

an access control decision. This must take place be-

SECRYPT 2023 - 20th International Conference on Security and Cryptography

536

Table 1: Computation time analysis.

Ref. No Costly cryptographic operations Computation time in milliseconds (ms)

(Wang et al., 2018) 3E

+ 4P

+ 3B

65.69 ms

(Liu et al., 2021) 3E

+ 1P

+ 3B

52.76 ms

(Zhang et al., 2022) 6E

+ 4B

67.1 ms

(Zhu et al., 2018a) 1E

+ 5B

75.75 ms

Proposed 3H

1.44 ms

fore an access control decision can be made. The size

of the messages exchanged can be measured to deter-

mine this, including the size of the attribute query, the

size of the attribute response, and the size of the ac-

cess control decision. Another method for determin-

ing the extent of the communication burden caused

by ABAC is to count the number of messages passed

back and forth between the various entities. Count-

ing the number of attribute queries, the number of at-

tribute responses, and the number of access control

decisions made is one way to determine this (Hussain

et al., 2020).

In this comparison, the designed approach is be-

ing evaluated against some of the previously sug-

gested schemes regarding communication overhead.

The proposed scheme is compared with (Wang et al.,

2018), (Liu et al., 2021), (Zhang et al., 2022), and

(Zhu et al., 2018a). The analysis assumes certain

variables such as bilinear pairing (G), hyperelliptic

curve cryptosystem (q), and message (m), as shown

in Table 1. The results, as shown in Figure 3 and Ta-

ble 2, indicate that the designed scheme substantially

improves communication overhead compared to the

previous schemes. The length of the elements in dif-

ferent cryptographic schemes are assumed to be spe-

ciﬁc values: bilinear pairing (G) is assumed to have

a length of 1024 bits, message (m) is considered to

have a length of 512 bits, and Hyperelliptic curve (q)

is considered to have a length of 80 bits.

The communication overhead for (Wang et al.,

2018) scheme is calculated to be 4|G| + |m|, and

for (Liu et al., 2021) scheme it is calculated to be

3|G| + |m|, for (Zhang et al., 2022) scheme it is cal-

culated to be 4|G| + |m|, Zhu et a. (Zhu et al., 2018a)

scheme is calculated to be 3|G| + |m| and for the pro-

posed is calculated to be 3|q| + |m|. The results, as

presented in Figure 4 and Table 2, indicate that the

designed scheme signiﬁcantly improved communica-

tion overhead compared to the previous schemes. We

suggest that the proposed scheme is a better option

than the previous ones based on the comparison re-

garding communication overhead and cost reduction.

Concluding Remarks: We have evaluated the commu-

nication overhead of several ABAC schemes for IoT

using blockchain technology. The results of our anal-

ysis are summarized in the table above, which shows

Figure 4: Communication overhead in bits.

the communication overhead in bits for four exist-

ing schemes and our proposed scheme. The existing

schemes considered in this research are (Wang et al.,

2018), (Liu et al., 2021), (Zhang et al., 2022), and

(Zhu et al., 2018a). The communication overhead of

these schemes ranges from 4608 bits to 3584 bits. It is

worth noting that two of the existing schemes, (Wang

et al., 2018) and (Zhang et al., 2022), have the same

communication overhead of 4608 bits.

On the other hand, our proposed scheme has a

communication overhead of 752 bits, which is signiﬁ-

cantly lower than the existing schemes. This indicates

that our proposed scheme is more efﬁcient in terms of

communication overhead.

9 CONCLUSION AND FUTURE

WORK

The Internet of Things (IoT) revolution has cre-

ated new security challenges, particularly managing

access control for connected devices. Traditional

methods may not be effective in addressing these

challenges. This paper proposes the use of IoT-

enabled attribute-based access control (ABAC) using

blockchain technology as a solution to this problem.

ABAC allows access to be granted or denied based on

the attributes of the user or device requesting access

rather than relying on a central authority. This allows

for more ﬂexible and dynamic access control policies

that can adapt to changing situations and minimize

the risk of unauthorized access. Blockchain technol-

ogy provides a secure and transparent way to manage

access control information for IoT devices. The au-

thors propose a new lightweight ABAC scheme for

A Lightweight Access Control Scheme with Attribute Policy for Blockchain-Enabled Internet of Things

537

Table 2: Communication overhead analysis.

Ref. No Costly operations for Communication overhead Communication overhead in bits

(Wang et al., 2018) 4|G| + |m| 4608 bits

(Liu et al., 2021) 3|G| + |m| 3584 bits

(Zhang et al., 2022) 4|G| + |m| 4608 bits

(Zhu et al., 2018a) 3|G| + |m| 3584 bits

Proposed 3|q| + |m| 752 bits

IoT using blockchain technology and conduct a se-

curity analysis to evaluate and compare the proposed

scheme to existing schemes. The results indicate that

the proposed scheme has the advantage of using mini-

mal computation time and communication bandwidth

compared to previous solutions. The authors suggest

that the ABAC scheme using blockchain combined

with a lightweight Hyperelliptic Curve Cryptosystem

(HCC) is well suited for secure deployment in IoT.

In the future, we will address the privacy concerns

that are raised by the practice of storing access histo-

ries on a blockchain. We will look into a more ad-

vanced privacy-preserving technique that offers the

beneﬁts of blockchain technology without compro-

mising user privacy.

REFERENCES

Alansari, S., Paci, F., and Sassone, V. (2017). A distributed

access control system for cloud federations. pages

2131–2136. IEEE.

Alniamy, A. and Taylor, B. D. (2020). Attribute-based

access control of data sharing based on hyperledger

blockchain. pages 135–139.

Arasi, V. E., Gandhi, K. I., and Kulothungan, K. (2022).

Auditable attribute-based data access control using

blockchain in cloud storage. Journal of Supercomput-

ing, 2022(1):1–27.

Ding, S., Cao, J., Li, C., Fan, K., and Li, H. (2019).

Novel attribute-based access control scheme using

blockchain for iot. IEEE Access, 7(1):38431–38441.

Dolev, D. and Yao, A. (1983). On the security of public key

protocols. IEEE Trans. Inf. theory, 29(2):198–208.

Dukkipati, C., Zhang, Y., and Cheng, L. C. (2018). Decen-

tralized, blockchain based access control framework

for the heterogeneous internet of things. pages 61–69.

ACM.

Ghorbel, A., Ghorbel, M., and Jmaiel, M. (2021). Account-

able privacy preserving attribute-based access control

for cloud services enforced using blockchain. Interna-

tional Journal of Information Security, 2021(1):1–22.

Hu, Vincent, C., Kuhn, D. R., Ferraiolo, D. F., and Voa,

J. (2015). Attribute-based access control. Computer,

48(2):85–88.

Hussain, S., Ullah, I., Khattak, H., Adnan, M., Kumari, S.,

Ullah, S. S., Khan, M. A., and Khattak, S. J. (2020).

A lightweight and formally secure certiﬁcate based

signcryption with proxy re-encryption (cbsre) for in-

ternet of things enabled smart grid. IEEE Access,

8(1):93230–93248.

Hussain, S., Ullah, I., Khattak, H., Khan, M. A., Chen, C.,

and Kumari, S. (2021). A lightweight and provable

secure identity-based generalized proxy signcryption

(ibgps) scheme for industrial internet of things (iiot).

Journal of Information Security and Applications,

58:102625.

Jiang, X. (2021). Decentralized, blockchain based access

control framework for the heterogeneous internet of

things. pages 21–23. J. Phys Conf. Ser. 1955.

Khan, M., Rehman, S., Uddin, M., Nisar, S., and Noor,

F. (2020). n online-ofﬂine certiﬁcateless signature

scheme for internet of health things. Journal of

Healthcare Engineering, 2020(1):1–10.

Kumar, S., Tiwari, P., and Zymbler, M. (2019). Internet

of things is a revolutionary approach for future tech-

nology enhancement: a review. Journal of Big data,

6(1):1–21.

Liu, X., Zheng, Y., and Li, X. Z. (2021). A re-

vocable attribute-based access control system using

blockchain. In Journal of Physics: Conference Series,

1971(1):012058.

Lu, X., Fu, S., Jiang, C., and Lio, P. (2021). A ﬁne-grained

iot data access control scheme combining attribute-

based encryption and blockchain. Security and Com-

munication Networks, 2021:1–13.

Omolara, Esther, A., Alabdulatif, A., Abiodun, I., Alaw-

ida, M., and Arshad, H. (2022). The internet of

things security: A survey encompassing unexplored

areas and new insights. Computers and Security,

112(1):102494.

Rouhani, S., Belchior, R., and Cruz, R. S. (2021). Dis-

tributed attribute-based access control system us-

ing permissioned blockchain. World Wide Web,

24(1):617–644.

Sandhu, S. R. and Samarati, P. (1994). Access control: prin-

ciple and practice. IEEE communications magazine,

32(9):40–48.

Sinha, Bahadur, B., and Dhanalakshmi, R. (2022). Recent

advancements and challenges of internet of things in

smart agriculture: A survey. Future Generation Com-

puter Systems, 126(1):169–184.

Swessi, D. and Idoudi, H. (2022). A survey on internet-

of-things security: threats and emerging coun-

termeasures. Wireless Personal Communications,

124(2):1557–1592.

Ullah, I., Khan, M. A., F.Khan, M.A.Jan, R.Srinivasan,

S.Mastorakis, S.Hussain, and Khattak, H. (2021). An

efﬁcient and secure multimessage and multireceiver

SECRYPT 2023 - 20th International Conference on Security and Cryptography

538

signcryption scheme for edge-enabled internet of ve-

hicles. IEEE Internet of Things Journal, 29(9):2688–

2697.

Ullah, S. S., Ullah, I., Khattak, H., Khan, M., Adnan, M.,

Hussain, S., Amin, N., and Khattak, M. (2020). A

lightweight identity-based signature scheme for mit-

igation of content poisoning attack in named data

networking with internet of things. IEEE Access,

8(1):98910–98928.

Wang, L., Wijesekera, D., and Jajodia, S. (2004). A logic-

based framework for attribute based access control.

pages 45–55. ACM.

Wang, S., Zhang, Y., and Zhang, Y. (2018). Blockchain-

based framework for data sharing with ﬁne-grained

access control in decentralized storage systems. IEEE

Access, 6(1):38437–38450.

Yang, Q., Zhang, M., Zhou, Y., Xia, T. W. Z., and Yang, B.

(2021). Non-interactive attribute-based access control

scheme by blockchain for iot. Electronics, 15:1–11.

Zaidi, S. Y. A., Shah, M. A., Khattak, H. A., Maple, C.,

Rauf, H. T., El-Sherbeeny, A. M., and El-Meligy,

M. A. (2021). An attribute-based access control for

iot using blockchain and smart contracts. Sustainabil-

ity, 13(21):10556.

Zhang, Y., Wei, X., Cao, J., Ning, J., Ying, Z., and

Zheng, D. (2022). Blockchain-enabled decentralized

attribute-based access control with policy hiding for

smart healthcare. Journal of King Saud University -

Computer and Information Sciences.

Zhu, Y., Qin, Y., Gan, G., Shuai, Y., and Chu, W. C. C.

(2018a). Digital asset management with distributed

permission over blockchain and attribute-based access

control. pages 535–544. IEEE.

Zhu, Y., Qin, Y., Zhou, Z., Song, X., Liu, G., and Chu, W. C.

(2018b). Digital asset management with distributed

permission over blockchain and attribute-based access

control. pages 193–200. IEEE.

A Lightweight Access Control Scheme with Attribute Policy for Blockchain-Enabled Internet of Things

539