Meteor M2 Satellite Image Decoder

Overview

The capstone project: a complete satellite ground station that receives and decodes weather images from Russian polar-orbiting Meteor M2 satellites. This project demonstrates mastery of digital communications, DSP, and system integration at a professional level.

Receiving images from space is compelling and memorable — this is the interview centerpiece. The project demonstrates the ability to design complete digital communications systems from theory to working hardware, directly relevant to aerospace and RF employers.

Signal Characteristics

Frequency	137.1 MHz or 137.9 MHz
Modulation	QPSK
Symbol Rate	72 ksps or 80 ksps
Bandwidth	~120 kHz
FEC (inner)	Convolutional K=7, R=1/2
FEC (outer)	Reed-Solomon (255,223)
Protocol	LRPT (Low Rate Picture Transmission)

System Architecture

137MHz QFH Antenna → LNA → SAW Filter → RTL-SDR
                                            ↓
                                      USB → Zynq ARM
                                            ↓
                                      Sample Buffer
                                            ↓
                                        FPGA PL
                                            ↓
    ┌————————————————————————————————┐
    ↓               ↓              ↓              ↓
  DDC → RRC → AGC → Costas → Gardner → Symbol → Viterbi → RS
  Filter    Loop    TED     Decision  Decoder     Decoder
    └————————————————————————————————┘
                                            ↓
                                      Frame Sync
                                            ↓
                                      LRPT Parser
                                            ↓
                                    Image Assembly → HDMI/File

Implementation Phases

Month 19-20: QPSK Demodulator

• Matched filter (root-raised-cosine)
• AGC (automatic gain control)
• Carrier recovery (Costas loop)
• Symbol timing recovery (Gardner TED)
• Soft decision output

Month 21-22: Forward Error Correction

• Viterbi decoder (K=7, R=1/2, 64 states)
• Convolutional deinterleaver
• Reed-Solomon decoder (use Xilinx IP initially)

Month 23-24: Protocol & Image

• CCSDS frame synchronization
• LRPT protocol parsing
• JPEG-like decompression
• RGB image assembly
• HDMI display output

Requirements

• Full QPSK demodulation chain
• Soft-decision Viterbi decoder
• Real-time processing of 72-80 ksps
• Frame synchronization and error handling
• Image output via HDMI or file

Skills Demonstrated

• Advanced DSP: Carrier and timing recovery loops
• FEC Implementation: Viterbi and Reed-Solomon decoders
• Complex Protocol Decoding: CCSDS and LRPT
• Complete System Integration: RF to image pipeline
• Space Communications: Domain knowledge

Hardware

Item	Est. Cost	Source
RTL-SDR V3/V4	(already have from ADS-B)	—
137 MHz LNA	$15-25	Amazon/AliExpress
137 MHz SAW filter	$10-15	Amazon/AliExpress
QFH Antenna (DIY)	$20-30	PVC pipe + coax
Coax cables, adapters	$15	Amazon
Total additional	~$60-85

Resources

• LeanHRPT — Reference decoder
• MeteorDemod — Another reference
• CCSDS 131.0-B-3 (TM Synchronization and Channel Coding)
• Costas loop and Gardner TED papers
• Viterbi decoder tutorials

Deliverables

• GitHub repo with complete VHDL source
• Individual testbenches for each module
• Simulation showing demodulation performance
• Gallery of received satellite images
• Video of live satellite pass reception
• Technical writeup explaining design decisions
• Resource utilization and timing reports

Implementation Notes

Implementation notes will be added during development.

Resource Utilization

To be measured after implementation.

Progress Log