Sharif Digital Repository

Convolutional neural networks (CNNs) are at the heart of several machine learning applications, while they suffer from computational complexity due to their large number of parameters and operations. Recently, all-optical implementation of the CNNs has achieved many attentions, however, the recently proposed optical architectures for CNNs cannot fully utilize the tremendous capabilities of optical processing, due to the required electro-optical conversions in-between successive layers. To implement an all-optical multi-layer CNN, it is essential to optically implement all required operations, namely convolution, summation of channels' output for each convolutional kernel feeding the... 

In this article we have acquired exact solitary wave solutions for the truncated Ω- fractional Lakshamanan–Porsezian–Daniel model with Kerr, parabolic, and anti-cubic nonlinear laws employing extended auxiliary technique. Diverse set of exponential function solutions acquired relying on a map between the considered equation and an auxiliary ODE. Obtained solutions are recast in several hyperbolic and trigonometric forms based on different restrictions between parameters involved in equations and integration constants that appear in the solution. A few significant ones among the reported solutions are pictured to perceive the physical utility and peculiarity of the considered model using... 

Optical Neural Networks (ONNs) can be promising alternatives for conventional electrical neural networks as they offer ultra-fast data processing with low energy consumption. However, lack of suitable nonlinearity is standing in their road of achieving this goal. While this problem can be circumvented in feed-forward neural networks, the performance of the recurrent neural networks (RNNs) depends heavily on their nonlinearity. In this paper, we first propose and numerically demonstrate a novel reconfigurable optical activation function, named ROA, based on adding or subtracting the outputs of two saturable absorbers (SAs). RAO can provide both bounded and unbounded outputs by facilitating an... 

Despite various optical realizations of convolutional neural networks (CNNs), optical implementation of nonlinear activation functions and pooling operations are still challenging problems. In this regard, this paper proposes an optical saturable absorption nonlinearity and its atomic-level model, as well as two various optical pooling operations, namely optical average pooling and optical motion pooling, by means of 4f optical correlators. Proposing these optical building blocks not only speed up the neural networks due to negligible optical processing latency, but also facilitate the concatenation of optical convolutional layers with no optoelectrical conversions in-between, as the... 

This paper is concerned with the problem of guidance filter design for a homing air defense missile using measurements of an on-board pitch-yaw gimballed seeker mechanism. This problem possess a highly nonlinear dynamic which is influenced by various error sources such as seeker's stabilization loop torque disturbances originated from pursuer's accelerations and angular velocities, significant channel couplings, time-delayed and noisy measurements of optical sensor besides the unknown target maneuvers. To handle the inherent nonlinearity of problem a wide variety of assumptions are made in literature to derive simplified engagement kinematics. In present work, performance degradations which... 

This paper introduces the concept of on-chip temporal optical computing, based on dispersive Fourier transform and suitably designed modulation module, to perform mathematical operations of interest, such as differentiation, integration, or convolution in time domain. The desired mathematical operation is performed as signal propagates through a fully reconfigurable on-chip photonic signal processor. Although a few numbers of photonic temporal signal processors have been introduced recently, they are usually bulky or they suffer from limited reconfigurability which is of great importance to implement large-scale general-purpose photonic signal processors. To address these limitations, this... 

Least squares estimation is a widely-used technique for range-based source localization, which obtains the most probable position of mobile station. These methods cannot provide desirable accuracy in the case with a non line of sight (NLOS) path between mobile station and base stations. To circumvent this drawback, many algorithms have been proposed to identify and mitigate this error; however, they have a large run-time overhead. On the other hand, new positioning systems utilize a large set of base stations, and a practical algorithm should be fast enough to deal with them. In this paper, we propose a novel algorithm based on subspace method to identify and eliminate the NLOS error.... 

We propose an energy-efficient procedure for transponder configuration in few-mode fiber-based elastic optical networks in which bit error rate and physical constraints are guaranteed and joint optimization of temporal, spectral, and spatial resources are addressed. We use geometric convexification techniques to provide convex formulations for the signal-to-interference plus noise ratio, transponder power consumption, and transponder configuration problem. Simulation results demonstrate that the proposed convex formulation for the transponder configuration problem is considerably faster than its mixed-integer nonlinear counterpart. We consider transmit optical power as a configurable... 

Compression of dissipative soliton (DS) and dissipative soliton resonance (DSR) have attracted considerable attention for generation of short pulse lasers. Generation of DS/DSR is investigated numerically in circular fiber array with optical central fiber. Parameter management can generate the DS and DSR in circular fiber array with central optical fiber and in normal or anomalous dispersion. The nonlinear circular fiber arrays can be used as an optical pulse compressor. In this paper, compression of DS and DSR versus the nonlinearity and dispersion parameters in circular fiber array with central fiber, are taken into investigation. © 2016 Elsevier Inc  

Photonic crystal fibers (PCFs) are widely used in optical communications systems, nonlinear devices, gas sensors, high power transmission, and so on. Photonic crystal fiber with minimum values of confinement loss, nonlinear effects, and chromatic dispersion is used in optical communication. But in some applications, high nonlinear coefficient is required. This paper presents new design of index-guiding photonic crystal fiber (IGPCF) with characteristics appropriate for nonlinear applications. In the present design with a square-hexagonal nano-structure having air holes with unequal diameters, nearly zero dispersion at the wavelength range of 1540 to 1550 nm is achieved. The simulation... 

In this paper, a spectral-phase-encoded ultrashort light pulse optical code division multiple access (SPE-OCDMA) system employing a novel nonlinear power-cubic optical preprocessor at its receiver's front end is theoretically investigated. The system's mathematical model and the statistical distribution of the decision variable Y, prior to the decision module of the receiver, are discussed. The first three moments of the random decision variable Y are obtained in the context of the above OCDMA system and subsequently used in an appropriate Log-Pearson type 3 (LP3) distribution to represent the random decision variable $Y$. Multiple access interference (MAI) and amplified spontaneous emission... 

In this paper, we present an analytical approach in obtaining the probability density function (pdf) of the random decision variable Y, which is formed at the output of the power-cubic all-optical nonlinear preprocessor followed by the photodetector, with applications in ultrafast optical time-division multiplexing and optical code-division multiple-access systems. Our approach can be used to accurately evaluate the performance of ultrafast pulse detection in the presence of Gaussian noise. Through rigorous Monte Carlo simulation, the accuracy of the widely used Gaussian approximation of decision variable Y is refuted. However, in this paper, we show that the so-called log-Pearson type-3... 

In this study, kernel least mean square (KLMS) algorithm with fractionally spaced equalizing structure is proposed for electrical compensation of chromatic dispersion (CD) and nonlinear phase noise (NLPN) in a dual polarization optical communications system with coherent detection. We consider single mode fiber channel. At the receiver, the additive optical noise is represented as additive white Gaussian noise. Phase modification is utilized at high signal powers to maintain the validity of Gaussian model of noise. We consider QAM and PSK modulations and evaluate the performance of the proposed method in terms of error rate, phase error, and error vector magnitude (EVM). The results are... 

A new method which is a combination of the harmonic balance and finite difference techniques (HBFD) is proposed for complete time-harmonic solution of the nonlinear wave equation. All interactions between different harmonics up to an arbitrary order can be incorporated. The effect of higher order harmonics on two important nonlinear optical phenomena, namely, the second harmonic generation (SHG) and frequency mixing is investigated by this method and the results are compared with well-known analytical solutions. The method is quite general and can be used to study wave propagation in all nonlinear media  

The resonant second harmonic generation in the presence of a wiggler magnetic field by twisted laser plasma interaction is surveyed. The wiggler magnetic field provides additional momentum required for the phase matching. The Laguerre-Gaussian modes can be used to control the self-focusing and improve the second harmonic generation in laser plasma interaction. The wave equations for the fundamental and the second harmonic fields have been solved in the paraxial approximation. The generation of the second harmonic considering self-focusing is investigated. Also, the dependence of the second harmonic power on the propagation distance for different values of initial fundamental beam intensity,... 

Superconducting materials are known to exhibit nonlinear effects and to produce harmonic generation and intermodulation distortion in superconductive circuits. In planar structures, these nonlinearities depend on the current distribution on the strip which is mainly determined by the structure of the device. This paper investigates the current distribution at the step-in-width discontinuity in superconducting microstrip transmission lines, which is computed by a numerical approach based on a 3-D finite-element method. This current distribution is used to obtain the parameters of the nonlinear circuit model for the superconducting microstrip step-in-width discontinuity. The proposed... 

A simple analytic formula is derived to extract tiny dispersion fluctuations along highly nonlinear fibers from distributed measurements of parametric gain. A refined BOTDA scheme, suitable to track Kerr processes, enables low noise measurements  

The performance of devices based on highly nonlinear fibers (HNLF) can be drastically impeded by even tiny fluctuations of the zero dispersion wavelength (ZDW) along the fiber. Being able to measure ZDW fluctuations along an HNLF is therefore essential for the design of efficient nonlinear optics based devices such as fiber optical parametric amplifiers (FOPA), regenerators and limiters. Different schemes using complex distributed sensing of localized nonlinear interactions have been used in order to derive the ZDW fluctuations along the fiber [1-2]. In [3], the distributed gain of a pulsed pump FOPA along an HNLF was measured using a Brillouin Optical-Time Domain Analysis (BOTDA) based... 

We present a formulation for the nonlinear optical response in gapped graphene, where the low-energy single-particle spectrum is modeled by massive Dirac theory. As a representative example of the formulation presented here, we obtain a closed form formula for the third harmonic generation in gapped graphene. It turns out that the covariant form of the low-energy theory gives rise to peculiar logarithmic singularities in the nonlinear optical spectra. The universal functional dependence of the response function on dimensionless quantities indicates that the optical nonlinearity can be largely enhanced by tuning the gap to smaller values  

In this paper, we study theoretically the generation of high repetition rate short pulses using fiber optical parametric amplification. We show that the pulse shape and duration depend on the signal location relatively to the pump frequency. We demonstrate that in order to get the shortest pulse width, the signal must be located at one of the extremities of the gain spectrum associated with the pump peak power. We derive the analytical expression of the pulse shape in this case and compare it to the exponential gain regime case. Using numerical simulations, we also analyze the impact of walk-off and pump phase modulation that is required to suppress Stimulated Brillouin Scattering and derive...