Spinal Codes Based Efficient Rateless Transmission Scheme for Massive MIMO System with Imperfect CSI

The potential performance gains promised by massive multi-input and multi-output (MIMO) rely heavily on the access to accurate channel state information (CSI), which are difficult to obtain in practice when channel coherence time is short and the number of mobile users is huge. Therefore, a critical question is how to make the system perform well with imperfect CSI. Rateless codes with adaptive code-rates can free the system from the accurate CSI requirements and guarantee it to approach the maximum achievable rate (MAR) as well as improve the achieved-rate over that based on the fixed-rate codes. In this paper, a recently proposed family of rateless codes, called spinal codes, will be involved in massive MIMO transmission scheme, which in general approaches to the MAR with sufficiently large encoding block-size. In addition, multi-level puncturing and dynamic block-size allocation (MPDBA) techniques are proposed, where the block-sizes are determined by user MAR to curb the average retransmission delay for successfully decoding the messages, which further enhancing the system retransmission efficiency. Multi-level puncturing, which is MAR dependent, limits the gap between the system MAR and the related achieved-rate. Theoretical analysis are provided to demonstrate that spinal codes with MPDBA can guarantee the system retransmission efficiency as well as achieved-rate. Numerical simulation results are presented to demonstrate these benefits.