MEKDA: Multi-Level ECC based Key Distribution and Authentication in Internet of Things

International Journal of Computer Networks & Communications (IJCNC) Vol.13, No.6, November 2021
DOI: 10.5121/ijcnc.2021.13602 19
MEKDA: MULTI-LEVEL ECC BASED KEY
DISTRIBUTION AND AUTHENTICATION IN
INTERNET OF THINGS
Padmashree M G, Mallikarjun J P, Arunalatha J S and Venugopal K R
Department of Computer Science and Engineering, University Visvesvaraya
College of Engineering, Bangalore University, Bengaluru, India
ABSTRACT
The Internet of Things (IoT) is an extensive system of networks and connected devices with minimal human
interaction and swift growth. The constraints of the System and limitations of Devices pose several
challenges, including security; hence billions of devices must protect from attacks and compromises. The
resource-constrained nature of IoT devices amplifies security challenges. Thus standard data
communication and security measures are inefficient in the IoT environment. The ubiquity of IoT devices
and their deployment in sensitive applications increase the vulnerability of any security breaches to risk
lives. Hence, IoT-related security challenges are of great concern. Authentication is the solution to the
vulnerability of a malicious device in the IoT environment. The proposed Multi-level Elliptic Curve
Cryptography based Key Distribution and Authentication in IoT enhances the security by Multi-level
Authentication when the devices enter or exit the Cluster in an IoT system. The decreased Computation
Time and Energy Consumption by generating and distributing Keys using Elliptic Curve Cryptography
extends the availability of the IoT devices. The Performance analysis shows the improvement over the Fast
Authentication and Data Transfer method.
KEYWORDS
Authentication, Elliptic Curve Cryptography, Internet of Things, Key Distribution, Security.
1. INTRODUCTION
The Internet of Things (IoT) has mainstreamed portable Internet applications and lead to the
standardization of Communication protocols [1]. IoT strategy is incorporated in modern
applications where the System exhibits limitations on communication overhead and information
transmissions [2], [3]. IoT device obtains access to the System through the System Network.
Authentication is an important and initial activity in security measures and data communication
functionalities. Many connected nodes that communicate with each other have limited
computational and battery power. The existing Authenticating and Key Distribution protocols to
validate the IoT devices and the systems lead to more resource utilization viz., Communication
cost, Memory, and Energy consumption further increases when multiple IoT devices initiate
concurrently.
The conventional information exchange framework uses the Cryptographic Key to secure
communicated information. When Data transfers through various nodes, the traditional mono
Key methodology is inappropriate. If this Key is compromised, then the complete communication
framework is compromised. The disclosure is enormous, as a high volume of information
transmission occurs in an IoT framework. Thus, constant verification and security are

20
predominant, complicated, and time-consuming. Robust security protection with identity privacy
preservation [34] and non-repudiation uses Bilinear Pairing.
IoT networks usually have triple-layer architecture, including the Physical Sensor, Network, and
Application layer. IoT nodes with different sensor characteristics are in the Physical layer. The
Network layer enables servers to receive the sensor data transmitted to them. In general,
gateways, routers, and packet-transfer devices connect the Application and the Physical layer.
Standard Authentication methods are efficient in gateways because these devices have high
computing power. Physical Sensor Layer nodes, conversely, need lightweight authentication
solutions.
The security of IoT depends mainly on the secrecy of the Group Key. In dynamic networks, the
devices frequently change the Groups leading to difficulty in securing shared crypto keys within
a group remarkably, a Symmetric-Key. Robust key management protocols require Forward and
Backward Secrecy and ensure immunity to Collusion attacks [4]. The compromised Group Key
results in IoT functionalities exposed to attacks viz., Man-In-The-Middle, Denial-of-Service, and
Replay attack [5], [6], [7].
Motivation: Authentication and Data transmission framework secure and validate the entire IoT
Group based on complete Signing and Encryption [1]. Key Generator Center generates partial
private Keys for IoT devices and avoids the Key Escrow problem [8]. The path among Devices,
Gateway and Server, is dependent on the network domain security. The inter Group secure data
transmission is not considered. A Single Key generated by the Key Generator Center for Group
Communication is vulnerable to IoT Group Key Compromise Attack. The proposed Multi-level
ECC-based Key Distribution and Authentication in IoT enhances the security by authenticating
the communicating resource-constrained devices using Dynamic Aggregate Cluster Key and
integrating Authentication and Data transmission.
Contributions: The main contributions of the proposed Multi-level Elliptic Curve Cryptography
(ECC) based Key Distribution and Authentication in IoT are:
i) Authenticate and enhance security using Multi-level Authentication when an IoT device
enters or exits the Cluster.
ii) Decrease the Computation Time and Energy Consumption and extend the availability of
the IoT devices by accomplishing Key distribution using Elliptic Curve Cryptography.
Organization: The paper is organized as follows: Section 2 presents the works Related to the
Cluster Key Distribution and Authentication; Section 3 abstracts the Background Work. Section 4
describes the Proposed Multi-level ECC-based Key Distribution and Authentication in IoT.
Section 5 analyses the Performance and security of the proposed MEKDA, and Section 6
concludes the work.
2. RELATED WORKS
The Authentication protocol using simple functions of Cryptography viz., the Advanced
Encryption Standard, and Hashing [9], the inter-network compatible security algorithm with
scalable and portable IoT device provides a solution to the issues above. Authentication deals
with the stolen devices without compromising the System services. The use of self-learning
Crossbar Adaptive Array Artificial Intelligence technique protects from data disclosure. An
Authentication and Data Sharing strategy [1] for an IoT environment allows multiple
authentications while transmitting the data between the IoT devices and the network. Mutual
Authentication is accomplished using the Aggregate Signing and Encryption process. The

21
Aggregated Sign Encryption method is resistant to Man-In-The-Middle and Replay attacks. The
use of a single Group Key leads to a vulnerable communication system. The Lattice Encryption
technology resists the Quantum Attack [9]. Mirai Attack, which controls the Actuators, can be
overcome using Authentication and Encryption [10].
Lin et al., [11] proposed a Twin layered Authentication approach using a lookup Attribute table.
A Topology Control Mechanism is used to distribute the Device and the Gateway Keys;
Authentication is performed using token or device information of the IoT device and the Gateway
managed by the Server withstanding the Denial-of-Service Attack. The IoT nodes are identified
by the Communicational attributes within the Group and by Physical Attributes for inter Group
device authentication. The approach authenticates for the IoT devices having identical Attributes
and provides security from the attacker while initializing the System.
Dammak et al., [12] presented a Decentralized Keys Exchange framework to provide secure
communication between the authenticated users of IoT environment using Master Token
Encryption in a Hierarchical Framework. The Single Group Key and multiple Member Keys are
updated when IoT devices change the Group association without the contribution of the
Members. Forward/Backward Secrecy is achieved with resistance to the Collusion Attack. Data
is accessible when the group member is compromised. Xu et al., [13] constructed a Searchable
Public Key Encryption method that generates the Ciphers comprising embedded structured
Keywords using a single multiply and pair functionality. The Cloud recognizes the Ciphers by
searching the Keyword List disclosed partial Keys.
Mansour et al., [4] introduced a Centralized Dynamic Group Symmetric and Asymmetric Key
Management approach using the Prime Number Factoring method. A single disconnected key is
generated simultaneously during registration. The computation time and the size of the Group
Key are directly proportional to the number of registered Devices in the Group. Aydin et al., [6]
propose a Centralized/Decentralized Group Authentication strategy with Symmetric Group Key
Encryption. An Elliptic Curve Diffie-Hellman Group Key Exchange distributes the Group Keys
among the Devices with Private Keys. Every Group Member shares the Partial Public Key with
all the Group members. The Authentication leads to an Encrypted Private Key sharing process
within the Group to obtain the Group Key. Thus security of the System depends on the
Authentication process.
Cheng et al., [14] designed a Group Member Authentication and Paired Key Establishment
strategy using Polynomial based Tokens. The scheme is resistive against Internal and External
Attacks. The use of XOR encryption and the Horner Polynomial Evaluation method reduces the
computation time. Zhang et al., [15] propounded an Anonymous Batch Authentication protocol
using Certificates generated using Simplex Hash Chain. The Device-wise Certificates are
maintained in a table for distribution while protecting the active Certificates. The authenticated
IoT devices are identified, and Dual Hash Seeds are generated for unauthenticated devices.
Mutual Authentication is achieved with resistance to Replay and Man-In-The-Middle Attack.
Wu et al., [16] proposed a Group based Encryption/Decryption Scheme for nested Groups with
Quad-Decryption using Single Key. The Group Key is generated and divided into Partial Keys to
distribute among the Group Members. The Partial Group Keys are shared among the Group
members using Second-degree Polynomial and recovered using Lagrange Interpolation. The
threshold value restricts the involvement of all the Group Members in Group Key generation.
Qiu et al., [17] proposed a Sign Encryption method to multicast the multiple data using Elliptic
Curve Cryptographic Scalar Point Multiplication functionality. This method secures data transfer
involving the Device and the Key Generating Authority. The Certificate-less Receiving Node

22
Identity is verified by the Gateway to reduce the computation cost. The Authentication
framework reduces the security of the data, as the Authenticity of the Receiver depends on the
Gateway. The authentication protocols use range [18], ECC [19], distributed [20], Unclonable
Function [21], Polynomial [22], Mono-Input Multi-Output [23], Cross-Domain-oriented [24] Key
management [25], [26]. Table 1 summarises the related works.
Table 1. Comparison of Related Works
Author Approach Advantages Disadvantages
Dammak et
al., [12]
(2020)
Decentralized Keys
Exchange, Hierarchical
Master Token
Encryption.
Single Group Key and multiple
Member Keys updated IoT
devices change the Group
association; Forward/Backward
Secrecy; resistance to the
Collusion Attack.
Data is accessible when
the group member is
compromised
Mansour et
al., (2020)
Centralized Dynamic
Group Symmetric and
Asymmetric Key
Management, Prime
Number Factoring
A single disconnected key is
generated simultaneously
during registration.
The computation time
and the size of the
Group Key are directly
proportional to the
number of registered
Devices in the Group.
Wu et al., [4]
(2019)
Group based Quad-
Decryption using Single
Key; Second-degree
Polynomial, Lagrange
Interpolation
Group Key into Partial Keys
distribute to Group Members.
The Threshold value
restricts the
involvement in Group
Key generation.
Qiu et al.,
[16] (2019)
multicast multiple data
using Elliptic Curve
Cryptographic Scalar
Point Multiplication
Secure data transmission
between the IoT Device and the
Key Generating Authority;
Certificateless Receiving Node
Identity is verified by the
Gateway reduce the
computation cost.
reduces the security of
the data: Authenticity of
the Receiver depends on
the Gateway.
Cao et al., [1]
(2019)
Aggregate Signing and
Encryption process.
resistant to Man-In-The-Middle
and Replay attacks.
single Group Key may
be vulnerable
3. BACKGROUND WORK
3.1. Low-Energy Adaptive Clustering Hierarchy
Low-Energy Adaptive Clustering Hierarchy (LEACH) is a routing protocol where nodes transmit
to Cluster Leader (CL), and the CL aggregate, compress and forward data to the Gateway. It
assumes each node consists of sufficient transmission power to reach the Gateway or the nearest
CL directly. But always using the transmission in its entirety dissipates energy. The repeated
participation of CL, q, is the preferred quantum of nodes selected as CLs. 1/q is the possibility of
each node that repeats as a CL. Every iteration ends with the non-CL nodes joining the Cluster
with nearby CL. The nodes transmit data following the time slice that CL provides. The nodes
use the least energy to reach the CL. The transmission power is switched on only in a given time
slice.
Characteristics: LEACH selects CL randomly using a Threshold value tn, or sensor with more
energy. Given n non-Leader devices in ith
iteration with q quantum of participation for CL
selection, the threshold 0 < tn < 1 is in equation (1).

23
tn = q/(1- q( i mod 1/q) (1)
Every non-Leader device not participated in the last 1/q iterations generates a random number r 
[0, 1]. If r < tn, then device n is designated as CL in the current iteration. The non-CL devices
attach to the nearest CL. Data aggregates at the CL. CL communicates directly with Gateway or
the User. Devices communicate with CL is via Time Division Multiple Access. Each Cluster uses
a different Code Division Multiple Access codes set to minimize intervention between clusters.
Drawbacks: CL selection ignores the residual energy of the devices. One-hop CL Gateway
transmissions increase the energy usage leading to the death of CL disconnecting the Cluster.
Time-based data transmission of LEACH protocol and heterogeneous and mobile characteristics
of IoT devices makes it inappropriate for IoT scenarios.
3.2. K-Nearest Neighbor Clustering
K-nearest neighbors (KNN) is a simple, supervised clustering Machine Learning algorithm that
compares attributes to infer values of new data points. i.e., the point assignment is derived from
its association with the other points. KNN clustering determines the distance from current data to
each row of existing data using Euclidean, Manhattan, or Minkowski distance. The Data is
distance-wise sort in ascending order and select the upper K rows. The new point categorizes into
a recurrent group of K rows. Given two points (Px, Py) and (Qx, Qy) at d dimension, Euclidean
Distance is the shortest distance between two points given by equation (2) for 2-dimension and
equation (3) for n dimensions.
distEuclidean = √ (Px - Qx)2
+ (Py - Qy)2
(2)
distEuclidean = √n
d=1 (Pd - Qd)2
(3)
Manhattan Distance is the sum of absolute dimensional distance between two points given by the
equation (4) and equation (5) for 2 and n-dimensions respectively.
distManhattan = | Px - Qx | + |Py - Qy | (4)
distManhattan = n
d=1 |Pd – Qd | (5)
Minkowski Distance is the universal representation of Euclidean and Manhattan Distance with
norm order N given by equation (6).
distMinkowski = N
√n
d=1 | Pd - Qd|N
(6)
It suits Clustering and regression, works efficiently on multi-cluster issues. Low K is responsive
to extremes, and a high K is resilient to extremes as it mediates more supporters to predict. The
storage and computation cost is more with inefficiency for vast unrelated and irregular data and
reduced scalability.
3.3. Problem Statement
Given the set of IoT devices, design an efficient Key Distribution and Authentication framework
for the Internet of Things applications to secure data of communicating resource-constrained
devices, the objectives of the proposed work are:
i) To enhance security and authenticate IoT devices by Multi-level Authentication when an
IoT device enters or exits the Cluster.

24
ii) To extend the availability of the IoT devices by decreasing the Computation Time and
Energy Consumption by accomplishing secure Key distribution using Elliptic Curve
Cryptography.
4. THE PROPOSED MULTI-LEVEL ECC BASED KEY DISTRIBUTION AND
AUTHENTICATION IN IOT
The Sending and Receiving Cluster components effectively encrypt and decrypt the message [30]
using Elliptic Curve Cryptography [29], [32].
4.1. System Architecture
The architecture of Multi-level ECC-based Key Distribution and Authentication in the IoT is
shown in Figure 1. The nodes group into Clusters in the network. Each Cluster consists of a
Cluster Leader. The Cluster Leader comprises more computation power compared to other
Cluster nodes. The Cluster Leader is used for inter Cluster communication to make the overall
network energy-efficient and efficiently use the energy of each node.
Figure 1. The Architecture of the Proposed MEKDA
4.2. System Block
The Block Diagram of the Multi-level ECC-based Key Distribution and Authentication
(MEKDA) in IoT is given in Figure 2.
The Initialization, Cluster Formation, Elliptic Curve Coordinators selection is processed before
the Cluster Head selection. The Coordinator is the Auxiliary Cluster Leader that overtakes the
group on Cluster Head energy below the threshold. The Authentication and Data Transmission
module is processed after the Cluster Key generation and Exchange between the communicating
devices.

25
Figure 2. Block Diagram of the Proposed MEKDA
4.3. The proposed Multi-level ECC based Key Distribution and Authentication
(MEKDA) Algorithm
MEKDA scheme consists of Setup, Signup, and Authentication modules. The Server (S)
performs Setup, Sign in, and Signup modules. Device D interacts with Gateway G to secure
mutual Authentication and session key distribution. Cluster Auxiliary Leader device
communicates with G via an intermediary Cluster Leader device CL. Cluster Leader device D
directly communicates with G by eliminating the intermediary device CL. The proposed MEKDA
protocol is based on ECC and a Hash function h : 0, 1* → 0, 1l
. Table-2 defines the notations.
Table 2. Table of Notations
Symbol Definition
Xk Exclusive Primary Secret Key
Xid Exclusive Secret Identity
Xid’ Diminutive Distinct Identity
Xa Identity with Hashed Secret
Xb Identity with Shared Secret
CAL Cluster Auxiliary Leader device
CL Cluster Leader devices
Xt Timeframe
Xtid Transient identity
t*
received Time of the Message
Δt Maximum Transmission Delay
h Hash
sk* Session Key
r Random Number
tn Threshold value
Dr, Dk Elliptic Curve Points
Setup module: The Server (S) configures the Gateway (G) in Function 1. The S selects a primary
secret key Gk for G and stores Gk in G. It configures the Elliptic Curve parameters viz., Generator
Point, Curve Coefficients etc., [32].

26
Function 1: Setup(), Setup
Input: Did
Output: store Gk
1: The S selects a primary secret key Gk for G
2: Stores Gk in G
4: return
Sign in module: The Server and the nodes share the ECC Encryption/Decryption Keys [31]
during Sign in to Encrypt and Decrypt messages during communication. The Elliptic-Curve-
based encryption/decryption keys are shared among the devices to provide confidentiality.
Sign up module: The Device interacts with S and exchanges the ECC and other system
parameters for Server authentication. The Server archives a device (D) as in Function 2:
Function 2: Signup(), Signup
Input: D
Output: Did, Did’, Da, Db
1: S selects an exclusive secret identity Did for D
2: S selects Dk for D
3: S computes Da = Did  h(Gk||Dk) and Db = Gk  Da  Dk
4: S selects an additional diminutive distinct identity Did’ for the Cluster Leader CL
5: S stores {Did’, Did, Da, Db} in Cluster Leader device CL
6: S stores {Did, Da, Db} in the Cluster Auxiliary Leader device CAL
7: S stores the diminutive distinct identity Did’ in Cluster Leader devices CL in G
8: return
Dk is exclusive to compute Da and Db. The identity Did, is the stable and secret Key for device D.
Cluster Leader device CL uses diminutive identity Did’ only when functions as an intermediary.
The module uses 3 Xor and a Hash function to store Did  h(Gk||Dk) and Gk {Did 
h(Gk||Dk)}Dk in CAL and CL.
Authentication module: Every node interacts within the Cluster via the Cluster Leader and the
Cluster communication via the Gateway. The use of a single cluster leader for authentication
along with the Auxiliary Leader that acts secondary node increases the lifetime of the existing
cluster. When the resources of the Cluster Leader fall below that of the Auxiliary Leader, the
Cluster Leader transfers the control to the Auxiliary Leader via the Gateway retaining the Cluster.
The use of ECC and Hashing improves the security level with irreversibility. The Cluster Leader
verifies the Auxiliary Leader and Gateway verifies the Cluster Leader. Thus the scheme provides
integrity, confidentiality in Cluster data communication.
Function 3: Authentication_D_CL()
Input: Did
Output: Dtid, Dy, Da, Db, Dt
1: D selects Dr
2: D generates a time-frame Dt
3: D computes Dx = Da  Did
4: D computes Dy = Dx  Dr
5: D computes the Transient identity Dtid = h(Did  Dt||Dr)
6: return
The device D authenticates with the Gateway G via an intermediary device CL as in equation (7)
and Function 3. It uses 3 Xor and 2 hash functions. h(Gk||Dk), h(Gk||Dk)  Dr and h(Did  Dt||Dr)
sent to CL with a freshness [33] entity Dt that addresses the Replay attack.

27
{Dtid, Dy, Da, Db, Dt}
D -------------------------> CL (7)
Equation (8) shows intermediate device CL forwarding data from device D to G after attaching
the identity CLid’ since G identifies CL using CLid’, not CLid. CL forwards h(Did  Dt||Dr),
h(Gk||Dk)  Dr, Did  h(Gk||Dk), Gk {Did  h(Gk||Dk)}Dk with authentication entity to G.
{Dtid, Dy, Da, Db, Dt, CLid’}
CL ---------------------------------> G (8)
The Gateway transfers the authentication details viz., {h(Gk||Dk)} Df, h(Dx||Dr||Df ||η||), { Dr
Df} {Did  h(Gk||D+
k )}, and { Dr Df}{ Gk  Did  h(Gk||D+
k) D+
k} to the Cluster Leader
as shown in equation (10) and Function 4 that uses 4 hash and 11 Xor. The validness of time-
frame (Dt) verify using equation (9) to mitigates the Replay attack.
t*
- Dt > δt (9)
where, t*
= receive time of the message and δt = max(transmission_delay).
{ α, β, η, μ, CLid’}
G --------------------------> CL (10)
Function 4: Authentication_G_CL()
Input: {Dtid, Dy, Da, Db, Dt, CLid’}
Output: α, β, η, μ, CLid’
1: if CLid’ unknown to G then
2: Abort
3: else
4: if (t*
- Dt > δt) then
5: Abort
6: else
7: G computes Dk* = Gk  Da  Db, Dx* = h(Gk||Dk* ),
Did* = Dx*  Da, Dr* = Dx*  Dy
8: G computes Dtid* = h(Did*  Dt||Dr* )
9: if Dtid <> Dtid* then
10: Aborts
11: else
12: G selects Df
13: G computes α = Dx  Df and  = Dr*  Df
14: G selects a new D+
k
15: G computes D+
a = Did  h(Gk||D+
k )
16: G computes D+
b = Gk  D+
a  D+
k
17: G computes η =   D+
a and  =   D+
b
18: G computes β = h(Dx||Dr||Df ||η||)
19: G computes and stores the session key, Session_Key = h(Did||Dr||Df ||Dx)
20: end if
21: end if
22: end if
23: return

28
CL forwards {h(Gk||Dk)} Df, h(Dx||Dr||Df ||η||), {Dr Df}{Did  h(Gk||D+
k )}, and { Dr
Df}{ Gk  Did  h(Gk||D+
k) D+
k} received from G to D, detaching identity CLid’ used for
authentication as shown in equation (11) and Function 5.
{ α, β, η, μ}
CL -----------------> D (11)
Function 5: Authentication_CL_D()
Input: Dtid, α, β, η, μ
Output: α, β, η, μ
1: D computes Df* = Dx  α
2: D computes β* = h(Dx||Dr||Df*|| η || μ)
3: if β <> β* then
4: Aborts
5: else
6: D computes  = Dr  Df*
7: D computes D+
a =   η
8: D computes D+
b =   μ
9: D computes and stores the session key sk*(= Session_Key) = h(Did||Dr||Df||Dx)
10: D updates (Da||Db) with (D+
a ||D+
b )
11: end if
12: return
Algorithm 1: MEKDA, Multi-level ECC-based Key Distribution and Authentication
Input: Set of Devices
Output: Cluster Key Distribution and Authentication of Cluster Members
1: Initialize the System using Multi-attributed Nearest Neighbour Clustering Algorithm
2: Selects the Cluster Leader CL and CAL using LEACH algorithm with maximum resources
within the Cluster
3: repeat
4: IoT devices connect to nearby Registered Cluster Server via Gateway using Setup()
5: S archives a device D using Signup()
6: D sends Transitory identity to CL using Authentication_D_CL()
7: intermediate CL forward Data with CLid’ using Authentication_CL_G()
8: G sends authentication details to CL using Authentication_G_CL()
9: CL forwards the data to D detaching CLid’ using Authentication_CL_D()
10: if IoT Device Exits the Cluster, then
11: ReKeying Triggered by CL and Distributed among the member IoT Devices
12: end if
13: until all Clusters obtain the Cluster Keys
14: The Cluster Leader Authenticates the IoT device via the Server
15: if valid. then
16: Sending IoT device Encrypts the data using Session Key
17: end if
18: The Cluster Leader Authenticates the IoT device via the Server
19: if valid, then
20: Receiving IoT Device with Cluster Key Session Key Decrypts the Data
21: end if
22: return
Algorithm 1 shows the process of Multi-level ECC-based Key Distribution and Authentication.
Few selected Cluster Leaders aggregate the data from nearby members and forward them to the
Gateway. Dynamic Cluster Leader aggregation and an Adaptive Clustering method reduce the
energy overhead. The Setup state selects the CLs. The Consequent state maintains the CL when
data transmits between the nodes. During the Setup state, deployment of nodes, broadcast input

29
message from the Gateway. The nodes generate an arbitrary Number r bounded by 0 and 1 and
compares with the threshold value tn. The node which generated the Random Number, if r < tn,
then becomes the CL; During the Consequent state, broadcasting of advertisement message to all
other nodes from the elected CLs. The nodes other than the CLs determine to which Cluster it
belongs, based on the strength of a received signal. After the Cluster Formation, Leaf nodes then
transmit the sensed information to their CL. Then CL transmits the collected data to the Gateway.
When the designated transmission time is over, start the setup phase to elect new CL, Advertising
message from CLs to other nodes, and Cluster formation. During the Consequent state, a new
cluster formation with a new node as the CL. Data aggregation at the CL and then to the Gateway
takes place, and the process repeats. The sequential flow of the messages and the computations
are depicted in Figure 3.
Figure 3. The sequential flow in the Proposed MEKDA
5. PERFORMANCE ANALYSIS
The proposed MEKDA for secure communication in IoT simplifies the authentication process
and alleviates the load of the network, ensuring strong security protection, user anonymity, and
non-repudiation. The simulation is conducted using java library SHA256Digest and EC256 in
Java IDE on a 32-bit Ubuntu Platform with 1GB RAM. The 100 nodes are grouped into multiple
Clusters. Every Cluster node interacts within the Cluster via the Cluster Leader and the Cluster
communication via the Gateway using 256-bit Hashing Digest, 256 bit Elliptic Curve Java library
as depicted in Figure 3. The parameters are stated in Table 3. The proposed MEKDA scheme
withstands a variety of security attacks with ideal efficiency. The performance of MEKDA is
compared with Fast Authentication and Data Transfer [1], Anonymous Mutual Authentication

30
and Key Agreement Scheme (AMAKS) [27], Uniform Privacy Preservation Group Handover
Authentication (UPPGHA) [28].
Table 3. Parameters
Nodes 100
Area 950x650m
Channel Wireless
Transmission range 150m
Initial Energy 100Joules
Timestamp, id , tid 64bits
Keylength 256 bits
Hash 256bits
Computation Time: The Computation Time of MEKDA is given in Figure 4, Figure 5, Figure 6,
and Table 4. The computation Time of MEKDA is 2 times of FADTS, 16 times of UPPGHA.
Key distribution with a reduction in the number of ECC, Hash, and Xor decreases Computation
Time. MEKDA uses 19 Xor, 8 Hash, and 2 scalar multiplications; FADTS uses 10 scalar
multiplications and 6 hash; UPPGHA uses 10 exponential operations. MEKDA Computation
Time is 4% more than AMAKS as it uses only 5 Hash and 17 Xor and ECC is irreversible but not
Xor. Thus MEKDA authenticates and reduces computation time extending the availability of IoT
devices providing confidentiality and integrity.
Figure 4. Computation Time of the Proposed MEKDA
Figure 5. Computation Time of the Proposed MEKDA with group size 5

31
Figure 6. Computation Time of the Proposed MEKDA with group size 10
Energy: Figure 7, Figure 8, Figure 9, and Table 5 depict the Energy Consumed by MEKDA. The
Energy Consumption of MEKDA is 2 times that of FADTS and 16 times that of UPPGHA. Key
distribution with a reduction in the number of ECC, Hash, and Xor decreases energy
consumption. MEKDA Energy Consumption is 4% more than AMAKS for 100 nodes of Group
Size 1, 5, 10 as AMAKS uses limited 3 Hash and 6 Xor and ECC is irreversible but not Xor. The
decrease in Energy Consumption extends the availability of the IoT devices by accomplishing
Key distribution using Elliptic Curve Cryptography enhancing security by authenticating
providing confidentiality and integrity.
Figure 7. Energy consumed by the Proposed MEKDA
Figure 8. Energy consumed by the Proposed MEKDA with group size 5

32
Figure 9. Energy consumed by the Proposed MEKDA with group size 10
Table 4. Computation Time (in milliSeconds)
FADTS AMAKS UPPGHA MEKDA
Groups 1 5 10 1 5 10 1 5 10 1 5 10
10 102.96 134.08 172.98 4.98 24.90 49.80 87.53 429.23 856.35 5.19 26.64 51.90
20 196.56 227.68 266.58 9.78 48.90 97.80 172.43 853.73 1705.35 10.19 51.06 101.93
30 290.16 321.28 360.18 14.58 72.90 145.80 257.33 1278.23 2554.35 15.20 76.12 151.96
40 383.76 414.88 453.78 19.38 96.90 193.80 342.23 1702.73 3403.35 20.20 101.18 201.98
50 477.36 508.48 547.38 24.18 120.90 241.80 427.13 2127.23 4252.35 25.20 126.24 252.01
60 570.96 602.08 640.98 28.98 144.90 289.80 512.03 2551.73 5101.35 30.20 151.30 302.04
70 664.56 695.68 734.58 33.78 168.90 337.80 596.93 2976.23 5950.35 35.21 176.36 352.06
80 758.16 789.28 828.18 38.58 192.90 385.80 681.83 3400.73 6799.35 40.21 201.42 402.09
90 851.76 882.88 921.78 43.38 216.90 433.80 766.73 3825.23 7648.35 45.21 226.48 452.12
100 945.36 976.48 1015.38 48.18 240.90 481.80 851.63 4249.73 8497.35 50.21 251.54 502.14
Table 5. Energy Consumed (in milliJoules)
Groups 1 5 10 1 5 10 1 5 10 1 5 10
10 12.23 15.93 20.55 0.59 2.96 5.92 10.40 50.99 101.73 0.62 3.16 6.17
20 23.35 27.05 31.67 1.16 5.81 11.62 20.48 101.42 202.60 1.21 6.07 12.11
30 34.47 38.17 42.79 1.73 8.66 17.32 30.57 151.85 303.46 1.81 9.04 18.05
40 45.59 49.29 53.91 2.30 11.51 23.02 40.66 202.28 404.32 2.40 12.02 24.00
50 56.71 60.41 65.03 2.87 14.36 28.73 50.74 252.71 505.18 2.99 15.00 29.94
60 67.83 71.53 76.15 3.44 17.21 34.43 60.83 303.14 606.04 3.59 17.97 35.88
70 78.95 82.65 87.27 4.01 20.07 40.13 70.91 353.58 706.90 4.18 20.95 41.83
80 90.07 93.77 98.39 4.58 22.92 45.83 81.00 404.01 807.76 4.78 23.93 47.77
90 101.19 104.89 109.51 5.15 25.77 51.54 91.09 454.44 908.62 5.37 26.91 53.71
100 112.31 116.01 120.63 5.72 28.62 57.24 101.17 504.87 1009.49 5.97 29.88 59.65
Availability: Cluster Leaders nodes aggregate the data from nearby nodes and forward them to
the Gateway. Dynamic Cluster Leader aggregation and adaptive Clustering reduce the energy
overhead. MEKDA enhances security using Multi-level Authentication when an IoT device
enters or exits the Cluster. The availability of the IoT devices extends by decreasing the
Computation Time and Energy Consumption accomplishing secure Key distribution using

33
Elliptic Curve Cryptography. Elliptic Curve cryptography consumes more time than Xor but
provides more security.
The randomly distributed IoT nodes group into clusters of varied k values using the kNN
algorithm. The priority is given towards the Authentication and Key distribution process while
maximizing the system availability node drain out energy due to clustering connectivity issues.
The mean values with 95% confidence interval are compared in Figure 10 and Figure 11,
respectively.
Figure 10. Mean Computation Time of the Proposed MEKDA
Figure 11. Mean Energy consumed by the Proposed MEKDA
The use of a single cluster leader for authentication along with the Auxiliary Leader that acts
secondary node increases the lifetime of the existing cluster. The use of 256bit ECC and hashing
improves the security level to 128 bits with irreversible encryption. The Cluster Leader verifies
the Auxiliary Leader and Gateway verifies the Cluster Leader. Thus the MEKDA scheme
enhances security by authentication and secure key distribution providing integrity,
confidentiality in Cluster data communication.

34
5.1. Security Analysis
The points on the ECC-256 curve are the keys used for Cryptic functions. Gk is the System Key
shared to all the nodes. The Elliptic curve provides resistance from Replay, DoS, and MiTM
attacks verified using SPAN Simulator. The formal verification proves the Key exchange and
authentication process.
The Server selects an arbitrary secret and computes the public key Gk, and shares it. Did Dk, are the
preshared keys provided during the Setup and Signing phase. Gateway computes Da, Db.
Da = Did  h(Gk||Dk)
Db = Gk  Da  Dk
=Gk {Did  h(Gk||Dk)}Dk
The device selects two arbitrary secrets, Dr and Dt, used as an intermediary for Key Exchange and
Authentication. It computes Dx, Dy, Dtid to derive later by the Gateway for Authentication.
Dx = Da  Did
= {Did  h(Gk||Dk)}  Did
=h(Gk||Dk)
Dy = Dx  Dr
= h(Gk||Dk)  Dr
Dtid = h(Did  Dt||Dr)
Gateway receives Dtid, Dy, Da, Db, Dt, CLid’ and initially, verifies the Cluster Leader by verifying
the known CLid’. Then if the message is received within the possible transmission delay (t*
-
Dt < δt), it proceeds. G computes a series of statements viz., Dk*, Dx*, Did*, Dr*, Dtid*
using known and received variables to authenticate the device.
Dk* = Gk  Da  Db,
= Gk  { Did  h(Gk||Dk) } { Gk {Did  h(Gk||Dk)}Dk }
=Dk
Dx* = h(Gk||Dk* )
Did* = Dx*  Da
= {h(Gk||Dk) } {Did  h(Gk||Dk)}
= Did
Dr* = Dx*  Dy
={ h(Gk||Dk )} {h(Gk||Dk)  Dr}
=Dr
Dtid* = h(Did*  Dt||Dr*) = h(Did  Dt||Dr)
The Gateway authenticates the device by comparing the received and derived Transient
identities Dtid and Dtid*, respectively. Gateway selects Df, a partial session key. Gateway
computes α, , η, μ used by the device to derive the partial keys.
α = Dx*  Df
= {h(Gk||Dk)} Df
 = Dr*  Df

35
= Dr Df
Gateway selects an intermediary secret D+
k. and computes D+
a, D+
b to compute η, μ, β.
D+
a = Did  h(Gk||D+
k )
D+
b = Gk  D+
a  D+
k
= Gk { Did  h(Gk||D+
k)} D+
k
η =   D+
a
={ Dr Df} {Did  h(Gk||D+
k )}
 =   D+
b
={ Dr Df}{ Gk  Did  h(Gk||D+
k) D+
k}
β = h(Dx||Dr||Df ||η||)
The session key is sk*=h(Did||Dr||Df ||Dx). Gateway sends {α, β, η, μ}={h(Gk||Dk) Df ,h(Dx||Dr||Df
||η||), Dr Df Did  h(Gk||D+
k ), Dr Df Gk  Did  h(Gk||D+
k) D+
k} to the Device via the
Cluster Head and the Auxiliary Cluster Head. The device, on receiving the required parameters,
computes Df*, β* derives the session key.
Df* = Dx  α
={h(Gk||Dk)} {h(Gk||Dk) Df}
=Df
β* = h(Dx||Dr||Df*|| η || μ)
The Device validates the message authentication by verifying recieved β and derived β*.
It computes , D+
a, and D+
b.
 = Dr  Df*
= Dr  Df
D+
a =   η
={ Dr  Df} {Dr Df Did  h(Gk||D+
k )}
= Did  h(Gk||D+
k )
D+
b =   μ
= {Gk  Did }{ h(Gk||D+
k) D+
k}
sk*= h(Did||Dr||Df||Dx)
Thus, the device receives the session key without revealing any secrets to the intermediary
Cluster Head of the Cluster Auxiliary Leader. When the Cluster Leader resource is depleted, the
Auxiliary Leader node acts as the prospective leader.
Table 6 compares the computation functionality of MEKDA with FADTS [1], AMAKS [27], and
UPPGHA [28]. FADTS requires Scalar Multiplications with Hashing, AMAKS uses XoR and
Hashing function, UPPGHA uses Exponential function. MEKDA uses XoR and scalar
multiplication leads to a reduction in time complexity, and Hashing provides integrity. The XoR
reduces the time complexity but the reversibility breaches security; the use of scalar elliptic curve
multiplication enhances the security by concealing the secrets in the network. The messages
received by an adversary cannot retrieve the keys due to DDH and CDH hypothesis being
unsolvable in polynomial time.

36
Table 6. Comparison of Computation
Device 5Tm + 2Th 6Tx + 3Th 5Te 7Tx + 3Th + Tm
Gateway 5Tm + 4Th 11Tx + 5Th 5Te 12Tx + 5Th + Tm
Time of Tm: Scalar Multiplication; Te: Modular Exponentiation; Tx: XoR; Th: Hashing;
Thus MEKDA shows better performance than other algorithms due to the use of ECC and
reduced communications but the bandwidth. However, the real-time implementations may vary
due to the heterogeneous IoT device characteristics.
6. CONCLUSIONS
In an IoT system, the deployed devices generate a significant number of sensitive data. The
Authentication of a wide range of devices deployed in the IoT environment, and Key Distribution
among them, is a security challenge. The security of IoT depends mainly on the secrecy of the
Cluster Key. The Multi-level Elliptic Curve Cryptography-based Key Distribution and
Authentication in the Internet of Things enhances the security of the System by Multi-level
Authentication of the IoT devices when enters or exits the Cluster. The availability of the IoT
devices extended by decreased Computation Time and Energy Consumption accomplishing
secure Key distribution using Elliptic Curve Cryptography. The Computation Time and Energy
Consumption of MEKDA are two times that of FADTS and 16 times that of UPPGHA. The
decrease in computation time and Energy Consumption extends the availability of the IoT
devices by accomplishing Key distribution using ECC and Hashing with Xor enhancing security
by authentication providing confidentiality and integrity. The performance of MEKDA may vary
in a hardware setup with a large number of real-time IoT devices. Hybrid Authentication can
further improve performance.
CONFLICTS OF INTEREST
The authors declare no conflict of interest.
REFERENCES
[1] J. Cao, P. Yu, M. Ma, and W. Gao, “Fast Authentication and Data Transfer Scheme for Massive NB-
IoT Devices in 3GPP 5G Network,” IEEE Journal on Internet of Things, vol. 6, no. 2, pp. 1561–
1575, 2019.
[2] U. Khadam, M. M. Iqbal, M. Alruily, M. A. Al Ghamdi, M. Ramzan, and S. H. Almotiri, “Text Data
Security and Privacy in the Internet of Things: Threats, Challenges, and Future Directions,” Hindawi
Journal on Wireless Communications and Mobile Computing, vol. 2020, pp.1-15, 2020.
[3] J. Dizdarevi´c, F. Carpio, A. Jukan, and X. Masip-Bruin, “A Survey of Communication Protocols for
Internet of Things and Related Challenges of Fog and Cloud Computing Integration,” ACM Journal
on Computing Surveys, vol. 51, no. 6, pp. 1–30, 2019.
[4] A. Mansour, K. M. Malik, A. Alkaff, and H. Kanaan, “ALMS: Asymmetric Lightweight Centralized
Group Key Management Protocol for VANETs,” IEEE Transactions on Intelligent Transportation
Systems, vol. 22, no. 3, pp. 1663 - 1678, 2021.
[5] S. Dey and A. Hossain, “Session-Key Establishment and Authentication in a Smart Home Network
Using Public Key Cryptography,” IEEE Sensors Letters, vol. 3, no. 4, pp. 1–4, 2019.
[6] Y. Aydin, G. K. Kurt, E. Ozdemir, and H. Yanikomeroglu, “A Flexible and Lightweight Group
Authentication Scheme,” IEEE Journal on Internet of Things, vol. 7, no. 10, pp. 10277 - 10287, 2020.

37
[7] J. Cao, M. Ma, and H. Li, “G2RHA: Group-to-Route Handover Authentication Scheme for Mobile
Relays in LTE-a High-Speed Rail Networks,” IEEE Transactions on Vehicular Technology, vol. 66,
no. 11, pp. 9689– 9701, 2017.
[8] J. Cao, P. Yu, X. Xiang, M. Ma, and H. Li, “Anti-Quantum Fast Authentication and Data
Transmission Scheme for Massive Devices in 5G NB-IoT System,” IEEE Journal on Internet of
Things, vol. 6, no. 6, pp. 9794–9805, 2019.
[9] S. Kavianpour, B. Shanmugam, S. Azam, M. Zamani, G. Narayana Samy, and F. De Boer, “A
Systematic Literature Review of Authentication in Internet of Things for Heterogeneous Devices,”
Hindawi Journal of Computer Networks and Communications, vol. 2019, pp.1-14, 2019.
[10] X. Jiang, M. Lora, and S. Chattopadhyay, “An Experimental Analysis of Security Vulnerabilities in
Industrial IoT Devices,” ACM Transactions on Internet Technology, vol. 20, no. 2, pp. 16:1–16:24,
2020.
[11] S. C. Lin, C. Y. Wen, and W. A. Sethares, “Two-Tier Device-based Authentication Protocol Against
PUEA Attacks for IoT Applications,” IEEE Transactions on Signal and Information Processing over
Networks, vol. 4, no. 1, pp. 33–47, 2018.
[12] M. Dammak, S. M. Senouci, M. A. Messous, M. H. Elhdhili, and C. Gransart, “Decentralized
Lightweight Group Key Management for Dynamic Access Control in IoT Environments,” IEEE
Transactions on Network and Service Management, vol. 17, no. 3, pp. 1742 - 1757, 2020.
[13] P. Xu, S. He, W. Wang, W. Susilo, and H. Jin, “Lightweight Searchable Public-Key Encryption for
Cloud-Assisted Wireless Sensor Networks,” IEEE Transactions on Industrial Informatics, vol. 14,
no. 8, pp. 3712– 3723, 2018.
[14] Q. Cheng, C. Hsu, and L. Harn, “Lightweight Noninteractive Membership Authentication and Group
Key Establishment for WSNs,” Hindawi Journal on Mathematical Problems in Engineering, vol.
2020, pp. 1–9, 2020.
[15] J. Zhang, H. Zhong, J. Cui, Y. Xu, and L. Liu, “An Extensible and Effective Anonymous Batch
Authentication Scheme for Smart Vehicular Networks,” IEEE Journal on Internet of Things, vol. 7,
no. 4, pp. 3462– 3473, 2020.
[16] Z. Y. Wu, “Group-Oriented Cryptosystem for Personal Health Records Exchange and Sharing,” IEEE
Access, vol. 7, pp. 146 495–146 505, 2019.
[17] J. Qiu, K. Fan, K. Zhang, Q. Pan, H. Li, and Y. Yang, “An Efficient Multi-Message and Multi-
Receiver Signcryption Scheme for Heterogeneous Smart Mobile IoT,” IEEE Access, vol. 7, pp. 180
205–180 217, 2019.
[18] H. Y. Chien, “Group-Oriented Range-Bound Key Agreement for Internet of Things Scenarios,” IEEE
Journal on Internet of Things, vol. 5, no. 3, pp. 1890–1903, 2018.
[19] A. K. Das, S. Member, M. Wazid, and Y. Park, “Provably Secure ECC based Device Access Control
and Key Agreement Protocol for IoT Environment,” IEEE Access, vol. 7, pp. 55 382–55 397, 2019.
[20] C. Esposito, M. Ficco, A. Castiglione, F. Palmieri, and A. De Santis, “Distributed Group Key
Management for Event Notification Confidentiality among Sensors,” IEEE Transactions on
Dependable and Secure Computing, vol. 17, no. 3, pp. 566 - 580, 2020.
[21] D. Liu, X. Liu, H. Zhang, H. Yu, W. Wang, L. Ma, J. Chen, and D. Li, “Research on End-to-End
Security Authentication Protocol of NB-IoT for Smart Grid based on Physical Unclonable Function,”
in Proceedings of the Eleventh IEEE International Conference on Communication Software and
Networks, pp. 239–244, 2019.
[22] Z. Mahmood, A. Ullah, and H. Ning, “Distributed Multiparty Key Management for Efficient
Authentication in the Internet of Things,” IEEE Access, vol. 6, pp. 29 460–29 473, 2018.
[23] Z. Gu, H. Chen, P. Xu, Y. Li, and B. Vucetic, “Physical Layer Authentication for Non-Coherent
Massive SIMO-Enabled Industrial IoT Communications,” IEEE Transactions on Information
Forensics and Security, vol. 15, pp. 3722–3733, 2020.
[24] Z. Qikun, G. Yong, T. Yu-An, Z. Quanxin, and W. Ruifang, “A Dynamic and Cross-Domain
Authentication Asymmetric Group Key Agreement in Telemedicine Application,” IEEE Access, vol.
6, pp. 24 064–24 074, 2018.
[25] P. Vijayakumar, M. S. Obaidat, M. Azees, S. H. Islam, and N. Kumar, “Efficient and Secure
Anonymous Authentication with Location Privacy for IoT-based WBANs,” IEEE Transactions on
Industrial Informatics, vol. 16, no. 4, pp. 2603–2611, 2020.
[26] C. S. Park and W. S. Park, “A Group-Oriented DTLS Handshake for Secure IoT Applications,” IEEE
Transactions on Automation Science and Engineering, vol. 15, no. 4, pp. 1920–1929, 2018.

38
[27] X. Li, M. H. Ibrahim, S. Kumari, A. K. Sangaiah, V. Gupta, and K. K. R. Choo, “Anonymous Mutual
Authentication and Key Agreement Scheme for Wearable Sensors in Wireless Body Area Networks,”
Elsevier Journal of Computer Networks, vol. 129, pp. 429–443, 2017.
[28] J. Cao, H. Li, M. Ma, and F. Li, “UPPGHA: Uniform Privacy Preservation Group Handover
Authentication Mechanism for mMTC in LTE-A Networks,” Hindawi Journal of Security and
Communication Networks, vol. 2018, pp.1-16, 2018.
[29] M. G. Padmashree, S. Khanum, J.S. Arunalatha, and K. R. Venugopal, “SIRLC: Secure Information
Retrieval using Lightweight Cryptography in HIoT,” in Proceedings of the IEEE International
Technical Conference of Region 10 (TENCON 2019), pp. 169-173, 2019.
[30] M. G. Padmashree, Ranjitha, J. S. Arunalatha, and K. R. Venugopal, “CKDAC: Cluster-Key
Distribution and Access Control for Secure Communication in IoT,” in Proceedings of the Seventh
IEEE Uttar Pradesh Section International Conference on Electrical, Electronics and Computer
Engineering (UPCON 2020), pp. 1-6, 2020.
[31] M. G. Padmashree, J. S. Arunalatha, and K. R. Venugopal, “ HPAKE: Hybrid Precocious
Authentication and Key Establishment in IoT,” in Proceedings of the FiftyThird IEEE International
Carnahan Conference on Security Technology (ICCST 2019), pp. 129-134, 2019.
[32] M. G. Padmashree, S. Khanum, J. S. Arunalatha, and K. R. Venugopal, ”ETPAC: ECC based Trauma
Plight Access Control for Healthcare Internet of Things,” Springer International Journal of
Information Technology, vol. 13, no. 4, pp. 1481–1494, 2021.
[33] B. Kapito1, M. Nyirenda1, and H. Kim, “Privacy-Preserving Machine Authenticated Key Agreement
for Internet of Things,” International Journal of Computer Networks & Communications (IJCNC),
vol. 13, no. 2, pp. 99-120, 2021.
[34] Lakshmi M and Prashanth C R, “Designing an Energy Efficient Clustering in Heterogeneous
Wireless Sensor Network,” International Journal of Computer Networks & Communications
(IJCNC), vol. 13, no. 1, pp. 75-92, 2021.
AUTHORS
Padmashree M. G. received the B.E. Degree in Computer Science & Engineering from
JNNCE, Shivamogga, Kuvempu University, Karnataka, India, in 1998 and the M.Tech.
Degree in Computer Science & Engineering from RVCE, Bengaluru, Visvesvaraya
Technological University, Karnataka, in 2011. She is currently pursuing a Ph.D. degree in
Computer Science and Engineering at Bangalore University, Bengaluru, India. She has
published 6 articles in refereed International Journals and Conferences. Her research
interest includes Scheduling Techniques in Operating Systems, Cryptography, Security in the Internet of
Things.
Mallikarjun received the B.E. Degree in Computer Science & Engineering from
Visvesvaraya Technological University, India, and the M.Tech. Degree in Information
Technology from Bangalore University, India.
Arunalatha J S is a Professor in the Department of Computer Science and Engineering at
University Visvesvaraya College of Engineering, Bangalore University, Bangalore, India.
She obtained her Bachelor of Engineering in Computer Science and Engineering, from
PES College of Engineering, Mandya, Mysore University, and she received her Master’s
degree in Computer Science and Engineering from Bangalore University, Bangalore. She
pursued her Ph.D. in the area of Biometrics and has published 17 articles in refereed
International Journals and Conferences. Her research interest is in Biometrics, Image Processing, IoT, Big
Data Analytics, and Web Mining.

39
Venugopal K R is currently the Vice-Chancellor of Bangalore University, Bangalore. He
obtained his Bachelor of Engineering from University Visvesvaraya College of
Engineering and received his Master’s degree in Computer Science and Automation from
Indian Institute of Science Bangalore. He was awarded a Ph.D. in Economics from
Bangalore University, and Ph.D. in Computer Science from Indian Institute of Technology,
Madras. He has a distinguished academic career and has degrees in Electronics,
Economics, Law, Business Finance, Public Relations, Communications, Industrial Relations, Computer
Science, and Journalism. He has authored and edited 77 books on Computer Science and Economics. He
has over 980 research papers to his credit. His research interests include Computer Networks, Wireless
Sensor Networks, Parallel and Distributed Systems, Digital Signal Processing, Data Mining, IoT, and
Cloud Computing. He received IEEE Fellow and ACM Distinguished Educator award from the USA for
his outstanding contributions to Computer Science and Engineering.

MEKDA: Multi-Level ECC based Key Distribution and Authentication in Internet of Things

More Related Content

What's hot (20)

Similar to MEKDA: Multi-Level ECC based Key Distribution and Authentication in Internet of Things (20)

More from IJCNCJournal (20)

Recently uploaded (20)

MEKDA: Multi-Level ECC based Key Distribution and Authentication in Internet of Things