Lightweight Universal Codec

Under the LUC program, DARPA IPTO will aim to develop a universal encoder-decoder (codec) capable of sending and receiving information using any known communication error correction code and any developed in the future. Leveraging the groundbreaking work of DARPA’s Guessing Random Additive Noise Decoding (GRAND) study, the LUC program will aim to create an adaptive error correction codec capable of multiple on-the-fly changes to error correction codes. The program will seek to achieve this by developing an encoder capable of utilizing a highly diverse suite of error correcting codes that are easily updatable. The encoder is expected to pair with a GRAND-implementing decoder and the resulting codec will be designed such that the encoder incorporates the decoder’s GRAND noise model as well as other operating factors into its code selection process for link budget and task/mission optimization. Integration of these codecs into various existing communication technologies would demonstrate improved efficiency, security, throughput, and interoperability achieved by record-setting low-power decoding and adaptive, dynamic forward error correction. The LUC program will evaluate the success of these codecs and integrated systems by measuring elements of code universality, link budget improvement solely through the implementation of the codec, and the communication systems’ performance in heterogeneous network environments. To maximize benefits of the LUC program and proliferation of its resulting technologies into both the commercial and military domains, DARPA will execute the program at the unclassified level with performers’ target implementations/integrations not to exceed Controlled Unclassified Information. Performers will be allowed to select their own target communication system(s) for integration and are expected to identify candidates that will demonstrate the greatest and/or most impactful improvement(s) through LUC implementation.