During the European collaborative project
OMEGA, two optical-wireless prototypes have been
developed. The first prototype operates in the near-infrared
spectral region and features Giga Ethernet connectivity, a
simple transceiver architecture due to the use of on-off
keying, a multi-sector transceiver, and an ultra-fast switch
for sector-to-sector hand over. This full-duplex system,
composed by one base station and one module, transmits
data on three meters.
The second prototype is a visible-light-communications
system based on DMT signal processing and an adapted
MAC sublayer. Data rates around to 100 Mb/s at the
physical layer are achieved. This broadcast system,
composed also by one base station and one module, transmits
data up to two meters.
In this paper we present the adapted optical wireless
media-access-control sublayer protocol for visible-light
communications. This protocol accommodates link
adaptation from 128 Mb/s to 1024 Mb/s with multi-sector
coverage, and half-duplex or full-duplex transmission.
This paper studies the non-linear tolerance of several modulation formats in a four-span WDM system (8x10Gb/s) using low chromatic dispersion fiber. The narrow spacing between the channels (50 GHz) makes FWM to be the most detrimental effect experienced in each span made of G.655 fiber (80 km, D=2ps/nm.km) compensated by 0.8 km of DCF. A particular attention to the Q factor computed in the simulations enables a fair comparison between IMDD (Intensity Modulated Direct Detection) and phase modulated formats. It is shown that the various amplitude modulation alternatives result in more or less the same performance. Phase modulation schemes drastically increase the system performance leading to an increase of the Q-factor by almost 3dB.
The performance of high-powered Wavelength Division Multiplexed (WDM) optical networks can be severely degraded due to the Four Wave Mixing (FWM) induced distortion. FWM distortion depends on the statistics of the signals carried by the WDM channels and hence the Gaussian approximation may not be valid. This implies that the well known Q-factor method can not be used to yield an accurate value for the performance of the system in terms of the Bit-Error Rate (BER) of the receiver. To evaluate the BER, one must determine the probability density function (PDF) of the decision variable in the presence of FWM noise, which is related to the signal statistics in a complex manner and can not be evaluated in closed form. In this paper, the Multi-Canonical Monte Carlo Method (MCMC) is used to calculate the PDF of the decision variable of a receiver, limited by FWM noise. Compared to the conventional Monte Carlo method previously used in the literature to estimate this PDF, the MCMC method is much faster and can accurately estimate very low Bit Error Rates. The method takes into account the correlation between the components of the FWM noise unlike the Gaussian model, which is shown not to provide accurate results. The impact of traffic burstiness in the performance of a FWM limited WDM receiver is also investigated using MCMC. It is shown that the traffic load can significantly affect the performance of the system.
Photonic Crystals (PCs) are a promising technology for the realization of high-density optical integrated
circuits. Photonic Crystal-based couplers have been proposed as a compact means of achieving Wavelength
Multiplexing and Demultiplexing. However, the performance of such devices can be limited by fabrication
imperfections such as rod size non-uniformities. In this paper, Coupled Mode Theory (CMT) is applied in order to study
the implication of the variation of the size of the rods. CMT can provide a useful insight in the effect of size variations,
and unlike other numerical methods such as the Finite Difference Time Domain (FDTD), it does not require excessive
computational time. Using CMT, the relation between the size non-uniformities and the coupler's insertion loss and
extinction ratio is analyzed. It is shown even a small size variation of the order of 2%-3% can degrade the performance
of the device.