Yannis Le Guennec

Bidirectional millimeter-wave radio-over-fiber system based on photodiode mixing and optical heterodyning

We demonstrate a bidirectional millimeter-wave (mmW) radio-over-fiber system based on the use of a p-i-n photodiode (PD) biased in the nonlinear regime. Since frequency down-conversion and photodetection are carried out by a single photodiode, the base station (BS) architecture is simplified. Down-link transmission employs an optical heterodyning technique for mmW generation and remote local oscillator (LO) delivery to BSs. The up-link scheme uses the electrical port of the PD as an input for optoelectronic mixing with the remotely delivered LO. Up-link and down-link have been successfully demonstrated for 397 Mbps and 794 Mbps binary phase-shift keying complying with the 60 GHz ECMA 387 standard. Error vector magnitude measurements and theoretical analysis are discussed as regards optimal PD bias conditions. It is shown that PD can be used in this bidirectional system without any switching of the PD bias.

Quasi-static approach to optimize RF modulation of vertical-cavity surface-emitting lasers

This paper presents a new experimental study of vertical-cavity surface-emitting lasers (VCSELs) in a low-cost radio-over-fiber (RoF) application. This study shows the achievable advantages of optimizing VCSELs modulation thanks to a new “quasi-static” laser characterization.

Impact of VCSEL nonlinearity on Discrete MultiTone modulation: Quasi-static approach

Vertical-Cavity Surface-Emitting Lasers (VCSELs) represent many advantages for high data rate and low cost communication systems. However, they have a nonlinear transfer function, which could drastically affect the multicarrier modulations performances. The static characteristic is usually used to estimate the VCSEL nonlinearity impact on the Discrete MultiTone (DMT) modulation performances. In this paper, we demonstrate that the use of the static characteristic to model the VCSEL nonlinearity, overestimates the nonlinearity distortions when a large dynamic signal is modulating the VCSEL. In order to correctly estimate the VCSEL impact on the DMT modulation performances, we use the quasi-static characteristic to model the VCSEL nonlinearity. We show that the proposed modeling provides accurate estimations of the nonlinear distortions. The error vector magnitude (EVM) simulation results are very close to the measured EVM.

Multistandard RoF bus for in-building networks

This paper presents an optical bus architecture for multistandard communications which is competitive in cost and performance as regards typical star architecture. A point-to-point optical link is characterized by means of noise figure measurements before providing link budget calculations for the bus architecture. It is demonstrated that more than 10 m radio coverage per room can be obtained for cellular and wireless connectivity for up to 20 access points under the condition of optimizing laser modulation operating conditions. These performances are sufficient for in-building distribution and can be extended to an hybrid star-bus network topology, suitable for large building network infrastructure.

Optimization of power coupling coefficient of a carrier depletion silicon ring modulator for WDM optical transmissions

The presented method allows to evaluate the impact of power coupling on the performances of a carrier depletion ring resonator transmitter. From a measurement-based modulator model, coupling coefficient is varied and optimized maximizing modulation amplitude and reducing the impact of its process variability.

OFDM for optical wireless systems under severe clipping conditions

In this paper, we analytically model the impact of clipping and quantization for optical wireless communication systems which employ direct-current biased optical orthogonal frequency division multiplexing (DCO-OFDM). Considering uniform quantization for both digital-to-analog converter (DAC) and analog-to-digital converter (ADC), analytical expression for signal-to-distortion-quantization ratio has been evaluated. A lower bound on required bit resolution for DAC/ADC, as a function of clipping ratio has also been determined for a target bit-error-rate (BER). Gain in modulation power at the output of the DAC under severe clipping conditions has been analytically evaluated. Furthermore, an iterative method for clipping mitigation is presented, which can be employed to relax the bit resolution required by the DAC, and further, to mitigate the clipping distortions.