With the widespread of new radio systems, it is essential that their performance is evaluated prior to auxiliary design and commercial development. Although there are many methods to model and estimate the performance of a given radio system with some accuracy, most of these methods require high computational resources or long simulation times. In this study, the authors present a general and efficient simulation method for estimating the performance of a wireless communication systems, with particular emphasis for its forward error correction sub-system when using block channel codes, for channels affected by additive white Gaussian noise, free space attenuation, shadowing, fading and interfering sources. Both soft-decision (SD) and hard-decision (HD) decoding are considered and the overall performance takes into account modulation, coding and channel conditions. This efficient simulation is compared with the Monte Carlo simulation method and its performance is assessed for the Tomlinson, Cercas and Hughes code family. This new simulation method has an temporal efficiency of two or three orders of magnitude for SD and HD and three or four orders, considering a bit-error-ratio of 0.01 and 0.001%, respectively.

The performance of orthogonal frequency division multiplexing ultra wideband (UWB) radio signals distribution in long-reach passive optical networks (LR-PONs) using conventional chirp-less Mach Zehnder (MZ) and linearized (L) Y-fed directional couplers electro-optic modulators (EOMs) is compared. Particularly, the optimum modulation index and the corresponding minimum optical signal-to-noise ratio (OSNR) required to achieve a bit error probability of 10-12are evaluated through numerical simulation for systems operating with single and three UWB sub-bands and different standard single-mode fiber (SSMF) distances indicated for LR-PONs. Both modulators are characterized experimentally and theoretical model parameters are adjusted to correctly describe the power and chirp characteristics of the electro-optic conversion in the simulation process. It is shown that the optimum modulation index and the tolerance to modulation index variations, considering either single or multi band UWB operation, is approximately two times higher for systems employing L-EOM than MZ-EOM. Additionally, with MZ-EOM, the required OSNR may increase considerably when the fiber length increases due to the power fading induced by fiber chromatic dispersion, achieving a penalty of almost 3 dB for systems with 100 km of SSMF and single UWB sub-band operation. Instead, with L-EOM, the required OSNR penalty (when compared with the back-to-back case) is lower for fiber lengths up to 100 km than it is for systems employing the MZ-EOM modulator due to the combined effect of the chirp generated by the modulator and the chromatic dispersion.

This paper presents experimental results on an electrical adjustable linear filter prototype for compensating impairments due to residual group velocity dispersion (GVD) and first-order static polarization mode dispersion (PMD) on 40 Gb/s systems. Optical single-side band (OSSB) modulation is considered and was verified to be responsible for an increase in system tolerance to GVD and superior performance of the electrical filter. The detrimental effect of packaging on the filter prototype electrical characteristics is analyzed. Simulation and measured results are compared for GVD and first-order static PMD compensation, showing the practical applicability of the developed prototype on 40 Gb/s OSSB systems.

This work aims at presenting simulation results of performance of various Tomlinson-Cercas-Hughes (TCH) codes, namely, TCH(16,6,2), TCH(32,6,6), TCH(64,7,13), TCH(128,9,27) and TCH(256,14,55)#1 for a land mobile satellite channel, where multipath fading and shadowing occur. Performance results are presented for two environments, a city and a highway. It is shown that, in a highway environment with a small elevation angle, TCH codes with low code ratio do not show significant advantage over TCH codes with greater code ratio for high bit error ratio (BER), mainly, for BER>10/sup -5/. For lower BER, the TCH codes with lower code ratio present the expected gain. These results are explained by the statistical properties of bits in error, such as error bursts size and intervals without errors, as a function of signal-noise ratio.