Francesco La Marca


Francesco La Marca is a Marie-Curie Early Stage Researcher at Nokia Bell Labs (Dublin, Ireland) and Pompeu Fabra University (Barcelona, Spain), where his Ph.D. research work focuses on the design of wireless networks for autonomous robots. Francesco received the Bachelor’s degree in Biomedical Engineering (2015) and the Master’s degree in Telecommunication Engineering (2019) from Politecnico of Milan University (Milan, Italy). He completed an internship with Samsung Electronics Italy in summer 2016 where he was part of the Technical Product Team supporting the development and the finalization of mobile products by testing all the functionalities and providing relevant Field Test Reports and feedback to R&D and HQ. In April 2018, he was one of the 10 Italian Master students selected by Huawei Technologies for participating in the Seeds for the Future Program for a two weeks study trip to China in Huawei’s global headquarters in Shenzhen, where he received a training on wireless and core network technologies. He was a visitor student with Nokia Bell Labs (Dublin, Ireland) from October 2018 to April 2019 where he conducted research on UAV cellular communications and wrote a quantitative master’s thesis on the topic.


Research updates – Period 1

Intelligent wireless network operation in unlicensed spectrum (ESR#7)

The main objective of this research activity is to design a hyper intelligent wireless network operating in the unlicensed spectrum for enabling high-capacity and/or low-latency communications in future enterprise and Industry 4.0 scenarios, like the one shown in Figure 1, in a cost-effective manner.

The next generation Wi-Fi—based on IEEE 802.11be Extremely High Throughput (EHT)— has proposed multiple new features, like 320 MHz bandwidth, 16 spatial streams, Multi-band/multi-channel aggregation and operation, Multi-Access Point (AP) Coordination to successfully support reliable low-latency communications (RLLC).

Nokia has been actively participating in the standardization process of IEEE 802.11be focusing on multi AP coordination in combination with the null-steering technique.

In normal operation the two Access Points (AP) take independent scheduling decisions at each Transmission Opportunity (TxOP). To optimally place the nulls towards the STAs some level of coordination is required between the APs. The proposed solution that has been developed is to deploy and train a central controller (CC), through Deep Reinforcement Learning (DRL), able to take real-time optimal decisions on whom to null in each TxOP.

The proposed framework outperforms the standard Wi-Fi operation in terms of number of users successfully served by the network by a factor of 2. This directly translates into a more reliable network where the central controller is able to intelligently manage Multi-Access Point (AP) null-steering technique and at the same time maximize the trade-off between beamforming gain and null placement.





David Lopez-Perez,

Contact Painless ITN

12 + 3 =

We’ll get back to you shortly.