250
T. Ismail et al.
domain needs the reinforcement of security measures because data are susceptible to lose, leakage or theft. Therefore, confidentiality and integrity of the stored
Electronic Health Records (EHR) are deemed as one of the major challenges
elevated by the external storage. Besides, the privacy of sensitive data must be
guaranteed. To overcome the above cited challenges, cryptographic techniques
for securing e-health systems are widely adopted. But the reliance on a single
cloud storage provider has shown many drawbacks like a single point of failure,
vendor lock-in and malicious insiders. To narrow down the listed disadvantages,
it is advisable to use multi-cloud architecture. One of the key concepts of this
model is to store data on different cloud server providers where an insider is not
able to reconstruct the original data from a single share [1].
In this context, several solutions have been proposed in the literature to
ensure secure multi-cloud storage in e-health systems [2–5]. They mainly have
two phases: storage and retrieval. They also all use cryptographic primitives to
ensure EHRs security. Authors of [2] use an Attribute Based-Encryption (ABE)
for selective access authorisation and cryptographic secret sharing. The EHRs
split and reconstruction is done through a proxy. In [3], ABE is used for selective data sharing with physicians without allowing them to know the precise
description of the patient’s illnesses. Biometrics based authentication and Kerberos tickets session are used in [4] to guarantee secure interaction with the
EHR system. In addition, a steganographic technique is used to store EHR. In
[5], authors propose the use of Shamir’s Secret Sharing not only to distribute
EHR shares among cloud servers but to retrieve the requested EHR from partial cloud servers. In summary, the main drawback of [2–5] is the reliance on a
trusted third party which may not be adequate for practical use as they show
security risks. Hence, a secure privacy-preserving data storage solution is still
needed to improve the patient role to monitor his data on the cloud.
In this paper, we present a Hybrid and Secure Data Sharing Architecture
(HSDSA), for secure and privacy-preserving storing and sharing of patient’s
sensitive data in a Multi-cloud environment without relying on a trusted third
party. In HSDSA, cloud providers are assumed to be semi-trusted: honest but
curious. HSDSA gives the patient total control over the generation and management of the decryption keys without relying on a trusted authority and thus
it is more applicable for public cloud environments. To protect the data from
external attackers, Rivest–Shamir–Adleman (RSA) encryption is applied before
outsourcing EHR. To secure data against cloud providers curiosity, Shamir’s
secret sharing is adopted. The resulted shares are distributed to multiple clouds.
To download an EHR, HSDSA recovers its shares using an outsourcing reconstruction operation based on the (t, n) strategy. To complete the file decryption,
a Schnorr-based technique is used to prove data possession and to verify the
requester identity. Then a session, using the Diffie Hellman (DH) algorithm, is
created to securely exchange the decryption key. Finally, the key is extracted
and the original EHR could be recovered. Outsourcing reconstruction operation
based on the (t, n) strategy is used.
T. Ismail et al.
domain needs the reinforcement of security measures because data are susceptible to lose, leakage or theft. Therefore, confidentiality and integrity of the stored
Electronic Health Records (EHR) are deemed as one of the major challenges
elevated by the external storage. Besides, the privacy of sensitive data must be
guaranteed. To overcome the above cited challenges, cryptographic techniques
for securing e-health systems are widely adopted. But the reliance on a single
cloud storage provider has shown many drawbacks like a single point of failure,
vendor lock-in and malicious insiders. To narrow down the listed disadvantages,
it is advisable to use multi-cloud architecture. One of the key concepts of this
model is to store data on different cloud server providers where an insider is not
able to reconstruct the original data from a single share [1].
In this context, several solutions have been proposed in the literature to
ensure secure multi-cloud storage in e-health systems [2–5]. They mainly have
two phases: storage and retrieval. They also all use cryptographic primitives to
ensure EHRs security. Authors of [2] use an Attribute Based-Encryption (ABE)
for selective access authorisation and cryptographic secret sharing. The EHRs
split and reconstruction is done through a proxy. In [3], ABE is used for selective data sharing with physicians without allowing them to know the precise
description of the patient’s illnesses. Biometrics based authentication and Kerberos tickets session are used in [4] to guarantee secure interaction with the
EHR system. In addition, a steganographic technique is used to store EHR. In
[5], authors propose the use of Shamir’s Secret Sharing not only to distribute
EHR shares among cloud servers but to retrieve the requested EHR from partial cloud servers. In summary, the main drawback of [2–5] is the reliance on a
trusted third party which may not be adequate for practical use as they show
security risks. Hence, a secure privacy-preserving data storage solution is still
needed to improve the patient role to monitor his data on the cloud.
In this paper, we present a Hybrid and Secure Data Sharing Architecture
(HSDSA), for secure and privacy-preserving storing and sharing of patient’s
sensitive data in a Multi-cloud environment without relying on a trusted third
party. In HSDSA, cloud providers are assumed to be semi-trusted: honest but
curious. HSDSA gives the patient total control over the generation and management of the decryption keys without relying on a trusted authority and thus
it is more applicable for public cloud environments. To protect the data from
external attackers, Rivest–Shamir–Adleman (RSA) encryption is applied before
outsourcing EHR. To secure data against cloud providers curiosity, Shamir’s
secret sharing is adopted. The resulted shares are distributed to multiple clouds.
To download an EHR, HSDSA recovers its shares using an outsourcing reconstruction operation based on the (t, n) strategy. To complete the file decryption,
a Schnorr-based technique is used to prove data possession and to verify the
requester identity. Then a session, using the Diffie Hellman (DH) algorithm, is
created to securely exchange the decryption key. Finally, the key is extracted
and the original EHR could be recovered. Outsourcing reconstruction operation
based on the (t, n) strategy is used.