Digital Communications Projects

Cairo University — Faculty of Engineering Electronics and Communications Dept. | 2nd Year

---

Project 1 — Statistical Analysis of Line Coding

Folder: Part1/

What it does

Simulates a digital transmitter generating an ensemble of 500 random waveforms (100 bits each) for three different line coding schemes, then statistically analyzes each one.

Line Codes Covered

Code	Logic 0	Logic 1
Unipolar	0 V	+A
Polar NRZ	−A	+A
Return to Zero (RZ)	−A (first half), 0 (second half)	+A (first half), 0 (second half)

Requirements Implemented

1. Statistical mean across the ensemble
2. Stationarity check (is the mean constant over time?)
3. Ensemble autocorrelation function R_x(τ)
4. Time mean and autocorrelation for a single waveform
5. Ergodicity check (does time mean ≈ ensemble mean?)
6. Bandwidth estimation via FFT

Files

• Statistical Analysis of Line Coding for Digital Communication.m — main MATLAB simulation
• Project Requirments.pdf — original project sheet
• Project1NotFinished.pdf / .docx — report (in progress)

---

Project 2 — Matched Filters, Correlators, ISI, and Raised Cosine Filters

Folder: Part2/
Status: In Progress

What it does

Simulates a binary PAM communication system over an AWGN channel, evaluating receiver performance using matched filters and correlators, then studies Inter-Symbol Interference (ISI) using raised cosine pulse shaping.

System Overview

• Transmitter generates binary polar signals (+1, −1) convolved with a pulse shaping function p(t)
• Signal passes through an AWGN channel
• Receiver uses either a matched filter h(t) = p(Ts − t) or a correlator to maximize SNR

Requirements

Part 1 — Noise-Free Analysis (10 bits)

• Generate 10 random bits and map to ±1
• Upsample and convolve with pulse shape p = [5 4 3 2 1]/sqrt(55)
• Filter with: (i) matched filter, (ii) a rectangular filter
• Compare filter outputs at sampling instants
• Compare matched filter output vs. correlator output

Part 2 — Noise Analysis (10,000 bits)

• Add scaled AWGN noise: variance = N0/2
• Calculate BER using matched filter vs. rectangular filter
• Sweep Eb/N0 from −2 dB to +5 dB in 1 dB steps
• Plot simulated BER curves vs. theoretical: BER = 0.5 × erfc(sqrt(Eb/N0))

Part 3 — ISI and Raised Cosine Filters (100 bits)

• Use square root raised cosine filters at transmitter and receiver
• Test 4 combinations of rolloff factor R and filter delay:
  • R=0, delay=2
  • R=0, delay=8
  • R=1, delay=2
  • R=1, delay=8
• Plot eye diagrams before and after the receive filter
• Comment on eye opening vs. sampling instant

Files

• Project Requirments.pdf — original project sheet

---

General Info

• Language: MATLAB
• Due date: 30 April 2026
• Max group size: 5 students