A comprehensive suite of four major digital hardware design projects implemented in Verilog / SystemVerilog, covering neuromorphic spiking neural networks and single-cycle, multi-cycle, and pipelined RISC-V processor architectures.
👨💻 Authors: Amirali Dehghani (
810102443) · Nazhin Nikkhahbahrami (810102530)
🎓 Course: Computer Architecture — Dr. Safari | University of Tehran
| Project | Topic | Target Architecture & Highlights | Reports & Schematics | Full Documentation |
|---|---|---|---|---|
| CA1 | LIF Neuron | Neuromorphic computing, Q4.8 signed fixed-point, 19-state FSM, Euler leakage ( |
CA01-Report.pdf | CA1 Documentation ➔ |
| CA2 | Single-Cycle RISC-V | 32-bit RV32I core ( |
CA02-Report.pdf | CA2 Documentation ➔ |
| CA3 | Multi-Cycle RISC-V | 16-bit core, 14-state FSM controller, shared ALU & Memory, Array Sum benchmark | CA03-Report.pdf | CA3 Documentation ➔ |
| CA4 | 5-Stage Pipelined RISC-V | RV32I core with Hazard Unit (EX & MEM forwarding, load-use stalls, branch flushes) | CA04-Report.pdf | CA4 Documentation ➔ |
Hardware implementation of a biological Leaky Integrate-and-Fire (LIF) neuron model in SystemVerilog. This module simulates biological spiking neuron behavior by integrating input currents over time using Euler discretization (
👉 Explore CA1 Detailed Specifications & RTL Modules ➔
A fully functional 32-bit single-cycle RISC-V (RV32I) processor core where every instruction executes within one clock cycle (add, sub, and, or, slt, addi, xori, ori, slti, lw, sw, beq, bne, jal, jalr, lui). It includes a custom Python toolchain and a hand-crafted assembly benchmark implementing a Bubble Sort algorithm on 10 integers in memory.
👉 Explore CA2 Full Datapath & Decoder Control Matrix ➔
An optimized 16-bit multi-cycle RISC-V processor core that breaks instruction execution into multiple clock cycles using a 14-state Finite State Machine (FSM). By sharing hardware resources across execution states (a single shared 16-bit ALU for PC increments and data calculations, and a single unified memory for instructions and data) and utilizing intermediate registers (IR, MDR, A, B, ALUOut), this design achieves high area efficiency while executing complex control flows and array summation algorithms.
👉 Explore CA3 Full FSM State Machine & Control Matrix ➔
A high-performance 32-bit 5-stage pipelined RISC-V processor core featuring Instruction Fetch (IF), Decode (ID), Execute (EX), Memory (MEM), and Write-Back (WB) stages. It includes a dedicated Hazard Handling Unit with EX-to-EX and MEM-to-EX data forwarding to eliminate RAW stalls, load-use interlocking stall mechanisms, and branch misprediction flushing logic. Verified via ModelSim waveforms executing an array minimum search algorithm (
👉 Explore CA4 Full Pipelined Datapath & Hazard Handling Architecture ➔
- Languages: SystemVerilog (IEEE 1800-2017) & Verilog HDL
- Simulator: ModelSim / QuestaSim EDA Simulator
- Mathematical Modeling: MATLAB (LIF Neuron potential tracing)
- Assembly Toolchain: Custom Python Assemblers (
assembly_to_machine.py)
Computer-Architecture-Course-Projects-Fall04/
├── README.md # Root Repository Landing Page
├── CA1/ # CA1: Leaky Integrate-and-Fire (LIF) Neuron
│ ├── README.md # CA1 Detailed Documentation
│ ├── LIF.sv / datapath.sv / controller.sv # SystemVerilog RTL Modules
│ └── CA01-810102443-810102530.pdf # Project Report & Schematics
├── CA2/ # CA2: Single-Cycle RISC-V Processor
│ ├── README.md # CA2 Detailed Documentation
│ ├── RiscV.sv / Datapath.sv / ControlUnit.sv # Processor Source Code
│ ├── datapath_diagram.png # Datapath Schematic Image
│ └── Assembly/ # Bubble Sort C/Assembly/Python Toolchain
├── CA3/ # CA3: Multi-Cycle RISC-V Processor
│ ├── README.md # CA3 Detailed Documentation
│ ├── MultiCycle.sv / Datapath.sv / ControlUnit.sv # Processor Source Code
│ ├── datapath_diagram.png / fsm_diagram.png # Hardware & FSM Schematics
│ └── Report.pdf # Project Report & FSM Tables
└── CA4/ # CA4: 5-Stage Pipelined RISC-V Processor
├── README.md # CA4 Detailed Documentation
├── datapath_diagram.png / waveform.png # Pipelined Schematic & Simulation Waveforms
└── src/ # Pipelined Datapath & Hazard Handling Unit
🎓 Computer Architecture Course — Fall 1404 (2025–2026)
Department of Electrical and Computer Engineering — University of Tehran
⭐ If you found this repository helpful, please consider giving it a star!