Dheeraj Raja Kumar

Centre Tecnològic de Telecomunicacions de Catalunya (CTTC), Spain - Researcher

Dheeraj’s doctoral dissertation (published in September 2025) titled “AI-Driven Multi-Antenna Designs for Next Generations of Wireless Communication Systems” focused on AI/ML techniques, specifically promoting Rate Splitting Multiple Access (RSMA), one of the candidate Multiple Access Techniques (MAT) considered in the ETSI ISG MAT. The knowledge and skills developed during his PhD studies enabled Dheeraj to actively participate in ETSI ISG MAT meetings and make significant contributions.

The scope of the ETSI ISG on Multiple Access Techniques (MAT) is on downlink multiple access techniques for the physical layer of the 3GPP radio interface that improve spectrum efficiency in the presence of inter-user interference. A key capability of the ITU-R framework IMT-2030 is improved spectrum efficiency, i.e. the average data throughput per unit of spectrum and per cell. To achieve high data throughput within limited bandwidth, smart use of multiple frequency bands and the application of advanced technologies to boost spectrum efficiency are essential. The novelty and originality of this research direction were emphasised at the 3GPP 6G Workshop held in March 2025, where spectral efficiency was identified as a critical requirement to support a massive number of connected devices under stringent spectrum constraints. The first 6G RAN study item approved at RAN#108 Plenary Meeting defined a clear target for spectral efficiency enhancement, highlighting the need to study new MATs as enablers.
The candidate downlink Multiple Access Techniques studied in ISG MAT were: power-domain NOMA (Non-Orthogonal Multiple Access), RSMA (Rate-Splitting Multiple Access) and Cache-Aided MU-MIMO (CA MU-MIMO).

Dheeraj’s doctoral dissertation (published in September 2025) titled “AI-Driven Multi-Antenna Designs for Next Generations of Wireless Communication Systems” specifically focused on symbol detection and precoding techniques for RSMA. The thesis showed that combining domain knowledge with machine learning delivered architectures that are robust, interpretable, and well-suited to the challenges of future 6G networks. The knowledge acquired and publications produced (2 journals, 5 conference papers, 6 technical papers) during his PhD studies equipped Dheeraj with a strong background to make significant contributions to ETSI ISG MAT’s work. Dheeraj actively participated in the ISG MAT meetings, independently implemented the 3GPP-specified MAT and the candidate MATs, and evaluated their performance in the considered scenarios. The independent implementation and cross-validation of results by several participating members/organisations of the ISG MAT showed how the candidate MATs (which included RSMA, NOMA) can provide spectral efficiency gains against MAT specified in 3GPP (OMA, MU-MIMO) in certain channel conditions.

RSMA provides gains when the channels between scheduled UEs are highly correlated (e.g. UEs with small angular separation) or when there is a power imbalance between scheduled UEs. Thus, the candidate MATs presented and studied under ISG MAT are potential enablers of enhanced spectral efficiency in 6G. The main output of ISG MAT is the publication of Group Report ETSI GR MAT 001 v1.1.1 (https://www.etsi.org/deliver/etsi_gr/MAT/001_099/001/01.01.01_60/gr_MAT001v010101p.pdf) in January 2026.

Dheeraj plans to actively participate and contribute to the ISG MAT work towards a follow-up group report in 2026.