Abstract
In this article, we study closed-loop digital predistortion (DPD) systems and associated learning algorithms. Specifically, we propose various low-complexity approaches to estimate and manipulate the inverse of the input data covariance matrix (CM) and combine them with the so-called self-orthogonalized (SO) learning rule. The inherent simplicity of the SO algorithm, combined with the proposed solutions, allows for remarkably reduced complexity in the DPD system while maintaining similar linearization performance compared to other state-of-the-art methods. This is demonstrated with thorough over-the-air (OTA) mmW measurement results at 28 GHz, incorporating a state-of-the-art 64-element active antenna array, and very wide channel bandwidths up to 800 MHz. In addition, complexity analyses are carried out, which together with the measured linearization performance demonstrates favorable performance–complexity tradeoffs in linearizing mmW active array transmitters through the proposed solutions. The techniques can find application in systems where the power amplifier (PA) nonlinearities are time-varying and thus frequent or even constant updating of the DPD is required, good examples being mmW adaptive antenna arrays as well as terminal transmitters in 5G and beyond networks.
Highlights
C ONTEMPORARY radio communication systems, such as the recently introduced 5G new radio (NR) mobile networks, build on multicarrier modulation—most notably orthogonal frequency-division multiplexing (OFDM)
For reference and comparison purposes, we address the complexity of a classical memory polynomial (MP) model with basis functions (BFs) prewhitening or orthogonalization while considering block-least mean squares (LMS) as the digital predistortion (DPD) coefficient update rule
The cost of updating the DPD coefficients in the block-LMS learning rule [50] is given by 4C K + 2C, which leads to 1 600 040 real multiplications—a number that is of the same complexity order as the SO solution when combined with any of the proposed ICM estimation methods
Summary
C ONTEMPORARY radio communication systems, such as the recently introduced 5G new radio (NR) mobile networks, build on multicarrier modulation—most notably orthogonal frequency-division multiplexing (OFDM). Fast DPD adaptation is required such that the nonlinear distortions can be suppressed, while the beam is steered This issue, along with the extremely high processing rates and channel bandwidths at mmW frequencies, calls for reduced complexity DPD approaches and associated parameter learning algorithms. Another relatively new DPD use case is mobile device linearization. The LUTs aimed at substituting the high-order polynomials in the DPD model, relaxing the overall complexity Their modeling capabilities are somewhat limited, depending on the size of the LUTs. To enhance the performance, the LUT sizes can be increased to better describe the nonlinear characteristics, but this leads to higher memory requirements and slower convergence speeds. The expected value, absolute value, floor, ceil, factorial, Hadamard product, and Kronecker product operators are written as E{·}, | · |, ∗ ·, ∗ ·, !, ◦, and ⊗, respectively
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
