Digital Communication Link Simulator
A completed complex-baseband communication study comparing BPSK, Gray QPSK, Gray 16-QAM, AWGN, flat Rayleigh fading, convolutional coding, pulse shaping, and receiver impairments.

Software compatibility
The portable source was executed in GNU Octave 11.3.0 and targets MATLAB R2026a syntax. No Communications Toolbox or Simulink licence is required.
Project definition
Problem statement
Communication results depend on modulation mapping, energy normalization, code rate, noise definition, pulse shaping, fading, synchronization assumptions, and finite sample counts. A visually smooth BER curve can still be invalid when any convention is inconsistent.
Project objectives
- Implement BPSK, Gray QPSK, and Gray 16-QAM without proprietary toolbox objects.
- Compare AWGN and coherent flat Rayleigh channels under an explicit information-bit E_b/N_0 definition.
- Implement a rate 1/2 convolutional encoder and hard-decision Viterbi decoder.
- Measure phase, frequency, timing, and rolloff sensitivity with controlled seeded experiments.
- Report error counts, Wilson confidence intervals, convergence behavior, analytical comparisons, and limitations.
Project structure
Project components
Configuration
Defines and validates modulation, channel, code, bit count, seed, impairments, waveform settings, and stopping rules.
Symbol link
Maps bits, applies coding and channel effects, performs coherent decisions, and records BER with confidence limits.
Waveform link
Adds root-raised-cosine transmission, matched filtering, oversampling, and timing-offset experiments.
Coding
Provides the rate 1/2 convolutional encoder and hard-decision Viterbi decoder with shared trellis arithmetic.
Evidence pipeline
Generates 126 cases, a baseline trace, summaries, validation results, and 13 paired figures.
Methodology
Project workflow
- 01Test
Run all 38 tests in MATLAB, GNU Octave, or the pinned container.
- 02Reproduce
Generate all retained cases and figures from explicit seeds and configurations.
- 03Validate
Run 49 checks across case counts, theory agreement, trends, schemas, and figure inventory.
- 04Compare
Read BER, confidence, coding, fading, impairment, timing, and convergence evidence together.
- 05Extend
Change one mechanism, add a focused test, retain new evidence, and revise the claim boundary.
Demonstration scenario
The student runs the test suite, reproduces the retained study, compares simulated and analytical BER, explains the coding crossover, inspects constellation and waveform evidence, and shows why timing, frequency offset, fading, and finite sample size change the result.
Engineering
Tools and method
- Tools
- The project uses MATLAB R2026a compatible source, GNU Octave 11.3.0, Python 3 for subject analysis, simulation, and results.
- Environment
- Portable MATLAB-compatible functions verified in GNU Octave 11.3.0, with no Communications Toolbox or Simulink dependency.
- Simulation
- Seeded vectorized Monte Carlo cases with explicit energy normalization, target-error stopping, and maximum-bit limits.
- Waveform model
- Root-raised-cosine transmit and receive filtering with declared rolloff, span, oversampling, and sampling offset.
- Verification
- Analytical BER references, exact mapping tests, measured-noise checks, deterministic replay, statistical intervals, and retained raw evidence.
Testing
Evaluation
Evaluation measures
- 36 modulation and channel BER cases
- 10 coded and uncoded comparison cases
- 12 phase and frequency impairment cases
- 20 pulse-shaping and timing cases
- 48 finite-sample convergence replicates
- Maximum retained AWGN absolute theory error of 0.003465
- Maximum retained Rayleigh absolute theory error of 0.021021
- All 38 source tests and 49 result checks passing
Project boundaries
- The project is a coherent complex-baseband simulation.
- No RF front end, analogue hardware impairment, automatic synchronization acquisition, channel estimation, or over-the-air measurement is claimed.
- The hard-decision convolutional decoder is one selected coding example and does not represent a modern complete coded link.
- The waveform model does not measure occupied bandwidth or implement a complete communication standard.
- The project is not regulatory, spectrum, safety-critical, or hardware certification evidence.
Included
- 01Portable BPSK, Gray QPSK, and Gray 16-QAM source
- 02AWGN and flat Rayleigh channel studies
- 03Rate 1/2 convolutional encoder and hard-decision Viterbi decoder
- 04Phase, frequency, timing, rolloff, coding, and convergence experiments
- 05126 retained simulation cases with raw CSV and JSON evidence
- 0613 labelled result figures in PNG and SVG formats
- 0738 automated numerical tests and 49 result-validation checks
- 0846 annotated references and three attributed literature images
- 09Complete project files, models, calculations, and analysis material in a private GitHub repository
- 1073-page project documentation in PDF and editable Word formats
- 1119-page setup and usage guide
Project record
No information is collected on this page.
- Permanent project ID
- GP-EC-1ECPN31
- Catalogued
- 22 Aug 2026
- Completed
- 28 Aug 2026
- Verified
- 28 Aug 2026
- Demonstration
- Included in repository
Handover
After purchase
- 01Payment is confirmed
The project is marked unavailable and cannot be purchased again.
- 02Repository access is granted
The buyer's submitted GitHub account receives access to the private repository.
- 03The purchase record is delivered
The certification sheet is prepared from the reviewed buyer details and sent privately by email.