Module 06: Broadband PHY/MAC Scheduling for a Starlink-Like System

Phase: 2 - Acceleration Builds on: Modules 02, 04, and 05

Math You’ll Learn

Calculus I: Derivatives, Chain Rule, Optimization, and Applications

Scheduling is optimization under changing constraints. You will use derivatives and piecewise functions to reason about marginal capacity, MCS thresholds, and beam-resource allocation.

• Chain rule, product rule, quotient rule - composed capacity and link-quality functions.
  • Starlink application: throughput depends on SNR, beam load, bandwidth, and scheduler policy.
• Related rates - changing elevation changes range, FSPL, margin, and capacity.
• Optimization - choose the best allocation under power, bandwidth, beam, and fairness constraints.
  • Starlink application: decide which terminal or beam gets the next unit of resource.
• Piecewise functions - MCS thresholds and outage behavior.

After this: You can simulate a Starlink-like broadband scheduler and explain fairness/throughput trade-offs without assuming Starlink’s proprietary MAC.

Resources:

• Stewart, Calculus: Early Transcendentals, Chapters 3-4
• Digital communications texts for MCS, FEC, and spectral efficiency
• DVB-S2X/RCS2 standards as public comparison material, not assumed Starlink implementation

What You’ll Learn

This module replaces DVB as the central topic with broadband PHY/MAC principles that map better to a proprietary LEO broadband system. DVB remains useful as a public comparison for MODCOD/ACM, but the goal is to reason about Starlink-like scheduling and capacity.

Broadband PHY Concepts

• OFDM/OFDMA, TDMA, SC-FDMA, and why multiple-access choice affects scheduling.
• Adaptive modulation and coding: MCS/MODCOD thresholds, spectral efficiency, outage.
• LDPC/FEC basics and coding gain.
• HARQ/ARQ trade-offs under LEO delay.
• SNR, SINR, interference, beam isolation, and frequency reuse.

MAC and Beam Scheduling

• Multi-beam scheduling and terminal-to-beam assignment.
• Proportional fairness, max-throughput, strict priority, and weighted fair scheduling.
• Return-link scheduling and demand-based allocation.
• QoS for voice/video/gaming/bulk traffic under variable capacity.
• Beam hopping and load balancing across satellites.
• How link budget outputs become scheduler inputs.

Public vs Proprietary Boundary

• Do not claim Starlink uses DVB, LTE MAC, or any specific waveform internally unless public documentation says so.
• Use public standards and algorithms to build defensible models.
• Document what is a generic broadband satellite concept vs what is Starlink-public.

C++ and Python Skills

C++ focus: templates, enum classes, state/strategy patterns, STL algorithms, deterministic simulation loops.

Python focus: animated plots, lookup tables, event simulation, comparing scheduler policies.

Projects

Project 1: Starlink-Like Beam and Capacity Scheduler (C++)

Build a simplified beam scheduler.

What you’ll build:

• Model terminals with demand, QoS class, SNR, current MCS, and queue backlog.
• Model beams with bandwidth, capacity, and frequency-reuse constraints.
• Implement max-throughput, strict-priority, and proportional-fair scheduling strategies.
• Track utilization, dropped demand, queue delay, and fairness metrics.
• Allow link-margin inputs from Module 05.

C++ skills used: templates, strategy pattern, enum classes, STL algorithms, tests.

Toolkit: Add BeamScheduler.

Project 2: MCS/ACM Simulator (Python)

Visualize adaptive link behavior.

What you’ll build:

• Create a public MCS-style threshold table with SNR to spectral-efficiency mappings.
• Simulate SNR variation from elevation and rain fade.
• Compare fixed coding, adaptive coding, and scheduler-aware adaptation.
• Plot throughput, outage, selected MCS, and queue backlog over time.
• Explain why a scheduler must optimize user experience, not just instantaneous throughput.

Python skills used: NumPy, matplotlib animation, lookup tables, event simulation.

Technology Reference

Where This Tech Is Used

Books and Resources