MIMO Design for Internet of Things: Joint Optimization of Spectral Efficiency and Error Probability in Finite Blocklength Regime

In this article, we consider a multiple-input-multiple-output (MIMO) system serving Internet of Things (IoT) devices. To satisfy stringent requirements on the latency of IoT communications, the IoT devices communicate in the finite blocklength regime, wherein the achievable spectral efficiency (SE) has a backoff factor and decoding error probability is nonnegligible. Aiming to jointly optimize the sum SE and the maximum error probability, we first express the achievable SE as a function of the channel coefficients, precoders, and error probabilities. Subsequently, we formulate a problem with respect to precoders and error probabilities. A straightforward approach for the formulated problem, however, is challenging as it not only has multiple objectives but also is nonconvex. To resolve these issues, we first transform the problem as a single-objective optimization by using a weighted sum approach. Based on the reformulation, we propose an algorithm, in which error probabilities and precoders are determined by alternating two phases. Via simulations, we demonstrate that the proposed method offers significant gains compared to baseline methods, in terms of the achievable SE and the maximum error probability. In particular, we show that the communication latency is greatly reduced by using the proposed method.