BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20260522T064958EDT-7428XgUlMd@132.216.98.100
DTSTAMP:20260522T104958Z
DESCRIPTION:Abstract\n\nOptics and photonics offer a compelling path to ove
 rcoming the power and speed limitations of the state-of-the-art electronic
  systems\, particularly for the specialized applications such as machine l
 earning acceleration and quantum information processing. By exploiting the
  intrinsic advantages of light\, such as high bandwidth and spatial parall
 elism\, photonic processors enable high-performance computation. Among the
 se\, coherent interferometric photonic processors have emerged as a promin
 ent architecture for implementing matrix-vector multiplication\, where a m
 esh of tunable Mach-Zehnder interferometers (MZIs) coherently manipulates 
 light to perform linear transformations. The scalability of such systems i
 s hindered due to several factors\, including calibration and reconfigurat
 ion complexity that increases with mesh size\, raising power consumption\,
  and reducing system uptime.\n\nThis thesis addresses the calibration chal
 lenge by analyzing the root causes of phase errors in rectangular mesh top
 ologies\, specifically the inability to directly access or monitor interna
 l phase shifters during calibration and programming. This leads to cumulat
 ive errors and demands iterative\, computationally expensive optimization.
  To overcome this\, an approach based on diagonal optical paths is employe
 d at the cost of using N²/4 additional blocks. By performing calibration a
 nd programming along diagonal paths\, the contributing effects of the adja
 cent MZIs are suppressed\, significantly reducing crosstalk and computatio
 nal overhead. Experimental results on a 4×4 SOI-based photonic mesh demons
 trate robust performance\, enabling up to a 79% error reduction compared t
 o an offline calibration method. This contributes directly to mitigating o
 ne of the key factors behind the scalability challenge as the reconfigurat
 ion complexity\, time\, and power consumption.\n\nTo further mitigate the 
 scalability bottleneck\, this thesis introduces a mode-selective thermo-op
 tic phase shifter (MS-TOPS)\, based on subwavelength grating structures. T
 his innovative device selectively tunes the phase of a higher-order transv
 erse electric mode (TE1) without requiring mode conversion to the fundamen
 tal (TE0)\, allowing dual transverse modes to be simultaneously processed 
 on-chip. A good selectivity is realized with a worst-case modal crosstalk 
 of -13.1dB over a 40nm wavelength range. The MS-TOPS facilitates independe
 nt modal control\, unlocking applications in spatial switching and multimo
 de optical computing. This effectively doubles the processing capacity\, c
 ontributing directly to scalable MZI-based architectures.\n\nComplementing
  the aforementioned contributions in coherent optical computing\, the thes
 is also explores the scalability challenge in noncoherent photonic process
 ors\, particularly those based on wavelength-division multiplexing (WDM) a
 nd microring resonators (MRRs) for multiply-and-accumulate (MAC) operation
 s. Two novel architectures are proposed that leverage mode-division multip
 lexing (MDM) to extend the functionality of MRR-based MAC units. A proof-o
 f-concept two-mode vector-vector multiplication is demonstrated at 1541.4n
 m with a modal crosstalk of 20dB by employing independent binary weighting
  of TE0 and TE1 MRRs. Experimental results on an SOI platform confirm the 
 feasibility of this approach at low speed\, with reasonable modal crosstal
 k over C-band. This study establishes a foundational framework for the exp
 ansion of on-chip photonic computing to incorporate additional higher-orde
 r modes through a hybrid WDM-MDM approach\, thereby permitting an n-fold s
 calability for an n-mode system.\n\nCollectively\, the contributions of th
 is thesis form a cohesive innovative strategy to address the scalability l
 imitations in photonic processors. By targeting both coherent and noncoher
 ent platforms and by leveraging multi-transverse mode operations alongside
  architectural innovations\, this work contributes to the development of m
 ore scalable and high-performance optical computing engines.\n
DTSTART:20251016T163000Z
DTEND:20251016T183000Z
LOCATION:Room 603\, McConnell Engineering Building\, CA\, QC\, Montreal\, H
 3A 0E9\, 3480 rue University
SUMMARY:PhD defence of Mohammad Reza Safaee Ardestani – Toward scalable pho
 tonic computing: single- and dual-transverse mode architectures
URL:/ece/channels/event/phd-defence-mohammad-reza-safa
 ee-ardestani-toward-scalable-photonic-computing-single-and-dual-367916
END:VEVENT
END:VCALENDAR