Summary:
UCLA researchers have developed Syndrome Sphere Decoding (SSD), a breakthrough algorithm that delivers near-optimal error correction for tail-biting convolutional codes (TBCCs) with dramatically reduced decoding complexity. By eliminating the need for large list decoders, SSD enables faster, low-power, and reliable communication across wireless, satellite, and embedded systems—without compromising frame error rate (FER) performance.
Unmet Need:
Modern wireless communication—whether in satellites, industrial IoT, or autonomous vehicles—depends on short, reliable message delivery in noisy environments.
TBCC + CRC schemes are widely used because of their high reliability in short block-length scenarios.
Conventional decoders often rely on list decoding, which is computationally expensive, power-intensive, and poorly suited for latency-sensitive or battery-constrained platforms.
As systems evolve toward 5G/6G URLLC, IoT networks, and embedded devices, there is an urgent need for efficient, lightweight decoders that deliver the same reliability without the cost of large list searches.
Innovation:
UCLA’s Syndrome Sphere Decoding (SSD) algorithm introduces a two-stage decoding approach that balances reliability and efficiency:
Initial decoding with Viterbi: SSD identifies the most likely decoding path using standard low-complexity methods.
Syndrome-guided refinement: By computing the “syndrome” (a measure of deviation from a valid codeword), SSD restricts its search to a small neighborhood, or sphere, ensuring candidate solutions satisfy both the TBCC and linear expurgating function (e.g., CRC) constraints.
This method avoids exponential list growth while preserving error correction strength. Hybrid modes allow fallback to small, fixed-size lists for added flexibility, giving system designers fine control over the latency–complexity tradeoff.
Key Advantages:
Near-optimal error correction without reliance on large list decoding.
Order-of-magnitude reduction in average computational complexity.
Low power and memory footprint, ideal for embedded and portable devices.
Latency-optimized for real-time, short-packet communication.
Scalable hybrid mode supports customizable performance tuning.
Seamless integration into existing TBCC/CRC architectures for satellite, IoT, and 5G/6G platforms.
Potential Applications:
- Satellite & deep-space links: Maximizing reliability under extreme bandwidth and power constraints.
- 5G/6G URLLC: Supporting ultra-reliable, low-latency communication in telecom networks.
- IoT & sensor networks: Short-packet, low-power communication for industrial and environmental monitoring.
- Autonomous systems: Reliable, real-time communications for drones, vehicles, and robotics.
- Healthcare devices: Secure, low-power transmission for wearables and implantables.
- Defense & aerospace: Energy-efficient, high-reliability communication in constrained environments.
State of Development:
New development: Current invention submitted for presentation at the International Symposium on Topics in Coding (Aug 2025).
Representative Publications:
- W. Sui, B. Towell, Z. Qu, E. Min, R. D. Wesel. Linearity-Enhanced Serial List Decoding of Linearly Expurgated Tail-Biting Convolutional Codes. ISIT 2024, Athens, Greece.
- Syndrome Sphere Decoding of Linearly Expurgated Tail-Biting Convolutional Codes. Submitted to ISTC 2025.
Reference:
UCLA Case No. 2025-285
Lead Inventor:
Rick Wesel, PhD – Department of Electrical and Computer Engineering