488
|
JAYAWICKRAMA
each other parasitically to give an antenna gain of 12.15 dBi, which is comparable to the gain of a 4-element rectangular patch antenna array fed by microstrip lines or feeding patch.5 Hence, T-junctions, microstrip feeding line and feeding patch are not necessary to build up an array having the gain of 12.15 dBi. Being a high gain antenna, the antenna is compact (76.3 3 63.1 mm or 1.82 kg 3 1.51 kg including the substrate) and low profile (1.575 mm thick substrate). The antenna has a high simulated efficiency of 95% computed from the simulated peak realized gain (12.16 dBi) and the simulated directivity (12.36 dBi) at 5.24 GHz. Since the measured antenna gain (12.15 dBi) is very similar to the simulated antenna gain (12.16 dBi), it is expected that the fabricated antenna also has a high efficiency. CM analysis has been performed on the antenna structure to find out the independent current patterns that can naturally exist on the 3 coupled patches. Four CMs that are the most significant has been extracted for analysis. Their eigenvalues, characteristic currents, MS, and modal excitation coefficients are analyzed. Since the CMs are the current patterns that can naturally exist on the 3 coupled patches, a different excitation of the CMs for different applications could be a potential future work of the research area. O R CI D Sai Ho Yeung
http://orcid.org/0000-0002-7515-1319
R EF ERE NC ES [1] Pozar DM. Microstrip antennas. Proc IEEE. 1992;80(1):79–91. [2] Huynh T, Lee KF. Single-layer single-patch wideband microstrip antenna. Electron Lett. 1995;31(16):1310–1312. [3] Wood C. Improved bandwidth of microstrip antennas using parasitic elements. IEE Proc Part H Microw Opt Antennas. 1980;127:231–234. [4] Aanandan CK, Mohanan P, Nair KG. Broad-band gap coupled antenna. IEEE Trans Antenna Propag. 1990;38(10):1581–1586. [5] Yeung SH, García-Lamperez A, Sarkar TK. Comparison of the performance between a parasitically coupled and a direct coupled feed for a microstrip antenna array. IEEE Trans Antenna Propag. 2014;62(5):2813–2818. [6] Chen Y, Wang C-F. Characteristic Modes: Theory and Applications in Antenna Engineering. John Wiley and Sons, Inc., Hoboken, NJ, 2015. [7] Vogel M, Gampala G, Ludick D, Jakobus U, Reddy CJ. Characteristic mode analysis: Putting physics back into simulation. IEEE Antennas Propag Mag. 2015;57(2):307–317. [8] FEKO is a Trademark of Altair Engineering, Inc.
How to cite this article: Yeung SH, Wang C-F. Study of a parasitic U-slot patch array antenna with characteristic mode analysis. Microw Opt Technol Lett. 2018;60:482–488. https://doi.org/10.1002/mop.30992
ET AL.
Received: 19 July 2017 DOI: 10.1002/mop.30991
Novel wideband chaotic approach LNA with microcontroller compatibility for 5G wireless secure communication Chamindra Jayawickrama | Sandeep Kumar
| Hanjung Song
Department of Nanoscience and Engineering, Centre for Nano Manufacturing, Inje university, Gimhae 621-749, Republic of Korea Correspondence Hanjung Song, Department of Nanoscience and Engineering, Centre for Nano Manufacturing, Inje university, Gimhae 621-749, Republic of Korea. Email:
[email protected] Funding information Korea Electric Power Corporation (Kepco) through Korea Electrical Engineering & Science Research Institute, Grant/Award Number: R15XA03-66
Abstract The tremendous growth of recent technologies toward secure communication networks requires two application goals: (1) encrypted message signal and (2) wide bandwidth operation. This work proposes a wide bandwidth disembodied polynomial chaos approach low-noise amplifier (LNA) with microcontroller compatibility for 5G wireless secure communication. Two different architectures are proposed for achieve desired goal. First architecture approach uses three-state variable Lorenz system which novitiate into discrete mode by using Euler method and make compatibility with microcontroller. The characteristics of output chaotic waveforms from first approach realized by using bifurcation and Lyapunov exponent methods. The propose first novel design methodology reproduce its dynamical behavior with conventional continuous-time system and provides three output chaotic states which is used to encrypt message at the physical layer and achieves security in the telecommunication technologies. While in second architecture, LNA is designed using disembodied polynomial chaos approach which reproduce chaotic output signal over wide bandwidth ranging from 2 to 10 GHz. The LNA is implemented and fabricated using TSMC 45 nm commercial process and reveals minimum noise figure up to 1.5 dB. The
JAYAWICKRAMA
ET AL.
fabrication chip of chaos approach LNA under the power consumption of 3.2 mW made good correlation with simulation ones. KEYWORDS
chaotic LNA, Lorenz system, microcontroller, control parameter, bifurcation
1 | INTRODUCTION Today researchers have attained lot of momentum toward fifth generation 5G but still 5G addresses the demands of 2020 and beyond and is expected to enable a fully mobile and connected society which empower socio-economic transformations in countless ways.1 However, 5G performance is expected to be provided along with the capability to control a highly heterogeneous environment and capability to among others ensure security, trust, identity, and privacy. Besides these expectations, three important security-related challenges would be needed: (1) confidentiality of user and device identity, (2) signaling data confidentiality and integrity, and (3) user data confidentiality (which is not in LTE).2 Chaotic systems could be possible solution for the 5G secure communication because chaotic system is rich instrument for signal design and generation with potential application to communication and signal processing.3 The typical chaotic signals are wideband, noisy, and difficult to predict and can be used for shielding information bearing waveforms in various contexts. Figure 1 shows the proposed block diagram of wideband chaotic encryption signal for 5G secure communication where three states chaotic outputs are merged using polynomial chaos technique. This technique solves LNA example and reproduce wideband and noiseless chaotic
FIGURE 1
|
489
signal. Some specialty classes of chaotic systems are reported where they carried self-synchronization property.4–6 If chaotic systems can break into subsystems (transmitter and receiver), then it forms self-synchronized for coupled with common signal. While capacity to synchronize is robust for some chaotic systems, the synchronization is heavily robust to perturbations for driving signal in the Lorenz system. For confidential communications, chaotic signal shielding techniques as potential useful approach is introduced in Refs. [7,8]. To quantify the meaningful estimates of chaotic systems for information encryption, Lorenz equations are useful approach which provides numerical ability toward dynamical systems. Ed. N. Lorenz9 is a metrologist who invented the Lorenz system which consists three sets of nonlinear differential equations with its some range of control parameters and shows the chaotic behavior. Several approaches where continuous-time chaotic Lorenz systems has been reported,10–12 whereas chaotic system is desired to be in discrete in mode and in real time. In Ref. [13], proposed design analog implementation chaos circuit for novel communication scheme where it is difficult to reproduce dynamical behavior due to inherent imperfections of each analog block. The implementation of electronic chaotic oscillator from single amplifier biquad high Q bandpass filter shows complex behaviors such as bifurcations and chaos for a certain range of circuit parameters.14 The new three-dimensional fractional-order chaotic system without equilibrium is implemented able to control and synchronize the system by using active control and unidirectional coupling.15 In this scenario, wideband LNA chaotic approach with microcontroller compatibility for 5G secure communication is proposed. A proposed algorithm brief methodology for Lorenz oscillator into microcontroller. The character tics of chaotic output waveforms is realized using 2D and 3D chaos attractors, bifurcation, and Lyapunov exponents method which reproduce
Proposed block diagram of chaos approach LNA with microcontroller compatibility [Color figure can be viewed at wileyonlinelibrary.com]
490
|
JAYAWICKRAMA
interesting dynamical behaviors. The particular frequency band is chosen from unidentified chaotic output signal and achieves wideband chaotic approach using LNA with noiseless signal. The organization of this article is as follows: Section 2 discusses the details of chaos implementation into microcontroller while LNA approach chaos are given in Section 3 and finally conclusion is followed in Section 4.
2 | CHOATIC IMPLEMENTATION In this section, implementation of Lorenz chaotic oscillator with microcontroller compatibility will be discussed. The purpose for implementation with microcontroller is to achieve chaotic encrypted signals which could be provides secure communication for 5G wireless next generation. The implementation of chaotic Lorenz oscillator into microcontroller and its behavior analysis will be discussed in next subsections.
2.1 | Lorenz chaotic oscillator The implementation commences with three-dimensional nonlinear Lorenz chaotic equations which are given below as9
FIGURE 2 time Ts 5 0.003)
dx 5pðy2xÞ dt dy 5xðr2zÞ2y dt dz 5xy2bz dt
ET AL.
(1)
Here p 5 10, r 5 30, and b 5 2.66 are standard choices which control dynamics of the nonlinear system using parameters. The chosen values for the Lorenz system given in Equation 1 shows chaotic nature. The frequency domain representation corresponding to Lorenz system is given Equation 2.
S0 S0 Y2 X jxX5a S1 S1
S0 S0 S0 (2) 2n Z 2 X jxY5a X S3 S4 S5
S0 S0 jxZ5a n XY2 Z S6 S7 where a is the dilation factor and keep >1 for controlling the frequency spreading. To implement oscillator into microcontroller, it is possible to convert Lorentz equation 1 into discrete mode by using Euler method that resulting into equation 3.
Proposed Lorenz chaotic oscillator with microcontroller compatibility with its (A) algorithm and (B) schematic diagram (where sampling
JAYAWICKRAMA
FIGURE 3
|
ET AL.
491
Chaotic output waveform with control parameter “r” 5 30
Xðk11Þ5XðkÞ1tsðpðYðkÞ2XðkÞÞÞ Yðk11Þ5YðkÞ1tsðXðkÞðr2ZðkÞÞ2YðkÞÞ
(3)
Zðk11Þ5ZðkÞ1tsðXðkÞYðkÞ2bZðkÞÞ where XðkÞ, YðkÞ, and ZðkÞ are the composite system states with same parameter values which are said in Equation 1. The proposed chaotic Lorenz oscillator with its algorithm and schematic diagram are shown in Figure 2A,B, respectively. The schematic diagram consists of low-cost microcontroller, limited resistors, and proposed algorithm. The proposed algorithm briefs working of chaotic Lorenz oscillator into microcontroller. For implementation, a microcontroller PIC18F4520 from MicrochipR is selected where three ports naming as B, C, and D are coupled with R–2R ladder networks. The R–2R ladder networks naming as X-state, Ystate, and Z-state are used for converting digital chaotic into analog ones (DAC). The unidentified signals from X-state, Y-state, and Z-state are merged using MATLAB programming and achieves combined unidentified chaotic signal (Cout). Owing to inexpensive and simple to configure feature, microcontroller and R–2R ladder network are chosen for the implementation of chaotic oscillator. The analysis of chaotic output waveforms would be discussing in next subsection.
shows waveform for chaotic output (Cout) with proper selection control parameter value of “r” 5 30. It is clear at the observation point that desired periodic states appear with control parameter “r.” We know that classifying the dynamical regimes of the bifurcation diagram is to obtain a global representation of the various regimes that are encountered as the control parameter which is varied. Figure 4 shows the bifurcation diagrams of chaotic Lorenz oscillator with high resolution. We clarify bifurcation picture with chosen control parameter “r” from 15 to 39 points. With taken few data points from 15 to 39, it could be possible to obtain desired bifurcation behavior for chaotic Lorenz oscillator when implemented into microcontroller. The Lyapunov exponent is a measure of quantities which constitute the exponential divergence or convergence of nearby initial points in the phase space of dynamical system. The Lyapunov exponent is often used for indication of sensitive dependences on initial conditions. The system with negative Lyapunov exponent is considered stable while system with positive Lyapunov exponent is unstable or chaotic. The way of quantify chaotic behavior in a system is a measure to the divergence between orbits of two points of small initial separation. Assume f n is the nth iteration of a function f. Then, for two differential initial conditions and x1E, the
2.2 | Performance evaluation It is well known that various encryption methods for communicate the message signal at the transmitter and receiver end has been already reported. However, in recent time, much attention requires for encrypting the message signal over wide bandwidth transmission in the 5G security communication as per the telecommunication standard survey. This implementation of Lorenz chaotic oscillator into microcontroller could be providing chaotic behavior for encryption the message signal. The combined chaotic temporal evolution for all state variables with control parameter “r” is shown in Figure 3. It is observed that control parameter “r” provides wide variation in chaotic behavior as compared to all rest such as p and b, respectively. Therefore, characteristics of Lorenz chaotic oscillator is realized with only control parameter “r” with keep constant p and b, respectively. Figure 3
F I G U R E 4 Typical trajectory of chaotic Lorenz oscillator into microcontroller with selected parameters “r” from 15 to 39 points [Color figure can be viewed at wileyonlinelibrary.com]
492
|
JAYAWICKRAMA
FIGURE 7 FIGURE 5
Lyapunov exponent of typical trajectory corresponding
to bifurcation
ET AL.
Microchip photograph for proposed LNA 0
0
0
jxðkÞ2yðkÞj 5 jf ðxðk21ÞÞjjf ðxðk22ÞÞj . . . :jf ðxð0ÞÞjjxð0Þ2yð0Þj
(7) separation between these orbits is given by jf ðx1EÞ2f ðxÞj as a function of the number of iterations. If it is assumed that the separation of trajectories grows exponentially, then it is expressed in Equation 4 as n
n
jf n ðx1EÞ2f n ðxÞj Eenk
(4)
where k is the Lyapunov exponent. If considering initial separation E between trajectories to be small, it could be achieving desired Lyapunov exponent parameter as f n ðx1EÞ2f n ðxÞ 1 df n 1 (5) k log log dx n E n Consider the trajectory starting x(k) and y(k) from x(0) and y(0). If both trajectories are until time k always in same linear region, we can rewrite as below: 0
jxðj11Þ2yðj11Þj5jf ðxðjÞÞjxðjÞ2yðjÞj; for j50; 1 . . . ; k21 (6) 0
where f ðxÞ denotes the derivative of point x. Thus,
FIGURE 6
Proposed schematic of LNA using polynomial chaos
The two trajectories beyond time k, they eventually will fall into different linear regions; however, the closer initial conditions x(0) and y(0) are longer in average it takes until x(k) and y(k) are in different regions. This justifies interpreting the limit of the expression as the average rate of divergence or convergence of nearby trajectories from the given ones. Figure 5 shows the Lyapunov exponent for Lorenz chaotic microcontroller corresponding to bifurcation diagram. Note that the nonpositive part obtained using maximal Lyapunov exponent while at bifurcation boundaries, it is equal to zero.
3 | LNA IMPLEMENTATION This section provides LNA implementation approach using disembodied polynomial chaos technique. The implementation commences with general polynomial framework where unidentified stochastic waveforms are approximated by expansions of number of random variables which can be expressed in Equation 8.16
JAYAWICKRAMA
|
ET AL.
F I G U R E 8 Variation of return loss with respect to frequency [Color figure can be viewed at wileyonlinelibrary.com]
xðt; nÞ 5
K X
xk ðtÞuk ðnÞ
F I G U R E 9 Variation of noise figure with respect to frequency [Color figure can be viewed at wileyonlinelibrary.com]
(8)
Cm
K51
here uk are orthonormal with respect to probability density function of random parameters. While circuit with stochastic parameters described its behavior in Equation 9, CðnÞ
dxðt; nÞ 1GðnÞxðt; nÞ 1 f ðxðt; nÞ; nÞ 5 uðtÞ dt
(9)
where ðnÞ5ðn1 ; . . . . . . ; nr Þ is r-dimensional variable collecting all random circuit parameters, f ðxðt; nÞ; nÞ ; uðtÞ comprise the nonlinear currents and independent stimulus, xðt; nÞ collects node voltages and currents flowing into circuit components, respectively. To implement combined approach for three chaotic outputs (X-state, Y-state, and Zstate), polynomial expansion (Equation 8) is substituted into Equation 9, which results in Equation 10. K K K X X X d xk ðtÞuk ðnÞ1 GðnÞ CðnÞ xk ðtÞuk ðnÞ1 f xk ðtÞuk ðnÞ; n dt k51 k51 k51
!
(10) Equation 10 is compulsory to hold at specific value n5 nm of the random parameters and which as follows: ! K K K X X X d xk ðtÞamk 1 Gm Cm xk ðtÞamk 1 fm xk ðtÞamk 5uðtÞ dt k51 k51 k51
(11) where amk 5uk ðnm Þ is introduced and m is the corresponding quantity. Now it is possible to introduce to change a variable and leading to T A BL E 1
493
d Ym ðtÞ1Gm Ym ðtÞ1fm ðYm ðtÞÞ5uðtÞ dt
(12)
This disembodied Equation 12 is applied to LNA circuit with repeatedly simulation for the r-dimensional variable. The LNA circuit is begin to analyses with chosen parameter “r” 5 30 which is shown in Figure 6. To obtain wideband approach, LNA circuit is implemented into advanced design system (ADS) platform where microstrip lines (ML1 to ML5) provides matching between components from input to output port. The LNA is fabricated using 45 nm TSMC commercial process while chip measurement is done by using vector network analyzer and RF probe station with test plate capability of 150 mm. The analysis of LNA circuit is given in Figures 8 and 9, respectively. It is clear from Figure 8, the 5G band is obtained over the wide bandwidth which ranging from 2 to 10 GHz when polynomial chaos approach is used for circuit simulation and also made good correlation with measurement. While Figure 9 shows the noise analysis where both observations before and after the LNA circuit implementation are shown. The zig-zag noise is produced after the microcontroller implementation from 5.8 to 5.2 dB while noise figure is reducing from 4.4 to 1.5 dB after LNA circuit using polynomial chaos technique. The microchip photograph of fabricated LNA is shown in Figure 7 and the dimensions and given in Table 1.
4 | CONCLUSION This article presents the implementation of chaotic-based LNA with microcontroller compatibility for 5G secure
Dimensions value of proposed LNA using polynomial chaos
Components
ML1 (mm)
ML2 (mm)
ML3 (mm)
ML4 (mm)
ML5 (mm)
C1 (pF)
C2 (pF)
C3 (pF)
C4 (pF)
C5 (pF)
C6 (pF)
R1 (kX)
R2 (X)
R3 (X)
Values
12
32
5.2
13.8
12
0.5
1.2
1.1
0.3
52
12
12
92
103
494
|
KWON
communication. Two major implementations are folded to whole work. The first implementation used three-state variable Lorenz differential set of equations and shows encrypted chaotic signal waveforms which could provide security at the physical layer in the fifth-generation communication. With the control parameter of Lorenz system, characteristics of Lorenz chaos oscillator is realized with bifurcation and Lyapunov exponent methods. While in second implementation achieves wideband and noiseless performance of LNA circuit using disembodied polynomial chaos. A CK N OW LED G M EN TS This research was supported by Korea Electric Power Corporation (Kepco) through Korea Electrical Engineering & Science Research Institute (grant number: R15XA03-66).
AND
LEE
[14] Banerjee T, Karmakar B, Sarkar BC. Chaotic electronic oscilla€ tor from single amplifier biquad. Int J Electron Commun (AEU). 2012;66(7):593–597. [15] Pham V-T, Kingni ST, Volos C, Jafari S, Kapitaniak T. A simple three-dimensional fractional-order chaotic system without equilibrium: Dynamics, circuitry implementation, chaos control and synchronization. Int J Electron Commun. doi.org/10.1016/j. aeue.2017.04.012. [16] Xiu D, Karniadakis GE. The Wiener-Askey polynomial choas for stochastic differntial euations. SIAM J Sci Computat. 2002; 24(2):619–644.
How to cite this article: Jayawickrama C, Kumar S, Song H. Novel wideband chaotic approach LNA with microcontroller compatibility for 5G wireless secure communication. Microw Opt Technol Lett. 2018;60: 488–494. https://doi.org/10.1002/mop.30991
O R CI D Sandeep Kumar
http://orcid.org/0000-0003-4658-4497
Received: 30 June 2017 DOI: 10.1002/mop.30999
R EF ERE NC ES [1] El Hattachi R, Erfanian J. Next Generation Mobile Network Alliance:5G White Paper. Version 1.0 2015; 1–124. [2] 3GPP TS 33.401. 3GPP System Architecture Evolution (SAE). Security Architecture realease 8 (2008-10). [3] Cuomo KM, Oppenheim AV, Strogatz SH. Synchronizations of lorenz based chaotic circuits with applications to communications. IEEE Trans Circuit Syst II Analog Digital Signal Process. 1993;40(10):626–633.
Asymmetric double QW superluminescent diodes for broadband and high-power use Oh-Kee Kwon
| Chul Wook Lee
[4] Pecora LM, Carroll TL. Synchronization in chaotic systems. Phys Rev Lett. 1990;64(8):821–824.
Photonic Convergence Components Research Group, ICT Material & Components Research Laboratory, ETRI, Republic of Korea
[5] Pecora LM, Carroll TL. Driving system with chaotic signals. Phys Rev A. 1991;44(4):2374–2383.
Correspondence Oh Kee Kwon, Photonic Convergence Components Research Group, ICT Material & Components Research Laboratory, ETRI, 218 Gajeong-dong, Yuseong-gu, Daejeon 34129, Republic of Korea. Email:
[email protected]
[6] Carroll TL, Pecora LM. Synchronizing chaotic circuits. IEEE Trans Circuits Sysf. 1991;38(4):453–456. [7] Cuomo KM, Oppenheim AV, Isabelle SH. Spread spectrum modulation and signal masking using synchronized chaotic systems. MIT Res Lab Electron. 1992. [8] Oppenheim SIAV, Womell GW, Isabelle SH, Cuomo KM. Signal processing in the context of chaotic signals. in Proc. IEEE ICASSP 1992. [9] Lorenz EN. Deterministic nonperiodic flow. J Atmos Sci. 1963; 20(2):130–141. [10] Jiang G-P, Tang WK-S, Chen G. A simple global synchronization criterion for coupled chaotic systems. Chaos Solitons Fractals. 2003;15(5):925–935. [11] Carroll TL, Pecora LM. Synchronizing chaotic circuits. IEEE Trans Circuits Syst. 1991;38:453–456. [12] Carroll TL, Pecora LM. Synchronizing nonautonomous chaotic circuits. IEEE Trans Circuits Syst. 1993;40:646–652. [13] García-Lopez JH, Jaimes-Reategui R, Pisarchik AN, et al. Novel communication scheme based on chaotic R€ossler circuits. J Phys Conf Ser. 2005;23:276–284.
Abstract We report the use of broadband and high-power InGaAs/ GaAs asymmetric double quantum-well (ADQW) superluminescent diodes (SLDs), which consist of a tilted, bent, and straight waveguide. To examine the effects of the arrangement of the ADQW and the gain provided by a straight waveguide on the output performances, we fabricated SLDs with different lengths for two types of ADQW, that is, a wide QW at the p-side for Sample 1, and a narrow QW at the p-side for Sample 2, and tested their spectral width and output power. The fabricated SLDs with a straight length (L3) of 80 lm show 23 dB spectral widths of 122 and 116 nm and output powers of 20 and 16 mW at an injection current of 250 mA for Samples 1 and 2, respectively. For both samples, it appears
