Circuits, Systems, and Signal Processing (2023) 42:3812–3834

https://doi.org/10.1007/s00034-023-02322-5

Design and Implementation of a Novel Circuit-Based

Memristive Non-autonomous Hyperchaotic System
with Conservative and Offset Boosting for Applications
to Image Encryption

Nune Pratyusha1 · Santanu Mandal1

Received: 21 July 2022 / Revised: 7 February 2023 / Accepted: 8 February 2023 /

Published online: 27 February 2023
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2023

Abstract
The research on dissipative-chaotic systems has exhibited plenty of remarkable results
for past few decades. But unfortunately, the attackers can quickly rebuild the attrac-
tors in dissipative systems, causing multiple information security issues. Compared to
the dissipative system, the conservative system can efficiently prevent such external
attacks due to the presence of flows without any attractors. Therefore, the study of
conservative chaotic systems is essential for secure communications and engineering
applications. Presently, a few studies are there for non-autonomous conservative sys-
tems based on simple memristor. For this reason, a circuit-based new simple memristor
is designed and using this, a non-autonomous four-dimensional hyperchaotic conser-
vative system without an equilibrium point is proposed in this paper. The hyperchaotic
behavior and multiple offset boosting are explored by phase portraits, bifurcation dia-
grams, and Lyapunov exponent plots. It is found that due to the presence of applied
stimuli, the proposed system also produces time-varying hyperchaos and offset boost-
ing. Furthermore, different torus structures, triggered by initial conditions and control
parameters are observed here as the prominent property of the conservative system.
An analog circuit of the chaotic system is developed and executed experimentally
to verify the correctness of the numerical model. Concurrently, a new algorithm for
image encryption using the proposed hyperchaotic map is stated and the efficiency of
this algorithm is confirmed by several security tests.

Keywords Memristor · Hyperchaos · Conservative · Non-autonomous · Offset

boosting · Image encryption

B Santanu Mandal
[email protected]
Nune Pratyusha
[email protected]

1 School of Advanced Sciences, VIT-AP University, Amaravati, India

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 Circuits, Systems, and Signal Processing (2023) 42:3812–3834 3813

1 Introduction

Memristors contain several excellent characteristics in terms of designing a circuit,

including high-switching speed, low power consumption, high endurance, non-volatile
aspects, and it shows synaptic behaviors as well. These special features make the
memristor a promising candidate in modelling chaotic systems. Nano-fabrication of
memristors is very difficult and expensive, so numerous researchers have developed
circuit-based memristor emulators [1, 32, 34, 42, 48, 49] to explore the characteristics
applied to promising real-life applications, which are varying from neuromorphic cir-
cuit systems [2, 3] to chaotic circuit systems [26, 33]. Therefore, it is essential to study
and develop a memristor model and its analog circuit design to implement in different
engineering and chaotic applications. Chaos is widely discovered in nature, such as
medicine [11], biology [16], physics [24], and geology [22]. Chaos theory evolves
into a topic of curiosity and plays a significant prospect in engineering applications,
like synchronization and secured communication [13], audio and image encryption
[4, 25], random number generation [21], and weak signal detection [14].
As per Jafari’s classification, a system becomes conservative, if the divergences
equal to zero for all state variables [18]. Such systems hold prosperous dynamical
characteristics. Researchers categorize chaotic systems as conservative and dissipa-
tive, based on the system’s divergence. On the other hand, for a dissipative system, a
summation of Lyapunov exponent values is always negative, and it can have attractors.
Conservative systems are more promising than dissipative systems due to randomness,
energy perseverance, and ergodicity. Compared with dissipative systems, conservative
systems do not contain attractors [12]. So, reconstructing the system by rebuilding the
attractors can prevent the security and reliability of the encryption domain. Therefore,
conservative chaotic systems are more convenient for secure communications, pseudo-
random number generation, image encryption, and decryption analysis [36]. A. Wu
et al. [44] built various conservative four-dimensional chaotic systems on the Sprott-
A model and analyzed their dynamics using different mathematical simulations. In a
subsequent year, the same group of researchers presented a new five-dimensional con-
servative hyperchaotic system. J. P. Singh et al. [35] developed another conservative
system in 2018, where distinct non-hyperbolic equilibriums were studied. G. Qi [29]
presented six different four-dimensional Hamiltonian conservative systems in 2019,
where the maximal Lyapunov exponent value reaches 78,194.27; this characteristic
made the system more potential for applications in secure transmission and various
engineering fields [12]. Wang et al. presented a four-dimensional conservative chaotic
system with capacitor-inductor circuits. This system exhibits quasi-periodic, coexis-
tence, and chaotic flows [39]. However, some problems still deserve in-depth analysis
due to the conservative system’s extraordinary characteristics and application poten-
tial. For instance, modeling a conservative chaotic system with hyperchaotic features,
hidden behaviors, and other dynamic properties require to be systematically explored.
Memristors usually contain complex nonlinear dynamics, making better control
and easier chaotic systems [30]. A conservative system with a memristor exhibits sig-
nificant benefits and characteristics in representations of nonlinear aspects. Yet, most
of the existing memristor-based chaotic systems are dissipative systems, rather than
conservative chaotic systems. So far, a few exclusive investigations have been reported

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 3814 Circuits, Systems, and Signal Processing (2023) 42:3812–3834

in this regard. A four-dimensional memristor-based hyperchaotic conservative system

was developed by Vaidyanathan [37] in 2017. Y. Deng et al. (2021) presented a chaotic
circuit design with one capacitor and one memristor for a three-dimensional conser-
vative system, which shows quasi-periodic characteristics and transient-chaos [10]. In
2021, Z. Wang et al. introduced the four-dimensional non-Hamiltonian chaotic conser-
vative system with a three-terminal memristor [38]. Though, more extensive studies
are essential on further important dynamics of the memristive hyperchaotic system.
Nowadays, the study on offset boosting is a hot topic in chaos-based research due
to its significant applications in secure communications [7–9, 15, 19, 23, 46]. A non-
autonomous system is more robust and real time in nature rather than an autonomous
system. In 2019, Wu et al. presented a non-autonomous chaotic system using a step-
wise control pulse, which presents multi-butterfly attractors [45]. In [6], the authors
proposed a two-dimensional non-autonomous chaotic conservative system and exam-
ined initial conditions depending on quasi-periodic and chaotic flows. But in these
communications, no use of a memristor, no offset boosting behavior, and no real-
life application, like image encryption, pseudo-random number generation, etc., are
reported. In previous literature, the memristor-based non-autonomous conservative
system with hyperchaos and offset boosting is hard to find.
Inspired by the recent advancement, the following purposes are extensively studied
and investigated in this paper:

1. A memristor-based new non-autonomous conservative four-dimensional hyper-

chaotic system with no equilibrium point is presented.
2. The dynamical characteristics of the proposed chaotic system, like conservative-
ness, hyperchaos, hidden flows, offset boosting, etc., are investigated through three
different control parameters.
3. The proposed system also exhibits torus characteristics and other highly com-
plex chaotic dynamics, which are investigated using bifurcation diagrams, phase
portraits, and Lyapunov exponents plots.
4. Further, a hardware-based chaotic circuit is developed and experimented in the
laboratory to understand the chaotic dynamics and to realize the system’s feasibility
along with the newly designed memristor’s important fingerprints. This novel
non-autonomous conservative system also exhibits significant roles of external
stimulus factors like applied signal amplitude and frequency on chaotic dynamics,
conservativeness, and offset boosting features.
5. Using obtained chaotic system, a new algorithm scheme based on pseudo-random
number generation, for image encryption, is developed here. Histogram analysis,
correlation coefficients analysis, information entropy, number of changing pixel
rate (NPCR), and unified averaged changed intensity (UACI) are performed to
investigate the efficiency, accuracy, and robustness of the algorithm.

The remaining arrangement of the paper is as follows: In Sect. 2, the modeling

of a new memristive four-dimensional, non-autonomous, conservative system and its
associated dynamical characteristics are investigated in detail. Offset-boosting behav-
ior and multistability are analyzed in Sect. 3. The analog chaotic circuit design and
hardware-based laboratory experiments are demonstrated in Sect. 4. A new algorithm

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 Circuits, Systems, and Signal Processing (2023) 42:3812–3834 3815

for image encryption and some differential and statistical analysis are executed in
Sect. 5. The final conclusions are presented in last section.

2 Memristor-Based Non-autonomous Conservative Hyperchaotic

System

In 1976, Chua and Kang proposed a generic memristor model (voltage-controlled) as

i = g(x, v)v
ẋ = h(x, v) (1)

where v and i represent the input signal and output current of the memristor, respec-
tively, and the state variable is x. The function g(x, v) represents memductance and
h(x, v) represents continued scalar function.
In this paper, a new generic voltage-controlled memristive system is proposed as
follows:
 
i = ξ x3 + η v
ẋ = αv (2)

where ξ , η, and α are the constant parameters. The mathematical model of the pro-
posed memristor is simple and easy to implement; the fingerprints of the memristor
are verified using Eq. (2) by applying a sinusoidal signal (v = A sin(2π f t)) as input.
The proposed memristor model exhibits a pinched hysteresis loop, which is pinched
at the origin, and the hysteresis lobe area decreases with an increase in frequency
(20 Hz to 3 MHz) (Fig. 1). The memristor emulator circuit design is also developed
here (Fig. 12). Multisim simulations and the breadboard analysis are done for hard-
ware implementation of the proposed mathematical model of a simple memristor (2)
(Fig. 1b, c).
A non-autonomous hyperchaotic conservative system is presented based on the
proposed voltage-controlled memristor (2) as follows:

ẋ1 = δ(x2 − x3 )
  
ẋ2 = 1 β (−x1 + (ξ x43 + η)A sin(2π f t))
  
ẋ3 = 1 ρ x1
ẋ4 = α A sin(2π f t) (3)

where x 1 , x 2 , x 3 , x 4 are the state variables of the non-autonomous chaotic system with
parameters δ = 1, ξ = 4.7 × 103 , η = 103 , ρ = 1 × 10−5 , α = 1/(22 × 103 ), β =
1 × 10−6 , f = 30, A = 1 (amplitude of the applied signal) and the initial condition
(1,1,1,1). Phase-portraits of the proposed chaotic system, which are depicted in Fig. 2,
exhibit a chaotic invariant-torus, which is highly useful for secure communication and
cyber security systems.

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 3816 Circuits, Systems, and Signal Processing (2023) 42:3812–3834

Fig. 1 v − i characteristics of the proposed memristor a MATLAB simulation result with different frequen-
cies for (2) b Multisim simulation results with frequency f = 300 Hz c Breadboard experimental results
using an oscilloscope (Tektronix TBS 1072B-EDU) with frequency f = 3 MHz

The proposed non-autonomous system does not contain an equilibrium point; there-
fore, the system may generate hidden flows.

2.1 Conservative Property

The divergence of the system (3) for any initial condition is,

∂ ẋ1 ∂ ẋ2 ∂ ẋ3 ∂ ẋ4

div v = + + + =0
∂ x1 ∂ x2 ∂ x3 ∂ x4

which confirms that the proposed chaotic system (3) is a conservative system [12].

2.2 Dynamical Properties of the System as the Parameter ˇ Varies

Altering the value of β (from 1 × 10−6 to 2), and fixing the other parameters, the
phase portraits of the system (3) are illustrated in Fig. 3, and β-dependent Lyapunov
exponent plot (Fig. 4) is obtained using MATLAB. The Lyapunov exponents of the

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 Circuits, Systems, and Signal Processing (2023) 42:3812–3834 3817

Fig. 2 Different dynamical behaviors with different pair of state variables a x 2 –x 3 b x 3 –x 1 c x 1 –x 2 d x 4 –x 3

chaotic system are estimated by the Wolf algorithm [43].

The Lyapunov exponent values of system (3) are L 1 = 1.7468, L 2 = 1.0873, L 3 ≈0,
L 4 = − 2.8341. So, the system is hyperchaotic in nature as it has two positive Lyapunov
exponents. Further, the total Lyapunov exponent values (which is zero) represent the
conservativeness of the system. Kaplan–Yorke dimension (DKY) [37] is derived to
m
Lm
describe the nonlinear dynamics of the system. DKY = m + |Ln=1 m+1 |
, where m is
m m+1
the largest integer which satisfies n=1 L m ≥ 0 and n=1 L m < 0. Substituting
Lyapunov exponents in the above equation of DKY , the Kaplan–Yorke dimension of
the proposed chaotic system is derived as 4, equal to the dimension of the system’s state
variable. This result also confirms the conservativeness of the proposed system [10,
38]. The bifurcation diagram (Fig. 4b) is compatible with the investigation outcomes
of the Lyapunov exponent diagram (Fig. 4a).

2.3 Dynamical Properties of the System as the Parameter  Varies

When the parameter ρ changes from 1 × 10−6 to 1, and fixing other parameters (δ =
1, ξ = 4.7 × 103 , η = 103 , β = 1 × 10−6 , α = 1/(22 × 103 ), A = 1, f = 30 and

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 3818 Circuits, Systems, and Signal Processing (2023) 42:3812–3834

Fig. 3 Different phase-portraits in x 1 –x 2 plane with fixed δ = 1, ξ = 4.7 × 103 , η = 103 , ρ = 10−5 , α =
1/(22 × 103 ), A = 1, f = 30 and varied β: a β = 10−1 b β = 10−4 c β = 10−5 d β = 10−7

Fig. 4 a Lyapunov exponent plot, b bifurcation plot with varying parameter β (from 1 × 10−6 to 2),
δ = 1, ξ = 4.7 × 103 , η = 103 , ρ = 10−5 , α = 1/(22 × 103 ), A = 1, f = 30

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 Circuits, Systems, and Signal Processing (2023) 42:3812–3834 3819

Fig. 5 Different phase-portraits in x 2 –x 1 plane with fixed δ = 1, ξ = 4.7 × 103 , η = 103 , β = 10−6 , α =
1/(22 × 103 ), A = 1, f = 30 and varied ρ: a ρ = 1 b ρ = 10−4 c ρ = 10−6 d ρ = 10−7

the initial conditions is (1, 1, 1, 1)), the change of phase-portraits in x 2 –x 1 plane is

depicted in Fig. 5.
Four different Lyapunov exponents spectrum are plotted in Fig. 6a, where L 1 =
1.8182, L 2 = 1.1251, L 3 ≈0, L 4 = − 2.9433. Here, the system also presents hyperchaos
and conservativeness (because DKY = 4); in addition, the total value of all Lyapunov
exponents reaches zero. The bifurcation diagram (Fig. 6b) represents the hyperchaotic
dynamics of the proposed system, which is similarly depicted in Lyapunov spectrum
plot (Fig. 6a).
It indicates that the proposed non-autonomous chaotic system is highly complex in
nature and is evidently easy to encounter the hyperchaotic state.

2.4 Dynamical Properties of the System as the External Stimulus Parameter f

Varies

The phase portraits of system (3) with varying the values of f is illustrated in Fig. 7.

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 3820 Circuits, Systems, and Signal Processing (2023) 42:3812–3834

Fig. 6 a Lyapunov exponent plot, b bifurcation plot with varying parameter ρ (from 1 × 10−6 to 1),
δ = 1, ξ = 4.7 × 103 , η = 103 , β = 10−6 , α = 1/(22 × 103 ), A = 1, f = 30

Fig. 7 Different phase-portraits in x 2 –x 1 plane with fixed δ = 1, ξ = 4.7 × 103 , η = 103 , β = 10−6 , α =
1/(22 × 103 ), A = 1, ρ = 10−5 and varied f : a f = 15 b f = 40 c f = 70 d f = 120

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 Circuits, Systems, and Signal Processing (2023) 42:3812–3834 3821

Fig. 8 a Lyapunov exponent plot, b bifurcation plot with varying parameter f (from 1 to 150) δ = 1, ξ =
4.7 × 103 , η = 103 , β = 10−6 , α = 1/(22 × 103 ), A = 1, ρ = 10−5

The Lyapunov exponent diagram with varying parameter f (1 to 150), when the other
parameters δ = 1, ξ = 4.7 × 103 , η = 103 , β = 10−6 , ρ = 10−5 , α = 1/(22 × 103 ),
A = 1, and initial condition (1, 1, 1, 1) is illustrated in Fig. 8a.
The Lyapunov exponent values for system (3) with varying f are L 1 = 2.6187, L 2
= − 6.1779, L 3 = 3.5992, L 4 ≈ 0. The proposed system (3) has two positive Lya-
punov exponent values. So, the proposed system shows hyperchaotic nature. Again,
the system is recognized as a conservative system due to the Kaplan–Yorke dimension
is 4. The bifurcation plot (Fig. 8b) of system (3) with varying the values of f , which
is compatible with Lyapunov spectrum plot (Fig. 8a).

3 Two Distinct Offset Boosting Behaviors

This section investigates another essential characteristic of the proposed system: offset
boosting behavior. The chaotic system exhibits two types of offset boosting behaviors.
First, offset boosting occurs due to initial conditions, and second, it generates offset
boosting with change of amplitude of the applied signal.

3.1 Offset Boosting Behavior Due to Initial Conditions

Another characteristic of the proposed chaotic system is observed as the initial offset-
boosting. This characteristic appears with the evolution of initial conditions. The
proposed system shows the coexistence of multiple flows (tours) in Fig. 9 with identical
states by altering initial conditions with the remaining other parameters as constant.
In system (3), the parameters are selected as δ = 1, ξ = 4.7 × 103 , η = 103 , β =
10 , ρ = 10−5 , α = 1/(22 × 103 ), f = 30, A = 1, and initial conditions are chosen
−6

to be (1,1,1,x 4 (0)). Initially, x 4 (0) is taken as 1, when the value x 4 (0) is changed to
− 1, the phase portraits on x 4 –x 1 plane move parallel to the x 1 (Fig. 9a). Similarly,
when the values of x 4 (0) changes from 1 to − 1, the phase trajectories on x 2 –x 4 plane

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 3822 Circuits, Systems, and Signal Processing (2023) 42:3812–3834

Fig. 9 Offset boosting of the system (3) in a x 4 –x 1 plane and b x 2 –x 4 with fixed δ = 1, ξ = 4.7 × 103 , η =
103 , β = 10−6 , α = 1/(22 × 103 ), ρ = 10−5 , f = 30, A = 1 and initial conditions (1,1,1,1) and (1,1,1, −
1), respectively

move parallel to x 4 (Fig. 9b). Figure 9 exhibits not only shifting of the phase-portraits
diagram with varying initial conditions, but also changes its shapes. Therefore, the
proposed system shows an initial offset boosting characteristic with homogeneous
multistability [17].

3.2 Offset Boosting Behavior Due to Amplitude of Applied Voltage

Offset boosting behavior is analyzed by changing the values of A. Here, A (amplitude

of applied signal) acts as an offset boosting controller. In system (3), the parameters
are considered as δ = 1, ξ = 4.7 × 103 , η = 103 , β = 10−6 , ρ = 10−5 , α =
1/(22 × 103 ), f = 30, and initial condition is (1,1,1,1). Changing the values of A
from 3 to 8 (Fig. 10a), the phase trajectories in x 2 –x 1 move parallel to x 1 . Further,
considering values of A as 1 and 3 (Fig. 10b), the phase trajectories in x 2 –x 4 move up
and down parallel to x 2 . Figure 10 also exhibits the homogeneous multistability as the
shapes are also changing with A.
The Lyapunov exponent plot is illustrated in Fig. 11a for offset controller A (1 to
15), and other parameters δ = 1, ξ = 4.7 × 103 , η = 103 , β = 10−6 , ρ = 10−5 , α =
1/(22 × 103 ), f = 30, and initial condition (1, 1, 1, 1). The Lyapunov exponent values
of system with varying A are L 1 = 3.6958, L 2 = − 3.6958, L 3 = L 4 ≈0. It is observed
that out of four Lyapunov exponents, only one value is positive. Therefore, Fig. 11a,
b produces chaos. Similar to above cases, here also system is conservative with DKY
as 4.

4 Circuit Implementation and Hardware-Based Experiments

The hardware-based implementation of the theoretical non-autonomous chaotic sys-

tem is essential for real-life applications. It is important to recast system (3) into the

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 Circuits, Systems, and Signal Processing (2023) 42:3812–3834 3823

Fig. 10 a Offset boosting of the system (3) in the x 2 –x 1 plane with fixed δ = 1, ξ = 4.7×103 , η = 103 , β =
10−6 , α = 1/(22×103 ), ρ = 10−5 , f = 30, and values of A are 3, 5, and 8, respectively. b Offset boosting of
the system (3) in the x 2 –x 4 plane with fixed δ = 1, ξ = 4.7 × 103 , η = 103 , β = 10−6 , α = 1/(22 × 103 ),
ρ = 10−5 , f = 30, and values of A are 1 and 3, respectively

Fig. 11 a Lyapunov exponent plot, b bifurcation plot with offset controller A (from 1 to 15), δ = 1, ξ =
4.7 × 103 , η = 103 , β = 10−6 , α = 1/(22 × 103 ), f = 30, ρ = 10−5

following system:

ẋ1 = x2 − x3

1 1
ẋ2 = −6 −x1 + −3 (4.7x43 + 1) sin(2π f t)
10 10
1
ẋ3 = −5 x1
10
1
ẋ4 = sin(2π f t) (4)
22 × 103

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 3824 Circuits, Systems, and Signal Processing (2023) 42:3812–3834

Fig. 12 Circuit design of the hyperchaotic memristive system

Hardware-based execution is more effective and sufficient way to confirm the fea-
sibility of chaos in an actual intent. Hence, an analog chaotic circuit is designed here
(Fig. 12) for the conservative chaotic system (4). An analog multiplier (AD633AN)
is used to execute the nonlinear representations of the proposed model (4). From the
circuit (Fig. 12), the corresponding state equations can be represented as

1 1
ẋ1 = x2 − x3
R1 C 1 R2 C 1
1 1
ẋ2 = (−x1 ) + W (x4 )v
R6 C 2 R9 C 2
1
ẋ3 = x1
R10 C3
1
ẋ4 = v (5)
R7 C 4

where W (x4 ) = x43 R4 + R5 . The entire circuit design (Fig. 12), which is equivalent
to system (5), is developed and represented with four different circuit blocks. In this
circuit, the capacitances of the circuit are C1 = 1 µF, C2 = C3 = 1 nF, C4 = 22 µF,
and the resistances are R1 = R2 = 1 M , R4 = 4.7 K , R5 = R6 = R7 = R9 =
1 K , R10 = 10 K . Here, the applied voltage v = 0.8 V.
In this circuit design, Op-amp (TL082CD) is used and denoted by UXX, where
U2A performs the integration operation and U6A is inverting the output of U2A in
the first block. The Op-amp U7A is used in the second block to implement the inte-
gral operation. In third block, two Op-amps U8A and U9A are placed to perform
an integration and inversion, respectively. In fourth block, Op-amp U1A place a role
of integration, and three AD633AN multipliers (U3, U4, and U5) are used for mul-
tiplication operation. The last block (bounded by green dotted lines) represents the
memristor part proposed in Eq. (2), and the oscilloscope display is shown in Fig. 1c.
The experimental outcome of the breadboard analysis using proposed circuit design
are verified in digital oscilloscope (TBS 1072B-EDU) (Fig. 13). The outcomes of the

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 Circuits, Systems, and Signal Processing (2023) 42:3812–3834 3825

Fig. 13 a Breadboard schematic and phase-portrait in x 2 –x 3 which is similar to numerical simulation Fig. 2a.
b phase-portrait in x 3 –x 1 , which is similar to numerical simulation Fig. 2b

oscilloscope (Fig. 13a, b) are matching with numerical simulations (Fig. 2a, b).
It is difficult to precisely assign the initial condition of the memristor in the hardware
circuit, which is crucial to the generation of the initial offset boosting behaviors.
So, introducing different voltage amplitudes (1 V, 3 V), the amplitude-based offset
boosting phenomena are observed through hardware-based experiments in Fig. 14a,
b, respectively. In Fig. 14, the time series analysis for x 2 and x 4 exhibits shifting of
their positions with time.
However, some minor deviations exist in corresponding outcomes, because no ini-
tial condition is specified for the circuit. A hyperchaotic circuit with complex chaotic
dynamics is quite sensitive to outer circumstances which may restrict it to perform cor-
rectly. But considering all such sensitivities, a perfect circuit design is exhibited here,
for proposed memristor-based conservative hyperchaotic system (3). The hardware-
based laboratory experiment represents the feasibility of the proposed system.

5 Hyperchaos-Based Image Encryption Scheme

Here, a new algorithm for image encryption with the proposed hyperchaotic map is
described as follows:
Step 1: A color image (256 × 256) is the input for the encryption algorithm. Storing
the dimension of the image in variables.
Step 2: Hyperchaotic system (3) is used to generate a list of keys. Initializing blank
index lists to store the index of pixels.
Step 3: Populating x 1 , x 2 and x 3 indices; Rearranging x 1 , x 2 and x 3 keys to increase
randomness.

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 3826 Circuits, Systems, and Signal Processing (2023) 42:3812–3834

Fig. 14 Breadboard schematic and phase-portrait in x 2 –x 4 with amplitude a A = 1 V, b A = 3 V. The

offset-boosting characteristics are depicted in c and d through time series analysis

x1 key[i] > x1 key[ j]

x1 key[i], x1 key[ j] = x1 key[ j], x1 key[i]
x1 index[i], x1 index[ j] = x1 index[ j], x1 index[i]

Step 4: Shuffling input image pixels and storing them in an empty image.
Step 5: Altering the pseudo-random numbers generated into the digits between 0
and 255.
Defining key as

x4k = int((x4 key[l] ∗ pow(10, 5))%256)

and performing the XOR operation

Encryptedimage[i, j] = Encryptedimage[i, j]x4k

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 Circuits, Systems, and Signal Processing (2023) 42:3812–3834 3827

As a result, the encrypted image is created by the proposed algorithm.

5.1 Decryption

The decryption is comparable to the encryption algorithm, in contrast to some phases

which are executed inversely.

5.2 Analysis Using Histogram

A histogram for an image displays the density of the pixels for every grey level of
that image. Individual image has an exhaustive histogram, which can instantly present
the statistical differences between before and after the encryption procedure. The
encrypted image histogram must be uniform in gray value allocation to prevent from
decoding of the pattern by attackers. In this analysis, the image [Lena (Fig. 15a)] with
sizes of 256 × 256 is utilized to get the encrypted image (Fig. 15b) using the proposed
algorithm.
Figure 16 illustrates the histograms for Fig. 15a (with R, G, and B channels). It is

Fig. 15 Results obtained by simulations using a newly developed encryption algorithm a original input
image. b Encrypted image

Fig. 16 Histograms of input image

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 3828 Circuits, Systems, and Signal Processing (2023) 42:3812–3834

Fig. 17 Histograms of encrypted image

noticed that the histogram figures (Fig. 16a–c) are abrupt and fluctuating. Figure 17
illustrates the histograms analysis of the encrypted image. These figures present the
uniform distribution of pixels of encrypted image compared to the same for input
image. This phenomenon confirms that the proposed algorithm can resist any external
statistical attacks.

5.3 The Analysis Using Correlations

Generally, in original image, the pixels are located close to each other. In this process a
good encryption algorithm leads to very less correlation coefficients of adjacent pixels
for the encrypted image.
Correlations between two pixels are measured as follows [41]:

cov(δ, σ )
rδσ = √ √
D(δ) D(σ )

where
⎛ ⎞⎛ ⎞
n n n
1 ⎝δi − 1 1
cov(δ, σ ) = δ j ⎠⎝σi − σj⎠
n n n
i=1 j=1 j=1
⎛ ⎞2
n n
1 ⎝δi − 1
D(δ) = δj⎠
n n
i=1 j=1

Figure 18a–c shows the correlations of the adjacent pixels of input image Fig. 15a
in the diagonal, vertical, and horizontal directions, respectively. Figure 19a–c depicts
the correlations of adjacent pixels in encrypted image of Lena in diagonal, vertical,
and horizontal directions, respectively. Here, the correlation coefficients for pair of
adjacent pixels are near to zero.
Table 1 highlights the comparisons of existing works [5, 20, 50] with the proposed
algorithm to demonstrate the accuracy and effectiveness of this new algorithm. Here,
correlation coefficients are listed for different algorithms. Comparatively, the proposed

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 Circuits, Systems, and Signal Processing (2023) 42:3812–3834 3829

Fig. 18 Correlations of input image a diagonal b vertical c horizontal

Fig. 19 Correlations plots a diagonal b vertical c horizontal for encrypted image

Table 1 Comparison table of correlation coefficients between adjacent pixels

Ref Diagonal Vertical Horizontal

[20] − 0.00220 − 0.01680 0.04450

[50] − 0.01930 − 0.02264 − 0.02457
[5] 0.03230 − 0.00110 − 0.02080
Proposed algorithm at initial value (1,1,1,1) − 0.003233 0.0012954 0.001043
Proposed algorithm at initial value (1,1,1, − 1) − 0.003233 0.0012954 0.001043

Bold represent the results received in this paper from our proposed system are comparatively better than
that of other existing works

algorithm produces very less correlation coefficient value (near to zero) for the same
image of Lena. Therefore, the proposed image encryption algorithm is comparatively
more promising.
To check the offset-boosting behaviors on encryption test results, here in Table 1, the
correlation coefficient test values are calculated with different initial values (1,1,1,1)
and (1,1,1, − 1), respectively. It is observed that the offset boosting criteria does not
affect the test results, but the obtained results still present efficiency of the proposed
algorithm compared to [20, 50] and [5].

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 3830 Circuits, Systems, and Signal Processing (2023) 42:3812–3834

Table 2 Comparison table for

information entropy of Lena’s Ref R G B
image
[28] 7.9973 7.9975 7.9977
[20] 7.9278 7.9744 7.9705
[47] 7.9974 7.9971 7.9972
Using proposed scheme 7.9993 7.9992 7.9992

Bold represent the results received in this paper from our proposed
system are comparatively better than that of other existing works

5.4 Analysis on Information Entropy

Information entropy describes the overall features of information conceptions. It is

defined [40] as

N
log2
H (w) = p(wi )
p(wi )
i=1

where p(wi ) is the ratio of image grey values, and w denotes the gray-scale of image,
and H(w) is the maximum information entropy. The estimated information entropy
will be slightly less than the outstanding value. Thus, the information entropy of
the encrypted image of a good encryption algorithm should incredibly be near to 8
(for 256 × 256 image). Table 2 shows the comparisons of the information entropy for
encrypted image (Fig. 16b) of Lena using proposed scheme in context of other existing
algorithm references. The empirical data acquired here are closer to the expected value
8 in comparison with [20, 28, 47], which establishes again the efficiency of the image
encryption algorithm, designed in this paper.

5.5 Analysis of Differential Attacks

A good image encryption algorithm can able to prevent multiple attacks. Therefore, an
assessment on feasibility and safety of the proposed encryption algorithm is necessary
using NPCR and UACI tests. Such analysis is applied to two different encrypted images
with the same input image and encryption algorithm. The mathematical expressions
[27, 31] of NPCR and UACI are

1
NPCR = D(i, j) × 100
M×N
i, j
N M
|c1 (i, j) − c2 (i, j)|
UACI = × 100
M × N × 255
i=1 j=1

Here, M × N represents a dimension of each image. D(i, j) = 1, for c1 =

c2 , D(i, j) = 0, otherwise.

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 Circuits, Systems, and Signal Processing (2023) 42:3812–3834 3831

Table 3 NPCR and UACI values

of encrypted image of Lena Encrypted image NPCR UACI

R 100 33.6739
G 99.997 33.6742
B 99.9996 33.6765

Table 4 Comparison table for chaotic systems

Ref Offset boosting Memristor Autonomous/non-autonomous Hyper chaos

[20] No No Autonomous Yes

[47] No No Autonomous No
[50] No No Autonomous No
Proposed system Yes Yes Non-autonomous Yes

Bold represent the results received in this paper from our proposed system are comparatively better than
that of other existing works

If the NPCR value is nearer to 100 and the UACI value for encrypted image is near
34, then the encryption algorithm becomes highly secure. Adequate protection toward
encryption demands larger values of NPCR and UACI. Table 3 displays R, G, and B
values NPCR and UACI, which are close to 100 and 34, respectively. Therefore, the
presented algorithm can effectively defend against various differential attacks, and it
is secure enough.
In Table 4, the proposed system is compared with different chaotic systems [20, 47,
50]. This is observed that the non-autonomous chaotic system with offset-boosting,
hyperchaos exhibits better performance in encryption process compared to others.

6 Conclusions

This work proposes a new memristor-based, non-autonomous, four-dimensional, con-

servative, hyperchaotic system. The system shows hyperchaos with three control
parameters β, ρ, and f . As the DKY is equal to the dimension of the system (four), the
system exhibits a continuous conservative behavior along with the change of control
parameters. A series of MATLAB-based numerical simulations are performed using
Lyapunov exponent plots, bifurcation diagrams, and phase-portraits, to study various
interesting dynamical behaviors of the system. Moreover, two different categories of
offset boosting behaviors are observed due to altering initial conditions and applied
signal amplitudes. It is noticed that the external stimulus factors—frequency and ampli-
tude of the applied signal cause significant complex behaviors like hyperchaos and
offset boosting to the system. As per the author’s knowledge, such a non-autonomous
conservative system is rare to find in present works, which produces both hyperchaos
and offset boosting in the presence of a memristor in the system. A hardware-based

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 3832 Circuits, Systems, and Signal Processing (2023) 42:3812–3834

chaotic circuit is designed and experimented with a voltage-controlled new memris-

tor, to validate the proposed mathematical model. Finally, a novel image encryption
algorithm is developed using this hyperchaotic map to generate random keys. The use
of memristor in the model, conservativeness, offset-boosting and hyperchaotic nature
make the encryption scheme as more efficient. Different statistical and differential
analysis confirm that the proposed encryption algorithm is secure enough to use in
different communications and engineering applications.

Data availability The datasets analyzed during the current study are available from the corresponding author
on reasonable request.

Declarations

Conflict of interest The authors report that there is no competing interest to declare.

References
1. M.T. Abuelma’atti, Z.J. Khalifa, A new floating memristor emulator and its application in frequency-
to-voltage conversion. Analog Integr. Circuits Signal Process. 86(1), 141–147 (2016)
2. Y. Babacan, F. Kacar, K. Gürkan, A spiking and bursting neuron circuit based on memristor. Neuro-
computing 203, 86–91 (2016)
3. H. Bao, Y. Zhang, W. Liu, B. Bao, Memristor synapse-coupled memristive neuron network: synchro-
nization transition and occurrence of chimera. Nonlinear Dyn. 100(1), 937–950 (2020)
4. U. Cavusoglu, S. Panahi, A. Akgul, S. Jafari, S. Kacar, A new chaotic system with hidden attractor and
its engineering applications: analog circuit realization and image encryption. Analog Integr. Circuits
Signal Process. 98(1), 85–99 (2019)
5. L. Chen, H. Yin, T. Huang, L. Yuan, S. Zheng, L. Yin, Chaos in fractional-order discrete neural networks
with application to image encryption. Neural Netw. 125, 174–184 (2020)
6. M. Chen, C. Wang, H. Wu, Q. Xu, B. Bao, A non-autonomous conservative system and its reconstitution
in integral domain. Nonlinear Dyn. 103(1), 643–655 (2021)
7. L. Chunbiao, J.C. Sprott, Y. Lu, Z. Gu, J. Zhang, Offset boosting for breeding conditional symmetry.
Int. J. Bifurc. Chaos 6, 66 (2018)
8. L. Chunbiao, T. Lei, X. Wang, G. Chen, Dynamics editing based on offset boosting. Chaos 6, 66 (2020)
9. L. Chunbiao, X. Wang, G. Chen, Diagnosing multistability by offset boosting. Nonlinear Dyn. 90,
1335–1341 (2017)
10. Y. Deng, Y. Li, A memristive conservative chaotic circuit consisting of a memristor and a capacitor.
Chaos 30(1), 66 (2020)
11. D. Dharminder, U. Kumar, P. Gupta, A construction of a conformal Chebyshev chaotic map based
authentication protocol for healthcare telemedicine services. Complex Intell. Syst. 7(5), 2531–2542
(2021)
12. E. Dong, M. Yuan, S. Du, Z. Chen, A new class of Hamiltonian conservative chaotic systems with
multistability and design of pseudo-random number generator. Appl. Math. Model. 73, 40–71 (2019)
13. M. Feki, An adaptive chaos synchronization scheme applied to secure communication. Chaos Solitons
Fract. 18(1), 141–148 (2003)
14. A. Gokyildirim, Y. Uyaroglu, I. Pehlivan, A weak signal detection application based on hyperchaotic
Lorenz system. Tehnicki Vjesnik. 25(3), 701–708 (2018)
15. S. Gu, S. He, H. Wang, B. Du, Analysis of three types of initial offset-boosting behavior for a new
fractional-order dynamical system. Chaos Solitons Fract. 143, 110613 (2021)
16. M.L. Heltberg, S. Krishna, M.H. Jensen, On chaotic dynamics in transcription factors and the associated
effects in differential gene regulation. Nat. Commun. 10(1), 1–10 (2019)
17. L. Huang, W. Yao, J. Xiang, Z. Zhang, Heterogeneous and homogenous multistabilities in a novel 4D
memristor-based chaotic system with discrete bifurcation diagrams. Complexity 6, 66 (2020)

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 Circuits, Systems, and Signal Processing (2023) 42:3812–3834 3833

18. S. Jafari, J.C. Sprott, S. Dehghan, Categories of conservative flows. Int. J. Bifurc.Chaos 29(2), 1–16
(2019)
19. Y. Jiang, F. Yuan, Y. Li, A dual memristive Wien-bridge chaotic system with variable amplitude and
frequency. Chaos 30, 66 (2020)
20. A. Kadir, A. Hamdulla, W.Q. Guo, Color image encryption using skew tent map and hyperchaotic
system of 6th-order CNN. Optik 125(5), 1671–1675 (2014)
21. İ Koyuncu, M. Tuna, İ Pehlivan, C.B. Fidan, M. Alçın, Design, FPGA implementation and statistical
analysis of chaos-ring based dual entropy core true random number generator. Analog Integr. Circuits
Signal Process. 102(2), 445–456 (2020)
22. X. Li, C. Jiang, R. Xu, W. Yang, H.H. Wang, Y. Zou, Combining forecast of landslide displacement
based on chaos theory. Arab. J. Geosci. 14(3), 1–10 (2021)
23. C. Li, J.C. Sprott, Variable-boostable chaotic flows. Optik 127, 10389–10398 (2016)
24. Y. Liao, A. Vikram, V. Galitski, Many-body level statistics of single-particle quantum chaos. Phys.
Rev. Lett. 125(25), 250601 (2020)
25. J.R. Mboupda Pone, S. Çiçek, S. TakougangKingni, A. Tiedeu, M. Kom, Passive–active integrators
chaotic oscillator with anti-parallel diodes: analysis and its chaos-based encryption application to
protect electrocardiogram signals. Analog Integr. Circuits Signal Process. 103(1), 1–15 (2020)
26. B. Muthuswamy, L.O. Chua, Simplest chaotic circuit. Int. J. Bifurc. Chaos 20(5), 1567–1580 (2010)
27. C. Nirmal, B.S. Tej, N. Arjun, Secure image encryption using chaotic, hybrid chaotic and block cipher
approach. J. Imaging 8(6), 167 (2022)
28. B. Norouzi, S. Mirzakuchaki, An image encryption algorithm based on DNA sequence operations and
cellular neural network. Multimedia Tools Appl. 76(11), 13681–13701 (2017)
29. G. Qi, Modelings and mechanism analysis underlying both the 4D Euler equations and Hamiltonian
conservative chaotic systems. Nonlinear Dyn. 95(3), 2063–2077 (2019)
30. K. Rajagopal, A. Bayani, A.J.M. Khalaf, H. Namazi, S. Jafari, V.T. Pham, A no-equilibrium memristive
system with four-wing hyperchaotic attractor. AEU Int. J. Electron. Commun. 95, 207–215 (2018)
31. C. Ran, S. Zhang, X. Gao, A novel 3D image encryption based on the chaotic system and RNA
crossover and mutation. Front. Phys. 6, 66 (2022)
32. R.K. Ranjan, N. Rani, R. Pal, S.K. Paul, G. Kanyal, Single CCTA based high frequency floating and
grounded type of incremental/decremental memristor emulator and its application. Microelectron. J.
60, 119–128 (2017)
33. M.E. Sahin, A.S. Demirkol, H. Guler, S.E. Hamamci, Design of a hyperchaotic memristive circuit
based on wien bridge oscillator. Comput. Electr. Eng. 88, 106826 (2020)
34. C. Sanchez-Lopez, J. Mendoza-Lopez, M.A. Carrasco-Aguilar, C. Muniz-Montero, A floating analog
memristor emulator circuit. IEEE Trans. Circuits Syst. II Express Briefs 61(5), 309–313 (2014)
35. J.P. Singh, B.K. Roy, Five new 4-D autonomous conservative chaotic systems with various type of
non-hyperbolic and lines of equilibria. Chaos Solitons Fract. 114, 81–91 (2018)
36. J.C. Sprott, Some simple chaotic flows. Phys. Rev. E 50(2), 66 (1994)
37. S. Vaidyanathan, C. Volos, A conservative hyperchaotic hyperjerk system based on memristive device.
Adv. Memr. Memr. Dev. Syst. 66, 393–423 (2017)
38. Z. Wang, G. Qi, Modeling and analysis of a three-terminal-memristor-based conservative chaotic
system. Entropy 23(1), 1–17 (2021)
39. N. Wang, G. Zhang, H. Bao, Infinitely many coexisting conservative flows in a 4D conservative system
inspired by LC circuit. Nonlinear Dyn. 99(4), 3197–3216 (2020)
40. L. Wang, S. Jiang, M.-F. Ge, C. Hu, J. Hu, Finite-/fixed-time synchronization of memristor chaotic
systems and image encryption application. IEEE Trans. Circuit Syst. I Regul. Pap. 66, 1–13 (2021)
41. X. Wang, X. Qin, C. Liu, Color image encryption algorithm based on customized globally coupled
map lattices. Multimedia Tools Appl. 78(5), 6191–6209 (2019)
42. M.H. Weik, Relaxation oscillator. Comput. Sci. Commun. Dictionary 61(10), 1461–1461 (2000)
43. A. Wolf, J.B. Swift, H.L. Swinney, J.A. Vastano, Determining Lyapunov exponents from a time series.
Phys. D 16(3), 285–317 (1985)
44. A. Wu, S. Cang, R. Zhang, Z. Wang, Z. Chen, Hyperchaos in a conservative system with nonhyperbolic
fixed points. Complexity 6, 66 (2018)
45. Q. Wu, Q. Hong, X. Liu, X. Wang, Z. Zeng, Constructing multi-butterfly attractors based on Sprott C
system via non-autonomous approaches. Chaos 29(4), 66 (2019)
46. H.G. Wu, Y. Ye, B.C. Bao, M. Chen, Q. Xu, Memristor initial boosting behaviors in a two-memristor-
based hyperchaotic system. Chaos Solitons Fract. 121, 178–185 (2019)

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 3834 Circuits, Systems, and Signal Processing (2023) 42:3812–3834

47. G. Ye, X. Huang, An efficient symmetric image encryption algorithm based on an intertwining logistic
map. Neurocomputing 251, 45–53 (2017)
48. A. Yesil, Y. Babacan, F. Kacar, Design and experimental evolution of memristor with only one VDTA
and one capacitor. IEEE Trans. Comput. Aided Des. Integr. Circuits Syst. 38(6), 1123–1132 (2019)
49. A. Yesil, Y. Babacan, F. Kacar, A new DDCC based memristor emulator circuit and its applications.
Microelectron. J. 45(3), 282–287 (2014)
50. Y. Zhang, Y. Tang, A plaintext-related image encryption algorithm based on chaos. Multimedia Tools
Appl. 77(6), 6647–6669 (2018)

Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps
and institutional affiliations.

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under
a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted
manuscript version of this article is solely governed by the terms of such publishing agreement and applicable
law.

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center
GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers
and authorised users (“Users”), for small-scale personal, non-commercial use provided that all
copyright, trade and service marks and other proprietary notices are maintained. By accessing,
sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of
use (“Terms”). For these purposes, Springer Nature considers academic use (by researchers and
students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and
conditions, a relevant site licence or a personal subscription. These Terms will prevail over any
conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription (to
the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of
the Creative Commons license used will apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may
also use these personal data internally within ResearchGate and Springer Nature and as agreed share
it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not otherwise
disclose your personal data outside the ResearchGate or the Springer Nature group of companies
unless we have your permission as detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial
use, it is important to note that Users may not:

1. use such content for the purpose of providing other users with access on a regular or large scale
basis or as a means to circumvent access control;
2. use such content where to do so would be considered a criminal or statutory offence in any
jurisdiction, or gives rise to civil liability, or is otherwise unlawful;
3. falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association
unless explicitly agreed to by Springer Nature in writing;
4. use bots or other automated methods to access the content or redirect messages
5. override any security feature or exclusionary protocol; or
6. share the content in order to create substitute for Springer Nature products or services or a
systematic database of Springer Nature journal content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a
product or service that creates revenue, royalties, rent or income from our content or its inclusion as
part of a paid for service or for other commercial gain. Springer Nature journal content cannot be
used for inter-library loans and librarians may not upload Springer Nature journal content on a large
scale into their, or any other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not
obligated to publish any information or content on this website and may remove it or features or
functionality at our sole discretion, at any time with or without notice. Springer Nature may revoke
this licence to you at any time and remove access to any copies of the Springer Nature journal content
which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or
guarantees to Users, either express or implied with respect to the Springer nature journal content and
all parties disclaim and waive any implied warranties or warranties imposed by law, including
merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published
by Springer Nature that may be licensed from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a
regular basis or in any other manner not expressly permitted by these Terms, please contact Springer
Nature at

[email protected]