SFP Series fifty-third Week News Abstract 12

Date:
2011-12-30 14:43:51
   Author:
10Gtek
  
Tag:

Week News Abstract For SFP Series in 10GTEK
The abstract is mainly about the optical communication related products,including: FTTH,GPON,EPON,SFPPLC,PTN,ODN,Optical module,Optical devices,optical communications,Optical transceiver module,Etc.

EXFO details measurement of 528-Gbps signals using 66-GBd PM 16-QAM
In a post-deadline paper delivered at last month’s ECOC conference in Geneva, researchers at EXFO Inc. (NASDAQ: EXFO, TSX: EXF) and Chalmers University in Goteborg, Sweden described the use of the company’s PSO-200 Optical Modulation Analyzer to measure a 528-Gbps transmission using polarization-multiplexed 16-ary quadrature amplitude modulation (16-QAM). The success of the demonstration highlights the capabilities of the company’s use of optical sampling for the PSO-200, EXFO asserts.While the industry has settled on coherent-enabled modulation formats such as dual-polarization quadrature phase shift keying (DP-QPSK) as the most common modulation format for 100-Gbps transmission, higher data rates will likely require a different modulation approach. 16-QAM has emerged as one candidate for data rates such as 400 Gbps. However, as complex as DP-QPSK can be, 16-QAM represents a higher level of complexity and a potentially major challenge for test and measurement, particularly in terms of handling the baud rates required.EXFO has long asserted that its optical sampling approach would enable measurement of higher baud rates than the real-time sampling strategy championed by competitors such as Agilent Technologies and Optametra (recently acquired by Tektronix) because its performance isn’t bounded by the current state of the art in analog-to-digital converters and digital signal processing (for example, see “Characterizing advanced modulation formats using optical sampling” from the February 2010 edition of Lightwave).The demonstration reported at ECOC illustrates this point, said Greg Shinn, R&D director at EXFO, in a conversation after the show. As described in the ECOC paper, technicians amplified a pair of binary electrical PRBS15 signals at 66 GBd and applied them to a 31-GHz bandwidth IQ modulator to create a QPSK signal on a 100-kHz linewidth optical carrier. The QPSK was then up-converted to 16-QAM via a QAM emulation device from Kylia. The emulator combined a delayed (500 psec) and scaled (1:4 power ratio) copy of the signal to itself. The splitting ratio was controlled by rotatable waveplates, and a variable phase shifter controlled the phase shift between the copies to align them properly.The resultant 66-GBd 16-QAM signal was then polarization multiplexed by splitting the signal and recombining the copies in orthogonal polarizations using a polarization beam splitter with a large delay between the polarizations. The output from the PM-16-QAM transmitter then passed through a variable attenuator/EDFA combination to vary the OSNR, then through a 2-nm optical filter before it reached the PSO-200.The PSO-200 then performed both waveform and bit-error-rate measurements of the signal (Chalmers University supplied the data pattern generator that helped create the traffic). Shinn said that the instrument is capable of handling signals of at least 70 GBd, which he asserted is much greater than could be measured using real-time oscilloscopes whose current bandwidth tops out at 35 to 40 GHz.
----------------------------------------------------
Modular vs. conventional oscilloscope approach for optical coherent modulation analysis
Modern day high-bandwidth conventional oscilloscopes are typically supplied with a maximum of four input channels (and with as few as one or two channels at a higher bandwidth rating), owing to practical design tradeoffs between bandwidth and channels. However, dual-polarization quadrature phase-shift keyed (DP-QPSK) or quadrature amplitude-modulated (16-QAM) coherent optical modulation systems analysis requires real-time analog-to-digital conversion of four electrical tributaries using very high real-time analog bandwidths of 30 GHz to 45 GHz. Other types of encoding may require many more channels at half of this bandwidth.To meet these needs, oscilloscope manufacturers have developed methods to connect multiple conventional oscilloscopes, each with their own acquisition timebase, using 10-MHz external clocking. However, this leads to system inaccuracies and requires significant operator time to connect the oscilloscopes. A modular oscilloscope system could provide much higher synchronization accuracy by using a single timebase for all acquisition modules and a single integrated trigger circuit to operate the entire system. The basic architecture for a single channel in a digital storage oscilloscope includes a 10-GHz clocked timebase that drives a sampler and an analog-to-digital converter (ADC), as well as an integrated trigger circuit. Input signals are digitized and stored into memory, and the trigger circuit instructs the memory chip when to cease updating and to transfer stored data for processing and display. The timebase portion consists of an oscillator operating at 10 MHz. The output of this oscillator passes through a phase-locked-loop (PLL) frequency multiplier to achieve the required 10 GHz to drive the sampler and trigger circuit. The clock module also accepts an external 10 MHz reference source and can output its own 10-MHz reference clock for use with an external device (which could be another oscilloscope). The timebase is the “heartbeat” of the acquisition system, and any variation in the phase of a clock cycle relative to an ideal clock is defined as the time interval error (TIE, or phase jitter). Connecting multiple conventional oscilloscopes requires synchronizing the phase of each timebase using the 10-MHz reference clock output from one oscilloscope to the reference clock input of another. This is shown in Figure 1 for two conventional oscilloscopes. This method is not perfect, and leads to an increase in the total TIE of the system timebase. A 10-MHz timebase clock signal also has a relatively slow slew rate, and such low slew rates are susceptible to small amplitude noise effects that manifest as time uncertainty, which increases the TIE further. In addition, the use of a PLL to multiply the frequency of the 10-MHz clock by 1000 times to 10 GHz can add a non-trivial amount of TIE. Lastly, since the 10 MHz reference must pass through a cable to reach other oscilloscopes, noise may infiltrate the timebase clock signal, causing its TIE to potentially increase. The total TIE from all of these sources results in inaccurate sample point placement, and phase errors in the measurement. The trigger systems in each conventional oscilloscope must also be integrated so that when a valid trigger condition is found, both oscilloscopes are able to trigger simultaneously. Typically, BNC cables connect auxiliary inputs/outputs of the conventional oscilloscopes to permit cross-triggering. These auxiliary inputs/outputs are typically low in bandwidth, which reduces the slew rate of the trigger signal and leads to an increase in the amount of time uncertainty in triggering. Furthermore, by integrating the triggers of two conventional oscilloscopes, the total trigger increases as the quadrature sum of each individual oscilloscope’s trigger jitter.Consider the impact of all of the combined TIE (phase) and trigger jitter errors with 28-Gbps signals on two conventional oscilloscopes connected using a 10-MHz reference clock. These TIE and trigger jitter errors cause dynamic skew between the signals, or skew that cannot be eliminated with calibration processes. At 28 Gbps, the unit interval (UI) is approximately 36 ps. If the conventional oscilloscopes are connected together to yield 1 psrms of total timebase phase error and trigger jitter (or 6 pspk-pk for three standard deviations) between acquisition systems, as is typically specified, the result becomes an added phase uncertainty of nearly 20% of the unit interval due solely to the oscilloscope acquisition system. This is more than enough to wreak havoc on the phase relationship of the captured signal, and is especially noteworthy when working with phase and frequency modulated signals, such as DP-QPSK or 16-QAM signals. All of the dynamic skew issues discussed with connecting multiple conventional oscilloscopes can be resolved by designing the acquisition system of the oscilloscope to be modular (see Figure 2). In the modular oscilloscope design, a single distributed 10-GHz clock (1000 times faster than a 10-MHz reference clock) is generated in a master acquisition module and then distributed to multiple slave acquisition modules. Since the modular oscilloscope uses a distributed 10-GHz timebase tied directly to all of the samplers and ADCs of each slave, there are not multiple timebases to synchronize. Furthermore, with a 10-GHz clock rate, the slew rate of this clock is very high and results in no measurable increase in the TIE as a result of amplitude noise. Finally, since the slaves do not generate their own timebases, there are no additional PLLs in-circuit that would add to the acquisition system TIE. Also eliminated with a modular oscilloscope system are the major problems associated with integrating the triggers of multiple conventional oscilloscopes. The master acquisition module performs all triggering for the entire system using a single integrated trigger circuit made possible with PCI Express synchronizing cable connects between the master and each slave. With this approach, the master directly controls the sampler, ADCs, and memory of each channel in the slaves. Thus, there is nothing to synchronize vis-à-vis the trigger. Moreover, since only a single trigger circuit exists, the trigger jitter of the system is only that arising from the single trigger circuit. If the acquisition is a single-shot acquisition then there is no trigger jitter present at all, since there is only a single trigger for all oscilloscope acquisition modules.Lastly, the modular design has the advantage of not requiring complicated and inaccurate connection schemes, and all oscilloscope captures are conveniently displayed on a single display in the master. A central processing unit with a server-class multi-core CPU operating at an effective clock rate of 33.6 GHz and accommodating up to 192 GB of RAM is also included.The complete modular oscilloscope synchronization approach results in many oscilloscope channels at very high bandwidths—as many as 5 channels at 45 GHz, 10 channels at 30 or 36 GHz, or 20 channels at 20 GHz—with one-fourth the jitter between all channels compared to connecting conventional oscilloscopes. In addition, there is no hassle associated with connecting and debugging the complete system. This approach is ideal for optical coherent modulation analysis of DP-QPSK, 16-QAM, optical orthogonal frequency-division multiplexed (OFDM) signals, or even multiple-input/multiple-output (MIMO) formats.
----------------------------------------------------
Agilent expands 86100D digital communications analyzer singlemode, multimode receiver bandwidth
Agilent Technologies Inc. (NYSE:A) has enhanced the ability of the Agilent 86100D digital communications analyzer (DCA) to perform optical transmitter compliance testing. Leveraging a new approach called system impulse response correction (SIRC), the instrument can now accommodate multimode optical receivers with bandwidths in excess of 25 GHz and singlemode receivers with near 100-GHz bandwidths.In addition to improving test accuracy for optical compliance testing in design and manufacturing applications where data rates don’t exceed 10 Gbps, the improvement positions the DCA as a tool that can help in next-generation 26-Gbps applications. As previously reported, technicians have already begun work on optical transmitters and receivers for 25-Gbps transmission over multimode fiber (see “VCSEL-based 25-Gbps transmission at 850 nm nearing readiness”). However, the large physical size of multimode-compatible photodetectors has prevented optical oscilloscope channels to provide the bandwidth necessary for accurate waveform analysis, Agilent asserts. The new SIRC approach increases the measurement bandwidth of the 86105D from 20 GHz to more than 25 GHz, Agilent says. SIRC also enables the 86105D to be configured as a multimode reference receiver for standards-based compliance testing at both 25 and 28 Gbps. The 86115D also can provide what Agilent touts as the industry’s only quad-port reference receiver for 4x25-Gbps 100-Gigabit Ethernet test.Agilent notes that while the frequency responses of optical reference receivers are usually defined within standards, some deviation from an ideal frequency response is typically allowed to enable their production at a reasonable cost. However, these deviations can cause variation in measurement results among test systems. By performing an impulse-response analysis, Agilent says its oscilloscope channel’s frequency response is precisely determined. This ability enables the 86100D DCA to correct frequency response deviations and provide waveform results as if the reference receiver were ideal.Using the SIRC calibration, the 86100D mainframe also can make real-time corrections to the raw waveforms, Agilent adds. The displayed signal thus appears as if it had been acquired with a system that has an ideal frequency response. In fact, the SIRC process can help create an ideal reference receiver for virtually any data rate within the physical limits of the system, Agilent asserts. Users also can increase or decrease the effective bandwidth of the measurement system by approximately 50 percent from the nominal hardware capabilities.The Agilent 86100 SIRC process accurately preserves random signal components such as jitter and noise. General signal processing techniques can incorrectly filter these signal components resulting in an incorrect waveform display, Agilent concludes.The Agilent 86100 family of optical receivers -- which includes the 86105C, 86105D, 86115D, and 86116C -- can be ordered with a special SIRC option.
****************************************
The above information is edited by 10GTEK.

10GTEK is a professional manufacturer engaged in the R&D,manufacturing and marketing of Fiber Optic Transceivers and High Performance Cables Assemblies.

10GTEK provides customers with top quality and cost effective products. The prompt response and excellent customer support contribute to clients' full satisfaction.

10GTEK passed quality inspection,ISO9001:2008,and our products passed by CE?FCC and RoHS certificates

10GTEK now has developed a full series of optical transceivers including:SFP+,XFP,X2,XENPAK,GBIC,SFP,COPPER SFP,BiDi SFP,CWDM SFP,DWDM SFP etc,from 125Mbps low rate module to 10Gbps full line products.

10GTEK now has developed a full series HPC cable assemblies including SFP/SFP+ Cable,XFP Cable,QSFP Cable,MiniSAS Cables,ect.

10GTEK with the ability of R&D,can customize the products to fully meet customers' satisfaction.

This article reader also like:SFP Series fifty-third Week News Abstract 11