Commun Nonlinear Sci Numer Simulat 14 (2009) 1590–1597

Contents lists available at ScienceDirect

Commun Nonlinear Sci Numer Simulat

journal homepage: www.elsevier.com/locate/cnsns

Synchronization of Colpitts oscillators with different orders

J.Y. Effa a,*, B.Z. Essimbi a, J. Mucho Ngundam b
a
Département de Physique, Faculté des Sciences, Université de Yaoundé 1, B.P. 8251 Yaoundé, Cameroun
b
Département des Génies Electrique et Télécommunication, ENSP, Université de Yaoundé 1, B.P. 8390 Yaoundé, Cameroun

a r t i c l e i n f o a b s t r a c t

Article history: In this paper, we consider chaos synchronization between chaotic Colpitts oscillators with
Received 12 February 2008 different orders, consisting of standard and improved version of Colpitts oscillators. Firstly,
Received in revised form 25 March 2008 the normalized state equation of the improved version of the Colpitts oscillator designed to
Accepted 25 March 2008
operate in the ultrahigh frequency range are presented. It is found that this version is
Available online 23 May 2008
described by fourth-order nonlinear differential equations. The equations of motion are
solved numerically using the Runge–Kutta algorithm and simulations demonstrate chaos
PACS:
in the microwave frequencies range. Secondly, the problem of synchronization dynamics
05.45.Xt
05.45.Pq
of third and fourth orders systems in the chaotic states is also investigated, and a controller
is proposed based on stability theory by constructing the Lyapunov function, to ensure syn-
Keywords: chronization between both oscillators. Computer experiments demonstrate the effective-
Microwave improved and standard chaotic ness and feasibility of the proposed technique for these oscillators.
Colpitts oscillator Ó 2008 Elsevier B.V. All rights reserved.
Chaotic systems with different orders
Synchronization
Computer experiments

1. Introduction

In the past two decades, it has been shown that the standard Colpitts oscillator, with special settings of the circuit param-
eters can exhibit chaotic behaviour. This circuit was investigated at the kHz frequencies [1], high (3–300 MHz) frequencies
[2] and ultrahigh (300–1000 MHz) frequencies [3] both numerically and experimentally. Later, an alternative to this stan-
dard version of Colpitts oscillator has been suggested [4,5] in order to exhibit higher fundamental frequencies in a chaotic
regime: a two-stage and an improved Colpitts oscillators.
The study of chaotic oscillations to information technologies [6] is under intensive investigation. In this frame, there is a
great interest behind the study of chaotic oscillator operating at high MHz frequencies, due to it potential application in fu-
ture communication systems [7] and in radar systems [8]. In order to recover the information, the chaotic generator (drive
system) in the transmitter must be synchronized with the chaotic generator (response system) in the receiver. Therefore,
chaos synchronization refers to a process wherein two (or many) chaotic systems (either identical or non-identical) adjust
a given property of their motion to a common behaviour due to coupling or forcing. To synchronize Colpitts oscillators, either
nonlinear Pecora–Caroll method [9,10] or linear coupling technique [7,11,12], adaptive synchronization [13], active control
synchronization [14], linear feedback control method [15], observer-based synchronization [16] and so on have been em-
ployed. Here, the two chaotic Colpitts oscillators to synchronize may be identical [7,8,10–12,14–16] or non-identical with
different parameters [12,13]. However, it is difficult to find identical chaotic systems in reality, especially there are mis-
matches in parameters of the systems. Recently, a lot of effort has been devoted for example to synchronize a class of chaotic
systems in the presence of system’s disturbances and unknown parameters [17]. And, the proposed adaptive observer-based

* Corresponding author.
E-mail address: [email protected] (J.Y. Effa).

1007-5704/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.cnsns.2008.03.019
 J.Y. Effa et al. / Commun Nonlinear Sci Numer Simulat 14 (2009) 1590–1597 1591

synchronization operated satisfactorily in the communication application. Thus, it is important and interesting to investigate
synchronization behaviour between non-identical Colpitts oscillators. Although the method of the synchronization of two
different chaotic systems is far from being straightforward, in this work, we address the problem of chaos synchronization
between two non-identical Colpitts chaotic oscillators with different orders designed to operate in the ultrahigh frequency
range. This is done numerically by means of integration of simplified (piecewise-linear) differential equations. Following Ref.
[18], we propose a controller based on stability theory by constructing the Lyapunov function, to ensure chaos synchroniza-
tion of non-identical standard and improved version of Colpitts oscillators.
The paper is organized as follows. In Section 2, the circuit model for the standard and improved chaotic Colpitts oscillators
are introduced and normalized state equations are presented while their chaotic behaviour is confirmed. Section 3 deals with
the synchronization state of chaotic oscillators with different orders. In Section 4, computer experiments are also undertaken
to validate this approach. Finally, Section 5 concludes the paper.

2. Circuits and basic equations of oscillators

In this section we briefly introduce the circuits under investigation and give their corresponding state space equations
using a piece-wise linear approximation of nonlinear I–V characteristic of the bipolar junction transistor (BJT). We also stud-
ied the continuous autonomous systems by depicting their typical phase portrait.

2.1. The improved Colpitts oscillator

As described in Ref. [5], we consider the configuration of the improved Colpitts oscillator which is schematically repre-
sented in Fig. 1. It contains a BJT as the gain element and a resonant network consisting of an inductor and a pair of capac-
itors. The difference from the standard circuit diagram of the usual Colpitts oscillator is that the inductor L is moved from the
collector circuit to the base where it is in series with resistor Rb. Thus this model cannot be attributed to any common-node
configuration like a standard version. The basic mechanism behind this configuration is to diminish negative influence of the
capacitor CCB (zero-bias collector-base capacitance) at ultrahigh frequencies. Indeed, capacitor CCB in standard version of Col-
pitts oscillator grounds the collector node and acts as a parasitic element destroying chaotic oscillations [19]. In this novel
version, L and Rb screen CCB from the ground and diminish its negative influence. In Ref. [5], improved chaotic Colpitts oscil-
lator was experimentally simulated by using the microwave transistor BFG520 and chaos was predicted in the microwave
frequencies range. However, basic equations of improved version of Colpitts oscillator have not been reported.
Now, let us consider the following state equations for the improved Colpitts oscillator in Fig. 1. given by
8
dV C1
>
> RC 1 dt ¼ V 0  V C1  V C2 þ RIL  RIB  RIC ;
>
>
>
< RC dV C2 ¼ V  V  V  RI þ RI ;
2 dt 0 C1 C2 0 L
ð1Þ
>
> L dIL
¼ R I  V  V ;
>
> dt b L BE C2
>
:
C CB dVdtBC ¼ IL  IB :

This system describes a set of four nonlinear differential equations as

8
>
> RC 1 dVdtC1 ¼ V 0  V C1  V C2 þ RIL  Rðb þ 1ÞIB ;
>
>
>
< RC 2 dVdtC2 ¼ V 0  V C1  V C2  RI0 þ RIL ;
ð2Þ
>
> L dIdtL ¼ Rb IL  V BE  V C2 ;
>
>    
>
: RC dV BE ¼ V  V  V þ R 1 þ C 1 I  R 1 þ C 1 þ b I ;
1 dt 0 C1 C2 C CB L C CB B

Fig. 1. The circuit diagram of improved chaotic Colpitts oscillator.

1592 J.Y. Effa et al. / Commun Nonlinear Sci Numer Simulat 14 (2009) 1590–1597

where IB is the base current and IC the collector current in the BJT which are described as follows [1]
(
V BE V 
r1
V BE > V  ;
IB ¼ ð3Þ
0 V BE 6 V  ;
IC ¼ bIB : ð4Þ

Here V* is the break-point voltage (V*  0.7 V) of I–V characteristics, r1 is the differential resistance of the base–emitter (B–E)
junction in the forward-active mode which is considered to be a constant parameter, and b is the forward current gain of the
device.
Due to the fact that there is no common-base configuration, the base current is not neglected.
For convenient numerical analysis let us introduce the following set of dimensionless state variables and parameters as

x1 ¼ VVC1 ; y1 ¼ qI
V
L
; z1 ¼ VV
C2
; v1 ¼ VVBE ;
qffiffiffiffi
q ¼ CL1 ; e ¼ CC 21 ; a ¼ rq1 ; b ¼ Rq ; c ¼ VV0 d ¼ qIV0 ; a ¼ Rqb ; c ¼ 1 þ CCCB1 :
pffiffiffiffiffiffiffiffi
The normalized time t is such that, t ! st with s ¼ LC 1
Therefore, the above equations are rewritten in the dimensionless form as
8
> x_ 1 ¼ bc  1b ðx1 þ z1 Þ þ y1  aðb þ 1ÞFðv1 Þ;
>
>
< y_ ¼ ay  z  v ;
1 1 1 1
ð5Þ
> ez_ 1 ¼ bc  1b ðx1 þ z1 Þ þ y1  d;
>
>
:
v_ 1 ¼ bc  1b ðx1 þ z1 Þ þ cy1  aðc þ bÞFðv1 Þ;

where the nonlinear function F(v1) is given by


v1  1 v1 > 1;
Fðv1 Þ ¼ ð6Þ
0 v1 6 1:
For numerical simulations, in Eq. (5) the constant term bc has been omitted for simplicity since it does not influence the over-
all dynamics of the system. From Eq. (5) and by using the fourth-order Runge–Kutta algorithm, Fig. 2 shows the projection of
the phase trajectory onto the plane [x1 + z1, z1] The phase portrait is obtained for the value of frequency f = 0.5 GHz, with the
following parameters, a = 1.15, b = 0.67, d = 0.96, a = 0.87, b = 100, e = 1 and c = 21. For the judicious choice of the initial con-
ditions, one finds that the improved Colpitts oscillator is able to exhibit chaotic attractor with the above mentioned set of
parameters. The attractor shown in Fig. 2 is obtained with the initial conditions (x1(0), y1 (0), z1(0), v1(0)) = (0.7, 0.2, 0.5, 0.2).

2.2. The standard Colpitts oscillator

The standard Colpitts oscillator (Fig. 3) which has a typical common-base configuration is described by a set of three
autonomous state-space equations [12]
8
< x_ 2 ¼ y2  a1 Fðz2 Þ;
>
y_ 2 ¼ c  x2  z2  by2 ; ð7Þ
>
:_
z2 ¼ eðy2  dÞ:

–0.6

–0.7

–0.8

–0.9
z1

–1

–1.1

–1.2

–1.3
–1.2 –1 –0.8 –0.6 –0.4 –0.2
x1+z1

Fig. 2. The simulated phase portrait in the plane [x1 + z1, z1] for a = 1.15, b = 0.67, d = 0.96, a = 0.87, e = 1, b = 100 and c = 21.
 J.Y. Effa et al. / Commun Nonlinear Sci Numer Simulat 14 (2009) 1590–1597 1593

Fig. 3. The circuit diagram of standard chaotic Colpitts oscillator.

1
z2

–1

–2
–8 –6 –4 –2 0 2 4
x2+z2

Fig. 4. The simulated phase portrait in the plane [x2 + z2, z2] for a1 = 15.5, b = 0.67, d = 0.96 and e = 1.

The variables and parameters of the system are defined as

x2 ¼ VV
C1
; y2 ¼ qI
V
L
; z2 ¼ VV
C2
;
qffiffiffiffi
q ¼ CL1 ; e ¼ CC 21 ; a1 ¼ qr ; b ¼ Rq ; c ¼ VV 0 ; d ¼ qI
V
0
:

The normalized time is defined as in the precedent Section 2.1.

Here r is the differential resistance of the forward-biased base–emitter junction and V* is the break-point voltage of its I–V
characteristics (for silicon transistors V*  0.7 V). In Eq. (7) the collector current is assumed to be equal to the emitter current,
i.e., the base current is neglected. The parameter c does not influence dynamical behaviour of the oscillator, it sets only the dc
components of the variables. Therefore, it can be omitted for simplicity. Two linear segments are used to approximate the I–V
characteristic of the base–emitter junction:

ð1 þ z2 Þ z2 < 1;
Fðz2 Þ ¼ ð8Þ
0 z2 P 1:
Under the following parameters, a1 = 15.5, b = 0.67, d = 0.96 and e = 1, the standard Colpitts oscillator exhibits chaotic oscil-
lations as shown in Fig. 4, with the initial conditions (x2(0), y2(0), z2(0)) = (0.2, 0.5, 0.5).
1594 J.Y. Effa et al. / Commun Nonlinear Sci Numer Simulat 14 (2009) 1590–1597

3. Chaos synchronization between two Colpitts oscillators with different orders

We reported in this paper, basic equations of improved version of Colpitts oscillator. We found that, for the suitable
choice of parameters and initial conditions, this oscillator can exhibit chaotic oscillator in the microwave frequencies range
as it has been shown experimentally in Ref. [5]. Our objective in this section is now to build a controller to ensure synchro-
nization between the improved and standard Colpitts oscillator operating in microwave frequencies range. The improved
model of Colpitts oscillator is physically interesting so long as they promise higher fundamental frequencies. In the litera-
ture, several methods to investigate the stability of the synchronization dynamics are proposed. We prefer in this work
the method developed by Lü Ling et al. [18]. Based on the stability theory, this method is more indicated to design simple,
efficient and cost-effective controller. In order to observe the synchronization behaviour in the two different Colpitts oscil-
lators, we assume that the improved Colpitts oscillator is the drive system, such that
8
> x_ 1 ¼  1b ðx1 þ z1 Þ þ y1  aðb þ 1ÞFðv1 Þ
>
>
< y_ ¼ ay  z  v
1 1 1 1
ð9Þ
>
> z_ 1 ¼  1b ðx1 þ z1 Þ þ y1  d
>
:
v_ 1 ¼  1b ðx1 þ z1 Þ þ cy1  aðc þ bÞFðv1 Þ

and the standard Colpitts oscillator is the response system, such that
8
< x_ 2 ¼ y2  a1 Fðz2 Þ þ uðtÞ;
>
y_ 2 ¼ x2  z2  by2 ; ð10Þ
>
:_
z2 ¼ y2  d:

where u(t) is the designed controller. One notes that, we use only one controller. Before u(t) is added, system (9) and (10) are
not synchronous because they have different nonlinear functions and different orders. Our goal is to determine the control
function u(t) to ensure chaos synchronization between the two systems.
Thus, the stability of the resulting dynamical states can be analyzed through the construction of the error equations and
the Lyapunov function as follows:
Let us introduce e1 as an error variable of the systems such that, e1 = z2  z1. According to systems (9) and (10), we obtain

1
e_ 1 ¼ y2 þ ðx1 þ z1 Þ  y1 : ð11Þ
b

Constructing the first partial Lyapunov function as

1 2
V1 ¼ e ð12Þ
2 1
the derivative of this function is obtained as follows:
1
V_ 1 ¼ e1 e_ 1 ¼ ke21 þ e1 ðy2 þ ke1 þ ðx1 þ z1 Þ  y1 Þ; ð13Þ
b
where k > 0 is a constant value introduced in order to construct a first negative part of derivative of Lyapunov function.
Now, introduce e2 as an another error variable of the systems such that, e2 = y2  k1. Therefore, one has from Eq. (13)
1
k1 ¼ ke1  ðx1 þ z1 Þ þ y1 ð14Þ
b
and one obtains
   
k 2 2 d a
e_ 2 ¼ x2 þ ðk  bÞy2  z2 þ  2 ðx1 þ z1 Þ þ  k þ a y1 þ w1 þ z1   ðb þ 1ÞFðv1 Þ: ð15Þ
b b b b b
The second partial Lyapunov function is constructed as
1
V 2 ¼ V 1 þ e22 : ð16Þ
2
One gets

V_ 2 ¼ V_ 1 þ e2 e_ 2
¼ ke21  ge22
     
k 2 2 d a
þ e2 x2 þ e1 þ ge2 þ ðk  bÞy2  z2 þ  2 ðx1 þ z1 Þ þ  k þ a y1 þ v1 þ z1   ðb þ 1ÞFðv1 Þ ; ð17Þ
b b b b b
where g > 0 is a second constant value introduced in order to construct a second negative part of derivative of Lyapunov
function.
 J.Y. Effa et al. / Commun Nonlinear Sci Numer Simulat 14 (2009) 1590–1597 1595

At last, introduce e3 as an another error variable of the systems such that, e3 = x2  k2. Therefore, one has from Eq. (17)
   
k 2 2 d a
k2 ¼ e1 þ ge2 þ ðk  bÞy2  z2 þ  2 ðx1 þ z1 Þ þ  k þ a y1 þ v1 þ z1   ðb þ 1ÞFðv1 Þ ð18Þ
b b b b b
and one gets

e_ 3 ¼ x_ 2  k_ 2 ¼ y2  a1 Fðz2 Þ þ uðtÞ  k_ 2 ¼ uðtÞ þ WðtÞ; ð19Þ

where

WðtÞ ¼ y2  a1 Fðz2 Þ  k_ 2
   
k 2 2 a
¼ y2  a1 Fðz2 Þ  e_ 1  ge_ 2  ðk  bÞy_ 2 þ z_ 2   2 ðx_ 1 þ z_ 1 Þ   k þ a y_ 1  v_ 1  z_ 1 þ ðb þ 1ÞF 0 ðv1 Þ ð20Þ
b b b b
are combinations of the state variables of the two systems and

v_ 1 v1 > 1
F 0 ðv1 Þ ¼ ð21Þ
0 v1 6 1
is the derivative of function F(v).
The last Lyapunov function is then constructed as
1 2
V ¼ V2 þ e : ð22Þ
2 3
Then, the derivative of V is

V_ ¼ V_ 2 þ e3 e_ 3 ¼ ke21  ge22  le23 þ e3 ðle3  e2 þ e_ 3 Þ ¼ ke21  ge22  le23 þ e3 ½le3  e2 þ uðtÞ þ WðtÞ; ð23Þ

where l > 0 is a third constant value introduced in order to construct a third negative part of derivative of Lyapunov function.
According to system (10), if the controller satisfies
uðtÞ ¼ le3 þ e2  WðtÞ; ð24Þ
we get

V_ ¼ ke21  ge22  le23 6 0: ð25Þ

According to Lyapunov stability theory, the synchronization of the two systems is then achieved and the error variables of
the system ei(t) approach zero with the lapses of times, i.e.
lim ei ðtÞ ¼ 0; ði ¼ 1; 2; 3Þ: ð26Þ
t!1

Now, assume
S1 ðx2 ; y2 ; z2 Þ ¼ z2
S2 ðx2 ; y2 ; z2 Þ ¼ y2 þ kz2
S3 ðx2 ; y2 ; z2 Þ ¼ x2 þ ðg  k þ bÞy2  gkz2
and
H1 ðx1 ; y1 ; z1 Þ ¼ z1
H2 ðx1 ; y1 ; z1 Þ ¼  1b x1 þ y1  1b  k z1
   
H3 ðx1 ; y1 ; z1 Þ ¼ bg þ bk  b22 x1 þ 2b þ a  k  g y1 þ bg þ bk  gk  b22 z1 þ v1  db  ab ðb þ 1ÞFðv1 Þ

to be observable variables of system (10) and (9). We should read

e1 ¼ z2  z1 ¼ S1 ðx2 ; y2 ; z2 Þ  H1 ðx1 ; y1 ; z1 Þ; ð27Þ
e2 ¼ y2  k1 ¼ S2 ðx2 ; y2 ; z2 Þ  H2 ðx1 ; y1 ; z1 Þ; ð28Þ
e3 ¼ x2  k2 ¼ S3 ðx2 ; y2 ; z2 Þ  H3 ðx1 ; y1 ; z1 Þ: ð29Þ

Then
lim ei ðtÞ ¼ lim Si ðx2 ; y2 ; z2 Þ  Hi ðx1 ; y1 ; z1 Þ ¼ 0; ði ¼ 1; 2; 3Þ: ð30Þ
t!1 t!1

The construction of observable variables completes the method.

4. Computer experiments

We theoretically showed that systems (9) and (10) which respectively represent improved and standard Colpitts oscilla-
tor can be synchronized if the controller is of the form (24). This controller has been derived by constructing progressively
1596 J.Y. Effa et al. / Commun Nonlinear Sci Numer Simulat 14 (2009) 1590–1597

50

–50
e1
–100 e2
–150 e3

–200

–250

–300

–350

–400

–450
0 1 2 3 4 5
t

Fig. 5. Graphs of synchronization errors, e1, e2, e3 with normalized time t (k = g = l = 20).

0.6 5

0.4
0
0.2

0 –5
y2, k1
z1, z2

–0.2
–10
–0.4

–0.6 –15

–0.8
–20
–1

–1.2 –25
0 1 2 3 4 5 0 1 2 3 4 5
t t

450
400
350
300
250
x2, k2

200
150
100
50
0
–50
0 1 2 3 4 5
t

Fig. 6. State variables with normalized time t (k = g = l = 20).

the error equations and the Lyapunov functions. To verify the effectiveness of the proposed control technique, let the drive
system with initial conditions (0.7, 0.2, 0.5, 0.2) and the response system with initial conditions (0.2, 0.5, 0.5). The parame-
ters a1 = 15.5, a = 1.15, b = 0.67, d = 0.96, a = 0.87, e = 1, b = 100 and c = 21 are keep constants during present computations.
Choosing appropriated values for k,g,l is very important. For k = g = l = 20, Figs. 5 and 6 display the good results for synchro-
nization. Computer experiments show that synchronization is achieved successfully and the error signals approach zero
 J.Y. Effa et al. / Commun Nonlinear Sci Numer Simulat 14 (2009) 1590–1597 1597

smoothly and quickly. We have solved numerically Eqs. (9) and (10) using the fourth-order Runge–Kutta algorithm with a
time step Dt = 0.01.

5. Conclusion

In this paper, we introduced the normalized state equations of the improved version of the Colpitts oscillator designed to
operate in the ultrahigh frequency range. Computer experiments show that this oscillator demonstrates chaos for some par-
ticular parameters in the microwave frequencies range. In addition, a controller is designed base on the stability theory to
achieve chaos synchronization between this oscillator represented by fourth-order nonlinear differential equations and a
standard version, represented by third-order nonlinear differential equations. Computer experiments are also given to val-
idate the synchronization approach.
Moreover, this synchronization technique using only one controller could be very useful in future communication
systems.

References

[1] Kennedy MP. Chaos in the Colpitts oscillator. IEEE Trans Circ Syst I 1994;41:771–4.
[2] Wegener C, Kennedy MP. RF chaotic Colpitts oscillator, in: Proceedings of an international workshop on nonlinear dynamics of electronic systems
NDES’95, Dublin Ireland; 1995, p. 255–8.
[3] Mykolaitis G, Tamasevicius A, Bumeliene S. Experimental demonstration of chaos from the Colpitts oscillator in the VHF and the UHF ranges. Electron
Lett 2004;40:91–2.
[4] Tamasevicius A, Mykolaitis G, Bumeliene S, Cenys A, Anagnostopoulos AN, Lindberg E. Two-stage chaotic Colpitts oscillator. Electron Lett
2001;37:549–51.
[5] Tamasevicius A, Bumeliene S, Lindberg E. Improved chaotic Colpitts oscillator for ultrahigh frequencies. Electron Lett 2004;40:1569–70.
[6] Tamasevicius A, Cenys A, Mykolaitis G, Namajunas A. Synchronization of chaos and its application to secure communication. Lithuanian J Phys
1998;38:33–7.
[7] Uchida A, Kawano M, Yoshomori S. Dual Synchronization of chaos in Colpitts electronic oscillators and its applications for communications. Phys Rev E
2003;68:056207:1–056207:11.
[8] Qiao S, Shi ZG, Chen KS, Cui WZ, Ma W, Jiang T, et al. A new architecture of UWB radar utilizing microwave chaotic signals and chaos synchronization.
Prog Electromagnetics Res PIER 2007;75:225–37.
[9] Tamasevicius A, Mykolaitis G, Bumeliene S, Cenys A, Lindberg E. Synchronization of VHF chaotic Colpitts oscillator, in: Proceedings of international
workshop on nonlinear dynamics of electronic systems NDES’97, Delft Netherlands; 2001, p. 223–6.
[10] Rubezic V, Ostojic R. Synchonization of chaotic Colpitts oscillator with applications to binary communications, in: Proceedings of international
conference on electronic circuits and systems ICECS’99, Paphos Cyprus; 1999, p. 153–6.
[11] Burykin VA, Panas AI. Chaotic synchronization of RF generators, in: Proceedings of international specialist workshop on nonlinear dynamics of
electronic systems NDES’97, Moscow Russia; 1997, p. 548–53.
[12] Baziliauskas A, Krivickas R, Tamasevicius A. Coupled chaotic Colpitts oscillator: identical and mismatched cases. Nonlinear Dynam 2006;44:151–8.
[13] Fotsin HB, Daafouz J. Adaptive synchronization of uncertain chaotic Colpitts oscillators based on parameter identification. Phys Lett A
2005;339:304–15.
[14] Li GH. Synchronization and anti-synchronization of Colpitts oscillators using active control. Chaos Soliton Fract 2005;26:87–93.
[15] Cenys A, Tamasevicius A, Mykolaitis G. Hyperchaos and synchronization in mean field coupled chaotic oscillator, in: Proceedings of international
symposium on nonlinear theory and its applications NOLTA’98, Crans-Montana Switzerland; 1998, p. 519–40.
[16] Li GH. Chaos and synchronization of Colpitts oscillators. Microwave Opt Technol Lett 2003;39:446–9.
[17] Liao TL, Tsai SH. Adaptive synchronization of chaotic systems and its application to secure communications. Chaos Soliton Fract 2000;11:1387–96.
[18] Lü L, Luan L, Guo ZA. Synchronization of chaotic systems with different orders. Chin Phys 2007;16:06–346.
[19] Mykolaitis G, Tamasevicius A, Bumeliene S, Lasiene G, Cenys A, Anagnostopoulos AN, et al. Towards microwave chaos with two-stage Colpitts
oscillator, in: Proceedings of international workshop on nonlinear dynamics of electronic systems NDES’97, Delft Netherlands; 2001, p. 97–100.