Strategies for Strengthening Data Security in IoT Cloud: An In-depth Review

Introduction:

The Internet of Things (IoT) enables interconnected devices to communicate and be controlled via the Internet, facilitating real-time problem-solving. With the advent of intelligent sensors embedded in various objects and a variety of connectivity protocols, IoT devices can seamlessly connect and communicate over the Internet. Key communication protocols such as CoAP, MQTT, AMQP, and XMPP play a crucial role in enabling efficient communication among these networked IoT devices [1, 2, 3].

Data generated by the IoT network is transmitted over the Internet using protocols like HTTP. The gateway plays a central role in facilitating this communication by translating requests from IoT devices into HTTP, ensuring they are comprehensible to remote servers. The data collected and translated by the gateway ultimately reaches the Cloud for processing, analysis, or storage.

Cloud Computing (CC) offers distributed services regardless of geographical location, providing hardware and software on a metered basis. Key transactions utilize protocols such as HTTP, POP, SMTP, FTP, tailored to specific purposes. IoT devices typically communicate using protocols like CoAP or MQTT, while Cloud services respond in HTTP, forming a bridge between these languages. However, this setup poses security risks, as eavesdroppers or malicious users could intercept data [4]. By offering IT infrastructure as a service, businesses can avoid substantial upfront investments and save time and space. Despite its advantages, the growth of Cloud Computing is hindered by security threats.

The security issues that arise as CC leads to lack of manual control and implements virtualization. This may lead to the integrity, consistency and other privacy issues. Several research’s going on in the field of Cloud and IoT are indenting to enhance the security of these technologies. Looking at the cloud at any point of application, the major issue to be fixed would be the storage issue. Cloud Storage (CS) is holding private data on the Internet, and it is nearer to any user on the Cloud. Therefore, proper security mechanism is to be devised to enhance the authentication, encryption and monitory schemes to prove the proof of ownership such as Third-Party Audit (TPA) should be employed [4, 5].

Authentication can be implemented at two levels: device and user. However, relying solely on authentication is not sufficient. Implementing monitoring mechanisms for both users and devices can significantly enhance security. Additionally, improving encryption schemes, such as recent advancements in Hybrid encryption, proves highly effective in both IoT and Cloud environments. Researchers have found that DNA coding-based encryption, combined with ECC (Elliptic Curve Cryptography), enhances security by increasing the substitution cycle and leveraging ECC's capabilities.

As data generated by IoT networks is stored in the Cloud for analytics and other purposes, it is crucial to employ robust security mechanisms that address multiple dimensions of security. This paper aims to provide a comprehensive review of specific security mechanisms utilized in IoT and Cloud environments.

The paper is organized as follows:

Section (ii) presents the detailed review of the literature, Section (iii) presents the detailed analysis on the existing mechanisms, Section (iv), concludes the work with a brief summary to enhance the security in IoT Cloud integrated platform.

 

Review of literature:

IoT constitutes a global network comprising devices of varying capabilities connected via interfaces, forming integrated networks. These networks, often referred to as IoT networks, involve communication and information processing technologies [6, 7]. They find applications in diverse fields such as healthcare, manufacturing, and surveillance, where the sensitivity of generated data is paramount. Any anomalies in this data could pose significant risks. Therefore, protecting this data from unauthorized access, replacement, or manipulation is crucial.

This review explores the latest security techniques designed to safeguard IoT networks against potential threats.

IoT Security:

After the study on the security concerns in IoT Cloud [8] have proposed a novel method to enhance security of data in IoT Cloud. Authors have used Huffman algorithm for generating keys along DNA coding for the security of the data. The proposal involves two level of encoding the data. Here the plain text is encoded by the proposed DNA algorithm and then by Huffman scheme which is symmetric in nature.

[2] Have proposed a method to enhance the security in IoT integrated Cloud platform using DNA and Huffman based hybrid scheme. DNA coding scheme have an advantage of authentication, storage, digital signatures and so on in a secure manner. Authors have used variable size key. As it was intended by the authors the symmetric crypto system is used as the calculation involves low memory while performing faster. The asymmetric RSA is utilized to encode the symmetric key.

To secure the IoT [9] proposes a cryptographic technique based on stream cipher. "The data between IoT’s were encrypted with One Time Pad (OTP) along with DNA codes." (“Security Enhancing Techniques for Data in IoT Cloud – Analysis”) Furthermore, the key generation was using a linear feedback shift register (LFSR). The data from the publisher is encrypted and the receiver decrypts the data back.

[10] have proposed a methodology for securing the data with digital signatures and DNA cryptography. The digital signature is a public-key cryptographic technique which adopts the scheme of both public and private keys. This public-key cryptographic algorithm is implemented by DNA cryptography in the form of DNA sequencing. This DNA sequencing can store the data and is transferred in a more secure way. The data from the IoT sensors is collected through the machine learning algorithms and is secured by this digital signature algorithm by DNA Cryptography and is stored and sent in a more secure way.

 The performance analysis is conducted evaluating the energy and time taken. Authors have proved that DNA combined with ECC provides strong security.

To provide to IoT environment another approach by [13], have used DNA encoded ECC to reduce processing time. The technique was also intended for reducing the memory size in IoT devices. "These two levels of hybridization add security to IoT Cloud." (“Security Enhancing Techniques for Data in IoT Cloud – Analysis”)

Security in Cloud:

The data generated by the IoT may travel inside the network communicating with other devices. The data then may have to reach a point where it will be stored for future processing. Here comes the Cloud which will hold data or services for a long time. Virtualization being the backbone can be seen as a virtual hard disk where the generations may be forgetting the use of magnetic disks for transferring the data. It can also be assumed that the recent IT infrastructures such as computing devices may be using cloud for processing and storage. In such a technology, security must be enhanced. As CC is already in use the future AI based generations may require data to be safer from various attacks. The remarkable problems comprise Data Integrity (DI) [14]. Here in the study, the analysis is made on the existing techniques to protect the data from various attacks.

[15] Twin MDS code is robust against passive eavesdroppers and data repair process. After the analysis the authors derive that using regenerating code-based scheme performs better than MSR/ MBR.

In [16], the authors present an architecture for an Object Storage (OSt) system deployed on the Cloud. This system is designed to manage static and unstructured data, potentially replacing traditional archived storage solutions. The primary goals of this architecture are to enhance scalability, flexibility, and security in handling such data.

In [17], the authors propose a pipelined approach for Cloud-of-Clouds (CoC) storage systems. This approach aims to accelerate dispersal algorithms by executing calculation operations concurrently with transmission operations in a pipeline fashion. The author’s objective is to optimize the performance of Cloud operations by leveraging parallel processing within the pipeline structure. Through their analysis, they demonstrate that pipelined dispersal algorithms in CoC storage systems significantly enhance operational speed and efficiency.

In [18] authors have introduced, private Cloud infrastructure-based design, for providing security in access and sharing of files and easy maintenance. It adds time efficient storage and sharing of files while nod disturbing easiness and security. For better functionality compression methods can be introduced. Design best suits in situations like unused storage centers over which a cloud is built.

For making up a secure and reliable system [19], proposes a twin code framework which is most suitable for distributed storage, by which it may efficiently handle data reconstruction and efficient node repair.

To secure the CS [20], introduces biometric based framework. Techniques such as chaotic maps, key generation and reed-Solomon decoding are used in various levels of security aspects.

DNA Based Hybrid Approach:

DNA based encryption is becoming popular for the recent IoT Cloud transactions. Here the data will be converted into a genomic sequence which will be a DNA of the IoT Cloud data. "Though it requires subsequent substitutions as caesar ciphers, they are reducing the computational complexity would keep the data more secure." (“Security Enhancing Techniques for Data in IoT Cloud – Analysis”) Hence mixing it with the classical cryptographic scheme such as symmetric or asymmetric would provide strong security [21, 22].

In [23] authors have reviewed security algorithms such as DNA, ECCand RSA algorithms in hybrid form. DNA with ECC makes difference in IoT framework as it reduces time and space. This also ensures the security as the length of the key is more modest. The framework proposed by the authors is employed with double layered security. The first layer is DNA and second layer is ECC. Authors after the study have revealed that such algorithms are more efficient with low capable IoT devices.

Authers on their work [14] have devised a data encryption scheme using bi-serial DNA. The scheme converts the data into hex codes and then into binaries. Data is then split in two parts by the authors. From thus separated keys one of them is used as key and other as actual data for transfer. XOR operation included along this scheme by the authors have increased the compression rate. The performance analysis indicates that the proposal increases the security, with the computational complexity as for amplification two prime numbers where used.

In [24] authors have used DNA along with Binary (OTP) to enhance the security of data. The random nucleotides are made to form a sequence of DNA, generating a key. The length depends on the data size. A binary sequence is used for OTP where the length is twice that of DNA key. Now the process of encoding is done which actually puts at the risk of finding the length of the key, only after this the actual data could be decrypted.

[25] DNA cryptography represents a cutting-edge field within cryptography, leveraging principles from DNA computing to enhance security. Various DNA-based algorithms have been explored, including Symmetric and Asymmetric Key crypto systems using DNA, DNA Steganography Systems, and Triple stage DNA Cryptography [30]. These algorithms utilize DNA sequences to encode and decode information, introducing unique security measures.

Authors to enhance the security in [29] have proposed an ECC based technique for hiding the data in DNA. This technique provides solution for secure communication between the nodes. Authors have implemented the scheme as two parts, where the first is encrypting the data with play fairs. And the second, is hiding the encrypted data behind DNA in a random location using ECC.

Therefore, the DNA based security schemes could serve the IoT Cloud by providing strong security irrespective of the data being transferred.

TPA based Security Approach:

TPA being a more predominant scheme where the users or data blocks may be continuously monitored. This can be seen of as a separate component independent of any DU or DO or CSP’s. The study reveals the major TPA based schemes proposed to monitor the DI.

In [11] authors have proposed a Trusted TPA (TTPA) based framework to enhance security in Cloud. "TTPA increases the security that is gained by using DNA cryptography and Digital Signature." (“Security Enhancing Techniques for Data in IoT Cloud – Analysis”) Cloud User and CSP gain confidence that their data is safe with the help of TTPA. Data confidentiality is ensured by DNA Cryptography, DI by Digital Signature and TTPA ensure data authenticity.

Analysis and discussion:

Security being the major concern, based on the review conducted the analysis is being conducted. The analysis is recorded on the major aspects such as, IoTsecurity, Cloud Security, TPA and DNA based schemes.

Security in IoT – Analysis:

IoT Cloud dominating every field such as science and engineering. It is very rare to find a field that does not use IoT or Cloud.

In this scenario it is the duty of these technologies to protect the sensitive of their clients/users. Though artificial intelligence is employed in most cases the security enhancement schemes are to be enhanced. As the study conducted, the analysis on the recent schemes proposed to enhance the security of IoT is reflected below.

Tabel 3.1. Analysis on Security in IoT

	Proposal	Components	Issues
[2]	Hybrid Scheme	DNA, RSA & Hufman	Transport
[8]	Hybrid Algorithm	DNA, Huffman	Security
[9]	Hybrid Algorithm	DNA & OTP	Security
[10]	Hybrid Scheme	DNA & Digital Signatures	Security
[12]	Hybrid Algorithm	DNA & ECC	Security
[13]	Hybrid Algorithm	DNA & ECC	Security

As IoT is highly exposed to attacks, a large number of works have been proposed to enhance the security. The Table 3.1, show few recent proposals, of which all includes DNA based hybrid algorithm. It is revealed that DNA is more efficient and would be predominantly used in IoT Cloud security.

Security in Cloud – Analysis:

Protecting the Cloud becomes another major task next to IoT. Here as data on the Cloud Storage is at rest may be exposed to various attacks. This may have to be seen of as in two major aspects. First is by the physical disasters. In order to overcome such risks a number of security proposals exists. Scattering of data to n different nodes may help safeguarding a portion of data with which the others can be recalculated.

     Tabel 3.2 Analysis on Security in Cloud

	Proposal	Problem	Approach
[11]	Algorithm	Security	DNA & TPA
[15]	Twin MDS	PassiveEavesdropping, Node Recovery	MDS / MSR
[16]	Architecture	Unstructureddigital static data	Object Storage
[17]	Architecture	Dispersal algorithms	Calculation &Transmission
[18]	Architecture	Security in file access and sharing	Compression
[19]	Framework	Reconstruction & Node Failure	Twin Code
[20]	Framework	Security	Read-solomon, chaotic maps

Encountering the first way of lost data from cloud, Table 3.2, shows the existing mechanisms that safeguard the data on Cloud. Among which it is understood that splitting the data and storing it in multiple nodes will serve more. The second way of threat that is encountered with the data on Cloud is by anonymous users, malicious CU’s or malicious CSP’s This is being handled with the help of various security algorithms and TPA based techniques, are discussed in the following sections.

conclusion:

"The analysis made on the study reflects various security techniques used in IoT Cloud." (“Security Enhancing Techniques for Data in IoT Cloud – Analysis”) Here security to handle data at rest is handled with a separate mechanism and the security of data in transit have been handled using different techniques. Among which Hybrid algorithms have been used in both the places IoT and Cloud. Along with the predominant security schemes such as ECC various other algorithms were used. Among which [2, 9, 10, 11, 12, 13] DNA based encoding is becoming popular with the classical schemes such as ECC.

In emergency-based applications such as medical and military as other such applications, the data integrity has to be preserved largely. The study reveals in-order to preserve the integrity a part in security authentication scheme is employed. Also, authors have introduced [20, 11, 21, 22, 24, 25] TTP based techniques where data at rest is monitored for their integrity. DNA cryptography can be used in CS, signatures and authentication. While authors have also used DNA based scheme for enhancing the integrity of data [26, 2]. Therefore, using schemes such as TTP and Hybrid encryption could enhance the security in IoT Cloud.

References:

1. Y. Wu, Q. Z. Sheng, S. Zeadally, “RFID: Opportunities and challenges,” in Next- Generation Wireless Technologies, N. Chilamkurti, Ed. New York, NY, USA: Springer, 2013, pp. 105–129.

2. Vidhya Vijayan, Eldo P Elias, “Hybrid Method for Securing Data in IoT Cloud”, International Journal of Innovative Technology and Exploring Engineering (IJITEE), 2019.

3. [Sowmya Nagasimha Swamy, Dipti Jadhav, Nikita Kulkarni, “Security Threats in the Application layer in IOT Applications”, International conference on I-SMAC, 2017.

4. Joel J. P. C., Dante B. R., Heres A., Murilo H., Rafael M., Jalal Al-Muhtadi, Victor Hugo C., “Enabling Technologies for the Internet of Health Things”, IEEE, 2018, ISSN: 2169-3536.

5. Samhita Kanthavar, “Design of an Architecture for Cloud Storage to Provide Infrastructure as a Service (IaaS)”, IEEE, 2017.

6. [R. van Kranenburg, "The Internet of Things: A Critique of Ambient Technology and the All-Seeing Network of RFID", Amsterdam, The Netherlands: Institute of Network Cultures, 2007.

7. L. Tan, N. Wang, “Future internet: The In future the work will be extended towards proposing a security enhanced IoT Cloud architecture.

8. internet of things,” in Proc. 3rd Int. Conf. [18]. Marina, Velkaska,   Paunkoska,   “Efficient

9. Adv. Comput.   Theory   Eng.   (ICACTE),  distribution and   improved   security   for

10. Chengdu, China, 2010, pp. 376–380.reliablecloud storage system”, IEEE

11. [Harish Kumar   N,   Rajshekhar   M   Patil, EUROCON 2017–17th International

12. Deepak G,   Murthy   B   M,   “A   Novel  Conference on Smart Technologies, 2017,

13. Approach for securing data in IoTCloud  pp727 – 732.

14. Using DNA Cryptography and Huffman [19]. Nakouri, Hamdi, Kim, “A New Biometic

15. Coding Algorithm”, 2017 International  Based Security Framework for Cloud

16. Conference on Innovations in Information,  Storage”, 13th International Wireless

17.  Embedded and   Communication   Systems  Communication and   Mobile   Computing

18.  (ICIIECS), 2017.  Conference, 2017, pp390 – 395.

19. Noor A. Hussein, Mohamed Ibrahim Shujaa, [20]. Carvalho, Castro, Andrade, “Secure Cloud

20.  “DNA computing-based stream cipher for  Storage Service for detection of security

21.  internet of things using MQTT protocol”,  violations”, 17th IEEE/ ACM International

22.  International Journal   of   Electrical   and  Symposium on   cluster,   cloud   and   grid

23.  Computer Engineering (IJECE), Vol. 10(1),  computing, 2017, pp715 – 718.

24.  February 2020, pp. 1035 – 1042. [21]. Kang Ning, “A Pseudo DNA Cryptography

25. [10]. Naga Saranya Cherukupalli, Sesha  Method”, arXiv:0903.2693 [cs.CR], Cornell

26.  ShayeeMaruvada, “Securing Data in IoT  University Library, 2009.

27.  Devices using DNA Cryptography”, [22]. Kritika Gupta, Shailendra Singh,   “DNA

28.  International Journal for Modern Trends in  Based Cryptographic Techniques: A

29.  Science and Technology, 6(8S), 2020.  Review”, International Journal of Advanced

30. [11]. Nayna Agarwal, Anand Mahendran,  Research in Computer Science and Software

31.  Ramanathan Lakshmanan, “Trusted Third  Engineering, Vol. 3 (3), 2013.

32.  Party Auditing for Cloud Security Using [23]. Malti Bansal, Shubham Gupta, Siddhant

33.  Digital Signature and DNA Cryptography”,  Mathur, Comparison of ECC   and   RSA

34.  IJSTR, Vol 8(12), 2019.  Algorithm with DNA Encoding for IoT

35. [12]. Harsh Durga Tiwari, Jae Hyung Kim, Novel  Security, Proceedings of the Sixth

36.  Method for   DNA-Based   Elliptic   Curve  International Conference on Inventive

37.  Cryptography for IoT Devices, ETRI  Computation Technologies [ICICT 2021],

38. [13]. Journal, Vol. 40 (3), 2018.

39. (http://wileyonlinelibrary.com/journal /etrij) Barman, P., Saha, B. “DNA encoded elliptic IEEE Xplore, 2021.

40. Shreyas Chavan, “DNA Cryptography Based on DNA Hybridization and One Time

41.  curve cryptography system for IoT security”.  pad scheme”, International Journal of

42.  International Journal of Computational  Engineering Research & Technology, Vol. 2 Intelligence & IoT. 2, 2019.  (10), 2013.

43. [D. Prabhu, M. Adimoolam, “Bi-serial DNA [25]. Mansi Rathi, Shreyas Bhaskare, Tejas Kale,

44.  Encryption Algorithm” [Online]. Available:  Niral Shah, Naveen Vaswani, “Data Security

45.  https://pdfs. Semantic scholar. org/ 1754/  Using DNA   Cryptograph”,   International

46.  f0eb5 85 25005 98 a70af 4002 e186 cd2f  Journal of Computer Science and Mobile

47.  3c6 ce.pdf  Computing, Vol.5, 2016, pg. 123-129.

48.  Samundiswary.S, Nilma M Dongre, “Object,  

49. Storage Architecture in Cloud for  

50.  Unstructured  Data”, International Conference on Inventive Systems and  Control, IEEE, 2017.