Gabriela Nicolescu

Polytechnique Montréal · Department of Computer Science and Software Engineering

Integrated photonics based on silicon photonics platform is driving several application domains, from enabling ultra-fast chip-scale communication in high-performance computing systems to energy-efficient optical computation in artificial intelligence (AI) hardware accelerators. Integrating silicon photonics into a system necessitates the adoption...

Model stealing attacks have become a serious concern for deep learning models, where an attacker can steal a trained model by querying its black-box API. This can lead to intellectual property theft and other security and privacy risks. The current state-of-the-art defenses against model stealing attacks suggest adding perturbations to the predicti...

Silicon photonics (SiPh) has different applications, from enabling fast and high-bandwidth communication for high-performance computing systems to realizing energy-efficient optical computation for AI hardware accelerators. However, integrating SiPh with electronic sub-systems can introduce new security vulnerabilities that cannot be adequately add...

The use of unmanned aerial aircrafts (UAVs) is governed by strict regulatory frameworks that prioritize safety. To guarantee safety, it is necessary to acquire and maintain situational awareness (SA) throughout the operation. Existing Canadian regulations require pilots to operate their aircrafts in the visual line-of-sight. Therefore, the task of...

Integrated photonics arises as a fast and energy-efficient technology for the implementation of artificial neural networks (ANNs). Indeed, with the growing interest in ANNs, photonics shows great promise to overcome current limitations of electronic-based implementation. For example, it has been shown that neural networks integrating optical matrix...

The use of unmanned aerial aircrafts (UAVs) is governed by strict regulatory frameworks that prioritize safety. To guarantee safety, it is necessary to acquire and maintain situational awareness (SA) throughout the operation. Existing Canadian regulations require pilots to operate their aircrafts in visual line-of-sight. Therefore, the task of acqu...

Microservices is a cloud-native architecture in which a single application is implemented as a collection of small, independent, and loosely coupled services. This architecture is gaining popularity in the industry as it promises to make applications more scalable and easier to develop and deploy. Nonetheless, adopting this architecture in practice...

The widespread use of sensor devices in IoT networks imposes a significant burden on energy consumption at the network’s edge. To address energy concerns, a prompt anomaly detection strategy is required on demand for troubleshooting resource-constrained IoT devices. It enables devices to adapt their configuration according to the dynamic signal qua...

Recent attacks on Machine Learning (ML) models such as evasion attacks with adversarial examples and models stealing through extraction attacks pose several security and privacy threats. Prior work proposes to use adversarial training to secure models from adversarial examples that can evade the classification of a model and deteriorate its perform...

Recent attacks on Machine Learning (ML) models such as evasion attacks with adversarial examples and models stealing through extraction attacks pose several security and privacy threats. Prior work proposes to use adversarial training to secure models from adversarial examples that can evade the classification of a model and deteriorate its perform...

Aerospace systems are safety-critical systems that need to respect tight constraints in terms of execution time, resource usage and predictability. This industry is currently transitioning from predictable single-core processors to less predictable multicore architectures. This transition reveals multiple challenges due to interferences. The conten...

Detecting and tracking vehicles in urban scenes is a crucial step in many traffic-related applications as it helps to improve road user safety among other benefits. Various challenges remain unresolved in multi-object tracking (MOT) including target information description, long-term occlusions and fast motion. We propose a multi-vehicle detection...

Detecting and tracking vehicles in urban scenes is a crucial step in many traffic-related applications as it helps to improve road user safety among other benefits. Various challenges remain unresolved in multi-object tracking (MOT) including target information description, long-term occlusions and fast motion. We propose a multi-vehicle detection...

Avionic software is the subject of critical real time, determinism and safety constraints. Software designers face several challenges, one of them being the estimation of worst-case execution time (WCET) of applications, that dictates the execution time of the system. A pessimistic WCET estimation can lead to low execution performances of the syste...

Memory simulation is now widely used in system design processes. This process allows debugging, profiling and sometimes validating all memory-related tasks. The step before memory profiling is memory tracing, which consists in collecting memory events. State-of-the-art solutions rely on well-known environments provided by operating systems. Such en...

Avionics are highly regulated systems and their criticality requires determinism. Single-core systems have a single processor running, hence only one execution flow, making them easily predictable in execution. However, manufacturers are moving towards multi-core processors manufacturing only. Avionics systems are thus compelled to transition from...

Avionic software is the subject of stringent real time, determinism and safety constraints. Software designers face several challenges, one of them being the interferences that appear in common situations, such as resource sharing. The interferences introduce non-determinism and delays in execution time. One of the main interference prone resources...

Animal behavior is greatly influenced by interaction between peers as well as with the environment. Understanding the flow of information between individuals can help decipher their behavior. This applies to both the microscopic and macroscopic levels, from cellular communication to coordinated actions by humans. The aim of this work is to provide...

On-chip integrated lasers are key devices to deliver the high bandwidth expected from nanophotonic interconnects. However, lasers are highly sensitive to temperature variation, which influences the lasing efficiency and the wavelengths of emitted optical signals, both of which are key factors in interconnect power efficiency. It is, thus, necessary...

For the last decades, Verification and Validation techniques have been well improved in order to make safer complex systems. Sophisticated algorithms and methodologies have been proposed in the domain of formal modelling, and simulation. However, like for testing approaches, all these methodologies suffer from a strong weakness, while they depend o...

Efficient mapping of tasks onto heterogeneous multi-core systems is very challenging especially in the context of real-time applications. Assigning tasks to cores is an NP-hard problem and solving it requires the use of meta-heuristics. Relevantly, genetic algorithms have already proven to be one of the most powerful and widely used stochastic tool...

Silicon photonic interconnects (SPIs) have emerged as a promising solution to outperform the communication infrastructure in multiprocessor systems-on-chip (MPSoCs). Routing a message from one node to another in an MPSoC integrating SPIs, several photonic components (e.g., switching elements) need to be configured to realize an optical path between...

This paper presents an efficient device-level design method to enhance the performance and reliability (DeEPeR) in optical interconnection networks (OINs) under fabrication process variations (PV). Considering different range of variations, DeEPeR explores the design space of fundamental optical components in OINs (e.g., microresonators (MRs)) to i...

Digital Radar Landmass Simulation (DRLMS) for the purpose of training radar operators is a challenging computationally-intensive task. To improve fidelity, the databases are continuously increasing in resolution and density. Moreover, the radar simulation involves increasingly sophisticated physics-based models. One of them is the application of th...

The effect of sharing the last-level cache (LLC) among cores in a multi-core system has not been thoroughly investigated especially in the design of efficient scheduling algorithms. And with the growing interest in random caches, which allow for an easier estimation of the worst-case execution time of tasks in critical real-time embedded systems, t...

The increase of interdependent components in avionic and automotive software rises new challenges for real-time system integration. For instance, most scheduling and mapping techniques proposed in the literature rely on the availability of the system’s DAG representation. However, at the initial stage of system design, a dataflow graph (DFG) is gen...

Designing complex embedded systems requires simultaneous optimization of multiple system performance metrics that can be addressed by applying Pareto-based multi-objective optimization techniques. At the end of this type of optimization process, designers always face Pareto Fronts including a large number of near-optimal solutions from which select...

An FPGA-based centralized controller architecture for silicon photonics switches is experimentally demonstrated achieving scheduling decision in one clock cycle. The FPGA simultaneously operates as the controller, and the traffic payload generator with error detection.

Efficient mapping of tasks onto heterogeneous multi-core systems is very challenging especially under hard timing constraints. Assigning tasks to processors is an NP-hard problem and solving it requires the use of meta-heuristics. Relevantly, genetic algorithms have already proven to be one of the most powerful and widely used stochastic tools to s...

Silicon photonic interconnect (SPI) is an attractive alternative for the power-hungry and low-bandwidth metallic interconnect in multiprocessor systems-on-chip (MPSoCs). When employing SPIs for wavelength-division multiplexing (WDM)-based applications, it is essential to precisely align the central wavelengths of different photonic devices (e.g., p...

Information systems are designed to present services and functionalities for multiple users. Thus, it is used to have on one information system different levels of privilege for different users. Privileges describe what a user is permitted to do such as viewing files, modifying or deleting data. Privilege escalation takes place when a user gets acc...

Silicon photonic devices and integrated circuits are vulnerable to fabrication-induced process variations (a.k.a. fabrication non-uniformity). Designing silicon photonic integrated circuits (PICs) for wavelength-division multiplexing (WDM)-based applications requires a careful consideration of fabrication-induced process variations. While numerical...

Full-mission simulators (FMSs) are considered the most critical simulation tool belonging to the flight simulator family. FMSs include a faithful reproduction of fighter aircraft. They are used by armed forces for design, training, and investigation purposes. Due to the criticality of their timing constraints and the high computation cost of the wh...

Modern device trends present great challenges to design because many of them integrate continuous and discrete sub-systems and therefore their design involves specific global modeling and validation approaches. This paper proposes the integration of the rollback based simulation in a generic methodology for the design of continuous/discrete simulat...

Real-time systems not only require functional correctness, but also specific timing properties. Correct timing is especially challenging for hard real-time systems such as in medicine, avionics, and space industries, where missing a deadline can lead to catastrophic failure. A number of theories tackled this issue to determine whether a set of task...

With the rise of software system complexity, developers rely more on a modular approach to system design to reduce development cost. However, as a result, integrating a real-time system becomes a challenge. To be able to properly integrate the system, software developers are required to provide software characteristics such as the execution times o...

Electrical interconnects will reach their physical limitations as we scale down components on systems' designs. At the same time, the demand for high throughput communication architectures rises. Optical integrated networks (OIN) are currently considered to be one of the most promising paradigm in this design context: they present high bandwidth an...

Nanophotonic is an emerging technology considered as one of the key solutions for future generation on-chip interconnects. However, silicon photonic devices are highly sensitive to temperature variation. Under a given chip activity, this leads to a lower laser efficiency and a drift of wavelengths of optical devices (on-chip lasers and microring re...

A low-latency centralized controller integrated to a Silicon photonic (SiP) switch is demonstrated. The FPGA-based controller sets the configuration of a 4×4 Mach-Zehnder Interferometer-based Spanke-Benes switch and generates traffic patterns.

Optical integrated network (OIN) is a promising alternative to overcome the restrictions that electrical networks-on-chip (eNoCs) will face in the next generation of multiprocessor integrated systems due to electrical interconnects' physical limitations. OINs present a higher bandwidth and lower power consumption but their full capabilities are cur...

Image processing and computer vision applications are usually complex in terms of the large amount of processed data and high computation loads. To cope with this, optimization techniques and high-performance hardware platforms are required. Since these applications present many opportunities for parallelism, heterogeneous parallel platforms (HPPs)...

The design, implementation, and validation of a low-latency look-up-table-based centralized controller (LUCC) is presented. The LUCC is applied to different Mach-Zehnder Interferometer (MZI)-based photonic interconnects under dynamic workloads. We obtain a response time of only one clock cycle when using the LUCC in which the scalability can be up...

Optical Network-on-Chip (ONoC) is an emerging technology considered as one of the key solutions for future generation on-chip interconnects. However, silicon photonic devices in ONoC are highly sensitive to temperature variation, which leads to a lower efficiency of Vertical-Cavity Surface-Emitting Lasers (VCSELs), a resonant wavelength shift of Mi...

The many cores design research community have shown high interest in optical crossbars on chip for more than a decade. Key properties of optical crossbars, namely a) contention-free data routing b) low-latency communication and c) potential for high bandwidth through the use of WDM, motivate several implementations. These implementations demonstrat...

Modern systems integrate components specific to different application domains. Frequently, these systems combine continuous and discrete sub-systems and therefore their design involves overcoming specific global modeling and validation challenges. In order to generate global simulation models of heterogeneous systems the designers need efficient to...

Developing a computationally efficient and accurate bottom-up method, the impact of process variations on passive silicon photonic interconnects is systematically studied. Comparisons with numerical methods indicate a considerable computational efficiency of our proposed method.

The many-core design research community has shown high interest in optical crossbar on chip for more than a decade. Key properties of optical crossbars, namely (1) contention-free data routing, (2) low latency communication, and (3) potential for high bandwidth through the use of wavelength division multiplexing, motivate several implementations of...

Image processing and computer vision applications are used intensively in several domains in particular multimedia and medicine. The main challenge in developing such applications is how to guarantee both high accuracy and low execution time. Accordingly, we observe two research directions: the first focuses on improving the algorithms and the seco...

Optical on-chip interconnects enable significantly increased bandwidth and decreased latency in MPSoC. They are characterized by high static energy and low dynamic energy consumption. Achieving high energy-efficient communication thus requires a certain adaptability degree to the executed application and its communication traffic in order to reduce...

The next generation of MPSoC points to the integration of thousands of IP cores, requiring high performance interconnect for high throughput communications. Optical on-chip interconnect enables significantly increased bandwidth and decreased latency in MPSoC. However, the interface between electrical and photonic devices implies strong layout const...

Three-dimensional integrated circuits (3D ICs) with advanced cooling systems are emerging as a viable solution for many-core platforms. These architectures generate a high and rapidly changing thermal flux. Their design requires accurate transient thermal models. Several models have been proposed, either with limited capabilities, or poor simulatio...

In top-down multi-level design methodologies, design descriptions at higher levels of abstraction are incrementally refined to the final realizations. Simulation based techniques have traditionally been used to verify that such model refinements do not change the design functionality. Unfortunately, with computer simulations it is not possible to c...

Trends in SoC design are leading to 3D integration of thousands of high-performance computing resources and high-throughput interconnects, opening up new research directions for hybrid electronic/photonic architectures. In this paper, we introduce how state of the art silicon-photonic devices can realize elementary operations that are traditionally...

Optical Network on Chip (ONoC) architectures are emerging as promising candidates to solve congestion and latency issues in future embedded systems. In this work, we examine how a scalable and fully connected ONoC topology can be reduced to fit specific connectivity requirements in heterogeneous 3D architectures. Through such techniques, it is poss...

The high heat flux and compact structure of three-dimensional circuits (3D ICs) make conventional air-cooled devices more subsceptible to overheating. Liquid cooling is an alternative that can improve heat dissipation, and reduce thermal issues. Fast and accurate thermal models are needed to appropriately dimension the cooling system at design time...

This book provides a broad overview of current research in optical interconnect technologies and architectures. Introductory chapters on high-performance computing and the associated issues in conventional interconnect architectures, and on the fundamental building blocks for integrated optical interconnect, provide the foundations for the bulk of...

Design trends for next-generation multi-processor systems on chip (MPSoC) point to the integration of a large number of processing elements onto a single chip, requiring high-performance interconnect structures for high-throughput communication. On-chip optical interconnect and 3D die stacking are currently considered to be the two most promising p...