Linearization of 1.3-μm MQW electroabsorption modulators usingan all-optical frequency-insensitive technique

A novel all-optical linearization technique has been developed for discrete multiquantum-well electroabsorption modulators by exploiting the wavelength dependence of the quantum-confined Stark effect. The correcting signal for a particular order of nonlinearity is generated by the same modulator at a detuned optical wavelength, canceling the undesired order of nonlinear distortion via coherent RF recombination at the photodetector. This dual-wavelength approach is simple and insensitive to the microwave subcarrier frequency. A dramatic 30-dB reduction in the third-order nonlinear distortion at 10 GHz and 8-dB increase in the suboctave link dynamic range were achieved experimentally     

A predistortion linearizer using envelope-feedback technique withsimplified carrier cancellation scheme for class-A and class-AB poweramplifiers

A predistortion linearization method using an envelope-feedback technique is proposed and implemented in this paper. This linearizer compensates the gain and phase nonlinearity of power amplifier (PA) simultaneously by controlling both variable attenuator and phase shifter with the feedback of only the difference signal between input and output envelopes. A new carrier cancellation scheme composed of a minimization circuit, log detector, and vector modulator is also presented. This circuit achieves adaptive control of the linearizer by enabling direct measurement of out-of-band power. It is well suited to a multichannel system where the allocated channels are time variant. The principle of the proposed linearizer is described and simple AM-AM distortion analysis is presented analytically and graphically based on the conceptual schematic diagram. A two-tone test for a class-A PA at 1.855 GHz with frequency spacing of 1 MHz showed intermodulation-distortion reduction of maximum 16 dB and stable operation over 5-dB output power variation up to 4-dB backoff from the saturation power level. The proposed linearizer is also applicable to class-AB PA's without further special adjustments. The adaptation circuit is fully implemented with analog integrated circuits, which can further extend its applicability with the integration technology     

Linearization of receivers using envelope signal injection

This paper presents a new linearization method for receivers employing envelope signal injection. In this technique, the third-order intermodulation distortion (IM3), at the output of a mixer in IF band, is cancelled by injecting the envelope of the RF input signal to both the low noise amplifier (LNA) and the mixer. By properly adjusting the amplitude and polarity of the injected envelope signal, up to 40-dB improvement of the IM3 and 11-dB improvement of the IM5 is obtained in a two tone test with 100-kHz separation at 1.9GHz. This method operates very well over a wide range of power up to the 1-dB compression point of the receiver. The noise performance of the receiver under this linearization technique is also investigated. The noise floor at the output of the receiver is increased by 0.8 dB only when the system is optimized for linearity.     

High-efficiency linear RF Amplifier - a unified circuit approach to achieving compactness and low distortion

The realization of a highly efficient linearized amplifier has emerged as a paramount issue in the design of advanced mobile handsets. In this paper, a new RF amplifier linearization scheme using a compensating transistor combination is proposed. The devised approach ably utilizes all terminals of an additional transistor that offers a unified pre-post-distortion and cubic distortion characteristic for performance improvement. Meticulous modeling along with a power-dependent Volterra series is performed to identify contributions on each mechanism under various power levels. An experimental four-tone test reveals a maximum 28-dB reduction for the intermodulation distortion at 1.95 GHz, which outperforms typical pre-distortions of 5-10 dB. A single-ended two-stage amplifier module demonstrates a state-of-the-art power efficiency of 55% with 27-dB transducer gain at 24-dBm output power. Meanwhile, the adjacent channel power ratio (ACPR/sub 1/) is maintained with good margins of -35dBc for a four-channel wideband code-division multiple-access signal under all output dynamics. Graceful degradations on modulation bandwidths, tone spacing, bias, and gain variations are also discussed, showing superb performance with virtually no dedicated retuning circuit parameters for multicarrier applications. By combining a bias control along with the proposed linearization technique, the average efficiency (12%) is 3/spl times/ higher than that of the fixed bias (3.94%), demonstrating the potential utility on further prolonging battery lifetime in practical scenarios.     

Semiconductor laser optical amplifiers for multichannel coherentoptical transmission

The application of semiconductor laser optical amplifiers in multichannel coherent optical transmission systems is investigated. The amplifiers considered (λ=1.3 μm) exhibit a gain of 24 dB at a grain ripple <2 dB and a 3-dB bandwidth of about 4000 GHz. The characteristics of these amplifiers and transmission experiments with these amplifiers are described. The investigations concern noise accumulation in an amplifier chain, generation of echoes due to backward gain in cascaded amplifiers, crosstalk in multichannel transmission, and the effect of gain saturation due to spontaneous emission. A good fit is shown between the advantages of multichannel coherent optical transmission systems and the properties of semiconductor laser optical amplifiers, which are very promising for future long-haul optical transmission systems     

Signal distortion and noise in AM-SCM transmission systemsemploying the feedforward linearized MQW-EA external modulator

In subcarrier multiplexing (SCM) optical video distribution systems, the nonlinear signal distortion generated by the combined action of laser chirp and fiber dispersion limits the transmission distance. This paper for the first time shows that low-chirp MQW (multiquantum well)-EA (electroabsorption) external modulators are applicable to AM-SCM transmission systems. The feedforward linearization technique is used to compensate the signal distortion due to the nonlinearity in the modulator's L-V (light power versus voltage) characteristic. The effectiveness of this type of modulator for suppressing dispersion-induced distortion is clarified experimentally. A feedforward linearized transmitter with composite second order (CSO) intermodulation distortion <-62 dBc, composite triple beat (CTB) intermodulation distortion <-59 dBc and a carrier-to-noise ratio (CNR) >50.5 dB is successfully constructed. It is shown that the transmitter can achieve a CSO <-59 dBc, a CTB <-57 dBc and a CNR >49.5 dB for a 32 channel AM signal and 10 km transmission. Furthermore, estimations of the signal distortion cancellation and the noise characteristic achieved with feedforward circuit modeling are shown. The results obtained here emphasize that MQW-EA external modulators are applicable to AM-SCM transmission systems     

Impact of laser linewidth on optical channel spacing requirements for multichannel FSK and ASK systems

The sensitivity penalty is evaluated for amplitude-shift-keyed and frequency-shift-keyed multichannel coherent systems that use lasers with linewidths which are a significant fraction of the bit rate. The study was conducted for both ASK and FSK systems using a single-filter receiver with nonsynchronous detection. For FSK systems, both NRZ (nonreturn-to-zero) and alternate mark inversion (AMI) signal formats were studied. The optical channel spacing is strongly determined by the laser linewidth. For example, with the FSK-NRZ data rate of 150 Mb/s, the optical channel spacing which gives 1-dB crosstalk penalty is 4 GHz when the intermediate frequency linewidth is 50 MHz (laser linewidth is 25 MHz), as opposed to 1.8 GHz when the linewidth is negligible.<>     

Comparative study on temperature-dependent multichannel gain andnoise figure distortion for 1.48- and 0.98-μm pumped EDFAs

Temperature dependence of multichannel gain flatness and noise figure (NF) was compared for different pump wavelengths of 1.48 and 0.98 μm on silica-based erbium-doped fiber amplifiers (EDFAs) through measurement-based numerical simulation. Owing to its temperature sensitive pump emission cross section, the 1.48-μm pumping showed greater temperature sensitivity (maximum 0.75-dB gain flatness distortion with 0.57-dB average gain level shift, 0.3-dB NF variation for 25°C change) than the 0.98 μm pumping (maximum 0.5-dB gain flatness distortion with 0.015-dB average gain level shift, 0.05-dB NF variation for 25°C change). However, it was also found that distortion ripple spectra mainly coming from the changes of signal cross sections and asymmetric gain temperature dependence necessitate compensation techniques in the EDFA link, irrespective of pump wavelength     

High-Power High-Linearity Uni-Traveling-Carrier Photodiodes for Analog Photonic Links

We have fabricated and characterized two high-power high-linearity uni-traveling-carrier photodiode (UTC-PD) structures. The UTC performances are compared regarding their respective collector design. A -3-dB bandwidth improvement (from 16-25 GHz to 19-32 GHz) is achieved when the collector layer thickness is increased (from 250 to 350 nm, respectively). The bandwidth improvement for large photocurrent is at the origin of a ldquosupra-linearityrdquo effect. Photocurrent saturation effects are investigated and -1-dB compression current measurements at 20 GHz show saturation currents as high as 70 mA at -4 V. We also report third-order intermodulation distortion measurements at 20 GHz. The ldquosupra-linearityrdquo effect enhances the PD linearity with increased photocurrent, leading to a record third-order intercept point of 35 dBm at 40 mA.     

The operation of multichannel feedforward adaptive systems

A generalized formulation describing the operation of multichannel feedforward adaptive systems is developed that is applicable to random as well as to deterministic signals. Expressions for the optimum multichannel adaptive-filter transfer functions and the resulting minimized cost function are derived. The cost function is a generalized, nondiagonal, quadratic function of the error signal and the expended power. Spectral shaping of the cost function is also considered, which enables varying the emphasis between error and power reduction as a function of frequency. The maximum achievable error-reduction performance of multichannel feedforward adaptive systems is derived. Results of computer experiments are presented, validating all analytical results     

Efficient frequency assignment scheme for intermodulationdistortion reduction in fibre-optic microcellular systems

An efficient and effective frequency assignment scheme for fibre-optic microcellular systems is presented. The scheme is suitable for intermodulation distortion reduction in multichannel transmission systems. The computation time reduction is quite substantial with good intermodulation distortion reduction     

The Monolithic Integration of a Wavelength-Demultiplexer With Evanescently Coupled Uni-Traveling-Carrier Photodiodes

In this letter, we discuss a novel integration of a wavelength demultiplexer with a pair of uni-traveling-carrier photodiodes (UTC-PDs). With this integrated module, we could separate incoming wavelengths of 1530 and 1550 nm, which are near the wavelengths utilized for digital and analog signal transmission in coarse wavelength-division multiplexing (CWDM) systems. The integrated UTC-PDs exhibited the following advantages: a wide 3-dB bandwidth (25 GHz), a reasonable responsivity (0.35 A/W), and a high-saturation-current (>17 mA). A low level of radio frequency crosstalk between the two photodiodes (less than -15 dB) could also be achieved for a wide frequency range (near DC to 40 GHz). This is low enough for the multichannel receivers in a WDM broadcast system. In addition, by use of our module, we demonstrated digital/analog signals co-transmission and demultiplexing with speeds of 2.5 Gb/s at optical wavelengths of 1530 and 1550 nm.     

K频段自适应前馈行波管放大器设计

空间通信系统高保真转发对大功率行波管放大器的线性化度提出了更高的要求。针对高频段和大功率应用,基于预失真和前馈线性化技术,设计了一种K频段50 W自适应前馈线性化行波管放大器,介绍了其基本原理和参数,研究了前馈双检波电路结构及并行变步长的组合自适应控制算法。在K频段400 MHz带宽内,实现了输入功率20 dB大动态调整情况下失真信号自适应对消抑制,在相对于单载波饱和输出功率回退3 dB点,三阶交调小于-35 dBc,验证了前馈线性化技术在线性度改善方面的巨大优势。     

Amplifier linearization using RF feedback and feedforwardtechniques

The performance of feedback as a distortion reduction technique is highly dependent on the integrity of the feedback path. Any error or noise generated in this path is directly reflected into the output of the amplifier. Linearized RF power amplifiers (PAs) using Cartesian feedback require a demodulator in the feedback loop, and this is a potential source of linear errors, nonlinear errors, and noise. RF feedback with Cartesian compensation is proposed as a technique for overcoming some of these problems. The scheme is most suited to systems requiring an RF input. In addition, the RF nature of the input, feedback, and error signals enables the addition of a feedforward loop to further improve the linearization capability while still maintaining good efficiency. Design equations and simulation results are given for such a system. Disadvantages include the limited bandwidth (estimated at 1 MHz) and the need for additional circuits to generate the RF input signal when included in an integrated transmitter     

A dual-band 5.15-5.35-GHz, 2.4-2.5-GHz 0.18-/spl mu/m CMOS transceiver for 802.11a/b/g wireless LAN

A single-chip dual-band 5.15-5.35-GHz and 2.4-2.5-GHz zero-IF transceiver for IEEE 802.11a/b/g WLAN systems is fabricated on a 0.18-/spl mu/m CMOS technology. It utilizes an innovative architecture including feedback paths that enable digital calibration to help eliminate analog circuit imperfections such as transmit and receive I/Q mismatch. The dual-band receive paths feature a 4.8-dB (3.5-dB) noise figure at 5.25 GHz (2.45 GHz). The corresponding sensitivity at 54 Mb/s operation is -76 dBm for 802.11a and -77 dBm for 802.11g, both referred at the input of the chip. The transmit chain achieves output 1-dB compression at 6 dBm (9 dBm) at 5 GHz (2.4 GHz) operation. Digital calibration helps achieve an error vector magnitude (EVM) of -33 dB (-31 dB) at 5 GHz (2.4 GHz) while transmitting -4 dBm at 54Mb/s. The die size is 19.3 mm/sup 2/ and the power consumption is 260 mW for the receiver and 320 mW (270 mW) for the transmitter at 5 GHz (2.4 GHz) operation.     

On the performance of Cartesian feedback and feedforward linearization structures operating at 28 GHz

Some performance results of the Cartesian feedback and feedforward linearization techniques applied to a class-A power amplifier operating at 28 GHz are presented. The performance of the combination of HMMC-5040 (driver) and HMMC-5033 (power amplifier) is used as benchmark for simulation analysis. This analysis is addressed to show the key aspects on spectral regrowth due to phase margin and loop gain for Cartesian feedback, and time delay, phase and gain mismatch for feedforward. A 16 QAM digital signal at 10 Mbits/s filtered with a squared raised cosine filter with /spl alpha/=0, 25 is used as test signal.     

Adaptive Volterra filters for active control of nonlinear noiseprocesses

This paper presents a Volterra filtered-X least mean square (LMS) algorithm for feedforward active noise control. The research has demonstrated that linear active noise control (ANC) systems can be successfully applied to reduce the broadband noise and narrowband noise, specifically, such linear ANC systems are very efficient in reduction of low-frequency noise. However, in some situations, the noise that comes from a dynamic system may he a nonlinear and deterministic noise process rather than a stochastic, white, or tonal noise process, and the primary noise at the canceling point may exhibit nonlinear distortion. Furthermore, the secondary path estimate in the ANC system, which denotes the transfer function between the secondary source (secondary speaker) and the error microphone, may have nonminimum phase, and hence, the causality constraint is violated. If such situations exist, the linear ANC system will suffer performance degradation. An implementation of a Volterra filtered-X LMS (VFXLMS) algorithm based on a multichannel structure is described for feedforward active noise control. Numerical simulation results show that the developed algorithm achieves performance improvement over the standard filtered-X LMS algorithm for the following two situations: (1) the reference noise is a nonlinear noise process, and at the same time, the secondary path estimate is of nonminimum phase; (2) the primary path exhibits the nonlinear behavior. In addition, the developed VFXLMS algorithm can also be employed as an alternative in the case where the standard filtered-X LMS algorithm does not perform well     

Linear distortion correction using a feedforward system

The application of the feedforward amplifier linearization technique to the removal of both harmonic and intermodulation distortions is well understood and has been discussed in the literature. An additional benefit of the feedforward technique, which has received relatively little attention, is that of so-called linear distortion removal. Linear distortion is a term used to describe the nonideal gain and phase response (e.g., frequency response ripple) present in all practical amplifiers, and it is possible to significantly improve these parameters, in a practical application, using the feedforward technique. This paper discusses the improvements in this area, which may be achieved by utilizing the feedforward technique, demonstrates mathematically how they arise, and presents both simulation and practical results to demonstrate the effect for various types of gain and phase response nonlinearity     

Super low-noise GaAs MESFET's with a deep-recess structure

Super low-noise GaAs MESFET's for replacement of parametric amplifiers have been successfully developed by adopting a deep-recess structure. The structure of a 0.5-µm gate in a deeply recessed region with a cylindrical edge shape has enabled reduction of the source resistance to a half of that of conventional flat-type MESFET's. The noise figure was improved by more than 0.5 dB by this reduction of the source resistance, and less than 2.0-dB noise figure has been reproducibly obtained at 12 GHz. The best noise figures were 0.7 dB (14.9-dB gain) at 4 GHz and 1.68 dB (10.7-dB gain) at 12 GHz. The developed MESFET's were applied to two-stage amplifiers of 11.7-12.2-GHz band, and the noise figure obtained was 2.16 dB (T_{e}: 185K) at room temperature and 1.94 dB (T_{e}: 163K) at 0°C. This performance is good enough to replace some of parametic amplifiers.     

A radio-over-fiber network for microcellular system application

This paper investigates the feasibility of a radio-over-fiber network, which employs a single high-power Nd:YAG laser shared among many microcells. External amplitude modulators are employed at both base station (BS) and radio ports (RPs) to form an optical fiber distribution network. The distortions induced by the uplink remodulation are analyzed. The optimal preamplifier gain in the uplink is estimated for given sensitivity and dynamic range (DR). Appropriate modulator linearization and thermal-noise suppression schemes are utilized together to improve the DR. The clipping distortion for the predistortion-linearized modulator is also considered. The result shows that the effects of clipping and higher order distortions have little impact on the estimation of the system performance. A numerical example shows that a 100-mW Nd:YAG laser can provide for a macrocell of 5-km radius up to 16 microcells and 1600 channels when a 20-dB modulator linearization and 10-dB thermal-noise suppression are employed. The experiment setup uses a two-tone test to verify the theoretical calculation. The measurement agrees with the theoretical estimation very well