18 February 2013
The Institute is a part of the Siberian Branch of the Russian Academy of Sciences and we are members of Photonics Laboratory. The research carried out in the Laboratory, which is now ten years old, is mainly focused on three areas: optical communications, theory of photonic crystals and Bragg gratings, nonlinear optics of crystals and fibres. In addition we teach students of the Novosibirsk State University in Mathematical Physics, Computational Physics and Fibre Optics.
Currently the main operating condition of fibre communication links is the amplitude modulation. The logical “1” is transmitted as an optical pulse, while the logical “0” corresponds to the absence of pulse in a bit interval. As the bit rate increases, the influence of nonlinearity, chromatic dispersion and the effect of spontaneous noise of amplifiers distort the signal, and then errors appear. In order to make the information flux greater, researchers study advanced data formats. Among them, phase modulation with differential phase shift, well known in digital radio, is promising and interesting - the phase modulation keeps the amplitude constant then the nonlinear effects do not distort the bit pulses, in contrast to amplitude modulation.
In our Letter, we compare two differential phase-shift keying formats: with π and π/2 phase shift. Each format has its own bit error rate. The recognition of “1” or “0” in the random bit sequence depends on the statistical dispersion of the signal. We show the dispersion to be essentially less for the π/2 shift, so the bit-error rate is lower for this format and its Q-factor is greater.
The advantage of π/2 shift was observed in experiment ten years ago, and then confirmed by numerical simulation. However, no explanation of this property was given. After the mathematical proof, the advantage becomes clear. Some aspects will be ready for application once they are studied experimentally and confirmed by theoretical calculation. Our format seems to be appropriate for the next generation of high-capacity optical networks, and now π/2 differential phase-shift keying can be used in telecommunications, as our proof guarantees its smaller bit error rate.
The problem of data communication by wavelength division multiplexing with phase-shift keying will be analysed where the multiplexing will be performed by using quasi-rectangular Bragg filters. We are also going to study how the mean dispersion of the link and the shape of the transfer function of filters provide the decrease in bit-error rate for each channel. For this purpose it is necessary to select the front steepness and width of filters that provide the minimal overlap of pulses in neighbour spectral channels. When we formulate the requirements of the filter shape, the Bragg grating can be reconstructed by the solving of an inverse scattering problem. We will also try to calculate the dispersion of high-order formats with π/n phase shift.
Other than the present project, we work on the optimisation of a multichannel (comb) filters, and the scattering problem for an evanescent wave. After scattering by metal or dielectric nanoparticles a propagating wave is generated, and this diverging wave is observable. It is registered in experiment that the problem is significant. However, at the same time, for evanescent waves, the scattering problem is far from its solution.
The traffic in telecommunication industry grows exponentially. In the next ten years the needs will exceed the limiting capacity of long-haul lines. The problem of maximum the capacity of optical communication lines becomes most important: Is it determined by the Shannon theorem? Does the nonlinearity allow one to overcome the Shannon limit? We hope the near future will give the answer.
A PDF version (new window) of this interview is available.
Browse or search all papers in the latest or past issues of Electronics Letters on the IET Digital Library.