Abstract
Outline Benefits of Microwave Photonics ○ Requirements for RF Sensing applications Fiber Optic Remoting Links ○ Performance metrics ○ Improving performance ▪ Low Biasing ▪ Linearization RF Photonic Frontends ○ Efficient integration of antennas Summary Microwave Photonics For RF Sensing W4B.1.pdf Multidisciplinary field bringing together worlds of microwave engineering and opto-electronics ○ Over four decades old Initially spurred by interest in exploiting benefits of optical fiber for transporting ○ Communications ○ Ultrafast measurement systems ○ THz photonics ○ Signal processing ○ Sensing, surveillance, and radar Microwave Photonic Systems For RF Sensing W4B.1.pdf Sensing and surveillance networks are becoming increasingly more complex Photonics can significantly enhance the overall performance of these networks W4B.1.pdf Fiber remoting of receive signals ○ High fidelity, linear transport of RF signals ○ Wide operating frequency range and instantaneous bandwidth ○ Optical distribution/switching Fiber remoting of transmit signals ○ RF signal generation techniques with low phase noise ○ Multiband RF signal generation ○ Control of wideband phased array signals Photonic RF front-ends ○ Low conversion loss over wide frequency range Digital receiver ○ High sampling rate, low jitter, analog-to-digital conversion over wide bandwidths W4B.1.pdf Benefits of optical fiber for transporting RF signals ○ Low loss and low distortion propagation ▪ Weak frequency dependence from MHz to multi-GHz ○ Reduced cabling size and weight ▪ Fiber: < 0.5 dB and 1.7 kg per km; Coax: 360 dB at 2 GHz and 567 kg per km ○ Signal isolation/EMI resistance ○ Design flexibility ▪ Increased bandwidth and flexible routing with WDM Photonic devices can support a wide range of RF frequencies N ISR apertures interfaced with multiple receivers High Performance Fiber Remoting Links for RF Sensing W4B.1.pdf RF signal can be encoded onto the amplitude or phase of optical carrier ○ Intensity Modulation/Direct Detection (IMDD) ○ Phase Modulation/Coherent I/Q Demodulation Coherent links more complex to implement but capable of achieving better performance ○ O/E process dominant nonlinearity W4B.1.pdf ○ Gain ○ Noise Figure ○ Linearity ○ Spurious Free Dynamic Range W4B.1.pdf Two main types of IMDD links Spurious-Free Dynamic Range (SFDR) is a key performance figure of merit Improving Linearity for Wideband Sensing Networks Challenge: Achieving high SFDR over a large operational (multi-band) and/or instantaneous bandwidth ○ Inherent non-linearity of E/O conversion process Variety of techniques demonstrated to linearize link ○ Pre-distortion, Post distortion compensation ○ Feedforward, Feedback linearization ○ Downconversion with digital linearization ○ Novel optical modulator designs W4B.1.pdf Demonstrated technique for improving Noise Figure and SFDR Moves EOM bias point away from quadrature towards null ○ Operating point defined by bias angle, φ ○ Reduces amplitude of optical carrier ○ For same EOM output optical power, link gain will be greater than quadrature bias ▪ Relaxes performance constraints on PD ○ Corresponding decrease in link NF and increase in SFDR ○ No impact on third-order intercept point 2nd-order distortion products increase in magnitude ○ Link operational bandwidth limited to single octave W4B.1.pdf Laser RIN noise and shot noise decrease with EOM bias angle ○ At some bias angle, receiver thermal noise dominates At high optical input powers, NF of quadrature-biased link is dominated by laser RIN ○ Cannot be further reduced with increasing optical power W4B.1.pdf SFDR improves moving away from quadrature bias point At some bias angle, receiver thermal noise dominates Low Biased Millimeter- Wave Fiber Link for RF Sensing High power DFB laser followed by fiber power amplifier ○ Link is shot noise limited Low biased, wideband electro-optic modulator ○ Improved thermal stability with GaAs semiconductor structure Input 3rd order intercept point = 21 dBm at 35 GHz W4B.1.pdf > 118 dB-Hz2/3 SFDR over 28 – 38 GHz 150 MHz instantaneous bandwidth No apriori knowledge of distortion required Can correct for noise as well as nonlinear distortions Provides transient and static nonlinearity suppression Can be applied to directly and externally modulated fiber links W4B.1.pdf Linearization performed with correction signal derived from distorted signal ○ Nonlinear E/O encoding process creates distorted signals (error) ○ Signal Cancellation Circuit isolates distortion/error ○ Error Cancellation Circuit cancels distortion Distortion corrected in real-time Can adapt to any instantaneous bandwidth for sub-octave as well as multi-octave applications Input signals ○ Pulsed X-band carrier at 10 kHz repetition rate and 50 % duty cycle ○ CW signal at 1 MHz offset from pulsed carrier W4B.1.pdf Directly modulated transmitter and error correction lasers Opportunity for reduction in form factor and power dissipation Linearity of FF Linearized Directly Modulated Link W4B.1.pdf > 20 dB suppression of distortion at 3.2 GHz > 7 dB improvement in SFDR to 113 dB-Hz2/3 W4B.1.pdf Downconversion of RF signal at end of fiber remoting link Digitization followed by digital signal processing (DSP) to improve linearity Opportunity to enhance link performance and relax component requirements ○ Dynamic range of ADCs reduces with bandwidth Downconverting Link Approaches W4B.1.pdf - Electronic ○ Also require wideband PD Link linearity increases with optical power into modulator Link linearity independent of modulator switching voltage ○ Dominated by RF mixer linearity W4B.1.pdf ○ Enables lower bandwidth PD ○ Perfect isolation between RF and LO ports Link linearity increases with optical power into modulator Link linearity improves with low biasing W4B.1.pdf Distortion in link primarily due to the encoding optical modulator (EOM1): 𝐼𝑜𝑢𝑡 𝑡 = 𝜂𝑃𝑙𝑎𝑠𝑒𝑟 𝑇 sin 𝜋 𝑉𝑖𝑛 + 𝜙 2 𝑉𝜋 Downconverting optical modulator (EOM2) provides mixing without introducing any in-band distortion Original undistorted input signal can be recovered by applying inverse sine to output current: 𝑉 = 𝜋 −1 𝐼𝑜𝑢𝑡 (𝑡) 𝑉𝑖𝑛 sin − 𝜙 𝜋 𝐼𝑎𝑣𝑔 Arcsine linearization suitable for arbitrary input signals W4B.1.pdf Input RF signal of 9 GHz downconverted to 100 MHz 39 dB suppression of third-order intermodulation distortion W4B.1.pdf W4B.1.pdf DSP Linearization Operating on Modulated Signals Input RF signal of 9 GHz with 1 MHz QPSK W4B.1.pdf W4B.1.pdf Phase modulation (PM) provides linear E/O encoding Coherent detection using Inphase (I) and Quadrature (Q) demodulation ○ Linear recovery of phase and amplitude information Detection and digitization of I and Q signals separately followed by DSP W4B.1.pdf 126.8 dB-Hz2/3 at 1 GHz Efficient RF Photonic Frontends for RF Sensing W4B.1.pdf Challenge for Photonic RF Frontends W4B.1.pdf Efficient integration to reduce insertion loss and maximize link budget ○ Optical → Antenna ○ Antenna → Optical High isolation between Transmit and Receive path in RF photonic frontend ○ Requires antennas with ultra-low return loss over a wide bandwidth Printed antenna technologies capable of efficient integration with LiNbO3 W4B.1.pdf Single layer radiator Based on ‘traveling wave’ concepts ○ Quasi Yagi printed antennas ○ Circular slot antennas ○ Tapered slot antennas Highly efficient, broad bandwidth responses ○ ‘Independent’ of material Difficult to integrate into modulator package ○ Require modifications to wafer and EOM package W4B.1.pdf ○ Can use slots and cavities Highly efficient, broadband ○ Multi-layering helps achieve this Modifications to wafer ○ Can be minor (or none) Modifications to packaging ○ Can be minor Two styles ○ Direct contact ▪ Require modifications to the wafer ○ Indirect contact W4B.1.pdf Based on proximity coupled patches Multiple dielectric layers and two vertically coupled patches W4B.1.pdf W4B.1.pdf Broadband X-band antenna integrated with LiNbO3 modulator W4B.1.pdf
© 2018 The Author(s)
PDF ArticleMore Like This
A. Bogoni
M2G.1 Optical Fiber Communication Conference (OFC) 2018
Xingyuan Xu, Jiayang Wu, Sai T. Chu, Brent E. Little, Roberto Morandotti, Arnan Mitchell, and David J. Moss
JW1I.3 Integrated Photonics Research, Silicon and Nanophotonics (IPR) 2018
Daniel Dolfi
M3B.1 Optical Fiber Communication Conference (OFC) 2016