Filipe Ferreira

DSP complexity of mode-division multiplexed receivers

The complexities of common equalizer schemes are analytically analyzed in this paper in terms of complex multiplications per bit. Based on this approach we compare the complexity of mode-division multiplexed digital signal processing algorithms with different numbers of multiplexed modes in terms of modal dispersion and distance. It is found that training symbol based equalizers have significantly lower complexity compared to blind approaches for long-haul transmission. Among the training symbol based schemes, OFDM requires the lowest complexity for crosstalk compensation in a mode-division multiplexed receiver. The main challenge for training symbol based schemes is the additional overhead required to compensate modal crosstalk, which increases the data rate. In order to achieve 2000 km transmission, the effective modal dispersion must therefore be below 6 ps/km when the OFDM specific overhead is limited to 10%. It is concluded that for few mode transmission systems the reduction of modal delay is crucial to enable long-haul performance.

A Comparison of Radio Over Fiber Link Types for the Support of Wideband Radio Channels

Three radio over fiber link types are compared to assess their relative performance for the optical transmission of next generation wireless signals having multiple wideband radio channels with high-level modulation. These links differ in their choice of modulation device; either a directly modulated laser (DML) or external modulation using a Mach-Zehnder modulator (MZM) or a reflective semiconductor optical amplifier (RSOA). The DML and RSOA link types are shown to suffer minimal degradation of the uplink wireless range compared to the baseline value without an optical link, using optimum components in terms of cost and performance. The optimum technology depends on the relative merits of simplicity (DML) or optical network architecture flexibility (RSOA).

Nonlinear Semi-Analytical Model for Simulation of Few-Mode Fiber Transmission

In this letter, a nonlinear semi-analytical model (NSAM) for simulation of few-mode fiber transmission is proposed. The NSAM considers the mode mixing arising from the Kerr effect and waveguide imperfections. An analytical explanation of the model is presented, as well as simulation results for the transmission over a two mode fiber (TMF) of 112 Gb/s using coherently detected polarization multiplexed quadrature phase-shift-keying modulation. The simulations show that by transmitting over only one of the two modes on TMFs, long-haul transmission can be realized without increase of receiver complexity. For a 6000-km transmission link, a small modal dispersion penalty is observed in the linear domain, while a significant increase of the nonlinear threshold is observed due to the large core of TMF.

Design of Few-Mode Fibers With M-modes and Low Differential Mode Delay

In this paper, we investigate the design of few-mode fibers (FMFs) guiding 2 to 12 linearly polarized (LP) modes with low differential mode delay (DMD) over the C-band, suitable for long-haul transmission. Two different types of refractive index profile have been considered: a graded-core with a cladding trench (GCCT) profile and a multi-step-index (MSI) profile. The profiles parameters are optimized in order to achieve: the lowest possible DMD and macro-bend losses (MBL) lower than the ITU-T standard recommendation. The optimization results show that the MSI profiles present lower DMD than the minimum achieved with a GCCT profile. Moreover, it is shown that the optimum DMD and the MBL scale with the number of modes for both profiles. The optimum DMD obtained for 12 LP modes is lower than 3 ps/km using a GCCT profile and lower than 2.5 ps/km using a MSI profile. The optimization results reveal that the most preponderant parameter of the GCCT profile is the refractive index relative difference at the core center, Δnco. Reducing Δnco, the DMD is reduced at the expense of increasing the MBL. Regarding the MSI profiles, it is shown that 64 steps are required to obtain a DMD improvement considering 12 LP modes. Finally, the impact of the fabrication margins on the optimum DMD is analyzed. The probability of having a manufactured FMF with 12 LP modes and DMD lower than 12 ps/km is approximately 68% using a GCCT profile and 16% using a MSI profile.

T-SLAM: Registering topological and geometric maps for robot localization in large environments

This article reports on a map building method that integrates topological and geometric maps created independently using multiple sensors. The procedure is termed T-SLAM to emphasize the integration of Topological and local Geometric maps that are created using a SLAM algorithm. The topological and metric representations are created independently, being local metric maps associated with topological places and registered at the topological level. The T-SLAM approach is mathematically formulated and applied to the localization problem within the Intelligent Robotic Porter System (IRPS) project, which is aimed at deploying mobile robots in large environments (e.g. airports). Some preliminary experimental results demonstrate the validity of the proposed method.

Design of Few-Mode Fibers With Arbitrary and Flattened Differential Mode Delay

This letter proposes the use of a refractive index profile with a graded core and a cladding trench for the design of few-mode fibers, aiming an arbitrary differential mode delay (DMD) flattened over the C+L band. By optimizing the core grading exponent and the dimensioning of the trench, a deviation lower than 0.01 ps/km from a target DMD is observed over the investigated wavelength range. Additionally, it is found that the dimensioning of the trench is almost independent of the target DMD, thereby enabling the use of a simple design rule that guarantees a maximum DMD deviation of 1.8 ps/km for a DMD target between -200 and 200 ps/km.

Multi-robot complete exploration using hill climbing and topological recovery

This article addresses the problem of autonomous map building and exploration of an unknown environment with mobile robots. The proposed method assumes that mobile robots use occupancy grid maps as the main representation model for the built maps and a hill climbing local search algorithm for exploring the environment without any kind of human intervention. It is demonstrated that hill climbing based exploration may recover from local minima and cover completely any environment, if a topological representation of the environment is created incrementally along the mapping and exploration mission. The approach is devised for either a single mobile robot or multiple cooperative mobile robots.

Maximum-Likelihood Detection in Few-Mode Fiber Transmission With Mode-Dependent Loss

In this letter, the performance of maximum-likelihood (ML) detection is evaluated and compared with the zero-forcing (ZF) and minimum mean square error (MMSE) equalizers for a 3 × 158-Gb/s mode-division multiplexed dual-polarization quadrature phase shift keying orthogonal frequency division multiplexing transmission impaired by mode-dependent loss. The receiver schemes are compared in terms of performance and complexity. The simulations show that the ML approach outperforms both ZF and MMSE equalization for strongly coupled modes. However, the complexity of the ML detection is significantly higher and increases faster with the number of modes used for transmission and the modulation format order.

Equalizer complexity of mode division multiplexed coherent receivers

One of the possible solutions is to use space division multiplexing in order to overcome the capacity crunch. Few mode fibers (FMF) have attracted a lot of attention in the recent years, however still a lot of research is required to enable transmission over FMF. One of the problems with the realization of FMF transmission is that these fibers have a modal dispersion which results in high number of equalization taps which might make the FMFs impractical. The complexities of common equalizer schemes are analyzed in this paper in terms of complex multiplications per bit. It is found that training symbol based equalizers have significantly lower complexity compared to blind approaches for long-haul transmission. Among the training symbol based schemes, OFDM requires the lowest complexity for crosstalk compensation in a mode-division multiplexed receiver. The main challenge for training symbol based approaches is the increased data rate due to extra overhead required for cyclic prefix and training symbols. In order to achieve 2000 km transmission, the modal dispersion must be below 6 ps/km when the OFDM specific overhead is limited to 10%. This shows that for few mode transmission systems the reduction of modal delay is essential to enable long-haul performance.

Few-mode fibre group-delays with intermediate coupling

We study the probability density function of the group-delay in few-mode fibres, validating for the first time an analytical estimation for the maximum group-delay spread as a function of linear mode coupling for fibres with more than three LP modes.