Input scaling for artificial intelligence based channel state information feedback compression

A baseband processor, configured to:

• receive a channel state information (CSI) report configuration that indicates a number of PRBs;
• receive a configuration of the number of PRBs in each subband;
• determine a number of subbands based on the configured number of PRBs in each subband and the number of PRBs indicated by the CSI report configuration;
• and
• generate one or more eigen-vectors for each sub-band, wherein one or more of the eigen-vectors summarize a channel state matrix across a respective subband, further wherein the number of eigen-vectors corresponds to an integer multiple of the number of inputs of an artificial intelligence (AI)-based encoder used for CSI feedback compression.

A method for a base station, comprising:

• transmitting a channel state information (CSI) report configuration that indicates a number of PRBs;
• transmitting a configuration of the number of PRBs in each subband;
• receiving, from a user equipment (UE), a set of AI-based CSI feedback compression encoder outputs corresponding to compressed CSI feedback;
• determining a number of subbands based on the configured number of PRBs and the number of PRBs indicated by the CSI report configuration;
• based on the number of subbands, decoding the received set of AI-based CSI feedback compression encoder outputs to generate a set of M estimated input eigen-vectors summarizing a channel state matrix, wherein M is a number of inputs to an AI-based CSI feedback compression encoder;
• and
• transmitting physical downlink control channel or physical downlink shared channel (PDCCH/PDSCH) based on the M estimated input eigen-vectors.