Remote sensing satellites image the Earth from orbits hundreds of kilometers above the ground. The raw payload data cannot be used directly; instead, a full set of ground station equipment is required to perform radio frequency reception, digitization, signal synchronization, demodulation and decoding, protocol parsing, and image reconstruction, ultimately producing standardized, usable image products. Among these, .the VPX-architecture high-speed baseband data processing device - i.e., the Remote Sensing Data Processing Equipment - serves as the core hub of the ground receiving system: it interfaces with the antenna' s RF receive chain at the front end and outputs complete, valid data to fast-viewing systems, storage systems, and industry-specific workstations at the back end. The following section provides a comprehensive breakdown of the device' s standardized processing workflow and hardware architecture, based on an actual satellite-to-ground signal transmission link.
1. Front-End RF Input and Digital Preprocessing Chain
When a satellite passes overhead, the parabolic antenna captures the satellite's downlink RF carrier. After being amplified by a low-noise amplifier (LNA) and filtered, the signal is sent to an outdoor downconverter, which downconverts the high-frequency RF signal to a standard intermediate frequency. The device supports two types of input specifications:
1) Standard IF inputs: Common 70 MHz and 140 MHz IFs; the device can be configured with a dedicated 720 MHz IF as a custom solution;
2) L-Band RF Direct Input: Supports direct input of L-band signals in the 950–1500 MHz range; a built-in secondary downconversion module converts the signal to the device's standard intermediate frequency.
After the intermediate-frequency signal is fed into the baseband equipment, it first undergoes automatic gain control (AGC) and intermediate-frequency filtering, and is then sampled by a high-speed analog-to-digital converter (ADC), converting the analog intermediate-frequency signal into a complex digital baseband sample stream. All synchronization, demodulation, and decoding operations are performed based on the digital sample signal.
Due to the high-speed movement of the satellite in orbit, the downlink signal carries a wide range of Doppler shifts. The device incorporates a carrier tracking loop within its FPGA, which adjusts the frequency and phase of the local numerically controlled oscillator (NCO) in real time to continuously compensate for Doppler shifts and maintain a stable signal lock. The synchronization process follows the standard engineering sequence: bit synchronization first, followed by precise carrier tracking. The bit synchronization module extracts a precise symbol timing clock from the sampled data stream, providing a unified time reference for the carrier loop and symbol decision, thereby preventing symbol sampling offset and loss of lock caused by Doppler shift.
2. Multi-format Demodulation and Channel Error Correction Decoding
After synchronization and locking are complete, the digital symbol stream enters the FPGA hardware demodulation engine. The device incorporates a demodulation IP core compatible with mainstream remote sensing satellite systems, supporting modulation schemes such as BPSK, QPSK, SQPSK, 16QAM, and GMSK. It can be adapted to different satellite models simply by switching software parameters, without the need to replace hardware boards.
The space transmission process introduces noise and fading, which cause bit errors. The on-board transmitter applies multi-level channel error-correcting codes, and the baseband equipment is equipped with corresponding decoding modules: it supports RS codes, convolutional Viterbi decoding, and LDPC soft-decision decoding. Hardware parallel acceleration enables high-throughput, real-time error correction, maximizing the recovery of transmission errors and outputting a clean, error-free bit data stream.
3. CCSDS Standard Deinterleaving, Frame Synchronization, and Packet Parsing
All civilian and commercial remote sensing satellites uniformly comply with the CCSDS Space Data System standard, and the equipment fully implements CCSDS end-to-end processing: Satellite downlink data is scrambled using a pseudorandom sequence to suppress carrier leakage; after decoding, a de-scrambling operation is first performed to restore the original randomized bitstream.
After descrambling, the module searches for and locks onto the configurable frame synchronization header to locate the start of each frame of data; it then parses the frame header fields to extract the VCID (Virtual Channel Identifier) and APID (Application Process Identifier), thereby distinguishing between multiple types of data streams: optical/infrared multispectral payload image data packets, satellite telemetry auxiliary data packets, and on-board attitude and orbit auxiliary data.
The system categorizes and routes load data and auxiliary information based on APID; auxiliary data is cached and archived separately, while compressed image data packets are sent to the image decompression unit for independent processing.
4. Real-Time Decompression and Reconstruction of Compressed Images on Board
To conserve downlink bandwidth, optical, hyperspectral, and infrared remote sensing satellites are generally equipped with on-board real-time compression modules that are compatible with three mainstream compression standards:
1) Space Non-Destructive Testing Standard: CCSDS 123 (Specifically for Multispectral/Hyperspectral 3D Imaging);
2) General-purpose lossless: JPEGLS;
3) Lossy high-fidelity compression: JPEG 2000.
The baseband device features a multi-core CPU paired with a dedicated decompression acceleration unit. It adaptively selects the appropriate algorithm based on the compression flags in the packet headers and dynamically adjusts decompression parameters to balance processing latency and image fidelity, fully restoring sub-meter-resolution original imagery to meet the demands of high-precision applications such as national land surveying and agricultural and forestry monitoring.
5. VPX Heterogeneous Real-Time Computing Architecture (FPGA + Multi-core CPU)
The Remote Sensing Data Processing Equipment utilizes a standard VPX ruggedized chassis and employs a heterogeneous parallel architecture combining FPGA hardware acceleration with multi-core CPU general-purpose computing. Unlike purely software-based multi-core solutions, it supports a maximum real-time throughput of 800 Mbps per channel:
1) FPGA logic layer: Handles high-real-time physical layer tasks-carrier tracking, bit synchronization, demodulation, LDPC/Viterbi decoding, frame synchronization, and CCSDS de-scrambling and de-framing-using a million-level parallel pipeline to process high-speed data streams with zero latency;
2) Real-time software layer for multi-core CPUs: Responsible for image decompression, quick-view generation, task scheduling, status monitoring, data routing and distribution, and remote operations and maintenance interaction;
3) Modular assembly line decomposition: Carrier synchronization, demodulation and decoding, protocol parsing, decompression and quick preview, and data distribution are divided into independent task units and scheduled in parallel across CPU cores and FPGA channels.
The system supports software-defined task topology reconfiguration: when switching between satellites with different parameters, operations and maintenance personnel can use the backend software to reconfigure module data streams, code rates, modulation and coding parameters, and frame synchronization words; in the vast majority of scenarios, no hardware modifications are required. Only in scenarios involving RF bandwidth or specific frequency band requirements is it necessary to replace the front-end RF filter module.
6. High-Speed Data Exchange and Scalable Storage Solutions
The device is equipped with multiple types of high-speed service interfaces to support end-to-end data exchange between the front and back ends:
Front-end inputs: 10-Gigabit optical ports and high-speed LVDS differential interfaces, which receive RF-digitized data streams from the front end;
Backend Output: Gigabit Ethernet, 10-Gigabit Ethernet, used to distribute decompressed video and QuickView data;
Cluster Interconnect: IB high-speed bus, primarily used for interconnecting multi-device clusters and high-capacity backend storage pools; not used for front-end RF signal access.
The standard processing rate for a single channel ranges from 10 Mbps to 800 Mbps; when operating in dual-channel parallel mode, throughput doubles. Storage is expanded via an external disk array, with storage capacity scaling linearly with the number of hard drives in the array and backplane bandwidth. However, due to constraints imposed by hardware slots and I/O bandwidth, capacity cannot be expanded indefinitely; it is sufficient to meet the requirements for storing massive amounts of remote sensing data on a daily basis.
7. Full-Linkage Real-Time Quality Monitoring and High-Reliability Operational Support
The system deploys hierarchical condition monitoring nodes throughout the entire signal processing chain to continuously record operational metrics:
1) Physical Layer: Demodulation Eb/N0, real-time bit error rate, carrier/bit synchronization status, Doppler shift;
2) Data Link Layer: Number of frame synchronization lock-ups, packet checksum results, and APID packet integrity statistics;
3) Business Layer: Decompression frame rate, QuickView output latency, and interface throughput load.
All alarm logs and operational metrics are displayed locally in real time and automatically archived and stored. The system supports both local console and remote web-based access for operations and maintenance. The device utilizes military-grade ruggedized boards and a wide-temperature power supply, with a Mean Time Between Failures (MTBF) of ≥25,000 hours, meeting the requirements for long-term, continuous 24/7 operation at ground stations.
8. Project Implementation and Industry Applications
This series of Remote Sensing Data Processing Equipment has been deployed in large quantities on multiple domestic civilian remote sensing satellites currently in orbit, covering all types of payloads-including visible light, multispectral, hyperspectral, and infrared-and consistently producing sub-meter high-resolution standard imagery products. It has been widely applied in civilian fields such as agricultural and forestry resource surveys, water conservancy and flood control monitoring, natural resource mapping, and smart city planning; At the same time, it has successfully completed multiple rounds of international cooperative satellite-to-ground link testing and is compatible with CCSDS-standard remote sensing missions from multiple countries, demonstrating stable and reliable processing performance even under complex Doppler and low signal-to-noise ratio conditions.
9. Description of Customized Solutions
If a ground station needs to set up a brand-new satellite reception system, this Remote Sensing Data Processing Equipment offers comprehensive, in-depth customization: the intermediate frequency (IF) center frequency, modulation/coding schemes, CCSDS frame parameters, high-speed interface types, number of channels, and storage capacity can all be configured as needed. It provides a full-chain solution ranging from RF reception, real-time baseband processing, and live display to back-end storage and distribution.
To obtain a customized configuration plan for your project or a white paper detailing technical specifications, please contact our technical team: info@satgroundapplication.com
