English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

The reconfigurable intelligent surface (RIS) is a promising technology thatenables wireless communication systems to achieve improved performance byintelligently manipulating wireless channels. In this paper, we consider thesum-rate maximization problem in a downlink multi-usermulti-input-single-output (MISO) channel via space-division multiple access(SDMA). Two major challenges of this problem are the high dimensionality due tothe large number of RIS elements and the difficulty to obtain the full channelstate information (CSI), which is assumed known in many algorithms proposed inthe literature. Instead, we propose a hybrid machine learning approach usingthe weighted minimum mean squared error (WMMSE) precoder at the base station(BS) and a dedicated neural network (NN) architecture, RISnet, for RISconfiguration. The RISnet has a good scalability to optimize 1296 RIS elementsand requires partial CSI of only 16 RIS elements as input. We show it achievesa high performance with low requirement for channel estimation for geometricchannel models obtained with ray-tracing simulation. The unsupervised learninglets the RISnet find an optimized RIS configuration by itself. Numericalresults show that a trained model configures the RIS with low computationaleffort, considerably outperforms the baselines, and can work with discretephase shifts.

RISnet: A Scalable Approach for Reconfigurable Intelligent Surface Optimization with Partial CSI