Performance Analysis of Intelligent Reflecting Surface Assisted Opportunistic Communications

Abstract

Intelligent reflecting surfaces (IRSs) are a promising technology for enhancing coverage and spectral efficiency, especially in the millimeter wave (mmWave) bands. Existing approaches to leverage the benefits of IRS involve a resourceintensive channel estimation step followed by a computationally expensive algorithm to optimize the reflection coefficients at the IRS. In this work, we present and analyze several alternative schemes, where the phase configuration of the IRS is randomized and multi-user diversity is exploited to opportunistically select the best user at each point in time for data transmission. We show that the throughput of an IRS assisted opportunistic system asymptotically converges to the optimal beamformingbased throughput under fair allocation of resources, as the number of users gets large. Further, for all the proposed schemes, we derive the scaling law of the throughput in terms of the number of users and IRS elements, as the number of users gets large. We also introduce schemes that enhance the rate of convergence of the opportunistic rate to the beamforming rate as the number of users is increased. Following this, we extend the setup to wideband channels via an orthogonal frequency division multiplexing (OFDM) system and discuss two opportunistic schemes in an IRS assisted setting that elucidate the superior performance that IRS aided systems can offer over conventional systems at very low implementation cost and complexity.

Cascaded Channel Estimation for Distributed IRS Aided mmWave Massive MIMO Systems

“…(21) 4 We consider larger N only for the sake of analytical tractability. We showed in [5] that this approximation works well even with N = 5.…”

“…The path 3 A real-valued random variable X is stochastically larger than, or stochastically dominates, the random variable Y , written X >st Y , if Pr(X > a) ≥ Pr(Y > a), for all a. losses are computed as β = 1/d α where d is the distance and α is the path loss exponent. We use α = 1.5, 2 and 3 in the BS X/Y-IRS, IRS-UE and BS X/Y-UE (direct) links, respectively, similar to [5]. The fading channels are randomly generated as per Sec.…”

“…In order to study the impact on the OOB performance, we consider the scheduling of UEs in a round-robin (RR) fashion at both BS-X and BS-Y. We note that the performance under opportunistic scheduling at either or both BSs can also be derived along similar lines, e.g., following the approach in [5].…”

“…In order to obtain the professed benefits of an IRS, every IRS element needs to be configured to introduce a phase shift in the signal that is optimal to the scheduled user (UE) in terms of a metric of interest, e.g., the signal-to-noise ratio (SNR). This, however, requires the knowledge of the channel state information (CSI) of all the links from the base station (BS) to the user through every IRS element [5].…”

“…where the factors 1/K and 1/Q account for the fact that every UE gets scheduled only for a fraction of total number of UEs in the system under RR scheduling. Closed-form expressions for the ergodic sum-rate is difficult to obtain due to the complicated nature of the distribution of the SNR terms in (5), and (6). Instead, we provide an approximate characterization of the achievable SE below by bounding the data rates using Jensen's inequality.…”