International Journal of Engineering Trends and Applications (IJETA) – Volume 7 Issue 5, Sep-Oct 2020

RESEARCH ARTICLE OPEN ACCESS

Malware Detection and Signature Generation

Prateek Nigam
Dean & Hod, Electrical And Electronics Engineering (Eee) Sarvepalli Radhakrishnan
University-Bhopal
ABSTRACT
Now a day, the malware detection is needful to enhance the performance of the systems and omit the effect of
malware to system. The conventional signature-based detection of malware did not detect a major of new
variants. This paper presented a hybrid technique for automatic malware signature generation and classification.
The hybrid method is named as a ANFIS-SSA approach. Using this observation, we present a hybrid method for
detection of malware using the correlation between the semantics of the malware and its API calls. Here,
develops a base signature for a whole malware class more than for a solitary specimen of malware. The
signature can able of find out even advanced variants and unknown which related to that class. Here,
demonstrates our method on some well-known malware classes and presented that any advanced variants classes
is detected from the base signature.
Keywords: Malware detection, signature generation, malicious

I. INTRODUCTION
The most famous PC assault is malware that Static examination arrangements are fundamentally
comprise of seventeen classes, for example, actualized utilizing two techniques: signature-based
infections, worms, Trojan horse, spyware and and heuristic-based. Mark put together strategies
furthermore different malicious software. Malware is depend with respect to the recognizable proof of
a program with pernicious aim intended to harm the novel strings in the binary code [52]. The heuristic
PC on which it executes or the system over which it strategies depend on rules, which are either
imparts [1-10]. Albeit a wide range of malware has controlled by specialists or by AI systems that
their particular target, the primary design is to break characterize a noxious or a kindhearted conduct so
the PC activity. Thus, the security component should as to identify obscure malware. This exploration
be actualized so as to ensure all code and centers around robotizing the way toward producing
information against change, substitution or sub- marks to be introduced on such apparatuses for
formed. Present day PC and correspondence known malware that should be sifted by the
frameworks are exceptionally helpless to different machines [53-55]. Different methods have been
sorts of assault [11-20]. A typical method for proposed to determine malware marks consequently,
propelling these assaults is by methods for malicious including among others: helplessness based marks;
software (malware, for example, worms, infections, payload-based marks; content filtering; semantic-
and Trojan horse, which, when spread, can make mindful marks; The Amd calculation; Honeypot-
serious harm private clients, business organizations, based marks and polymorphic substance based
and governments [21-24]. marks [56-60]. Moreover, the deep learning is
acquainted with create the programmed marks. The
The ongoing development in fast Internet above strategies are not ready to create the best
associations gives a stage to making and quickly outcomes for programmed signature age and
spreading the new malware. A few examination malware identification. To conquer the current
strategies for recognizing malware have been techniques disadvantages, the proposed strategy will
proposed [25-34]. They are named to whether they be structured and created in this paper
are static or dynamic. In dynamic analysis
(otherwise called social based investigation), II. LITERATURE REVIEW
identification depends on data gathered from the
working framework at runtime (i.e., during the A wide range of strategies for malware identification
execution of the program, for example, system calls, and signature generation are created by the analysts.
network access and records, and memory alterations A portion of the techniques are explored here,
[35-50]. In static analysis, the location depends on
data removed expressly or verifiably from the Dina Saif et al. [61] have exhibited building up an
executable twofold/source code. The fundamental bit effective computational system dependent on Deep
of leeway of static examination is in giving fast Belief Networks for malware discovery. The
order. Since antivirus sellers are confronting every structure combines significant level static
day a mind-boggling measure of suspect documents investigation, dynamic examination and system calls
for review, rapid detection is basic [51]. in feature extraction so as to accomplish the highest

ISSN: 2393-9516 www.ijetajournal.org Page 15

International Journal of Engineering Trends and Applications (IJETA) – Volume 7 Issue 5, Sep-Oct 2020

accuracy. The assessment looks at the most Agency was generally good and its exhibition
commonplace AI approaches that were applied in overhead was sensible.
malware recognition with the proposed system. The
acquired outcomes are exhibit that Deep Belief III. AUTOMATIC SIGNATURE
Networks procedure can understand 99.1% accuracy GENERATION
with the introduced dataset. Far beyond that, here
build up our total static examination container which The word malware is a portmanteau of "malicious
embraces distinctive productive techniques trying to software" and means programming intended to
encourage and accelerate the static investigation by invade or harm a PC framework, without the
dealing with all the Android applications in just one proprietor's assent. Malware is an aggregate name
stage instead of thinking about each application in for adware, spyware, Trojan horses, worms, viruses
turn. and their preferences. Usually the word" viruses” is
utilized to depict all the above mentioned, despite
A Multi-Level Deep Learning System (MLDLS) the fact that genuine infections make out an entirely
[62] those sorts out various deep learning models modest number of the current malware [16]. The
utilizing the tree structure. Each model in the tree need to group malignant projects as per a bound
structure of MLDLS was not based in general together terminology is about as old as PC infections
dataset. Rather, every deep learning model spotlights themselves. Clearly this isn't a simple work with
on learning a particular information circulation for a numerous classes covering or being firmly related.
specific gathering of malware and all deep learning Now and again specialists additionally differ on the
models in the tree cooperate to settle on an ultimate arrangement, even inside CARO (Computer
conclusion. Thusly, the learning viability of every Antivirus Researchers Organization). The a portion
deep learning model worked for one bunch can be of the viruses are viruses, zoo viruses, Hoaxes, Root
improved. Trial results demonstrate that framework kits, key loggers, downloader’s, spyware, exploits,
performs superior to the conventional methodology. Trojan Horses (trojans), worms and so on. The detail
To safeguard against an expanding number of depiction of the a portion of the infections are
advanced malware assaults, deep learning based displayed underneath,
Malware Detection Systems (MDSs) have turned
into an indispensable segment of our financial and ❖ Viruses: a PC Virus is code that recursively
national security. Customarily, analysts construct the recreates a conceivably advanced duplicate of
single deep learning model utilizing the whole itself. Viruses taint a host record or framework
dataset. In any case, the single profound learning territory, or they essentially change a reference
model may not deal with the undeniably mind to such articles to take control and afterward
boggling malware information appropriations increase again to shape new generations.
successfully. Since various example subspaces ❖ Worms are ordinarily independent
speaking to a gathering of comparative malware may applications without a host program. They
have remarkable information conveyance. So as to fundamentally duplicate on systems, for the
further improve the presentation of deep learning most part with no assistance from a client. Be
based MDSs. that as it may, a few worms additionally spread
as a record infector virus and contaminate host
Yuan Xue et al. [63] have created root privilege programs, which is the reason they are
management agency named Root Agency, which frequently named a unique subclass of viruses.
receives a digital signature scheme to ensure the ❖ Trojan Horses (trojans) depict themselves as
selective root-privilege-granting chances of an option that is other than what they are at the
confirmed applications. Root Agency verifies an purpose of execution. In spite of the fact that it
application by checking whether it holds the mark might promote its action in the wake of
produced by the secret key, and awards the root propelling, this data isn't evident to the client
privilege when a marked application presents the in advance. A Trojan neither imitates nor
solicitation. In addition, it checks the application's duplicates itself, yet purposes harm or bargains
uprightness to keep it from repackaging. Hence, the the security of the PC.
clients were not engaged with decision making while ❖ Exploits are projects or procedures that exploit
defying root requests. The plan guarantees the defenselessness in programming. Endeavors
security of established Android gadgets, and can be utilized for breaking security or
upgrades the security of portable terminal gadgets. generally assaulting hosts over the system.
This reduces the risk to cloud foundation from root- ❖ Spyware are programs that can examine
misused Android gadgets. What's more, a model was frameworks or screen movement and pass this
executed to assess its viability, proficiency, and data on to the aggressor. Regular data that
overhead. The trial results demonstrate that Root might be effectively or latently assembled is
passwords, sign in subtleties, account numbers,

ISSN: 2393-9516 www.ijetajournal.org Page 16

International Journal of Engineering Trends and Applications (IJETA) – Volume 7 Issue 5, Sep-Oct 2020

individual data, singular records or other malware of not through program examination.
individual archives Regularly, each malware spoke to by at least one
mark designs which are novel to portray it. At the
The diverse malwares identification is basic to point when a program is executed, hostile to
improve the presentation of the PC and evade the malware programming will look through bytes of
effect of the framework by malware. Throughout the information stream. A large number of signatures
years malware essayists have been cunning in will be place on database and filtering procedure will
developing their manifestations. Malware have search for every signature to contrast and the
advanced as to replication and spreading program code that execute. Scanning calculation will
instruments, just as strategies used to avoid be utilized to contrast substance of program code
examination or potentially discovery. Such and the signature on database. In this structure,
procedures incorporate enemy of troubleshooting, signature-based strategy will actualize as the main
encryption, utilizing exe-packers, entry point guard from malware assault that will taint PC
obscuring and so forth. Despite the fact it has been activity. This method was picked in light of the fact
demonstrated that there is no calculation that can that this sort of procedure has been compelling in
flawlessly distinguish all future viruses in finite recognizing understood malware. So as to improve
time. It is critical to take note of that not all methods the effectiveness of PC activity, this system was
can be applied to all malware and this ought to proposed in this structure. The definite proposed
likewise not be required. Because one strategy can't technique architecture is introduced in the beneath
be utilized all the time doesn't mean it is totally segment.
incapable. It is sufficient to have a wide range of
procedures, one of which will be a decent answer for IV. PROPOSED ARCHITECTURE
square, recognize or sterilize a specific malware.
In this section, first, we briefly outline our approach
A wide range of methods are created by the for malware signature generation and classification.
specialists yet it not ready to deliver the best answer Next, we describe our program behavior model used
for stay away from the malware in the framework. for signature generation and the statistical
Here, signature based malware detection is presented comparison technique. Then, we present our
in this paper. Signature based recognition is one of malware detection algorithm using our program
the static investigation strategies that regularly behavior model [17]. Finally, we describe our
utilized on business antimalware programming. The prototype implementation in detail and show a
static investigation strategy would look over the sample signature of a malware extracted using our
program code for detection reason and some of the approach. The proposed model is presented in the
time called output strings. This system utilizes it figure 1.
portrayal of the malicious code to conclude that is

Trojan Trojan
Bengin

Backdoors Backdoors ANFIS-

IDA Pro SSA Malicious and
Diassembler Proposed signature generation
Worms Worms classifier

Test file Test file

Figure 1: Architecture of the proposed method.

We make signatures dependent on the attributes of a calls use. For example, an infection attempting to
whole malware class as opposed to a solitary look for executable records will commonly utilize
example of malware. Malware classes are API calls, for example, KERNEL32.DLL,
characterized dependent on comparative conduct. FindClose, FindNextFileA and FindFirstFileA. The
The conduct of a malware class can be determined conduct of looking through records is caught by the
dependent on the application program interface utilization of these API calls. As opposed to
(API) calls that the individuals from the malware considering all API calls, we consider just basic API

ISSN: 2393-9516 www.ijetajournal.org Page 17

International Journal of Engineering Trends and Applications (IJETA) – Volume 7 Issue 5, Sep-Oct 2020

calls. Basic API calls incorporate all API calls that files are opened with an interior instrument that
can prompt security bargain, for example, considers demonstrates a hexadecimal portrayal of the
that change the manner in which the working machine code and features regular code fragments
framework carries on or those utilized for between the open files. Which block from an
correspondence, for example, WinSock, File I/O example to contrast and which square from another
API, File I/O API Registry API and so on. must be picked carefully. Ordinarily the size of the
first records together with the size of the produced
Signature Generation block files can be useful in discovering which blocks
may contain basic code. For a malware where just a
Subsequent to arranging and making nearby couple of tests exist or where all variations are about
duplicates of the malware tests the scan engine is indistinguishable discovering normal code should be
kept running over them to part the examples into possible before long. Be that as it may, it isn't hard
block files. The block files speak to the segments of observing that when numerous examples exist and
the documents the scanner regard prone to contain they share almost no practically speaking this can be
malignant code. It is from these block files that a monotonous work to do physically. In the worst
mark at that point is produced [18]. To conceivably outcome imaginable every single imaginable pair of
produce generic signatures the documents are records must be analyzed just to discover that no
contrasted with discover normal code. The block likenesses exist.

Figure 2: Number of training samples with false positive, false negative values

We found that few benign projects share conduct malware utilizing the mark developed from wide
(for example, looking through records, replicating malware classes with sensible exactness. Once the
documents to network drives and so on.) with certain new malware is identified, its base signature can
noxious projects. The watched false positive rate is without much of a stretch be developed to recognize
because of such common conduct. The outcomes its future variations. The proposed technique is
demonstrate that even without the base signature, contrasted and the current strategies, for example,
our procedure had the option to identify another SAFE.
Table 1: Comparison analysis of the proposed method

Malware API call extractor/Annotator Detector

SAFE proposed SAFE proposed
Hare 9.142 1.665 1.604 0.0282
F0sf0r0 4.900 1.781 0.923 0.0256
Zombie-6.b 4.600 1.718 1.149 0.0314
Chernobyl 1.444 2.172 0.535 0.0138

ISSN: 2393-9516 www.ijetajournal.org Page 18

International Journal of Engineering Trends and Applications (IJETA) – Volume 7 Issue 5, Sep-Oct 2020

We tested the time it requires to order a given record Surveillance Robot for the Real Time
as malicious or benign. We consider the time Monitoring and Control System in
adopted by our strategy to extricate the API calls and Environment and Industrial Applications,
to arrange it as malicious or benign. We contrast our Advances in Intelligent System and
methodology with SAFE. SAFE makes a Computing, pp 229-243, Springer
deliberation example of the pernicious code and [2] Ezhilarasu, P., & Krishnaraj, N. (2015).
changes over it into an inward portrayal. Given a test Applications of Finite Automata in Lexical
program, it makes a control stream chart (CFG) of Analysis and as a Ticket Vending Machine–A
the test program, and checks whether the interior Review. Int. J. Comput. Sci. Eng.
portrayal of malevolent code is available in the CFG. Technol, 6(05), 267-270.
SAFE has been tried uniquely on a not many [3] Agrawal, U., Arora, J., Singh, R., Gupta, D.,
malware tests. Table 3 looks at the time adopted by Khanna, A., & Khamparia, A. (2020). Hybrid
our strategy with that of SAFE for four examples of Wolf-Bat Algorithm for Optimization of
malware. Plainly, our methodology is a lot quicker Connection Weights in Multi-layer
than SAFE. Perceptron. ACM Transactions on
Multimedia Computing, Communications,
V. CONCLUSION and Applications (TOMM), 16(1s), 1-20.
This paper proposes a hybrid approach for automatic
[4] Prasanna, S., & Ezhilmaran, D. (2016).
Association rule mining using enhanced
generation of signatures for malware executable of
apriori with modified GA for stock
all sizes with an aim to be utilized by rapid malware
prediction. International Journal of Data
sifting gadgets. We consider the way that huge
Mining, Modelling and Management, 8(2),
executables are involved generous measures of code
195-207.
that begins from the fundamental standard
improvement stages and is therefore duplicated [5] Pustokhina, I. V., Pustokhin, D. A., Gupta,
crosswise over different cases of both benign and D., Khanna, A., Shankar, K., & Nguyen, G.
malware created by these stages. So as to limit the N. (2020). An effective training scheme for
danger of false positive characterization of favorable deep neural network in edge computing
executables as malware, we propose and assess a enabled Internet of medical things (IoMT)
strategy to dispose of signature up-and-comers that systems. IEEE Access, 8, 107112-107123.
contain such duplicated lumps of code. The primary [6] Shankar, K., Zhang, Y., Liu, Y., Wu, L., &
advantage of the proposed strategy is that it Chen, C. H. (2020). Hyperparameter tuning
empowers examination at the binary level and deep learning for diabetic retinopathy fundus
doesn't require a semantic interpretation of code into image classification. IEEE Access, 8, 118164-
capacity blocks utilizing systems, for example, code 118173.
markers, dismantling, state-machines and so forth. [7] Joshi, G. P., Perumal, E., Shankar, K., Tariq,
This advantage implies that the procedure is U., Ahmad, T., & Ibrahim, A. (2020). Toward
conventional and isn't influenced by changes in CPU Blockchain-Enabled Privacy-Preserving Data
or presentation of new improvement stages. By and Transmission in Cluster-Based Vehicular
by, undertakings which might want acknowledge Networks. Electronics, 9(9), 1358.
Auto-Sign in creating signatures for high-throughput [8] Saračević, M. H., Adamović, S. Z., Miškovic,
arrange security machines need to pursue an V. A., Elhoseny, M., Maček, N. D., Selim, M.
increasingly thorough and orderly strategy for M., & Shankar, K. (2020). Data Encryption
structure their CFL vault. Thinking about the for Internet of Things Applications Based on
worldwide assortment of advancement stages and Catalan Objects and Two Combinatorial
the portability of dangers encouraged by the Internet, Structures. IEEE Transactions on Reliability.
guaranteeing the outside legitimacy of this [9] Namasudra, S., & Roy, P. (2017). Time
examination depends significantly on arriving at a saving protocol for data accessing in cloud
minimum amount of CFL records which speaks to computing. IET Communications, 11(10),
inexhaustible improvement stages. Besides, it 1558-1565.
frequently doesn't get the job done for a signature to
be accessible sent signature must be overseen,
[10] Elsir, A., Elsier, O., Abdurrahman, A., &
Mubarakali, A. (2019). Privacy Preservation
dispersed and stayed up with the latest by security
in Big Data with Data Scalability and
executives.
Efficiency Using Efficient and Secure Data
Balanced Scheduling Algorithm.
REFERENCE
[11] Ezhilarasu, P., Krishnaraj, N., & Babu, S. V.
[1] Anand Nayyar, Vikram Puri, (2015). Applications of finite automata in text
Nhu Gia Nguyen, Dac Nhuong Le, Smart search-a review. International Journal of

ISSN: 2393-9516 www.ijetajournal.org Page 19

International Journal of Engineering Trends and Applications (IJETA) – Volume 7 Issue 5, Sep-Oct 2020

Science, Engineering and Computer [22] Sankhwar, S., Gupta, D., Ramya, K. C., Rani,
Technology, 5(5), 116. S. S., Shankar, K., & Lakshmanaprabu, S. K.
[12] Huyen, D.T.T., Binh, N.T., Tuan, T.M., (2020). Improved grey wolf optimization-
Nguyen, G.N, Dey, N., Son, L.H, Analyzing based feature subset selection with fuzzy
trends in hospital-cost payments of patients neural classifier for financial crisis
using ARIMA and GIS: Case study at the prediction. Soft Computing, 24(1), 101-110.
Hanoi Medical University Hospital, Vietnam, [23] Namasudra, S., & Deka, G. C. (2018).
Journal of Medical Imaging and Health Introduction of DNA computing in
Informatics, 7(2), pp. 421-429. cryptography. In Advances of DNA
[13] Prasanna, S., & Maran, E. (2015). Stock computing in cryptography (pp. 1-18).
Market Prediction Using Clustering with Chapman and Hall/CRC.
Meta-Heuristic Approaches. Gazi University [24] Mubarakali, A., Srinivasan, K., Mukhalid, R.,
Journal of Science, 28(3). Jaganathan, S. C., & Marina, N. (2020).
[14] Pustokhina, I. V., Pustokhin, D. A., Security challenges in internet of things:
Rodrigues, J. J., Gupta, D., Khanna, A., Distributed denial of service attack detection
Shankar, K., ... & Joshi, G. P. (2020). using support vector machine‐based expert
Automatic Vehicle License Plate Recognition systems. Computational Intelligence.
using Optimal K-Means with Convolutional [25] Le Nguyen Bao, Dac-Nhuong Le, Gia Nhu
Neural Network for Intelligent Transportation Nguyen, Vikrant Bhateja, Suresh Chandra
Systems. IEEE Access. Satapathy (2017), Optimizing Feature
[15] Namasudra, S. (2018). Cloud computing: A Selection in Video-based Recognition using
new era. Journal of Fundamental and Applied Max-Min Ant System for the Online Video
Sciences, 10(2). Contextual Advertisement User-Oriented
[16] Uthayakumar, J., Elhoseny, M., & Shankar, System, Journal of Computational Science,
K. (2020). Highly Reliable and Low- Elsevier ISSN: 1877-7503. Vol.21, pp.361-
Complexity Image Compression Scheme 370.
Using Neighborhood Correlation Sequence [26] Ezhilarasu, P., Thirunavukkarasu, E.,
Algorithm in WSN. IEEE Transactions on Karuppusami, G., & Krishnaraj, N. (2015).
Reliability. Single substring based classification for
[17] Deepalakshmi, P., & Shankar, K. (2020). nondeterministic finite
Role and Impacts of Ant Colony Optimization automata. International Journal on
in Job Shop Scheduling Problems: A Detailed Applications in Information and
Analysis. Evolutionary Computation in Communication Engineering, 1(10), 29-31.
Scheduling, 11-35. [27] Bhateja, V., Gautam, A., Tiwari, A., Nhu,
[18] Ashwin, M., Kamalraj, S., & Azath, M. N.G., Le, D.-N, Haralick features-based
(2019). Multi objective trust optimization for classification of mammograms using SVM,
efficient communication in wireless M Advances in Intelligent Systems and
learning applications. Cluster Computing, Volume 672, 2018, Pages 787-
Computing, 22(5), 10687-10695. 795.
[19] Ezhilarasu, P., & Krishnaraj, N. (2015). [28] Latha, A., Prasanna, S., Hemalatha, S., &
Double Substring based Classification for Sivakumar, B. (2019). A harmonized trust
Nondeterministic Finite Automata. Indian assisted energy efficient data aggregation
Journal Of Science And Technology, 8, 26. scheme for distributed sensor
[20] Amira S. Ashour, Samsad Beagum, Nilanjan networks. Cognitive Systems Research, 56,
Dey, Ahmed S. Ashour, Dimitra Sifaki 14-22.
Pistolla, Gia Nhu Nguyen, Dac-Nhuong Le, [29] Krishnaraj, N., Elhoseny, M., Lydia, E. L.,
Fuqian Shi (2018), Light Microscopy Image Shankar, K., & ALDabbas, O. (2020). An
De-noising using Optimized LPA-ICI Filter, efficient radix trie‐based semantic visual
Neural Computing and Applications, indexing model for large‐scale image retrieval
Vol.29(12), pp 1517–1533, Springer, ISSN: in cloud environment. Software: Practice and
0941-0643. Experience.
[21] Prasanna, S., Govinda, K., & Kumaran, U. S. [30] Namasudra, S., Roy, P., Vijayakumar, P.,
(2012). An Evaluation study of Oral Cancer Audithan, S., & Balusamy, B. (2017). Time
Detection using Data Mining Classification efficient secure DNA based access control
Techniques. International Journal of model for cloud computing
Advanced Research in Computer environment. Future Generation Computer
Science, 3(1). Systems, 73, 90-105.

ISSN: 2393-9516 www.ijetajournal.org Page 20

International Journal of Engineering Trends and Applications (IJETA) – Volume 7 Issue 5, Sep-Oct 2020

[31] Lakshmanaprabu, S. K., Shankar, K., Rani, S. Cryptography (pp. 37-52). Chapman and
S., Abdulhay, E., Arunkumar, N., Ramirez, Hall/CRC.
G., & Uthayakumar, J. (2019). An effect of [42] Mubarakali, A., Ramakrishnan, J., Mavaluru,
big data technology with ant colony D., Elsir, A., Elsier, O., & Wakil, K. (2019).
optimization based routing in vehicular ad A new efficient design for random access
hoc networks: Towards smart cities. Journal memory based on quantum dot cellular
of cleaner production, 217, 584-593. automata nanotechnology. Nano
[32] Namasudra, S., & Deka, G. C. (Eds.). Communication Networks, 21, 100252.
(2018). Advances of DNA computing in [43] Ramakrishnan, J., Mavaluru, D., Sakthivel, R.
cryptography. CRC Press. S., Alqahtani, A. S., Mubarakali, A., &
[33] Mubarakali, A., Ashwin, M., Mavaluru, D., & Retnadhas, M. (2020). Brain–computer
Kumar, A. D. (2020). Design an attribute interface for amyotrophic lateral sclerosis
based health record protection algorithm for patients using deep learning
healthcare services in cloud network. NEURAL COMPUTING &
environment. Multimedia Tools and APPLICATIONS.
Applications, 79(5), 3943-3956. [44] Van, V.N., Chi, L.M., Long, N.Q., Nguyen,
[34] Dey, N., Ashour, A.S., Chakraborty, S., Le, G.N., Le, D.-N, A performance analysis of
D.-N., Nguyen, G.N, Healthy and unhealthy openstack open-source solution for IaaS cloud
rat hippocampus cells classification: A neural computing, Advances in Intelligent Systems
based automated system for Alzheimer and Computing, 380, pp. 141-150.
disease classification, Journal of Advanced [45] Sinha, A., Shrivastava, G., Kumar, P., &
Microscopy Research, 11(1), pp. 1-10 Gupta, D. (2020). A community‐based
[35] Krishnaraj, N., Ezhilarasu, P., & Gao, X. Z. hierarchical user authentication scheme for
Hybrid Soft Computing Approach for Industry 4.0. Software: Practice and
Prediction of Cancer in Colon Using Experience.
Microarray Gene Data. Current Signal [46] Namasudra, S., Devi, D., Kadry, S.,
Transduction Therapy, 11(2). Sundarasekar, R., & Shanthini, A. (2020).
[36] Namasudra, S., Deka, G. C., Johri, P., Towards DNA based data security in the
Hosseinpour, M., & Gandomi, A. H. (2020). cloud computing environment. Computer
The revolution of blockchain: State-of-the-art Communications, 151, 539-547.
and research challenges. Archives of [47] Mubarakali, A., Durai, A. D., Alshehri, M.,
Computational Methods in Engineering. AlFarraj, O., Ramakrishnan, J., & Mavaluru,
[37] Goel, N., Grover, B., Gupta, D., Khanna, A., D. (2020). Fog-Based Delay-Sensitive Data
& Sharma, M. (2020). Modified Grasshopper Transmission Algorithm for Data Forwarding
Optimization Algorithm for detection of and Storage in Cloud Environment for
Autism Spectrum Disorder. Physical Multimedia Applications. Big Data.
Communication, 101115. [48] Reshmi, T. R., & Azath, M. (2020). Improved
[38] Prasanna, S., Narayan, S., NallaKaruppan, M. self-healing technique for 5G networks using
K., Anilkumar, C., & Ramasubbareddy, S. predictive analysis. Peer-to-Peer Networking
(2019). Iterative Approach for Frequent Set and Applications, 1-17.
Mining Using Hadoop Over Cloud [49] Namasudra, S., Chakraborty, R., Majumder,
Environment. In Smart Intelligent Computing A., & Moparthi, N. R. (2020). Securing
and Applications (pp. 399-405). Springer, multimedia by using DNA based encryption
Singapore. in the cloud computing environment. ACM
[39] Le, D.-N.a, Kumar, R.b, Nguyen, G.N., Transactions on Multimedia Computing
Chatterjee, J.M.d, Cloud Computing and Communications and Applications.
Virtualization, [50] Patro, K. K., Reddi, S. P. R., Khalelulla, S. E.,
DOI: 10.1002/9781119488149, Wiley. Kumar, P. R., & Shankar, K. (2020). ECG
[40] Raj, R. J. S., Shobana, S. J., Pustokhina, I. V., data optimization for biometric human
Pustokhin, D. A., Gupta, D., & Shankar, K. recognition using statistical distributed
(2020). Optimal Feature Selection-Based machine learning algorithm. The Journal of
Medical Image Classification Using Deep Supercomputing, 76(2), 858-875.
Learning Model in Internet of Medical [51] Rajagopal, A., Joshi, G. P., Ramachandran,
Things. IEEE Access, 8, 58006-58017. A., Subhalakshmi, R. T., Khari, M., Jha, S., ...
[41] Namasudra, S., & Deka, G. C. (2018). & You, J. (2020). A Deep Learning Model
Taxonomy of DNA-based security models. Based on Multi-Objective Particle Swarm
In Advances of DNA Computing in Optimization for Scene Classification in

ISSN: 2393-9516 www.ijetajournal.org Page 21

International Journal of Engineering Trends and Applications (IJETA) – Volume 7 Issue 5, Sep-Oct 2020

Unmanned Aerial Vehicles. IEEE Access, 8, Notes in Networks and Systems. Springer,
135383-135393. 2019
[52] Chakchai So-In, Tri Gia Nguyen, Gia Nhu [62] Prasanna, S., & Narayanan, V. (2017). A
Nguyen: Barrier Coverage Deployment Novel Approach for Generation of All-
Algorithms for Mobile Sensor Networks. Optical OFDM Using Discrete Cosine
Journal of Internet Technology 12/2017; Transform Based on Optical Couplers in a
18(7):1689-1699. Radio-Over-Fiber Link. International Journal
[53] Mubarakali, A., Bose, S. C., Srinivasan, K., of Advanced Research in Engineering and
Elsir, A., & Elsier, O. (2019). Design a secure Technology, 8(3).
and efficient health record transaction [63] Sanjeevi, P., Prasanna, S., Siva Kumar, B.,
utilizing block chain (SEHRTB) algorithm for Gunasekaran, G., Alagiri, I., & Vijay Anand,
health record transaction in block R. (2020). Precision agriculture and farming
chain. Journal of Ambient Intelligence and using Internet of Things based on wireless
Humanized Computing, 1-9. sensor network. Transactions on Emerging
[54] Devaraj, A. F. S., Murugaboopathi, G., Telecommunications Technologies, e3978.
Elhoseny, M., Shankar, K., Min, K., Moon, [64] Rathi, V. K., Chaudhary, V., Rajput, N. K.,
H., & Joshi, G. P. (2020). An Efficient Ahuja, B., Jaiswal, A. K., Gupta, D., ... &
Framework for Secure Image Archival and Hammoudeh, M. (2020). A Blockchain-
Retrieval System Using Multiple Secret Share Enabled Multi Domain Edge Computing
Creation Scheme. IEEE Access, 8, 144310- Orchestrator. IEEE Internet of Things
144320. Magazine, 3(2), 30-36.
[55] Mubarakali, A. (2020). Healthcare Services [65] Khanna, A., Rodrigues, J. J., Gupta, N.,
Monitoring in Cloud Using Secure and Swaroop, A., & Gupta, D. (2020). Local
Robust Healthcare-Based BLOCKCHAIN mutual exclusion algorithm using fuzzy logic
(SRHB) Approach. MOBILE NETWORKS for Flying Ad hoc Networks. Computer
& APPLICATIONS. Communications.
[56] Namasudra, S. (2019). An improved
attribute‐based encryption technique towards
the data security in cloud
computing. Concurrency and Computation:
Practice and Experience, 31(3), e4364.
[57] Kathiresan, S., Sait, A. R. W., Gupta, D.,
Lakshmanaprabu, S. K., Khanna, A., &
Pandey, H. M. (2020). Automated detection
and classification of fundus diabetic
retinopathy images using synergic deep
learning model. Pattern Recognition Letters.
[58] Govinda, K., & Prasanna, S. (2015,
February). Medical dialysis prediction using
fuzzy rules. In 2015 International Conference
on Soft-Computing and Networks Security
(ICSNS) (pp. 1-5). IEEE.
[59] Sujatha, R., Navaneethan, C., Kaluri, R., &
Prasanna, S. (2020). Optimized Digital
Transformation in Government Services with
Blockchain. In Blockchain Technology and
Applications (pp. 79-100). Auerbach
Publications.
[60] Govinda, K., & Prasanna, S. (2015,
February). A generic image cryptography
based on Rubik's cube. In 2015 International
Conference on Soft-Computing and Networks
Security (ICSNS) (pp. 1-4). IEEE.
[61] Anand Nayyar, Vikram Puri,
Nhu Gia Nguyen, BioSenHealth 1.0: A Novel
Internet of Medical Things (IoMT) Based
Patient Health Monitoring System, Lecture

ISSN: 2393-9516 www.ijetajournal.org Page 22