Performance degradations of multigigabit-per-second NRZ/RZlightwave systems due to gain saturation in traveling-wave semiconductoroptical amplifiers

A nonlinear model for a travelling-wave semiconductor optical amplifier has been used to determine eye closure degradations for 2.4 and 10 Gb/s NRZ/RZ lightwave systems due to gain saturation effects in the optical amplifier. At 10 Gb/s, with a carrier lifetime of 300 ps, the results indicate that the penalty is less than 1 dB for both NRZ and RZ systems provided that the ratio of the input power (P_in ) to the saturation output power (P_sat) is less than -17 dB. The NRZ system penalty is slightly larger than the RZ penalty when P_in/P_sat is larger than -17 dB. For example, with P_in/P_sat=-10 dB, the NRZ system penalty is about 2.8 dB versus 2 dB for the RZ system. The system penalty at 2.4 Gb/s is slightly less than that at 10 Gb/s. At P _in/P_sat=-10 dB, the NRZ system penalty is about 2.5 dB versus 1.5 dB for RZ     

High-speed multiple-quantum-well optical power amplifier

A multiple-quantum-well optical amplifier is used at 8 Gb/s with high input power such that the average gain is compressed by 2.6 dB. Under these conditions, the output signal level is 35 mW and there is negligible intersymbol interference (ISI) penalty at the receiver. This is possible because of a rapid (7 ps) gain recovery process in the amplifier. A conventional semiconductor amplifier operating at a similar level of gain compression shows 2 dB ISI penalty     

Threshold IBO of HPA in the predistorted OFDM communication system

Orthogonal frequency division modulation (OFDM) system is very useful for the multi-path channel and highly bandwidth-efficient system. However, OFDM signal shows the high peak to average power ratio (PAPR) so that nonlinear distortion can happen in the high power amplifier (HPA). This paper studies BER performance variation according to input back off (IBO) values when the predistorter is used for the compensation for the HPA nonlinearity. In the case of the solid state power amplifier (SSPA) of nonlinearity parameter p=0.5 and IBO=0 dB, the system with predistorter is poorer than the one without predistorter. But this situation is inverted, if IBO is increased from 0 dB to 3 dB. So, we can find that there is a threshold value of IBO for the predistorter to effectively compensate for the nonlinear SSPA when the sub-carrier number and p value are changed. This threshold IBO value is proportional to the nonlinearity parameter p value but it is nearly independent of the sub-carrier number because amplitude and distribution are normalized. The OFDM system with predistorter can be improved only when thee IBO value is higher than the threshold value.     

Theory of signal degradation in semiconductor laser amplifiers indirect detection systems

Theoretical studies are carried out on long haul direct detection optical fiber communication systems, with inline optical semiconductor amplifier repeaters. Calculations are made of the noise, eye diagrams, and bit-error-rate characteristics of lightwave systems with optical amplifiers. Indications are given of the effect of amplifier characteristics such as spontaneous noise and signal distortion due to gain saturation on the system performance. The nonlinear process within semiconductor laser amplifiers leads generally to pulse amplitude-temporal distortions due to gain saturation. This theoretical study demonstrates that the system penalty caused by these nonlinear effects appears progressively as the optical input power at each amplifier is increased. For example, nonregenerated fiber transmission using traveling wave semiconductor laser amplifiers was simulated, and results obtained at 0.5 and 2.5 Gb/s are presented. In order to improve the system performance, the influence of structure and bulk dimensions of the amplifier cavity is also considered     

HBT power amplifier with dynamic base biasing for 3G handset applications

In this paper, a performance optimization of a digital signal processing driven and dynamically biased 3G handset's power amplifier technique is proposed, simulated and implemented. This technique uses a new dynamically base biasing heterojunction bipolar transistor which reduces the dc power consumption at low level drive and at the same time compensates the nonlinear distortion at high power drive of the PA in the transmitter of a universal mobile telecommunications system (UMTS) system with a high integrability. With the UMTS system, at low level drive, the dc power reduction is about 60% and at high emission power, the nonlinearity of the PA is corrected to respect the adjacent channel power ratio (ACPR) and error vector magnitude constraints imposed by the UMTS. With our system, the ACPR and efficiency of the power amplifier are improved, respectively, by 5 dB and 8%.     

OFDM link performance with companding for PAPR reduction in the presence of non-linear amplification

Use of companding for peak-to-average-power ratio (PAPR) control is explored for a link involving a nonlinear transmit power amplifier with orthogonal frequency division multiplexing (OFDM). Specifically, the objective of the study was to determine if companding using u-law compression/expansion at the transmitter/receiver, respectively, provides end-to-end performance gains relative to a system without companding. We consider the use of companding to ameliorate the impact of nonlinearities in the transmit amplifier. In the absence of companding, transmitter operation near saturation raises the signal level at the receiver but, because of the nonlinearities in the amplifier response, also results in distortion that impacts overall link performance. As the transmit power is backed-off from saturation, amplifier distortion is reduced, but error components due to lower SNR at the receiver become more significant. When companding is introduced in the system, the system is able to operate closer to saturation without substantial transmit distortion. However, requisite expansion of the compressed signal at the receiver yields noise amplification which can counteract any of the performance gains that would otherwise accrue from the increased SNR at the receiver. At issue is whether or not operating conditions exist (e.g., backoff, SNR, amplifier linearity model, etc) for which companding enhances the end-to-end performance relative to the link performance without companding. System simulation models were employed using Rapp's nonlinear power amplification models, where average symbol distance errors were used as performance metrics. We found that companding can provide very modest performance gains in comparison to systems that do not employ companding. Performance trends were corroborated in a hardware testbed with an amplifier chain, where average bit error rates were experimentally determined.     

Laser diode pumped Er3+-doped fiber amplifier in a 565Mbit/s DPSK coherent transmission experiment

A laser-diode-pumped erbium-doped fiber amplifier, exhibiting 9-dB gain, has been operated as an in-line optical repeater in a 565-Mb/s coherent optical communications system. A sensitivity penalty of 0.4 dB was observed when the amplifier was positioned 35 dB away from the receiver, thus indicating a system improvement of 8.6 dB. By progressively reducing the coupling loss between amplifier and receiver, the noise figure of the contradirectionally pumped amplifier was calculated to be 5.4 dB, a value which is consistent with simple noise theory     

改进负载调制的GaN Doherty功率放大器及其线性化

在分析传统Doherty负载调制的基础上,通过选取合适的峰值放大器负载阻抗和采用较高的偏置电压,增强了Doherty功率放大器的负载调制,使其适用于大范围（9dB）回退情况下的应用。为了验证分析的有效性,设计和实现了一个具有100MHz瞬时带宽的2.55GHz GaN Doherty功率放大器。测试结果表明：在工作带宽内饱和功率约为49.4dBm,平均峰值效率为64%,9dB回退时的平均效率约为40%。当使用5载波100MHz带宽LTE-advanced信号激励时,在平均输出功率为40.2dBm时效率可达40.3%,经过数字预失真校正过的邻道泄漏比（ACLR）低于-48dBc,达到较好的线性度。     

FM microwave multiplexed broad-band distribution systems using Er3+ fiber amplifiers and preamplifiers

The use of an Er³⁺ fiber preamplifier for microwave multiplexed systems and the use of an inline Er³⁺ amplifier in microwave multiplexed systems for signal distribution are reported. The improvement in receiver sensitivity as a preamplifier, without optical filtering, was 9 dB. No power penalty due to amplified spontaneous emission was found when the amplifier was used in a 30-channel signal distribution system     

基于维纳模型的非线性信道接收端均衡方案

针对功率放大器的非线性问题和多径信道引起的时间色散,提出了一种接收端联合补偿方案。由基于维纳模型的均衡器实现,该均衡器由一个线性滤波器级联一个多项式滤波器组成,所有运算均在接收端完成。仿真结果显示,经过非线性均衡的误码率性能比无非线性均衡时有很大改善,在同一误码率下,信噪比至少降低了3.5 dB。     

The search for optimized shaping filters suitable for vsat systems

The satellite communications industry has seen the emergence of VSAT (very small aperture terminal) systems with the promise of high growth in the next decade. The system design of the VSAT ground station has generally followed traditional satellite communication system design, namely featuring data transmission using QPSK (quadrature phase shift keying) or BPSK (binary phase shift keying) and 50 per cent raised cosine shaping filters with quasilinear transmission. Transmitter power is strictly limited and is at a premium for the VSAT transmitter which usually consists of a non-linear solid-state power amplifier (SSPA). Consequently it is important to operate the SSPA so that the maximum power is obtained, which means that the SSPA should be operated close to its saturation point. In so doing considerable distortion can be introduced, which leads to intersymbol interference (ISI) in the receiver and an associated degradation in error rate. Although constant envelope BPSK or QPSK solves the intersymbol interference problem and enables operation at the saturating point of the SSPA, the sinx/x spectral shape has unacceptable sidelobe levels. This paper gives results of a search for a pulse shaping filter characteristic that is bandlimited and yet has improved performance over the traditional root 50 per cent cosine roll-off filter when used in a VSAT ground station with a SSPA. The performance improvement due to improved shaping filtering is maintained over a range of output levels of the SSPA. Performance evaluation results using computer simulation are presented.     

Optimum phase deviations between mark-and space-state bits inlong-haul coherent communication systems with balanced PLLreceivers

In this paper, we report the importance of the optimum phase deviation between mark-state and space-state bits in a long-haul PSK homodyne system with a balanced phase-locked-loop (PLL) receiver. The laser phase noise, data-to-phase-lock crosstalk, receiver shot noise and amplifier noise are considered in analyzing the systems with cascaded optical amplifier chains. The signal power penalty incurred by improper use of phase deviations in a long-haul system is evaluated. It is found that such power penalty can amount to larger than 1.5 dB in a 10-Gbt/sec system     

55-mW 200-MSPS 10-bit pipeline ADCs for wireless receivers

A new power reduction technique for analog-to-digital converters (ADCs) is proposed in this paper. The power reduction technique is a kind of amplifier sharing technique and it is suitable for ADCs in a wireless receiver. A test chip, which contains two ADCs, is fabricated in 90-nm 1-poly 7-metal CMOS technology. The 10-bit ADC dissipates 55 mW from 1.2-V supply, when the ADC operates at 200 mega-samples per second (MSPS). The 10-bit, 200-MSPS ADCs achieve maximum differential nonlinearity (DNL) of 0.66 least significant bit (LSB), maximum integral nonlinearity (INL) of 1.00 LSB, a spurious-free dynamic range (SFDR) of 66.5 dB and a peak signal-to-noise plus distortion ratio (SNDR) of 54.4 dB that corresponds to 8.7 effective number of bits (ENOB). The active area is 1.8 mm /spl times/ 1.4 mm.     

High-repetition-rate pulsed-pump fiber OPA for amplification of communication signals

The use of a high-repetition-rate pulsed-pumped fiber optical parametric amplifier (OPA), followed by a narrow optical filter for transparent signal amplification, was proposed. Theory and simulations predict larger gain and gain bandwidth compared to a continuous-wave pump with the same average power. Experimentally, when using a pump with 0.63 W of average power in a 500-m-long highly nonlinear fiber, the gain increased from 19.7 to 29.2 dB, and the bandwidth increased when a CW pump was changed to one that is modulated by a 20-GHz cosine-squared function. Clear eye openings were demonstrated for the amplification of a 10-Gb/s NRZ signal, with a power penalty of 1.5 dB.     

Power penalty assessment in optically preamplified receivers witharbitrary optical filtering and signal-dependent noise dominance

The assessment of the power penalty of optically preamplified receivers with signal-dependent noise (SDN) dominance is often accomplished by neglecting the influence of the optical filtering of the amplified spontaneous emission (ASE) noise on the signal-ASE beat noise. In this paper, it is shown that the optical filtering of the ASE noise can have a strong impact on the signal-ASE beat noise and remarkably affect the power penalty, even for optical filter bandwidths five times wider than the signal bandwidth. A simple expression to analytically evaluate the power penalty due to optical filtering, which describes reasonably well the influence of the optical filter on the signal-ASE beat noise variance, is proposed. The accuracy of the new expression is investigated, in the case of assessment of the optical filter detuning impact on receiver performance and the case of optical filter bandwidth optimization, and its predictions are satisfactory in comparison with rigorous estimates. Two new expressions of power penalty due to extinction ratio and to eye closure are also presented. It is shown that the power penalty due to eye closure depends on the extinction ratio and vice versa. Our results show also that the power penalty due to eye closure is remarkably dependent on the eye closure asymmetry     

1.3 mu m semiconductor laser power amplifier

The performance of a 3.4-Gb/s system using a low-power 1.318- mu m distributed-feedback (DFB) laser transmitter and a traveling-wave semiconductor laser power amplifier is studied. The -14.5-dBm, input from a directly modulated DFB laser is boosted to +10.3 dBm, of which +4.8 dBm is coupled into the transmission fiber. The penalty, caused by amplifier noise and pattern effects due to gain saturation, is less than 0.5 dB.<>     

A 1.7 Gb/s ASK and a 622 Mb/s incoherent FSK transmissionexperiment using a 1.55-μm multiquantum well distributed feedbacklaser

A multiple-quantum-well distributed-feedback (MQW-DFB) laser with narrow linewidth and low frequency chirp at low output power may experience linewidth rebroadening at high output power. the rebroadening is mostly due to a large carrier-induced change of refractive index, which also causes a large frequency modulation response for the MQW-DFB lasers. Using a 1.55-μm MQW-DFB laser, a 622-Mb/s amplitude-shift-keying (ASK) transmission experiment employing 200-km of fiber and an erbium-doped fiber amplifier has been demonstrated having a dispersion power penalty less than 9.8 dB. The receiver sensitivities at BER=10^-9 of the ASK system are -34.5 dBm and -42.5 dBm for 1.7-Gb/s and 622-Mb/s modulation, respectively. A 622-Mb/s incoherent frequency-shift-keying (FSK) transmission experiment using the same laser has also achieved a receiver sensitivity of -42.5 dBm     

Intersymbol interference due to linear and nonlinear distortion

The paper provides a recursion model for the calculation of the probability density function (pdf) of intersymbol interference (ISI) which is caused by the combined effect of linear channel dispersion and of nonlinear distortion. The nonlinearity introduces statistical interdependencies between the interfering symbols and these dependencies are implicitly taken into account in a trellis-structured recursion rule. The results were verified by time consuming Monte Carlo simulation and show, e.g., that the nonlinear characteristic of a high power amplifier (HPA) reduces in some cases the error probability caused by the linear dispersion. Surprisingly, the ISI due to the nonlinear amplifier is increased in the case of offset modulation     

An optical power equalizer based on one Er-doped fiber amplifier

A device which exhibits zero or slightly negative differential optical gain over a broad input power range is demonstrated. The scheme is based upon saturation enhancement by bidirectional signal input in an Er-doped fiber amplifier. Power equalization of better than 1.2 dB over a 14-dB dynamic input power range is achieved at the 1530-nm signal wavelength. The noise figure penalty due to the equalization is below 3 dB as evaluated by a device simulation     

一种简单RF预失真电路的研究

文章介绍一种基于肖特基二极管的预失真器及其工作原理。预失真器主要由一个肖特基二极管和直流偏置电阻组成,利用二极管的非线性,产生与功放相反的幅度失真与相位失真。通过调节二极管的偏压来调整预失真器的工作状态,使功率放大器在不同输出功率下得到最好的线性度改善。设计实际电路经实验发现,对于应用在C波段的某功率放大器,在不同的输出功率下IMD3的改善不同。在功率放大器1dB增益压缩点处IMD3改善了5dB,而在其他功率点处IMD3最大甚至可以改善20dB。