KEYWORDS: Semiconductor lasers, Data processing, Systems modeling, Mathematical optimization, Laser systems engineering, Education and training, Computing systems, Transceivers, Signal processing, Photonics systems
Photonic systems, exhibiting multi-gigahertz bandwidth, facilitate data transmission at gigabit-per-second rates. While traditionally used in optical communication for data transfer, semiconductor lasers are now being explored for their potential in optical computation and signal processing. Injecting information into these lasers leads to nonlinear transformations and high-speed processing. Experimentally, a single semiconductor laser shows essential features for versatile computation, such as high-dimensional and nonlinear responses within sub-nanoseconds. To boost computational power, we study numerically the training of delay-coupled laser networks. The objective is, akin to training artificial neural networks, optimizing laser network's to improve performance and computational efficiency in challenging machine learning tasks. However, relying on offline optimization methods and physical models raises challenges due to device variability and limited system observability. Here, we propose evolutionary strategies to optimize physical systems without needing precise model knowledge, offering a promising approach for online system optimization.
Convolutional Neural Networks (CNNs) are employed in a plethora of fields, including computer vision, natural language processing, and speech recognition. We present an integrated photonic accelerator for CNNs based on the temporal-spatial interleaving of signals. This architecture supports 1D kernels, and can be extended to 2D convolutional kernels, providing scalability for complex networks. A supervised on-chip learning algorithm is employed to guarantee a reliable setting of convolutional weights against fabrication tolerances, thermal cross-talks, and changes in operating conditions. Overall, by leveraging photonics technology, the proposed accelerator significantly reduces hardware complexity while enabling high-speed processing and parallelism.
Proposed for analog computing, multimode fibers have limitations due to slow spatial-domain encoding. Our work showcases instead the computational prowess of a scheme employing a step-index few-mode fiber (FMF) segment, for high-speed spatiotemporal coincidence detection by leveraging the FMF’s dispersive optical properties. The FMF is a custom-made fabrication, with NA = 0.15, a core diameter of 22 μm, and a length of 13 m, introducing delay to temporal input pulses through the supported propagation of higher-order fiber modes. The temporal mixing of these modes creates short-term memory for time-encoded information which we exploit for coincidence detection. By slightly misaligning the input beam with the FMF’s longitudinal axis, we can modify the impact of the different modes on the overall spatial pattern distribution. Our experimental system operates at 1550 nm and encodes 6-bit header patterns with 35.1 ps pulses per bit. With four distinct 40 GHz photodetected points at the output speckle pattern of the FMF, we capture four different time series that correspond to different power integrals and use them to train a logistic regression classifier. Eventually, every header classification is performed with the sampling of only one pulse time window, thus our system operates at 28.5 Gb/s. Remarkably, under various input misalignment conditions, our system demonstrates error rates below 1/5000. This level of performance could not be obtained with a standard step-index multimode fiber of the appropriate length.
In this work, we consider a hybrid-coherent time-delay photonic reservoir, with a coherent input layer and an incoherent output layer, for post-processing signals from a 200 km, 28 GBaud PAM-4 transmission link. The amplitude and the phase of the transmission link are obtained through a coherent receiver and introduced as two independent encoding signals at the input of our photonic reservoir. We use the photodetected intensity of the reservoir’s response to train a linear classifier and perform the data recovery task. This hybrid-coherent reservoir exhibits a bit error rate of 10−4 , three orders of magnitude lower compared to the performance of the same photonic reservoir that processes only the amplitude information at the input.
We show the computational power of few-mode fibers (FMF) in a 40 Gbps spatiotemporal coincidence detector scheme. We consider a 5.5 m step-index FMF, with a 16.6 μm core diameter, as the medium that introduces various delays to a temporal input pulse, via the supported propagation fiber modes. In our representation, the different group velocities of the excited fiber modes define equivalent optical dendritic branches. A 1550 nm laser’s optical output is modulated by a 40 Gbps binary sequence and coupled to the FMF. The output optical pattern is photodetected by a 3×3 array and used to solve successfully a 6-bit header classification task.
Time-delay Reservoir Computing (TRC) are neural networks which thanks to their hardware simplicity are suited for photonic implementation. Here, we numerically investigated and experimentally tested a TRC based on a nonlinear single silicon microring resonator (MRR) coupled to an optical fibre loop which provides an external tuneable optical feedback. The work is inspired by the key role that MRRs demonstrate in integrated optical devices, together with the significant advantage they provide due to their fully passive nonlinearity. Insights about the computational properties of the system are provided as a result of the performance on multiple benchmark tasks.
Bidirectional coupling of semiconductor lasers (SLs) through optical injection is a well established method to generate chaotic signals which, through their dynamics, may give rise to several applications from sensing to monitoring and from communication to security. Recent works have shown the capability of joint behavior or complete synchrony of mutually coupled networks of SLs. In these works, the coupling architecture, the operational conditions and the properties of the active elements determine the types of dynamics of the emitted optical signals, through which the network can potentially be synchronized. In this experimental work, a network of mutually coupled semiconductor lasers has been synchronized through chaotic optical signals that spectrally extend over 10GHz. The synchronization among the lasers that participate in the coupled network is affected, besides the structural and operational conditions, by the signals' bandwidth that circulates optically. Here we show that the synchronization performance of the detected signals when monitoring the network nodes through optoelectronic conversion is in direct dependence on the signal bandwidth. Smaller signal bandwidth at the GHz range may result in synchronization with cross-correlation values over 0.97 in most of the SL nodes, rejecting higher frequencies that are not optimally synchronized. Another source of improving the synchronization of the network that has been recorded in this experimental setup is by harnessing the de-synchronization events that are almost always apparent, especially when emitted signals include power dropouts.
Semiconductor lasers (SL) have been proven to be a key device in the generation of ultrafast true random bit streams. Their potential to emit chaotic signals under conditions with desirable statistics, establish them as a low cost solution to cover various needs, from large volume key generation to real-time encrypted communications. Usually, only undemanding post-processing is needed to convert the acquired analog timeseries to digital sequences that pass all established tests of randomness. A novel architecture that can generate and exploit these true random sequences is through a fiber network in which the nodes are semiconductor lasers that are coupled and synchronized to central hub laser. In this work we show experimentally that laser nodes in such a star network topology can synchronize with each other through complex broadband signals that are the seed to true random bit sequences (TRBS) generated at several Gb/s. The potential for each node to access real-time generated and synchronized with the rest of the nodes random bit streams, through the fiber optic network, allows to implement an one-time-pad encryption protocol that mixes the synchronized true random bit sequence with real data at Gb/s rates. Forward-error correction methods are used to reduce the errors in the TRBS and the final error rate at the data decoding level. An appropriate selection in the sampling methodology and properties, as well as in the physical properties of the chaotic seed signal through which network locks in synchronization, allows an error free performance.
The potential of conventional semiconductor lasers to generate complex and chaotic dynamics at a bandwidth that extends up to tens of GHz turns them into useful components in applications oriented to sensing and security. Specifically, latest theoretical and experimental works have demonstrated the capability of mutually coupled semiconductor lasers to exhibit a joint behaviour under various conditions. In an uncoupled network consisting of N similar SLs - representing autonomous nodes in the network - each node emits an optical signal of various dynamics depending on its biasing conditions and internal properties. These nodes remain unsynchronized unless appropriate coupling and biasing conditions apply. A synchronized behaviour can be in principle observed in sub-groups of lasers or in the overall laser network. In the present work, experimental topologies that employ eight SLs, under diverse biasing and coupling conditions, are built and investigated. The deployed systems incorporate off-the-shelf fiber-optic communications components operating at the 1550nm spectral window. The role of emission wavelength detuning of each participating node in the network - at GHz level - is evaluated.
Two multi-semiconductor-laser (SL) topologies, based on mutually coupled semiconductor lasers - representing a startype and a mesh-type network - are evaluated in terms of their synchrony potential and their sensitivity towards critical SLs' intrinsic and operational parameters. The coupling topology, the coupling conditions and the values of key SL parameters determine the type of dynamics of the emitted optical signals. The number of nodes and the detuning in their fundamental properties have been assessed to be decisive in terms of efficiency and quality of synchronized outputs, as wells as for the overall dynamical map of the network. Our investigation mainly focuses on discrepancies in SL parameter values and their effect on the efficiency of synchronized dynamics. This type of investigation will provide preliminary guidelines on building experimentally large scale networks of coupled SLs under various coupling matrices that could support optical sensing or cryptographic applications.
A thorough study of an all-optical chaotic communication system, including experimental realization real-world testing
and performance characterization through bit-error-rate analysis, is presented. Pseudorandom data that are effectively
encrypted in the chaotic emitter and sent for transmission are recovered at the receiver with bit-error-rate (BER) values
as low as 10-7 for 1 Gb/s data rate. Different data code lengths and bit-rates at the Gb/s region have been tested. Optical
transmission using 100km fiber spools in an in-situ experiment and 120km in an installed optical network showed that
chaotic communication systems does not act as a considerably deteriorating factor in the final performance.
The performance of an all-optical closed-loop chaotic communication system in a transmission link consisting of single mode fibers (SMF) applying two different dispersion management techniques is numerically studied. The first technique is implemented by the usage of dispersion compensating fibers (DCFs), while the second utilizes optical phase conjugators (OPCs). The latter is implemented by means of four wave mixing (FWM) in a dispersion shifted fiber (DSF), where the chaotic carrier corresponds to the signal wave and a high power continuous wave corresponds to the pump wave. Calculation of the recovered message Q-factor values obtained from the corresponding eye diagrams has been carried out applying chaotic modulation (CM) and chaos shift keying (CSK) encryption techniques for two repetition rates (2.4Gbps, 5Gbps). It is shown that the optical phase conjugation is an effective dispersion and non-linear effects compensation technique even if high-bit rate message encoding is applied. The superiority of a transmission system including OPCs to that utilizing (DCFs) is presented. The influence of key system parameters such as optical power, OPC spacing, pump power level, etc. to the transmission performance has been investigated. Acceptable system performance is presented for approximately 600Km at 2.4Gbps and 400Km at 5Gbps.
A detailed investigation of the decoding properties of different receiver configurations in an all-optical chaotic transmission system is presented for two data-encoding techniques and for various dispersion compensation maps. A semiconductor laser subjected to optical feedback generates the chaotic carrier while data is encoded either by Chaotic Modulation (CM) or Chaotic Shift Keying (CSK) methods. The complete transmission module consists of various dispersion management maps, in-line amplifiers and Gaussian optical filters. The receiver, employing a high facet reflectivity laser, is either forming a closed-loop configuration operating at the non-amplification regime or a strongly injected open-loop one. For the latter configuration the possibility of utilizing an anti-reflection (AR) coated laser is also investigated. System's performance is numerically tested by calculating the Q-factor of the eye diagram of the 1 Gb/s received data. The influence of the optical power launched into fibre or the transmission distance to the quality of the decoded message has been investigated. The closed-loop scheme had better performance relative to the open-loop, while CSK method and maps utilizing Dispersion Shifted Fibres are superior to CM and that employing Dispersion Compensating Fibres respectively. When an AR-coated laser is used in the open-loop receiver setup, improved decoding performance occurs.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.