██████╗ ██████╗ ███╗ ███╗███╗ ███╗███████╗
██╔════╝██╔═══██╗████╗ ████║████╗ ████║██╔════╝
██║ ██║ ██║██╔████╔██║██╔████╔██║███████╗
██║ ██║ ██║██║╚██╔╝██║██║╚██╔╝██║╚════██║
╚██████╗╚██████╔╝██║ ╚═╝ ██║██║ ╚═╝ ██║███████║
╚═════╝ ╚═════╝ ╚═╝ ╚═╝╚═╝ ╚═╝╚══════╝
channel coding · spread spectrum · OFDM · LEO
A hands-on simulation library covering the full digital communications stack — from binary channel coding and error correction to spread-spectrum CDMA systems and live LEO satellite link budgets. Every result is derived analytically and validated by Monte Carlo simulation.
Files: A1/EX1.m, A1/EX2.m, A1/EX3.m
Linear block codes characterised by generator matrix G and parity-check matrix H — built from scratch. Derived the full distance spectrum, computed undetected-error probability analytically, then cross-validated against BSC Monte Carlo runs across six decade-spaced BEP values.
Key concepts implemented:
- Systematic (8,4) linear block codes — two competing code designs compared head-to-head
- Hamming weight distributions and minimum distance extraction
- CRC generator polynomial design and detection-capability evaluation
- Syndrome-based error detection over a Binary Symmetric Channel
Files: A2/EX1.m, A2/EX2.m, A2/EX3.m
Side-by-side analytical and simulated comparison of hard-decision vs. soft-decision decoding. Derived the union bound and asymptotic bound on codeword error rate for soft decoding and demonstrated the 2–3 dB gain over hard decoding empirically.
Key concepts implemented:
- Maximum Likelihood hard decoding with syndrome lookup table
- Soft-decision decoding: union bound and asymptotic approximation vs. simulation
- Linear Feedback Shift Register (LFSR) design — maximal-length m-sequences generation
- Synchronisation marker insertion and detection
- Autocorrelation and power spectral density via FFT
Files: A3/EX1.m, A3/EX2.m, A3/EX3.m
From sequence design theory to a full CDMA system simulation and adaptive OFDM bit allocation.
Key concepts implemented:
- Gold sequence generation (degree-7 polynomials, N=127 chips) — autocorrelation and cross-correlation characterised and plotted
- Welch and Sidelnikov bounds on sequence sets, evaluated across family sizes
- CDMA multi-user system: interference analysis and capacity limits
- OFDM adaptive bit loading via the Hughes-Hartogs algorithm — per-subcarrier SNR-aware allocation to maximise throughput over a frequency-selective channel
File: A4/A4.m
End-to-end satellite communication system modelled in MATLAB's Satellite Communications Toolbox. Simulated a sun-synchronous LEO orbit (500 km, 97.4° inclination) over a 7-day scenario with a 60-second propagation timestep.
Key concepts implemented:
- Orbital mechanics: satellite scenario creation with Keplerian parameters
- Four global ground stations (Inuvik, Svalbard, Awarua, Troll)
- Gaussian dish antenna sizing (0.18 m satellite, 3.7 m ground) with aperture efficiency
- Link budget: EIRP, G/T, free-space path loss, Doppler, Eb/N₀ margin
- Access interval and visibility window analysis across the station network
- Channel degradation: ITU-R P.618 rain attenuation model
| Tool | Role |
|---|---|
| MATLAB | Core simulation & visualisation engine |
| Communications Toolbox | BSC, AWGN channels, Gold sequence generator |
| Satellite Communications Toolbox | Orbital mechanics, link budget, ground station modelling |
| Signal Processing Toolbox | FFT-based correlation, PSD analysis |
Channel Coding · Error Correction · Monte Carlo Simulation · Spread Spectrum · CDMA · OFDM · LFSR / m-sequences · Gold Sequences · Satellite Link Budget · Orbital Mechanics · Analytical-Simulation Cross-Validation
M.Sc. Communications Engineering — Politecnico di Torino