I. Introduction
In recent years, linearly-precoded Rate-Splitting (RS) has been recognized as a promising transmission strategy to enhance rate, robustness and Quality of Service (QoS) for future generations of wireless communication systems. Inspired by the Han-Kobayashi scheme for the two-user interference channel [1], in RS, the message of each receiver is split into a common part and a private part at the transmitter [2]–[7]. The common parts of all the receivers are jointly encoded into a common stream required to be decoded by all the receivers while the private parts are independently encoded into private streams for the corresponding receivers only. All the streams are linearly precoded and simultaneously transmitted to the receivers. By allowing each receiver to first decode the common stream and use Successive Interference Cancellation (SIC) to remove the common stream before decoding the intended private streams, receivers are enabled to partially decode the interference and partially treat the remaining interference as noise. A more general framework of RS, namely Rate-Splitting Multiple Access (RSMA), is proposed in [8]. RSMA has been shown to outperform Multi-User Linear Precoding (MU-LP) where each receiver directly decodes the intended message by fully treating the interference as noise and power-domain Non-Orthogonal Multiple Access (NOMA) (simply referred to as NOMA in the sequel) where Superposition Coding (SC) and SIC are enabled respectively at the transmitter and receivers (SC-SIC) such that receivers with stronger channel strength are required to decode the messages of the receivers with weaker channel strength. However, most existing works on RS only consider Wireless Information Transfer (WIT) via Radio-Frequency (RF).