Bit Error Rate Floor Research Articles

Interferometric crosstalk reduction by phase scrambling

Interferometric crosstalk, arising from the detection of undesired signals at the same nominal wavelength, may introduce large power penalties and bit-error rate (BER) floor significantly restricting the scalability of optical networks. In this paper, interferometric crosstalk reduction in optical wavelength-division multiplexing (WDM) networks by phase scrambling is theoretically and experimentally investigated. Enhancement of 7- and 5-dB tolerance toward crosstalk is measured in a 2.5-Gb/s transmission link of 100 km and 200 km of SSMF, respectively. This result proves the feasibility of optical networking in the local area network/metropolitan area network (LAN/MAN) domain while tolerating the relatively high crosstalk levels of present integrated optical switching and cross-connect technology. Experiment is in good agreement with theory. Recommendations on the use of phase scrambling to reduce crosstalk in WDM systems are given.

Maximum likelihood sequence detection using a pilot tone

This paper derives, analyzes, and simulates a maximum likelihood (ML) sequence detector (MLSD) for a linearly modulated signal transmitted with a pilot tone (PT-MLSD). The transmitted signal is distorted by a time-varying frequency-selective Rayleigh fading channel and corrupted by additive Gaussian noise. The received signal is unsynchronized in that there are residual carrier frequency, carrier phase, and symbol timing offsets. The PT-MLSD treats the channel as a stochastic process, and so symbol sequences are distinguished by their autocovariances. Coherent communication is possible even in overspread channels. As the sequences' autocovariances explicitly account for the channel's time variation, the PT-MLSD's bit error rate (BER) floor is normally lower than the BER floor suffered by receivers that estimate the channel impulse response conventionally. Both the data-bearing signal and pilot tone are used together for synchronization, equalization, and detection. The pilot tone is only needed to remove the constellation's phase ambiguity and provide a stable amplitude reference for QAM constellations. It is not needed for estimating the channel impulse response. The pilot tone does not require a spectral null for its insertion, and it does not significantly degrade the peak-to-average or maximum-to-minimum power ratios. Thus, many of the disadvantages of other pilot tone systems are avoided, as there is no bandwidth expansion, and linear amplification is not made appreciably more difficult.

Performance of Optically Preamplified Receivers in WDM Systems Disturbed by Interferometric Crosstalk

Interferometric crosstalk is a performance limiting factor of major concern in all-optical WDM transmission networks. Interferometric crosstalk arising from performance imperfections in optical components may introduce large power penalties and bit-error rate floors. Optical amplifiers are often used to increase the signal level incident on a detector so that high receiver sensitivity can be obtained. We investigate theoretically and experimentally the performance of optically preamplified, direct detection receivers in the presence of interferometric crosstalk. The model includes an accurate description of filtered interferometric crosstalk by using a so-called maximum entropy approach. Experimental results, using both directly and externally modulated light sources, are found to be in good agreement with theory.

List decoding of turbo codes

List decoding of turbo codes is analyzed under the assumption of a maximum-likelihood (ML) list decoder. It is shown that large asymptotic gains can be achieved on both the additive white Gaussian noise (AWGN) and fully interleaved flat Rayleigh-fading channels. It is also shown that the relative asymptotic gains for turbo codes are larger than those for convolutional codes. Finally, a practical list decoding algorithm based on the list output Viterbi algorithm (LOVA) is proposed as an approximation to the ML list decoder. Simulation results show that the proposed algorithm provides significant gains corroborating the analytical results. The asymptotic gain manifests itself as a reduction in the bit-error rate (BER) and frame-error rate (FER) floor of turbo codes.

Error floors of digital FM in simulcast and Rayleigh fading

A digital FM system in the presence of simulcast interference and/or Rayleigh fading exhibits performance floors, where increased signal strength does not reduce the error rate. Results obtained both in the lab, using a radio channel simulator, and in the field show this performance. Observation of signal waveforms indicates that the bit-error rate (BER) floor is a result of FM clicks during deep fades. We present a mathematical model that can be used to calculate the BER floors. The model is based on the calculation of the click shape as a function of amplitude imbalance, frequency offset, and postdetection filter impulse response. The Rayleigh fading is approximated with a two-signal simulcast by substituting the RMS Doppler spread for the simulcast frequency offset. A good match between calculated and measured results is obtained. It is found that the symbol errors caused by simulcast and Rayleigh fading can be treated as independent, and their effect is cumulative.

Homodyne crosstalk in WDM ring and bus networks

We examine the combined effects of coherent and incoherent crosstalk in wavelength-division multiplexed (WDM) ring and bus networks which use add-drop multiplexers. It is widely understood that incoherent crosstalk in these networks causes a fixed bit-error-rate (BER) door at the receiver. We show that coherent crosstalk causes a range of possible received powers. In combination with incoherent crosstalk, this leads to a range of possible BER floor positions. We use a Monte Carlo simulation to show that when both coherent and incoherent crosstalk are considered, BER floors are likely to be higher than previously appreciated. Satisfactory network performance can be ensured by careful network management, or by reducing component leakage.

Selective serial concatenation of turbo codes

A serial concatenation scheme consisting of a Bose-Chaudhuri-Hocquenghem (BCH) outer code and a turbo inner code is proposed. We first establish that only a small number of bit positions at the turbo decoder output are likely to be in error at high signal-to-noise ratios. A double-error correcting BCH outer code is used to protect these particular error prone bits. Simulation results for an additive white Gaussian noise (AWGN) channel show that the bit-error rate (BER) floor of the turbo code can be lowered by using this serial concatenation scheme. The proposed technique offers higher throughput efficiency and lower complexity than other serial concatenation schemes.

The equivalent transmission-path model-a tool for analyzing error floor characteristics due to intersymbol interference in Nakagami-Rice fading environments

Land mobile radio systems such as car telephones and handy personal terminals used outdoors have enjoyed a remarkable evolution. To design reliable mobile radio systems, however, it is vital to have a good understanding of the impact of wave propagation characteristics on digital transmission quality in a wide variety of mobile radio environments. A very simple but general scheme for calculating irreducible bit error rate (BER) (namely, BER floor) due to intersymbol interference in frequency-selective Nakagami-Rice fading environments has been developed. The scheme, which we call the equivalent transmission-path model, plays a role in connecting wave propagation with digital transmission characteristics in a general manner. Through computer simulations assuming various types of delay profiles, we first identify key parameters of Nakagami-Rice fading dominating principally the occurrence of irreducible errors and we develop a simple method for the calculation of irreducible BER utilizing the nature or the key parameters. Then we examine the accuracy of the scheme for various types of phase-shift keying (PSK) transmission systems. Finally, based on the scheme, we show calculation examples of BER floor characteristics in line-of-sight fading environments.

BER Floor Due to Modulation Instability in a 2.5 Gbit/s Field Trial with +22 dBm Booster EDFA

We present the results of a field trial with a commercial 2.5 Gbit/s transmission system at 1545 nm on dispersion-shifted fiber and a +22 dBm booster erbium-doped fiber amplifier (EDFA). It was found that the amplified spontaneous emission of the EDFA was parametrically amplified by the process of modulation instability, which caused a bit-error-rate (BER) floor of 10−10. There is a good agreement between the theoretically predicted and the measured optical spectrum and BER floor.

Performance analysis of CDMA optical communication systems with avalanche photodiodes

The performance of optical code-division multiple-access (CDMA) communication systems with avalanche photodiodes (APD) is analyzed. The bit error rate (BER) can be accurately computed by using the saddlepoint method. The effects of the multiple-user interference (MUI), signal-dependent shot noise, and receiver thermal noise are investigated. Results of the numerical integration illustrate the non-Gaussian property of the receiver output distribution. Exact means and variances are derived for the Gaussian approximation. It is found that when the MUI increases, the saddlepoint approximation yields satisfactory results, but the Gaussian approximation yields higher BER floors. Prime codes and on-off keying (OOK) are considered. Examples illustrate the effects of the system parameters such as the APD gain, threshold, prime code length, and the number of simultaneous users. >

Adaptive two-stage maximum likelihood estimation for cellular radio

The problem of optimal detection over nonselective fading channels for cellular radio is treated from the viewpoint of adaptive two-stage maximum likelihood estimation (MLE). Data for transmission are divided into blocks, each preceded by one or more reference symbols. For selective fading, the receiver consists of an adaptive filter (AF) and a two-stage maximum likelihood estimator (TSMLE). For nonselective fading, TSMLE is preceded by a fixed filter. For quadrature modulation, the complex output of the filter contains two sets of unknowns, the quadrature data symbols (QDS) and the quadrature gains (QG) between transmitter and filter output. Stage I generates MLE estimates of the QG during the reference period and Stage II generates QDS estimates during the data intervals. AF coefficients are adjusted using a QR-decomposition least squares (QRD-LS) algorithm. A likelihood ratio test is performed to produce noise-free data estimates which Stage I in turn can use, in a decision aided manner, to update its gain estimates for the next interval. The quadrature gain estimates can be used to further correct reference carrier phase errors. Nonrecursive and recursive estimation procedures are developed. Bit error rate (BER) expressions are presented for 4-PSK and pi /4-DQPSK. Numerical results show that the scheme meets the North American Digital Cellular (NADC) system performance specifications (IS-54) with a margin. The BER floor is an order of magnitude lower than those of other schemes reported in the literature. For nonselective fading the BER floor is practically eliminated. This, and results reported elsewhere (Haeb and Meyr, 1989), lead us to conclude that schemes performing quadrature gain estimation can be used to lower error probabilities, attractively.

Error‐Correction Coding for High‐Speed Optical Transmission Systems Based on the Synchronous Digital Hierarchy

AbstractThe quality objectives given in the CCITT recommendations concerning fiber optics transmission over the Synchronous Digital Hierarchy (SDH) network are difficult to achieve in high‐speed systems. Equipment imperfections often result in a Bit Error Rate (BER) floor above the recommended BER value of 1CH4. In this paper we propose a simple Reed Solomon (RS) coding scheme for the Synchronous Transport Module level‐1 (STM‐1) frame which leads to a substantial BER performance improvement over uncoded systems. The RS code proposed is a shortened one‐symbol‐correcting code. Experimental results obtained with a laboratory prototype are reported which show that this code reduces the BER floor by four orders of magnitude.

An asymmetrical pulse shaping technique to combat delay spread

Residual error floors caused by delay spread are reduced with a new pulse-shaping (or filtering) technique described in the present paper. Delay spread, which causes frequency-selective fading, is a major cause of error floors in high-speed digital mobile radio systems. The authors propose a new class of nonlinear-phase intersymbol interference (ISI) free filters that lead to asymmetrical pulse shapes and are more robust against frequency-selective multipath fading. The nonlinear-phase filters do not require redundant signalling bandwidth in comparison to raised-cosine filters which are more robust against delay spread. A piece-wise linear-phase filter with cosine roll-off is designed to demonstrate the robustness of nonlinear-phase filters against delay spread. Coherent quarternary phase-shift keying (QPSK) modems and the two-ray Rayleigh fading model are assumed in the analyses. For the filter roll-off factor alpha =1, the delay tolerance increases from tau =0.5 T (where tau is the relative delay of two paths and T is the symbol duration) for the raised-cosine filter to 0.65 T for the nonlinear-phase filter if bit error rate floors (BERF)=3% is used as a benchmark and the equal-power two-ray Rayleigh fading model is assumed.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Operation of a passive optical network with subcarrier multiplexing in the presence of optical beat interference

When two lasers carrying subcarrier-multiplexed data operate with very closely spaced wavelengths, beating between the lasers can occur which increases the noise at the detector. This effect, optical beat interference (OBI), can greatly interfere with the operation of these systems. The authors have observed OBI with DFB lasers in a system carrying two 155.5-Mb/s data streams. With an optical modulation index of m=0.8, a bit-error rate floor of approximately 10/sup -2/ was observed due to OBI, which is consistent with the noise measured by an RF power meter. When m is increased to 1.8, a significant reduction in OBI was observed, and it was possible to achieve transmission at the 10/sup -9/ level with a penalty of only 1.8 dB due to the presence of OBI. >

Elusive bit-error-rate floors resulting from transient partitioning in 1.5- mu m DFB lasers

Some 1.5- mu m distributed feedback (DFB) lasers modulated at 1.7 Gb/s exhibit bit-error rate (BER) floors for certain fiber lengths, and hence these floors are difficult to detect. For example, a laser that does not exhibit a BER floor for 43 km of fiber may exhibit a BER floor after only 27 km. It is shown that the source of such BER floors is a narrow pulse resulting from transient partitioning between the DFB TE and TM modes. The narrow TM pulses cause false bits if the fiber delay between the modes positions the TM energy within a zero time slot. It is shown that for the 11 lasers studied, TM partitioning floors rarely occur in lasers with adequate threshold gain difference. When the duration of the TM partitioning is only a small fraction of a bit period, the BER floor is effectively removed using a polarized film in the transmission path to suppress the TM mode. Similarly, increased suppression of the TM mode has been achieved by changing the laser structure to reduce the TM mode gain. >

Performance of lightwave systems incorporating multilongitudinal mode laser and optically preamplified receiver combinations

The performance of optical communication systems employing multilongitudinal mode laser and semiconductor laser preamplified receiver combinations is investigated theoretically, emphasis being placed on the influence of mode partition noise (MPN). Two types of MPN are considered in the analysis: one owing to wavelength-dependent fibre dispersion and the other owing to the wavelength-dependent attenuation mechanism associated with both the amplifier gain and optical filter transmission characteristics. To reduce the influence of amplifier gain ripple and to ensure high receiver performance the mode spacing and the centre mode wavelength of the laser source may need to match closely those of the amplifier, especially for high internal amplifier gains and substantial facet reflectivity. For example, for 1% amplifier facet reflectivity, 30 dB internal optical gain, and 3 nm half-spectral-width laser the receiver sensitivity degrades by 5.1 dB when the lengths of the laser and the amplifier differ by 5%. In contrast with 0.1 and 0.01% reflectivity this reduces to 1.95 dB and 0.4 dB, respectively. We also show that for multimode laser operation there is an optimum bandwidth for the bandpass optical filter at which the required detectable optical power for a given bit error rate (BER) is minimum. Operating with filter bandwidth less than this optimum value will degrade dramatically the sensitivity and may, even in the absence of fibre dispersion, lead to a BER floor in the presence of MPN. The analysis presented here may be used to provide guidelines for the design of optical systems based on laser amplifiers and multimode or nearly single mode laser sources.

Influence of the non-Lorentzian emission lineshape of laser diodes on the BERfloor in DPSK heterodyne optical systems

A bit-error rate floor for DPSK heterodyne optical systems is evaluated, considering the actual shape of the emission line of single-mode semiconductor lasers. Considering a strictly Lorentzian shape can give rise to relevant errors in the design of the above systems. An ‘equivalent linewidth’ is introduced which, when accounted for in the formalism usually adopted, allows the correct results to be obtained.

Frequency-noise cancellation in semiconductor lasers by nonlinear heterodyne detection

The bit-error-rate (BER) performance of conventional noncoherent, heterodyne frequency-shift-keyed (FSK) optical communications systems can be surpassed by the use of a differential FSK modulation format and nonlinear postdetection processing at the receiver. A BER floor exists for conventional frequency-shift keying because of the frequency noise of the transmitter and local oscillator. The use of differential frequency-shift keying with nonlinear postdetection processing suppresses this BER floor for the semiconductor laser system considered here.