Mohamed El-Hadedy
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Mohamed El-Hadedy has a total of 18 co-authors. Here are the top co-authors by number of shared publications:
- Benny N. Cheng
Naval Surface Warfare Center, Corona Division, Corona, CA, USA
2 shared publications Kevin Skadron
2 shared publications- Wen-Mei Hwu
Coordinated Science Laboratory, University of Illinois at Urbana-Champaign, Urbana, IL, USA
2 shared publications - Amit Kulkarni
Systems Group, Department of Computer Science, ETH Zurich, Zurich, Switzerland
1 shared publication Andrea Abelian
Department of Electrical and Computer Engineering, California State Polytechnic University, Pomona, CA, USA
1 shared publication- Deming Chen
University of Illinois Urbana-Champaign, Urbana, IL, USA
1 shared publication
Latest Publications
- Open Access
- 17 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 59-67
KHONSU: Kernel Hash Offload Network of Swappable Units for Post-Quantum Signatures
Hash-based post-quantum signatures such as LMS, XMSS, and SPHINCS+ offer a conservative foundation for long-lived secure systems, but their heavy reliance on repeated hash and extensible-output function (XOF) computations creates a significant efficiency challenge. Recent work has shown that a single FPGA or SoC can host unified accelerators covering all three families. A different question remains open at the system level: how to structure unified acceleration as a distributed substrate, where orchestration cost across multiple worker FPGAs becomes a first-class concern alongside primitive compute. KHONSU addresses this question through a reconfigurable hash-offload architecture and runtime that separates a stable streaming shell from swappable XOF units in FPGA fabric. The prototype uses two Zynq UltraScale+ MPSoC (AUP-ZU3) workers coordinated by a ClusterHAT controller and exposes eight total hardware lanes for LMS, XMSS, and SPHINCS+-oriented execution under a unified sign/verify interface. The design sustains correct operation at 300–330 MHz on known Keccak test vectors. Worker-side phase times remain stable at 2.3–2.7 ms across all three schemes, while end-to-end controller time is much larger at 7.4 s for LMS and 14.7–14.8 s for XMSS and SPHINCS+. A controller-side decomposition attributes roughly 75–80% of the wall time to communication and synchronization on the RPC path, identifying orchestration—rather than primitive throughput—as the dominant end-to-end bottleneck of the present prototype.
- 17 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 59-67
Open AccessKHONSU: Kernel Hash Offload Network of Swappable Units for Post-Quantum Signatures
Hash-based post-quantum signatures such as LMS, XMSS, and SPHINCS+ offer a conservative foundation for long-lived secure systems, but their heavy reliance on repeated hash and extensible-output function (XOF) computations creates a significant efficiency challenge. Recent work has shown that a single FPGA or SoC can host unified accelerators covering all three families. A different question remains open at the system level: how to structure unified acceleration as a distributed substrate, where orchestration cost across multiple worker FPGAs becomes a first-class concern alongside primitive compute. KHONSU addresses this question through a reconfigurable hash-offload architecture and runtime that separates a stable streaming shell from swappable XOF units in FPGA fabric. The prototype uses two Zynq UltraScale+ MPSoC (AUP-ZU3) workers coordinated by a ClusterHAT controller and exposes eight total hardware lanes for LMS, XMSS, and SPHINCS+-oriented execution under a unified sign/verify interface. The design sustains correct operation at 300–330 MHz on known Keccak test vectors. Worker-side phase times remain stable at 2.3–2.7 ms across all three schemes, while end-to-end controller time is much larger at 7.4 s for LMS and 14.7–14.8 s for XMSS and SPHINCS+. A controller-side decomposition attributes roughly 75–80% of the wall time to communication and synchronization on the RPC path, identifying orchestration—rather than primitive throughput—as the dominant end-to-end bottleneck of the present prototype.
- Open Access
- 27 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in TechRxiv
SESHAT: Systematic Energy & Seed-Provenance Harness for Algorithmic Tradeoffs
SESHAT is a reproducible benchmark suite for modular exponentiation on embedded Arm and RISC-V systemon-chip (SoC) platforms typical of Internet of Things (IoT) and unmanned aerial vehicle (UAV) gateways. SESHAT evaluates binary square-and-multiply, the constant-time Montgomery ladder, and fixed-window methods under pinned CPU governors on Raspberry Pi 5 (Cortex-A76), PYNQ-Z1 (Cortex-A9), and VisionFive 2 (RISC-V JH7110), reporting latency (median, 95% confidence intervals), idle-subtracted energy, and hardware counters. This work targets the single-operation regime (B=1, where B is the batch size) and introduces a compact model with threshold B ⋆ = ⌈tpre/(t bin − t fw)⌉ that predicts when fixedwindow amortizes its precomputation. Measurements and the model agree: across 1024-4096-bit moduli, there is no crossover at B=1; fixed-window becomes preferable only once B ≥ B ⋆. A brief microarchitectural rationale indicates that algorithmic precomputation costs primarily explain the ordering, with branch prediction and cache locality playing secondary roles across architectures. Seed-provenance logging using Global Positioning System (GPS) time and a true random number generator (TRNG) is audit-oriented and performance-neutral.
- 27 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in TechRxiv
Open AccessSESHAT: Systematic Energy & Seed-Provenance Harness for Algorithmic Tradeoffs
SESHAT is a reproducible benchmark suite for modular exponentiation on embedded Arm and RISC-V systemon-chip (SoC) platforms typical of Internet of Things (IoT) and unmanned aerial vehicle (UAV) gateways. SESHAT evaluates binary square-and-multiply, the constant-time Montgomery ladder, and fixed-window methods under pinned CPU governors on Raspberry Pi 5 (Cortex-A76), PYNQ-Z1 (Cortex-A9), and VisionFive 2 (RISC-V JH7110), reporting latency (median, 95% confidence intervals), idle-subtracted energy, and hardware counters. This work targets the single-operation regime (B=1, where B is the batch size) and introduces a compact model with threshold B ⋆ = ⌈tpre/(t bin − t fw)⌉ that predicts when fixedwindow amortizes its precomputation. Measurements and the model agree: across 1024-4096-bit moduli, there is no crossover at B=1; fixed-window becomes preferable only once B ≥ B ⋆. A brief microarchitectural rationale indicates that algorithmic precomputation costs primarily explain the ordering, with branch prediction and cache locality playing secondary roles across architectures. Seed-provenance logging using Global Positioning System (GPS) time and a true random number generator (TRNG) is audit-oriented and performance-neutral.
- Open Access
- 21 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Ptah: Teaching Computer Architecture through NLP-Controlled Drones with Edge AI and Kubernetes
- 21 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Open AccessPtah: Teaching Computer Architecture through NLP-Controlled Drones with Edge AI and Kubernetes
- 9 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Embedded Systems Letters
- Vol. 17 (3) , 164-167
ANUBIS: Hybrid FPAA-FPGA Architecture for Entropy-Based True Random Number Generation in Secure UAV Communication
Field-Programmable Gate Arrays (FPGAs) and Field-Programmable Analog Arrays (FPAAs) are reconfigurable circuits that enable flexible digital and analog implementations post-manufacturing. FPGAs are widely used in telecommunications, mixed-signal, and embedded systems due to their parallel processing and reconfigurability. Meanwhile, FPAAs provide flexibility for analog systems, which is crucial for modern mixed-signal processing. This study introduces ANUBIS, a hybrid system combining FPGA and FPAA technologies to generate true random numbers (TRNGs) for secure UAV communication. Due to its reliability and cost efficiency, ANUBIS leverages a thermistor circuit as an entropy source. The FPAA amplifies the analog noise generated by the thermistor, while the FPGA digitizes and processes the signal using Von Neumann Whitening (VNW) to remove bias. The ASCON hash function is applied to the whitened bitstream to generate cryptographically secure keys. These keys are utilized in a DHKE to enable secure communication via Bluetooth Low Energy (BLE), an ideal protocol for energy-constrained UAV applications. ANUBIS demonstrates reconfigurability, power efficiency, and ease of implementation, showcasing its potential for secure communication applications. It achieves robust randomization, setting a new standard for UAV communication security and addressing applications requiring reliable TRNG solutions. The system consumes 1.615 W in total, with 1.54 W consumed by the FPGA and 75 mW by the FPAA. Resource utilization on the PYNQ-Z1 board includes 5,186 LUTs (9.75%), 549 units of memory (3.15%), and 5.5 units of BRAM (3.93%), indicating moderate resource usage with room for future enhancements. By integrating reliable analog noise harvesting with efficient digital post-processing, ANUBIS offers a novel approach to TRNG design, demonstrating the potential for broader cryptographic applications in resource-constrained environments.
- 9 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Embedded Systems Letters
- Vol. 17 (3) , 164-167
ANUBIS: Hybrid FPAA-FPGA Architecture for Entropy-Based True Random Number Generation in Secure UAV Communication
Field-Programmable Gate Arrays (FPGAs) and Field-Programmable Analog Arrays (FPAAs) are reconfigurable circuits that enable flexible digital and analog implementations post-manufacturing. FPGAs are widely used in telecommunications, mixed-signal, and embedded systems due to their parallel processing and reconfigurability. Meanwhile, FPAAs provide flexibility for analog systems, which is crucial for modern mixed-signal processing. This study introduces ANUBIS, a hybrid system combining FPGA and FPAA technologies to generate true random numbers (TRNGs) for secure UAV communication. Due to its reliability and cost efficiency, ANUBIS leverages a thermistor circuit as an entropy source. The FPAA amplifies the analog noise generated by the thermistor, while the FPGA digitizes and processes the signal using Von Neumann Whitening (VNW) to remove bias. The ASCON hash function is applied to the whitened bitstream to generate cryptographically secure keys. These keys are utilized in a DHKE to enable secure communication via Bluetooth Low Energy (BLE), an ideal protocol for energy-constrained UAV applications. ANUBIS demonstrates reconfigurability, power efficiency, and ease of implementation, showcasing its potential for secure communication applications. It achieves robust randomization, setting a new standard for UAV communication security and addressing applications requiring reliable TRNG solutions. The system consumes 1.615 W in total, with 1.54 W consumed by the FPGA and 75 mW by the FPAA. Resource utilization on the PYNQ-Z1 board includes 5,186 LUTs (9.75%), 549 units of memory (3.15%), and 5.5 units of BRAM (3.93%), indicating moderate resource usage with room for future enhancements. By integrating reliable analog noise harvesting with efficient digital post-processing, ANUBIS offers a novel approach to TRNG design, demonstrating the potential for broader cryptographic applications in resource-constrained environments.
- 1 November 2019
- Published by Elsevier in Journal of Parallel and Distributed Computing
- Vol. 133, 420-431
Reco-Pi: A reconfigurable Cryptoprocessor for -Cipher- 1 November 2019
- Published by Elsevier in Journal of Parallel and Distributed Computing
- Vol. 133, 420-431
Reco-Pi: A reconfigurable Cryptoprocessor for -Cipher- 12 October 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. 68 (3) , 331-346
ASAP: Accelerated Short-Read Alignment on Programmable Hardware
The proliferation of high-throughput sequencing machines ensures rapid generation of up to billions of short nucleotide fragments in a short period of time. This massive amount of sequence data can quickly overwhelm today's storage and compute infrastructure. This paper explores the use of hardware acceleration to significantly improve the runtime of short-read alignment, a crucial step in preprocessing sequenced genomes. We focus on the Levenshtein distance (edit-distance) computation kernel and propose the ASAP accelerator, which utilizes the intrinsic delay of circuits for edit-distance computation elements as a proxy for computation. Our design is implemented on an Xilinx Virtex 7 FPGA in an IBM POWER8 system that uses the CAPI interface for cache coherence across the CPU and FPGA. Our design is 200x faster than an equivalent Smith-Waterman-C implementation of the kernel running on the host processor, 40-60x faster than an equivalent Landau-Vishkin-C++ implementation of the kernel running on the IBM Power8 host processor, and 2x faster for an end-to-end alignment tool for 120--150 base-pair short-read sequences. Further the design represents a 3760x improvement over the CPU in performance/Watt terms.
- 12 October 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. 68 (3) , 331-346
ASAP: Accelerated Short-Read Alignment on Programmable Hardware
The proliferation of high-throughput sequencing machines ensures rapid generation of up to billions of short nucleotide fragments in a short period of time. This massive amount of sequence data can quickly overwhelm today's storage and compute infrastructure. This paper explores the use of hardware acceleration to significantly improve the runtime of short-read alignment, a crucial step in preprocessing sequenced genomes. We focus on the Levenshtein distance (edit-distance) computation kernel and propose the ASAP accelerator, which utilizes the intrinsic delay of circuits for edit-distance computation elements as a proxy for computation. Our design is implemented on an Xilinx Virtex 7 FPGA in an IBM POWER8 system that uses the CAPI interface for cache coherence across the CPU and FPGA. Our design is 200x faster than an equivalent Smith-Waterman-C implementation of the kernel running on the host processor, 40-60x faster than an equivalent Landau-Vishkin-C++ implementation of the kernel running on the IBM Power8 host processor, and 2x faster for an end-to-end alignment tool for 120--150 base-pair short-read sequences. Further the design represents a 3760x improvement over the CPU in performance/Watt terms.
- Open Access
Posting with credibility in Micro-blogging systems using Digital Signatures and Watermarks: A case study on Twitter
Micro-blogs are contemporary broadcasting services for exchanging small elements of content, including video and images. Despite its popularity, micro-blogging is not without issues. So far, various security concerns, such as: privacy and confidentiality of micro-blogging systems have attracted the interest of the scientific community. Nevertheless, in this document we refer to a security issue that is concerned with the posting and circulation of fake messages. With the aim to make micro-blogs a credible source of information, in this work we propose a protocol, as a solution, that can be easily adapted to existing services such as: Twitter and Facebook. With the main keys of the solution being the employment of digital signatures and watermarks, the protocol can serve postings of various types of content. Finally, the solution serves the dual purpose of proofing the fake messages as well as the repudiation of postings.
Open AccessPosting with credibility in Micro-blogging systems using Digital Signatures and Watermarks: A case study on Twitter
Micro-blogs are contemporary broadcasting services for exchanging small elements of content, including video and images. Despite its popularity, micro-blogging is not without issues. So far, various security concerns, such as: privacy and confidentiality of micro-blogging systems have attracted the interest of the scientific community. Nevertheless, in this document we refer to a security issue that is concerned with the posting and circulation of fake messages. With the aim to make micro-blogs a credible source of information, in this work we propose a protocol, as a solution, that can be easily adapted to existing services such as: Twitter and Facebook. With the main keys of the solution being the employment of digital signatures and watermarks, the protocol can serve postings of various types of content. Finally, the solution serves the dual purpose of proofing the fake messages as well as the repudiation of postings.
- 9 November 2016
- Published by Springer Nature in Journal of Signal Processing Systems
- Vol. 88 (2) , 167-184
Dual-Data Rate Transpose-Memory Architecture Improves the Performance, Power and Area of Signal-Processing Systems
- 9 November 2016
- Published by Springer Nature in Journal of Signal Processing Systems
- Vol. 88 (2) , 167-184
Dual-Data Rate Transpose-Memory Architecture Improves the Performance, Power and Area of Signal-Processing Systems