About

I am a Ph.D. student in Computer Science at the University of Maryland, College Park, working in the Computer Architecture and Systems Lab (CASL) with Prof. Bahar Asgari. My research focuses on resilient and efficient systems for IoT-driven smart environments, with a particular emphasis on lightweight Transformer-based methods for real-time fault detection and diagnosis of smart-home sensors and devices at the edge.

Research Interests

  • Resilient and dependable IoT systems for smart homes and smart environments
  • Lightweight Transformer architectures and ML models for real-time edge deployment
  • Hardware/software co-design for efficient ML and sensing workloads
  • High-performance computing systems and parallel programming frameworks
  • FPGA-based acceleration and reconfigurable computing

Publications

  1. Tureis: Transformer-based Unified Resilience for IoT Devices in Smart Homes
    A. Borhani, V. Andalibi, B. Asgari
    Submitted to the 53rd International Symposium on Computer Architecture (ISCA), 2026.
    Under review
  2. ThingsDND: IoT Device Failure Detection and Diagnosis for Multi-User Smart Homes
    A. Borhani, H. R. Zarandi
    18th European Dependable Computing Conference (EDCC), pp. 113–116, 2022.
    Paper Slides
  3. FAST: FPGA Acceleration of Neural Networks Training
    A. Borhani, M. H. Goharinejad, H. R. Zarandi
    12th International Conference on Computer and Knowledge Engineering (ICCKE), pp. 492–497, 2022.
    Paper Slides Recording

Selected Projects

Tureis: Unified Resilience for Smart-Home IoT Devices
Transformer-based fault detection & diagnosis · Smart homes · Edge deployment
  • Lightweight BERT-style Transformer for real-time sensor failure detection and diagnosis in smart homes.
  • Bit-level sensor encodings and confidence-based fault scoring designed for edge devices.
  • Targets robust, low-latency inference on resource-constrained platforms.
Transformers Edge ML IoT Fault Detection
ThingsDND: IoT Device Failure Detection & Diagnosis
Smart-home sensing · Dependability · Anomaly detection
  • Context-aware method for detecting and diagnosing sensor failures in multi-user smart homes.
  • Models correlations between heterogeneous sensors to localize the source and type of faults.
Dependability Smart Homes IoT
Comparative Study of Parallel Programming Frameworks
HPC architectures · Intel MPI · OpenMPI · Intel TBB · PSTL
  • Evaluated Intel MPI, OpenMPI, Intel TBB, and PSTL on HPC platforms.
  • Analyzed scalability, communication efficiency, and performance bottlenecks.
HPC MPI TBB Parallel Programming
Hardware/Software Co-design of an Arithmetic Core
Reconfigurable systems · Xilinx Vivado · Arithmetic design
  • Designed and implemented an arithmetic core using Xilinx Vivado Design Suite.
  • Explored resource–performance trade-offs and performance optimization for FPGA-based realizations.
FPGA Vivado Hardware/Software Co-design
Error-Detection Techniques in Python
CRC · Checksum · Hamming code
  • Implemented CRC, checksum, and Hamming code algorithms in Python.
  • Compared detection capability and overhead across different error models.
Python Error Detection

Teaching & Experience

Research Assistant · University of Maryland
Computer Architecture and Systems Lab (CASL)
  • Research on resilient and efficient smart-home IoT systems, including the Tureis framework.
  • Focus on Transformer-based modeling, sensor encodings, and real-time edge deployment.
Teaching Assistant · CMSC 351 (Algorithms)
University of Maryland, College Park
  • Lead discussion sections, hold office hours, and help design/grade exams and assignments.
  • Assist students with algorithm design, complexity analysis, and problem-solving skills.

Skills

Programming
C, Python, CUDA, VHDL, Verilog, MPI, OpenMP, MATLAB

Tools
Xilinx Vivado, Synopsys Design Compiler, SOC Encounter, gem5, Keil, STM32CubeMX, Microsoft Visual Studio, Jupyter Notebook, LaTeX, Microsoft Office, Git

Platforms
Windows, Linux