Sparse Coding for URLLC Transmission

The patent extracts only the essential error detection function by transmitting a compact CRC checksum instead of using complex channel coding schemes. The sparse coding structure inherently provides error detection capability, allowing the system to achieve high reliability without the overhead of complex forward error correction codes, thereby reducing signal processing complexity and latency

The patent changes the fundamental parameter of code rate by using sparse coding with very low effective code rates (only K non-zero symbols out of N total symbols). This parameter change enables the system to achieve high transmission reliability with significantly reduced processing complexity compared to traditional channel coding methods that require low code rates but result in long encoded sequences

The patent segments the encoded signal into sparse representations where only K out of N symbols are non-zero. This segmentation allows the receiver to focus processing only on the essential information carriers rather than processing the entire long encoded sequence, thereby achieving high reliability with reduced decoding latency despite using low effective code rates

The patent uses spreading codes to create multiple copies of the encoded information across different time-frequency resources. This copying approach provides redundancy for error correction while maintaining the ability to decode with low latency through simple correlation operations, avoiding the need to process long encoded sequences