Efficient Scheduling for the Massive Random Access Gaussian Channel

On Compressed Sensing of Binary Signals for the Unsourced Random Access Channel

“…Gaussian sensing matrix and AMP recovery. Both implementations for Phase 1 correspond to similar performance, with slight preference for the latter, and substantially improve the state-of-the-art for unsourced random access with (a small amount of) feedback [12]. End-to-end required Eb/N0 vs k AMP (dense Gaussian i.i.d.)…”

“…Total E b /N 0 required to achieve end-to-end PUPE ≤ 0.05. We see that, by using a better compressed sensing algorithm for binary signals, significant gains can be achieved over the current state of the art [12].…”

“…We slightly deviate from the mainstream literature on unsourced random access, by allowing for some feedback to be sent from the receiver to all potential users through a broadcast channel. This option was mostly avoided until now, with the exception of Facenda and Silva [12], as it was believed that the large number of potential users and the small payloads for each active users renders scheduling too wasteful. Recently, Kang and Yu [71] established a connection between scheduling for the unsourced random access channel and perfect hashing and demonstrated that in fact scheduling for the unsourced random access channel can be attained with a very small cost.…”

“…End-to-end required Eb/N0 vs k AMP (dense Gaussian i.i.d.) Proposed: sparse LDPC; Glauber, NNLS init Facenda-Silva [12] Figure 3. Total E b /N 0 required to achieve end-to-end PUPE ≤ 0.05.…”

“…Over the last few years, there has been great interest in developing efficient lowcomplexity schemes for the unsourced random access channel [2][3][4][5][6][7][8][9][10][11][12][13][14][15]. A natural approach for this setup is for all users to transmit codewords from the same codebook.…”