- Cryptography and Data Security
- Cryptography and Residue Arithmetic
- Cryptographic Implementations and Security
- Coding theory and cryptography
- Chaos-based Image/Signal Encryption
- Complexity and Algorithms in Graphs
- Privacy-Preserving Technologies in Data
- Low-power high-performance VLSI design
- Advanced Data Storage Technologies
- Parallel Computing and Optimization Techniques
- Security and Verification in Computing
- VLSI and Analog Circuit Testing
- Quantum Computing Algorithms and Architecture
- Advanced Steganography and Watermarking Techniques
- Cloud Data Security Solutions
- Embedded Systems Design Techniques
- Ferroelectric and Negative Capacitance Devices
- Neural Networks and Reservoir Computing
- User Authentication and Security Systems
- Cellular Automata and Applications
- Advanced Memory and Neural Computing
- Physical Unclonable Functions (PUFs) and Hardware Security
- Interconnection Networks and Systems
Boston University
2019-2025
Intel (United States)
2023
Adapti (Slovenia)
2020
Fully Homomorphic Encryption (FHE) offers protection to private data on third-party cloud servers by allowing computations the in encrypted form. To support general-purpose computations, all existing FHE schemes require an expensive operation known as "bootstrapping". Unfortunately, computation cost and memory bandwidth required for bootstrapping add significant overhead FHE-based limiting practical use of FHE.In this work, we propose FAB, FPGA-based accelerator bootstrappable FHE. Prior...
Homomorphic Encryption (HE) enables users to securely outsource both the storage and computation of sensitive data untrusted servers. Not only does HE offer an attractive solution for security in cloud systems, but lattice-based systems are also believed be resistant attacks by quantum computers. However, current implementations suffer from prohibitively high latency. For become viable real-world it is necessary key bottlenecks—particularly polynomial multiplication—to highly efficient. In...
Fully Homomorphic Encryption (FHE) enables the processing of encrypted data without decrypting it. FHE has garnered significant attention over past decade as it supports secure outsourcing to remote cloud services. Despite its promise strong privacy and security guarantees, introduces a slowdown up five orders magnitude compared same computation using plaintext data. This overhead is presently major barrier commercial adoption FHE.
Fully Homomorphic Encryption (FHE) is a rapidly developing technology that enables computation directly on encrypted data, making it compelling solution for security in cloud-based systems. In addition, modern FHE schemes are believed to be resistant quantum attacks. Although offers unprecedented potential security, current implementations suffer from prohibitively high latency. Finite field arithmetic operations, particularly the multiplication of high-degree polynomials, key computational...
Today, edge devices commonly connect to the cloud use its storage and computing capabilities. This leads security privacy concerns about user data. Homomorphic encryption (HE) is a promising solution address data problem as it allows arbitrarily complex computations on encrypted without ever needing decrypt it. While there has been lot of work accelerating HE in cloud, small attention paid message-to-ciphertext ciphertext-to-message conversion operations edge. In this work, we profile...
Computer architecture lies at the intersection of electrical engineering, digital design, compiler programming language theory and high-performance computing. It is considered a foundational segment an computer engineering education. RISC-V new open ISA that gaining significant traction in academia. Despite it being used extensively research, more RISC-V-based tools need to be developed order for gain greater adoption organization classes. To end, we present BRISC-V Platform, design space...
The following topics are dealt with: field programmable gate arrays; learning (artificial intelligence); convolutional neural nets; logic design; system-on-chip; power aware computing; cryptography; cloud reconfigurable architectures; nets.
Cloud computing has made it easier for individuals and companies to get access large compute memory resources. However, also raised privacy concerns about the data that users share with remote cloud servers. Fully homomorphic encryption (FHE) offers a solution this problem by enabling computations over encrypted data. Unfortunately, all known constructions of FHE require noise term security, grows during computation. To perform unlimited on data, we need periodic reduction step as...
As more and edge devices connect to the cloud use its storage compute capabilities, they bring in security data privacy concerns. Homomorphic Encryption (HE) is a promising solution maintain by enabling computations on encrypted user cloud. While there has been lot of work accelerating HE computation cloud, little attention paid optimize en/decryption edge. Therefore, this paper, we present RACE, custom-designed area- energy-efficient SoC for HE. Owing similar operations en/decryption, RACE...
A prevalent issue in the residue number system (RNS) variant of Cheon-Kim-Kim-Song (CKKS) homomorphic encryption (HE) scheme is challenge efficiently achieving high precision on hardware architectures with a fixed, yet smaller, word-size bit length W , especially when scaling factor satisfies log Δ >
While the notion of achieving "quantum supremacy" may be debatable, rapid developments in field quantum computing are heading towards more realistic computers. As practical computers start becoming feasible, requirement to have secure cryptosystems becomes compelling. Due its many advantages, lattice based cryptography has become one key candidates for designing systems post-quantum era. The security lattice-based is governed by small error samples generated from a Gaussian distribution....
Permissioned blockchain platforms heavily depend on cryptography to provide a layer of trust within the network, thus verification cryptographic signatures often becomes bottleneck. ECDSA is most commonly used scheme in permissioned blockchains. In this work, we propose an efficient implementation signature FPGA, order improve performance blockchains that aim use FPGA-based hardware accelerators. We several optimizations for modular arithmetic (e.g., custom multipliers and fast reduction)...
With billions of devices connected over the internet, rise sensor-based electronic have led to cloud computing being used as a commodity technology service. These are often small and limited by power, storage, or compute capabilities, hence, they achieve these capabilities via services. However, this gives data privacy issues sensitive is stored computed cloud, which at most times, shared resource. Homomorphic encryption can be along with services perform computations on encrypted data,...
Random number generator (RNG) is a core component in many applications such as scientific research, testing and diagnosis, gaming, cryptosystems (e.g., obfuscation, encryption, authentication). Although, there are various RNG designs targeting specific application goals low-power, high-throughput, stronger security guarantees, universal programmable design has remained elusive. Indeed, it challenge to have only one unit system with multiple compute modules different randomness requirements....
Due to the rapid advances in development of quantum computers and their susceptibility errors, there is a renewed interest error correction algorithms. In particular, correcting code-based cryptosystems have reemerged as highly desirable coding technique. This due fact that most classical asymmetric will fail computing era. However, are still secure against computers, since decoding linear codes remains NP-hard even on these systems. One such cryptosystem was proposed by McEliece. The...
Permissioned blockchain platforms heavily depend on cryptography to provide a layer of trust within the network, thus verification cryptographic signatures often becomes bottleneck. ECDSA is most commonly used scheme in permissioned blockchains. In this work, we propose an efficient implementation signature FPGA, order improve performance blockchains that aim use FPGA-based hardware accelerators. We several optimizations for modular arithmetic (e.g., custom multipliers and fast reduction)...
Homomorphic Encryption (HE) enables users to securely outsource both the storage and computation of sensitive data untrusted servers. Not only does HE offer an attractive solution for security in cloud systems, but lattice-based systems are also believed be resistant attacks by quantum computers. However, current implementations suffer from prohibitively high latency. For become viable real-world it is necessary key bottlenecks - particularly polynomial multiplication highly efficient. In...
In heterogeneous distributed systems, computing devices and software components often come from different providers have security, trust, privacy levels. many of these the need frequently arises to (i) control access services resources granted individual or in a context-aware manner (ii) establish enforce data sharing policies that preserve critical information on end users. essence, is authenticate anonymize an entity device simultaneously, two seemingly contradictory goals. The design...
Novel sensor processing algorithms face many hurdles to their adoption. Sensor environments have become increasingly difficult with an ever increasing array of threats. These threats have, in turn, raised the bar on deploying new capabilities. Many novel exploit or induce randomness boost algorithm performance. Co-designing this cryptographic features could be a powerful combination providing both improved performance and increased resiliency. The emerging field signal encrypted domain has...
Achieving high accuracy, while maintaining good energy efficiency, in analog DNN accelerators is challenging as high-precision data converters are expensive. In this paper, we overcome challenge by using the residue number system (RNS) to compose operations from multiple low-precision operations. This enables us eliminate information loss caused limited precision of ADCs. Our study shows that RNS can achieve 99% FP32 accuracy for state-of-the-art inference with only $6$-bit precision. We...