Brave-beyond5g

CentraleSupélec and the SCEE Research Group in Beyond-5G

An academic laboratory with an industrial tempo

At CentraleSupélec, the SCEE research group approaches Beyond-5G as a system problem rather than a single-layer puzzle. The team works on physical-layer modelling and embedded electronics side by side, because a waveform that cannot be computed in real time on a realistic chip is, for their purposes, not yet a waveform. This dual identity shapes every project the group accepts.

The laboratory sits inside a French engineering school with strong ties to industry, so doctoral work is expected to survive contact with hardware constraints. Students move between simulation environments and measurement benches, and they learn early that a model is only as useful as the assumptions it makes explicit.

Physical-layer modelling as an engineering discipline

On the PHY side, SCEE researchers study waveforms, channel models, and transceiver architectures for future radio generations. Rather than treating modelling as a purely mathematical exercise, they ask how each block maps onto silicon and how imperfect components alter performance. This perspective pushes the group toward hybrid approaches that combine analytical models with measured data.

Typical activities in this area include:

  • Waveform and modulation studies for sub-THz and massive-MIMO scenarios.
  • Channel and impairment modelling validated against testbed measurements.
  • Algorithm-hardware co-design for baseband and RF front-ends.
  • Energy and complexity budgets treated as first-class design metrics.

Because the group publishes and prototypes, results circulate between theory papers and demonstrators. A receiver algorithm may begin as a MATLAB script and end as an FPGA bitstream, with the intermediate assumptions documented along the way.

Embedded electronics where theory meets silicon

The embedded electronics side of SCEE focuses on implementing and validating signal processing close to the antenna. That means custom boards, data converters, clocking, and the unglamorous work of power integrity and thermal behaviour. The group argues that Beyond-5G performance will be limited as much by packaging and synchronisation as by abstract spectral efficiency.

Researchers therefore build platforms that let them inject real signals into real chains. A blockquote often heard in the lab captures the philosophy: "If it does not run on the board, it is not a result yet." That sentence is less a slogan than a workflow rule, and it explains why measurement campaigns occupy a large share of the group's calendar.

From testbench to Beyond-5G scenarios

The combination of modelling and electronics gives SCEE a distinctive role in collaborative projects. The group can propose a PHY concept, estimate its cost, implement a reduced version, and report where the model diverged from reality. That feedback loop is valuable for standards discussions, where optimistic assumptions can hide expensive implementation problems.

Current interests span sub-THz links, reconfigurable intelligent surfaces, and energy-efficient transceivers. In each case the same questions return: which part of the chain dominates the error budget, and which part dominates the power budget?

Training the next generation of radio engineers

CentraleSupélec provides a pipeline of engineers and doctoral candidates who are comfortable in both domains. SCEE deliberately mixes profiles so that a digital designer learns channel modelling and a theorist learns to read a schematic. The result is a laboratory culture that values reproducible measurements as much as elegant derivations.

For Beyond-5G, that culture matters. The next generation of networks will not be built by simulation alone, nor by hardware alone, but by teams that can move between the two without losing rigour. SCEE is one of the places where that movement is practised daily.

Latest articles