Data Communications to Optical Beamforming

Hajimiri, “2D sparse aperture scalable phased optical array,” in Conference on Lasers and Electro-Optics, Optical Society of Amer-. Hajimiri, “High sensitivity flat optical phased array receiver with a two-dimensional aperture,” Opt.

LIST OF ILLUSTRATIONS

26 3.15 Radiation pattern cross-section for transmitter and receiver and. effective transceiver array including the effect of the grid links. On-chip low-loss delay light compensates for the path mismatch between different blocks of the aperture.

NOMENCLATURE

INTRODUCTION

Contributions

This architecture is inspired by the diffraction pattern of the annular ring and offers low side loop levels. Three electro-optical modulators are proposed that reduce the footprint of the integrated high-speed modulator.

Thesis Outline

Conventional integrated high-speed silicon photonics modulators require large driving voltage and have a large chip footprint. Finally, a multi-ring resonator architecture is proposed that reduces the driving voltage requirements for high-speed amplitude modulation in a small footprint.

SILICON PHOTONICS INTEGRATION

Silicon Photonics Platform

Silicon Photonic Devices

Chip-side terminated tapered waveguides [34] can be used to efficiently interface with the lens fiber. The phase of the photonic wave can be modulated in various ways including the thermo-optical effect [37] and electro-optical effects.

High-Density Photonics Integration

Electro-optical effects can be divided into two groups: those acting based on the plasma scattering effect [38] and those acting based on the Pockel effect [39]. The type of photodiode device determines the sensitivity as well as the speed of these modulators [40], [41].

COHERENT OPTICAL TRANSCEIVER BASED ON CO-PRIME SAMPLING

Introduction

This design choice mainly arises from the dielectric nature of the optical waveguides and radiating elements in SiP OPAs. The non-uniform placement of the radiating elements in this array architecture achieves a narrow beamwidth without reducing the field of view with a smaller number of elements compared to the same number of evenly distributed radiators.

Figure 3.1: Routing optical signal to radiating elements in a 2D grid requires a larger inter-element spacing

Background Formulation for Co-Prime Sampling

Using sopranumbers 3 and 5 and a unit spacing of 𝜆/2, the beam pattern for such transmitter-receiver pairs and the resulting sopranumber array of transmitters is visualized in the figure. It is shown in [70] that lattice beams are two matrices overlap only at one point (𝜃𝑀 𝑡 𝑥 =𝜃𝑀 𝑟 𝑥) for the same prime number of values 𝑃in𝑄.

Figure 3.3: Co-prime sampling in spatial domain and the equivalent spectral sam- sam-ples.

Design

The distribution waveguide is included in the design and optimization of the radiating element to account for their effect. The spacing between the transmitter and receiver elements and the beam pattern of the radiators result in the beamforming capability of this architecture over a 30◦ field of view.

Measurement Result

The lattice lobes are spaced 9.55◦, consistent with a distance of 9.2𝜇𝑚 between the radiating elements. a-d) Beam shaping heat map optimized for four directions. e) Perspective of the normalized optical power for (b). The lattice lobes are spaced 7.2◦ apart, consistent with a distance of 12.4◦ between the radiating elements. a-d) Beam shaping heat map optimized for four directions. e) Perspective of the normalized optical power for (c). 3.20 (blue curves) shows a 1D scan of the formed transmitter and receiver beam over a 16◦ field of view with the expected grating lobes.

The AC performance of the TO phase shifter is measured and shows a bandwidth of approximately 1 kHz.

Figure 3.17: Electrical programming nodes of the row-column phase shifters and sniffer detectors for the cascaded amplitude modulator block

Discussion

In current implementations, the two apertures are separated by only 100𝜇𝑚 to reduce aperture crosstalk through the substrate without affecting the projected beam patterns. This analysis indicates a trade-off between the size of the aperture, which translates to higher resolution and received power versus transceiver efficiency. For larger arrays, more current is collected by the receiver, but the distance between the elements increases, resulting in power loss in the lobes of the transmitter grid, which reduces the system SNR.

Multibeam receiver architectures can simultaneously collect current from different transmitter grid lobes and break this trade-off.

LARGE-SCALE BEAMFORMING BASED ON ANNULAR-RING DIFFRACTION PATTERN

Introduction
Theory
Design
Measurement Result
Discussion

The combination of multiple annular apertures with different radii improves the beam efficiency and converges to the far-field pattern of the circular aperture at the boundary. The pair (𝑟𝑚, 𝑁𝑚) gives the radius of the ring and the number of elements on 𝑚that ring as shown in the figure. The 255 aperture radiating elements are arranged on five concentric circles with the innermost circle 𝑛.

Low-loss on-chip delay light compensates for the path mismatch between different blocks of the diaphragm.

Figure 4.1: Comparison of circular and annular ring diffraction patterns. (a) A circular ring aperture of diameter D illuminated by a plane wave

LARGE-SCALE COHERENT IQ IMAGER

Introduction
Background
Design
Setup and Measurement

Each measurement point should be sampled continuously until the amplitude of the received signal is maximized (the relative phase error between the transmitter and receiver paths becomes zero and the mixed signal is maximized). In addition, the strongest signal is the intensity of the reference signal, which allows the signal to be saturated in the electrical amplification chain (Fig. 5.2). This topology outputs both the in-phase and quadrature forms of the mixed optical signal, and the contributions of the random phase fluctuations are canceled in the sum-squared of the two outputs.

Balanced detectors cancel out a strong reference signal and allow a higher. a) IQ unit cell layout.

Figure 5.1: In this coherent imaging system, the target is illuminated using a collimated beam, the returned signal is captured via a lens, and the image is formed on the sensor array.

A COMPACT SPIRAL MACH-ZEHNDER INTERFEROMETER MODULATOR ON SOI PROCESS

Introduction

Design

The partial cross-section of the MZI structure for two turns of the spiral is shown in figure. A coplanar electrical transmission line, which is capacitively loaded by the PN junction, supplies the electrical driving voltage to the modulator. This capacitive load on the transmission line reduces the speed of propagation of the electrical signal.

The electrical wave is terminated in the center of the chip via an integrated 50Ω resistor.

Measurements

It is considered in the design to reduce the speed mismatch between electrical and optical waves. Additionally, to reduce crosstalk between adjacent coplanar signals in the coil, each signal is shielded by two bases on each side. Finally, the ground plane extends above the supply line (VDD) to act as a distributed bypass capacitor that reduces supply voltage ripple.

Figure 6.2: (a) Measured DC response of the modulator. (b) Measured frequency response of the modulator.

MONOLITHIC MACH-ZEHNDER INTERFEROMETER MODULATOR IN AN UNMODIFIED CMOS PROCESS

Introduction

Design

The process's poly-silicon layer helps contain the mode in the core of the waveguide. Forward biasing the PIN diodes causes current to flow through the silicon which changes the refractive index of the interferometer arm for modulation. The common node of the modulators is grounded while the other two terminals are powered via two external RF sources.

For this structure, we measured a propagation loss of 14 dB/cm for the 500nm-wide waveguides, 10 dB/cm for the 600nm-wide waveguides, and 8 dB/cm for the 1.5𝜇𝑚-wide waveguides.

Measurements

The polysilicon layer on the silicon structure limits the optical mode in the PIN structure. Equal length modulator arms of the interferometer result in less sensitivity to the operating wavelength. Without prior emphasis, we demonstrate the NRZ modulation eye diagram at 900 Mbps, as shown in Fig.

With the pre-emphasis and adjustment of the first and second post-touch coefficients, we demonstrate an open eye at 1.25 Gbps, as shown in Fig.

HIGH-SPEED SLOW-WAVE MODULATOR USING CORRUGATED WAVEGUIDES

Introduction
Design
Measurements
Discussion

The traveling wave is terminated using integrated parallel resistors to prevent back reflection of the electric wave. A calibrated 40GHz photodiode in combination with an electrical VNA is used to measure the AC response of the modulator 8.2(b). The electro-optical bandwidth of the modulator was measured to be 14 GHz, as shown in Figure 8.2(c).

The modulation efficiency was improved by 67% in the slow-wave operation region of the modulator at 1533.7 nm due to the extended array index.

Figure 8.1: (a) Schematic of the proposed corrugated-base slow-light modulator, modulator cross-section view, and schematic of corrugated waveguide

HIGH-EFFICIENCY POLYMER-BASED MODULATORS

Introduction
SOH Modulator Implementations and Modulation Efficiency
Strip-to-Slot Mode Adapter
Sources of Insertion Loss
High-Speed SOH Modulator Design
Dual-SSB LIDAR Signal Generator
SOH 1D OPA Design

At the end of the strip-to-slot mode adapter, there was a significant fraction of extraneous propagating mode in the waveguide. The insertion loss of the strip-to-slot mode converter can be as low as 0.2 dB [122]. Furthermore, manufacturing imperfections in the slot regions, such as incomplete etching of the slot region or contamination can contribute more loss.

Die Photo of the design is shown in fig. a) Row-column programming of the capacitive modulators.

Figure 9.1: High-bandwidth modulation. (a) Trends in high-bandwidth optical modulation [100]

HIGH-SPEED NESTED-RING-ASSISTED MZI MODULATOR

Introduction

Previously, ring-assisted MZI modulators as a hybrid architecture of resonance and MZI structures were used to improve the linearity of modulators. Improvements to this ring-supported architecture can be made by using multiple ring resonators mounted on the modulator in a cascaded fashion. This new architecture, called nested ring assisted MZM (NRAMZM), uses flat bandpass ring filters while maintaining a linear phase response in the operating range.

A nested ring-assisted MZM using a maximally flat three-stage filter response is implemented in a standard silicon photonic process based on the parameters of the three-stage flat ring in [135].

Background and Theory

The steady-state ring resonator modulator response at the output of the two ports is given by (10.5) and (10.6), where 𝛼 and 𝜙 are the return loss and phase accumulation; 𝜅. Here the intensity and phase response of the ring drop port and changes in intensity and phase response with respect to wavelength are plotted for different drop port coupling coefficients (Fig. 10.2) under critical coupling conditions (𝛼 = 𝜎. 10.9( b) the slope takes of phase response increases with increasing number of rings, reducing drive voltage requirements.

10.9) Backward waves for the same coupler[𝐴. 1]leaves the coupler as a result of the input waves [𝐵0.

Figure 10.1: Basic ring resonator and interferometric modulators.

Integrated NRAMZM Design

Quality factor, insertion loss and maximum group delay for an increasing number of rings are shown in Fig. To compare the performance of this nested ring-assisted MZI modulator with single-ring and cascaded MZI modulators, the required phase shift in the ring resonance (change in effective length of the resonator) for 𝜋/2 phase change is calculated. The resulting phase and intensity response of the output is plotted in Fig. Compared to the cascading structure, this nested ring structure experiences much lower amplitude fluctuations for the same total induced phase shift.

The modulated time domain output intensity of the differential nested ring supported MZI modulator is shown in Fig.

Figure 10.16: Performance metrics for nested ring-assisted MZI for different number of rings for maximally flat response.

BIBLIOGRAPHY

Hajimiri, “Scalable optical phased array with sparse 2D aperture,” in Conference on Lasers and Electro-Optics, Optical Society of America, 2018, STu4B.6. Watts, “An 8192-element optical phased array with 100° control area and flip-chip cmos,” in Conference on Lasers and Electro-Optics, Optical Society of America, 2020, JTh4A.3. Stojanović, “Single-chip optical phased array in a wafer-sized silicon photonics/cmos 3D integration platform,” IEEE Journal of Solid-State Circuits, vol.

Freude, “High-speed low-voltage electro-optical modulator with a polymer-infiltrated silicon photonic crystal waveguide,” Opt.