{"id":17479,"date":"2026-03-30T17:22:48","date_gmt":"2026-03-30T15:22:48","guid":{"rendered":"https:\/\/www.uni.lu\/research-en\/?post_type=events&p=17479"},"modified":"2026-03-30T17:23:18","modified_gmt":"2026-03-30T15:23:18","slug":"doctoral-defence-ashish-kumar-meshram","status":"publish","type":"events","link":"https:\/\/www.uni.lu\/research-en\/events\/doctoral-defence-ashish-kumar-meshram\/","title":{"rendered":"Doctoral Defence: Ashish Kumar MESHRAM"},"content":{"rendered":"\n
<\/div>\n\n\n

The Doctoral School in Science and Engineering<\/a> is happy to invite you to Ashish Kumar MESHRAM \u2019s defence entitled<\/p>\n\n\n\n

\nSynchronization and Precoding Techniques in 5G Non-Terrestrial-Networks<\/h2>\n\n\n\n

Supervisor<\/strong>: Dr. Jorge QUEROL<\/a><\/p>\n\n\n\n

Non-terrestrial networks (NTN) are central to the Fifth generation (5G) Advanced and the International Mobile Telecommunications (IMT)-2030 vision, enabling direct-to-device (D2D) broadband beyond terrestrial coverage. Compared to terrestrial systems, satellite access faces harsher physical-layer conditions. High mobility and long propagation delay introduce large Doppler shifts, rapid channel time variation, and stringent constraints on channel state information (CSI) acquisition. These effects degrade both the user equipment (UE) initial access reliability and the effectiveness of multi-user (MU) downlink transmission when CSI feedback is delayed or outdated. This thesis develops a unified set of Third-Generation Partnership Project (3GPP) 5G New Radio (NR) standard-compliance techniques that require no modifications to NR procedures or reference signals. The proposed techniques completely rely on signal processing and system-aware parameterization to enable (i) robust, reduced-complexity initial synchronization in multibeam regenerative NTN and (ii) practical MU downlink precoding under CSI limitations. The primary focus is on low Earth orbit (LEO) satellites, where impairments are most severe; the proposed frameworks can be adapted to other orbital regimes through parameterization.<\/p>\n\n\n\n

The first part addresses initial access under Doppler and uncertainty. The core concept is beam-centric Doppler pre-compensation, where the satellite on-board gNB uses its instantaneous state to pre-compensate Doppler relative to a reference point (typically the beam center), so the UE observes only a bounded residual carrier frequency offset (CFO). Building on this work, a reduced-complexity initial synchronization strategy is proposed that integrates pre-compensation with practical CFO estimation and detection. The processing complexity is further lowered by shrinking the search space via aggregated correlations over NR primary synchronization signals (PSS), and adaptive thresholds are used to control false-alarm probability at low SNR under imperfect Doppler knowledge. Thus, yielding practical guidance for the synchronization design of multibeam NTN.<\/p>\n\n\n\n

The second part of the thesis targets MU transmission, where precoding gains are limited by stale downlink CSI. To mitigate feedback delay and time variation, again without altering NR signaling, the thesis investigates an approach that uses uplink sounding reference signals (SRS) at the regenerative gNB to infer and extrapolate downlink CSI across frequency and time, enabling feedback-free MU-MIMO precoding. An inter-frequency extrapolation method suited to FDD-like operation is formulated and evaluated in terms of normalized mean-squared error and sum spectral efficiency relative to perfect-CSI and outdated-CSI baselines. Furthermore, a low-complexity MMSE-like estimator (LCMMSE) is developed to improve robustness without matrix inversion. Besides, an NR-consistent precoding granularity is studied through precoding resource block group (PRGB) bundling and subband-level operation, clarifying trade-offs among accuracy, overhead, and achievable MU spectral efficiency.<\/p>\n\n\n\n

In its final phase, the thesis investigates ray-tracing-based channel modeling for D2D NTN to complement simplified fading assumptions with environment-specific blockage and multipath structure. Ray-traced channels are integrated with 5G NR link-level simulation to stress-test synchronization and SRS-based precoding and to quantify sensitivity to deployment environments.<\/p>\n\n\n\n