2-Step RACH and Unsourced Massive Random Access in Cell-Free User-Centric Networks

In this work we consider a scheme for unsourced random access (uRA) in cell free (CF) wireless networks, which is conceptually reminiscent of the 2-step RACH scheme defined in 3GPP for cellular networks. During the designated random access channel (RACH) slots, users engaged in random access transmit codewords drawn from a shared codebook. The distributed and CF nature of the network is adressed by partitioning the network coverage area into zones, referred to as “locations,” and assigning a location-specific access codebook to each location. We use the recently proposed multisource Approximated Message Passing (AMP) algorithm for joint detection of the access codewords and channel estimation. In particular, AMP generates two outputs:a list of active codewords and an estimate of the corresponding channel vectors. The aforementioned outputs facilitate the formation of a user-centric cluster. The impact of random access users and the overall system performance are assessed based on the sum spectral efficiency for uplink (UL) and downlink (DL) of the data transmission slots, under limited random activity of the users.