Multimode Links Research Articles

On Continued Significance of Multimode Links in Data Centers

On Continued Significance of Multimode Links in Data Centers

Self-supervised multimodal learning for group inferences from MRI data: Discovering disorder-relevant brain regions and multimodal links

In recent years, deep learning approaches have gained significant attention in predicting brain disorders using neuroimaging data. However, conventional methods often rely on single-modality data and supervised models, which provide only a limited perspective of the intricacies of the highly complex brain. Moreover, the scarcity of accurate diagnostic labels in clinical settings hinders the applicability of the supervised models. To address these limitations, we propose a novel self-supervised framework for extracting multiple representations from multimodal neuroimaging data to enhance group inferences and enable analysis without resorting to labeled data during pre-training. Our approach leverages Deep InfoMax (DIM), a self-supervised methodology renowned for its efficacy in learning representations by estimating mutual information without the need for explicit labels. While DIM has shown promise in predicting brain disorders from single-modality MRI data, its potential for multimodal data remains untapped. This work extends DIM to multimodal neuroimaging data, allowing us to identify disorder-relevant brain regions and explore multimodal links. We present compelling evidence of the efficacy of our multimodal DIM analysis in uncovering disorder-relevant brain regions, including the hippocampus, caudate, insula, - and multimodal links with the thalamus, precuneus, and subthalamus hypothalamus. Our self-supervised representations demonstrate promising capabilities in predicting the presence of brain disorders across a spectrum of Alzheimer’s phenotypes. Comparative evaluations against state-of-the-art unsupervised methods based on autoencoders, canonical correlation analysis, and supervised models highlight the superiority of our proposed method in achieving improved classification performance, capturing joint information, and interpretability capabilities. The computational efficiency of the decoder-free strategy enhances its practical utility, as it saves compute resources without compromising performance. This work offers a significant step forward in addressing the challenge of understanding multimodal links in complex brain disorders, with potential applications in neuroimaging research and clinical diagnosis.

The Role of Language in Organizational Sensemaking: An Integrative Theoretical Framework and an Agenda for Future Research

Despite the rapid growth of research on organizational sensemaking and an acknowledgment of the critical role of language in the sensemaking process, the literature on sensemaking and language is fragmented. In particular, there is no systematic review that explains the roles and functions of different linguistic elements in meaning construction. The purpose of this review paper is to develop an integrative theoretical framework that organizes existing research in a way that allows us to better understand how different linguistic processes shape the construction of meaning in sensemaking. First, we explain how a cognitive linguistic perspective elucidates how language provides the cognitive associations, schema, and frames used in sensemaking. Second, we discuss how a focus on the social practices of language use, as in rhetorical, narrative, or interactionist approaches, illuminates the patterns of meaning-making among organizational members in social interaction. Third, we explain how a focus on discourse helps us to understand how discursive structures enable or constrain sensemaking and thereby reproduce or transform systems of thought. This leads us to turn to the differences in these perspectives and suggest how they can be brought together in an integrative theoretical framework. Finally, we discuss the contributions of our framework and propose an agenda for future research focusing on the multifaceted role of language in sensemaking, intertextual and multimodal links in the language of sensemaking, the agentic and structural role of language in sensemaking, episodic and latent aspects of language use in sensemaking, and the role of power in discursive sensemaking.

Grief in Times of Corona (Envoi).

This autoethnographic poem tells of personal grief happening in a time of lockdown. It draws on the concept of chronotope, a discrete time and space unit, a parenthesis of sorts, which I have chosen to illustrate as a bubble. In our daily speech, we see bubbles as related to both time and space, now with the added meaning of close relationship of people, those who belong to the same COVID bubble. In this autoethnographic piece, relationships are mediated by technology which anchors our bubbles together, with multimodal links carrying affect and emotion.

A high speed 100 Gbps MDM-SAC-OCDMA multimode transmission system for short haul communication

In this work, a cost-effective spectrum amplitude coding optical code division multiple access (SAC-OCDMA) multimode transmission system is designed with the integration of mode division multiplexing (MDM) scheme in order to transmit 100 Gbps of data over 11 km multimode link. OCDMA scheme is carried out by using random diagonal (RD) code whereas MDM scheme is carried out by using modal multiplexing of Hermite Gaussian (HG) modes 00 and 01. The reported results show the successful transmission of 100 Gbps data over multimode link. Furthermore, the performance of multimode fiber and few mode fiber is investigated in terms of bit error rate which shows that few mode fiber performs better than multimode fiber link for the proposed MDM-OCDMA system. Moreover, the decomposition of HG modes into linearly polarized (LP) modes at receiver side is presented.

Multimodal obstacle detection in unstructured environments with conditional random fields

AbstractReliable obstacle detection and classification in rough and unstructured terrain such as agricultural fields or orchards remains a challenging problem. These environments involve large variations in both geometry and appearance, challenging perception systems that rely on only a single sensor modality. Geometrically, tall grass, fallen leaves, or terrain roughness can mistakenly be perceived as nontraversable or might even obscure actual obstacles. Likewise, traversable grass or dirt roads and obstacles such as trees and bushes might be visually ambiguous. In this paper, we combine appearance‐ and geometry‐based detection methods by probabilistically fusing lidar and camera sensing with semantic segmentation using a conditional random field. We apply a state‐of‐the‐art multimodal fusion algorithm from the scene analysis domain and adjust it for obstacle detection in agriculture with moving ground vehicles. This involves explicitly handling sparse point cloud data and exploiting both spatial, temporal, and multimodal links between corresponding 2D and 3D regions. The proposed method was evaluated on a diverse data set, comprising a dairy paddock and different orchards gathered with a perception research robot in Australia. Results showed that for a two‐class classification problem (ground and nonground), only the camera leveraged from information provided by the other modality with an increase in the mean classification score of 0.5%. However, as more classes were introduced (ground, sky, vegetation, and object), both modalities complemented each other with improvements of 1.4% in 2D and 7.9% in 3D. Finally, introducing temporal links between successive frames resulted in improvements of 0.2% in 2D and 1.5% in 3D.

Solutions for 100/400-Gb/s Ethernet Systems Based on Multimode Photonic Technologies

In this paper, we experimentally demonstrate the transmission of 112 Gb/s four-level pulse amplitude modulation over 100-m OM4 multimode fiber employing a multimode 850-nm vertical-cavity surface-emitting laser (VCSEL) at the transmitter side and equalization techniques at the receiver's digital signal processing (DSP). The penalties imposed by the strong bandwidth limitations due to the optical components as well as the low modal bandwidth of the fiber are compensated by three variant DSP schemes at the receiver, i.e., 1) a finite-impulse response (FIR) filter, 2) a maximum likelihood sequence estimation equalizer (MLSE), and 3) an FIR filter followed by an MLSE equalizer (FIR/MLSE) in a cascaded form. We evaluate all three aforementioned equalization schemes under two different transmitter implementations, i.e., employing a 30-GHz arbitrary waveform generator and a lower bandwidth 15-GHz commercially available digital-to-analog converter and we infer about the applicability of each DSP scheme under these implementations. We show that the hybrid implementation of an FIR followed by a 16-state MLSE can enable the 100-m transmission below the 7% hard decision (HD) forward error correction (FEC) threshold limit and outperforms its other two counterparts for the back-to-back case as well as after 100-m transmission for the high-bandwidth transmitter implementation. On the other hand, lower bandwidth DAC implementations, i.e., 15 GHz, require an increased state MLSE without the need for a preceding FIR filter to bring the bit error rate (BER) below the HD-FEC limit after 100-m OM4 fiber transmission. DSP complexity versus BER performance is assessed for all the aforementioned scenarios evaluating the impact of the transmitter's bandwidth on the overall system's performance. Our proposed solutions show that 112 Gb/s 100-m OM4 multimode links based on VCSELs and standard OM4 fiber can enable next generation 100 and 400 Gb/s wavelength division multiplexed optical interconnects.

4 × 10 GBPS cost effective hybrid OADM-MDM short haul interconnects

This work is focused on high-speed multimode link for short haul communication by adopting hybrid optical add/drop multiplexing and mode division multiplexing schemes. 4× 10 GBps are successfully transported over 1 km multimode link by using Hermite Gaussian (HG) modes at 850 and 850.5 nm. Moreover, mode decomposition at receiver side is also reported for each channel. © 2016 Wiley Periodicals, Inc. Microwave Opt Technol Lett 58:1613–1617, 2016

When it comes to container port efficiency, are all developing regions equal?

In this paper we carry out a container port performance analysis of the developing world between 2000 and 2010, using both parametric and nonparametric approaches. From a unique dataset – our sample covers 70 developing countries, 203 ports, and 1750 data points–, we examine the evolution and drivers of productivity and efficiency changes across developing regions. We show that productivity growth rates between 2000 and 2010 vary significantly and that this heterogeneity is explained by pure efficiency changes rather than scale efficiency of technological changes. Therefore, we carry out a detailed efficiency analysis to determine the drivers of port efficiency. Time series results show an upward trend for port efficiency in developing regions, as it increased from 51 percent in 2000 to 61 percent in 2010. Our analysis indicates that private sector participation, the reduction of corruption in the public sector, improvements in liner connectivity and the existence of multimodal links increase the level of port efficiency in developing regions.

Mode division multiplexed (MDM) waveguide link scheme with cascaded Y-junctions

A potentially high throughput waveguide mode multiplexer/demultiplexer (MUX/DEMUX) exemplified by two cascaded Y-junctions is shown to be capable of efficiently transforming modes from single mode waveguides into a two-dimensional set of the orthogonal ones in a multi-mode waveguide, and back. The design parameters of the cascade and the mode transformations are found by accurate beam-propagation modeling for possible realization with optical polymers. The proposed MUX/DEMUX can enable multifold increase of the data rates of multimode links used in datacom and optical computer communications.

Asymmetric Y junctions in silicon waveguides for on-chip mode-division multiplexing

Silicon waveguide asymmetric Y junction mode multiplexers and demultiplexers are demonstrated for applications in on-chip mode-division multiplexing (MDM). We measure demultiplexed crosstalk as low as -30 dB, <-9 dB over the C band, and insertion loss <1.5 dB for multimode links up to 1.2 mm in length. The frequency response of these devices is shown to depend upon Y junction angle and multimode interconnect length. Interference effects are shown to be advantageous for low-crosstalk MDM, even while using compact Y junctions designed to be outside the mode-sorting regime.

Phase-modulated radio over fiber multimode links

We present the first experimental demonstration of a phase-modulated MMF link implementing high-frequency digital transmission in a cost-effective solution based on direct detection. Successful subcarrier transmission of QPSK, 16-QAM and 64-QAM data channels for bit rates up to 120 Mb/s through a 5 km MMF link is achieved over the microwave region comprised between 6 and 20 GHz. The overall capacity of the proposed approach can be further increased by properly accommodating more passband channels in the operative frequency range determined by the phase-to-intensity conversion process provided by the dispersive nature of the optical fiber. In this sense, our results show the possibility of achieving an aggregate bit rate per length product of 144 Gb/s · km and confirm, in consequence, the possibility of broadband phase-modulated radio over fiber transmission through MMF links suitable for multichannel SCM signal distribution.

A Versatile Link for High-Speed, Radiation Resistant Optical Transmission in LHC Upgrades

Abstract The Versatile Link project is developing a general purpose physical layer optical link with high bandwidth, radiation resistance and magnetic-field tolerance that meets the requirements of LHC upgrade experiments. This paper presents recent work on system specifications, front-end transceiver prototypes, passive components studies and commercial back-end transceiver evaluations. System optical power budgets are specified for single mode (1310nm) and multi-mode (850nm) links, with a target data rate of 4.8 Gbps and a transmission length of 150 meters. Noise and interference penalties are simulated using the 10GbE link model and verified by bit error ratio measurement on reference links. The power margin is particularly constrained by radiation degradation of the front-end receivers. We report the power budgets for all link variants where at least 1.8 dB safety margins are maintained. The Versatile Transceiver (VTRx) - the front-end module to be installed on-detector - is based on a commercial small form pluggable (SFP+) package, modified to optimize size and mass, assembled to host a qualified laser, PIN photodiode, custom-designed radiation tolerant laser driver and receiving amplifier. A set of VTRxs with validated components have been prototyped and compliance tested. We also present the radiation test results on front-end components and passive components. The total fluence tests for lasers and PINs have been carried out with pions and neutrons up to 4 x 1015/cm2. SEU tests have been performed on PIN photodiodes and the full receiver optical subassembly. Radiation induced absorption in a number of single mode and multi-mode fibers, at -25iC and up to 500 kGy, have been measured and high performance candidates identified. Commercial off-of-the-shelf parts have been examined for use as back-end transceivers. Compliance tests on SFP+, 4+4 parallel optical engines and SNAP 12 transmitter/receivers have been completed.

Highly reliable and high-yield 1300-nm InGaAlAs directly modulated ridge fabry-Perot lasers, operating at 10-gb/s, up to 110/spl deg/C, with constant current swing

Transceivers for 300-m multimode links, based on a serial 10-Gb/s laser source and incorporating a receiver based on electronic dispersion compensation (EDC), are creating the first high-volume application for a 10-Gb Fabry-Perot (FP). A highly reliable and high-yield uncooled ridge FP laser is presented. The device shows excellent power characteristics in the 25/spl divide/150/spl deg/C temperature range with very high T/sub 0/ (95 K in the temperature range 0/spl divide/85/spl deg/C and still 78 K at 150/spl deg/C). Outstanding dynamic performances are also shown: 6 dB of extinction ratio can be achieved up to 110/spl deg/C by using a constant current swing of 50 mA. Because of their enhanced performances, these devices have enabled single temperature setting of the optical module, leading to a significant test cost reduction.

Dynamic range of vertical-cavity surface-emitting lasers in multimode links

We report spurious-free dynamic-range measurements of 850-nm vertical-cavity surface-emitting lasers in short multimode links for radio frequency communication. For a 27 m fiber link, the dynamic range at optimal bias was greater than 95 dBHz/sup 2/3/ for modulation frequencies between 1 and 5.5 GHz, which exceeds the requirements for antenna remoting in microcellular networks. In a free-space link, we have measured the highest dynamic range in an 850-nm vertical-cavity surface-emitting laser of 113 dB Hz/sup 1/3/ at 900 MHz. We have also investigated the effects of modal noise and differential mode delay on the dynamic range for longer lengths of fiber.

Optical performance of low-cost self-aligned MCM-D based optical data links

We discuss the details of the design and performance of the optics proposed for a MCM-D based low-cost multimode optical data link. The optical system is completely self-aligned, taking advantage of flip-chip bonding, solder-bump self-alignment, mechanical self-alignment using silicon micromachining, and precision plastic components. Preliminary results demonstrate that the integral-lens design of the opto electronic device significantly effects the coupling efficiency and the requisite tolerances of the components that comprise the coupling optics. We show that coupling efficiencies of 70% or better are readily achievable with piece parts made using standard practices. These coupling efficiencies compare favorably with those in actively-aligned optical subassemblies used in current optical data link products.

Comparaison de la fiabilité opérationnelle des modules optiques équipant les systémes à 34 Mbit/s de France télécom

The purpose of the joint research programme conducted by the CNET and Alcatel CIT on various France Telecom 34 Mbit/s multimode optical links (0.85 and 1.3 μm) is to assess their behaviour versus time, compare their operational reliability and the different types of optoelectronic components used. Traceability on the entire product life cycle was performed using a micro-computer data base. Data was recorded from the begin-ning (January 1985) until the end of the first quarter of 1990. Analysis of the results obtained has taken into account the influence of storage and the failure analyses carried out on rejected components. The results obtained on 0.85 μm optical transmit modules have been used to elaborate a modelling procedure for reliability predictions.

LED bandwidth of multimode fibers as a function of laser bandwidth and LED spectral characteristics

Equations have been developed that relate the bandwidth performance of 62.5-μm core diameter ( <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">NA = 0.29</tex> ) graded-index multimode fiber when measured with LED sources to the LED's spectral characteristics and to the bandwidth measured with laser sources. Because the fiber's spectral loss characteristic filters the LED's power spectrum by shifting its center wavelength and narrowing its spectral width, these equations suggest that chromatic bandwidth scales more favorably than inverse linearly with length. The equations developed here have been used to predict hundreds of LED bandwidths measured independently both in the laboratory and in field installations. With typical agreement between these predictions and the measurements having an rms error of about 11 percent, these equations allow for system design and prediction of the bandwidth performance of LED-based multimode links.