Massive Connectivity With Massive MIMO—Part I: Device Activity Detection and Channel Estimation

This two-part paper considers an uplink massive device communication scenario\nin which a large number of devices are connected to a base-station (BS), but\nuser traffic is sporadic so that in any given coherence interval, only a subset\nof users are active. The objective is to quantify the cost of active user\ndetection and channel estimation and to characterize the overall achievable\nrate of a grant-free two-phase access scheme in which device activity detection\nand channel estimation are performed jointly using pilot sequences in the first\nphase and data is transmitted in the second phase. In order to accommodate a\nlarge number of simultaneously transmitting devices, this paper studies an\nasymptotic regime where the BS is equipped with a massive number of antennas.\nThe main contributions of Part I of this paper are as follows. First, we note\nthat as a consequence of having a large pool of potentially active devices but\nlimited coherence time, the pilot sequences cannot all be orthogonal. However,\ndespite the non-orthogonality, this paper shows that in the asymptotic massive\nmultiple-input multiple-output (MIMO) regime, both the missed device detection\nand the false alarm probabilities for activity detection can always be made to\ngo to zero by utilizing compressed sensing techniques that exploit sparsity in\nthe user activity pattern. Part II of this paper further characterizes the\nachievable rates using the proposed scheme and quantifies the cost of using\nnon-orthogonal pilot sequences for channel estimation in achievable rates.\n