摘要:

Semiconductor laser amplifiers can exhibit either temporal pulse compression or broadening. A transmission-line laser model was used to isolate the mechanisms responsible and to investigate the conditions required for this form of pulse distortion. Results show that gain saturation is the dominant temporal compression mechanism in travelling wave laser amplifiers, and that filtering may account for some compression in Fabry-Perot laser amplifiers. Compression is highly dependent on the input pulse shape, being zero for Gaussian-like pulses. This means that it is unlikely to be useful for shortening pulses produced by highly modulated injection lasers. However, compression can occur with pulses with a slowly decaying trailing edge, but only over a limited range of input powers. Pulse broadening is also caused by gain saturation and is again dependent on the input pulse shape, being zero for square-edged pulses. For Gaussian pulses, the broadening is negligible for input pulse energies below the amplifier's saturation energy. This suggests that TWLAs may be used to boost the power of pulses from mode-locked semiconductor lasers. These results are in excellent agreement with other workers' experimental results.

DOI:10.1049/ip-j.1989.0025

出版商:IEE    

年代:1989

数据来源: IET

摘要:

A theoretical study of radiation from bends in multimode optical fibres is developed on the basis of ray theory, and the results are compared with experimental measurements. The same theory is also used to predict the power in the core of the fibre as a function of angle around the bend. On the basis of the agreement with experiment for the radiated power from multimode fibres, the same technique is then applied to a single mode fibre using a single axial ray to represent this case. A solution is obtained which is close to that previously obtained experimentally, and also close to alternative theoretical solutions.

DOI:10.1049/ip-j.1989.0026

出版商:IEE    

年代:1989

数据来源: IET

摘要:

A design of an integrated 2×2 directional-coupler switch for λ = 1.3 μm in electro-optic polymers is presented. It is shown that the switch configuration with electro-optically induced waveguides is feasible in these materials. Both switching states of the proposed device are tunable. The parameters predicted are as follows: the polarisation voltage is lower than 10 V in both states, the coupling length is smaller than 1 mm, and the lowest on/off ratio is 26.4 dB.

DOI:10.1049/ip-j.1989.0027

出版商:IEE    

年代:1989

数据来源: IET

摘要:

Weakly coherent optical communication systems, in which the receiver measures the signal power within a passband determined by the IF filter without using phase information in the signal, can be designed to accommodate severe laser phase noise. This paper compares theoretical predictions for the effect of phase noise on these systems, using the models of Garrett and Jacobsen, Kazovsky, and of Foschini and co-workers, and models derived from them. For total line-widths of less than about the bit rate, all these models agree reasonably well in predicting the IF bandwidth required to accommodate the phase noise. For larger linewidths, it is shown that significantly narrower IF filters may be used than has been predicted by any model published previously. As examples, it is predicted that effective IF bandwidths of about six and eleven times the data rate can accommodate total linewidths of as much as three and ten times the data rate, respectively.

DOI:10.1049/ip-j.1989.0029

出版商:IEE    

年代:1989

数据来源: IET

摘要:

The surface integral equations method, capable of treating a guiding structure composed of bounded homogeneous dielectric cylinders, is generalised to treat unbounded (in a transverse plane) waveguides in integrated optics, such as the slab-coupled waveguide. The approximation made in the present method is to truncate associated infinite integrations. Justification of such a truncation is given.

DOI:10.1049/ip-j.1989.0030

出版商:IEE    

年代:1989

数据来源: IET

摘要:

In a digital optical transmission system, the decision threshold, the sampling instant and the gain of the avalanche photodiode need to be maintained at their optimum values in order to maintain a minimum bit error ratio. In the paper the authors consider in detail the theory of steering these three parameters while pseudoerror monitoring is used to monitor the system performance, and they show that this decision-point steering can allow the optical link to maintain its performance under system variations, thus allowing a reduction in the system power margin that needs to be allocated.

DOI:10.1049/ip-j.1989.0031

出版商:IEE    

年代:1989

数据来源: IET

摘要:

Waveguide photodetectors, in which the direction of the incident light is perpendicular to direction of the carrier collection, are characterised by their high speed and high quantum efficiency, but also the possibility of bias-free operation. Basic design equations for waveguidepinheterojunction photodetectors are derived, and criteria for optimum operation in terms of external quantum efficiency and bandwidth are discussed. The analysis shows that bandwidths of more than 150 GHz with a quantum efficiency of 50%, limited by the fibre-to-detector coupling efficiency, is achieved at zero-bias operation.

DOI:10.1049/ip-j.1989.0032

出版商:IEE    

年代:1989

数据来源: IET

摘要:

Developments of optical fibre communication systems have until recently concentrated on increasing both the distances between repeaters and the intensity modulation bandwidths for single channel optical transmission. Tremendous achievements have been made in both areas such that, in the latter case, it is electronic component limitations which are starting to restrict the speed of operation of experimental systems. In this context, there is renewed interest in wavelength division multiplexing as a technique to upgrade the transmission bandwidth capacities of existing systems together with the possible future provision of broadband multichannel systems. A critical review of the component technology to enable wavelength division multiplexing is therefore useful at this stage. The major implementations of wavelength multiplexer/demultiplexer devices are discussed and compared with particular emphasis placed on their optical efficiency, spectral stability, crosstalk performance and the number of channels that may be provided. Passive devices ranging from the bulk optic implementations (i.e. prisms, gratings, filters) to integrated waveguide and fibre directional coupler realisations are considered. In addition developments associated with active devices including integrated source and detector arrays, wavelength tunable devices and optical amplifiers which may be employed within wave length multiplexed systems are outlined to provide an insight into possible future directions within the field.

DOI:10.1049/ip-j.1989.0033

出版商:IEE    

年代:1989

数据来源: IET