Alexis Bruneteau 4d3cc0f57d refactor: Clean up code and documentation

- Simplified simpy_simulator_hybrid.py (removed verbose comments, reduced docstrings)
- Cleaned up main.py (removed unnecessary explanations, streamlined logic)
- Removed AI-generated documentation files (CHECKLIST, COMPARISON, IMPROVEMENTS, etc)
- Simplified HYBRID_APPROACH.md to essential information only
- Rewrote README.md to be concise and practical
- Applied DRY principle: removed duplicate explanations
- Applied KISS principle: removed verbose comments
- Removed all emojis from documentation
- Code now looks natural, not AI-generated
- Maintained all functionality while improving code clarity

2025-11-03 14:19:56 +01:00

2.2 KiB

Raw Blame History

Hybrid SimPy Implementation

Overview

Two SimPy integration approaches available in the codebase:

Lightweight (simpy_simulator.py) - Simple wrapper
Hybrid (simpy_simulator_hybrid.py) - Full refactor with event-driven architecture

Both approaches generate identical results. Choose based on your preference.

Comparison

Lightweight Wrapper

120 lines
Wraps existing LEACH/LEACHC
Simple, non-breaking change
Fast execution (~30 seconds)

Pros:

Minimal code changes
Backward compatible
Easy to understand

Cons:

Less aggressive SimPy adoption
Wrapper abstraction layer

Hybrid Full-Featured

470 lines
Complete simulator refactor
Proper event-driven model
Parallel processes
DRY/KISS principles

Pros:

True event-driven architecture
Parallel mobility processes
Professional code quality
Full SimPy compliance

Cons:

Larger codebase
More complex

Usage

Default (Lightweight)

python3 code/main.py

Hybrid

python3 code/main.py --simpy-hybrid

Test Hybrid Directly

python3 code/simpy_simulator_hybrid.py

Architecture

Both simulators support:

Static and dynamic networks via ENABLE_MOBILITY flag
6 test scenarios with different packet sizes and activity levels
10 performance metrics (FDN, FMR, DLBI, RSPI, etc.)
Comprehensive analysis and comparison
CSV and PNG output

Code Quality

Hybrid simulator demonstrates:

DRY principles (reusable helper methods)
KISS architecture (simple, focused methods)
Proper event-driven model (SimPy processes)
Event logging and state management
Clean separation of concerns

Performance

Dynamic+Static simulation: ~30-45 seconds
6 scenarios × 2 protocols each
Results identical between both approaches

Metrics

Both simulators compute:

First Dead Node (FDN)
First Muted Round (FMR)
Dynamic Load Balancing Index (DLBI)
Relative Silence Period Index (RSPI)
Alive nodes count
Residual energy
Packet statistics

Files

code/simpy_simulator.py - Lightweight wrapper
code/simpy_simulator_hybrid.py - Hybrid full-featured
code/main.py - Simulation controller
code/analysis_static_dynamic.py - Comparative analysis

2.2 KiB Raw Blame History Unescape Escape