Cisco University Research Program fund (Project CG No. 890107, Towards Deterministic Channel Access in High-Density WLANs), acorporate advised fund of Silicon Valley Community Foun-dation.

The goal of this project was to study the performance of next-generation WLANs based on the upcoming IEEE 802.11ax amendment in large, dense and dynamic environments, with emphasis in understanding the dependencies between multiuser transmissions, transmit power and sensitivity adaptation, and the use of channel bonding. Our main goal is to provide insights on the IEEE 802.11ax performance limits, as well as feasible network configurations to achieve them in practice.

We also focus on the mechanisms and functions included in the IEEE 802.11ax amendment to achieve a deterministic operation, in which stations’ transmissions are scheduled by the AP taking into account the traffic characteristics, delay constraints, and energy consumption requirements.

Finally, to deal with the changing conditions in the considered scenarios, machine learning techniques are used to develop a self-configuration module at each WLAN able to gather data from the environment, detect and predict changes in the channel, traffic loads and number of users, in order to update its configuration accordingly.

In detail, the five specific objectives of this project were:

    1. Develop a simulation platform to study the performance of IEEE 802.11ax WLANs in dynamic and dense scenarios.

    2. Characterize the IEEE 802.11ax performance in large, high-density WLAN and dynamic scenarios.
    3. Develop a solution to achieve contention-free / deterministic operation in future WLANs based on IEEE 802.11ax new capabilities.
    4. Propose self-configuration strategies using machine learning techniques able to dynamically adapt the WLAN operation parameters based on gathering contextual information from the environment, as well as using the information available at each network.
    5. Evaluate experimentally several key future scenarios, including multi-WLAN technology  aggregation (11ac/ax and 11ad) and the use of multipath transport protocols.


    • IEEE 802.11ax: Overview of the development of the 11ax amendment. Performance evaluation (system level) of MU-MIMO and OFDMA features.
    • Improving Spectrum Efficiency: Performance evaluation of Dynamic Channel Bonding (DCB) and Preamble Puncturing (PP) in High Density (HD) WLANs under different traffic load conditions. Characterization of the interrelations among WLANs in the spatial domain. Results reveal that, while always selecting the widest available channel in DCB normally maximizes the individual long-term throughput, it often generates unfair situations where other WLANs starve. Moreover, new channel selection solutions are required along with channel bonding to account for the potential bonds. Finally, PP clearly outperforms DCB in scenarios with sparse spectrum occupancy.
    • Improving Spatial Reuse (SR): Performance evaluation of the potential gains of adjusting both the transmission power and sensitivity levels dynamically, showing how a proper set-up of those parameters is able to improve WLANs’ operation in dense scenarios. We have also studied the performance of IEEE 802.11ax Spatial Reuse mechanisms, providing a detailed tutorial, and, to the best of our knowledge, the first detailed performance evaluation of the SR mechanisms included in the 11ax amendment.
    • Scheduled / Deterministic Access: Evaluation the Target Wake Time (TWT) mechanism included in the IEEE 802.11ax amendment as it provides an extremely simple but effective mechanism to schedule transmissions in time. Moreover, we describe how using TWT may also contribute to taking full advantage of other novel mechanisms, such as MU transmissions, multi-AP cooperation, spatial reuse and coexistence in HD WLAN scenarios.  The use of TWT has been considered in different scenarios, such as for video streaming and IoT.
    • Machine Learning: The use of machine learning have been used for network optimization through all activities in this project. In detail, Reinforcement Learning has been used to find (learn) the optimal configurations (tx power, CCA, channel width) in adversial scenarios. Similarly, Supervised Learning has been used for predicting future user’s satisfaction given some initial collected parameters.

We have also developed the Komondor 11ax simulator which allows us to simulate efficiently large WLANs deployments. Other contributions include: new empirical 5 GHz indoor path-loss model, performance evaluation of MP-TCP for HD WLANs, and AP-selection in dense deployments


We have developed an IEEE 802.11ax simulator able to reproduce the operation of dense WLAN deployments. It also provides support for decentralized, distributed and centralized ‘agents’ able to apply machine learning policies. Last stable version of the IEEE 802.11ax simulator can be found here (, including a tutorial version (inside the Documentation folder).

We have developed a tool for the mathematical analysis of next-generation WLANs. This tool is mainly used to provide insight on the interactions of dense WLAN scenarios.  Last available version of the SFCTMN framework can be  found here (

Submitted Papers

We are currently preparing several papers with the last results from the project. These papers include topics such as AP-selection in high-density scenarios, performance evaluation of Dynamic Channel Bonding and Preamble Puncturing using real spectrum occupancy traces, and the definition of a management architecture that supports machine learning functionalities.

Konstantinos Dovelos, Boris Bellalta; "Optimal Resource Allocation in IEEE 802.11ax Uplink OFDMA with Scheduled Access"; arXiv:1811.00957. 2019. Submitted to IEEE Transactions on Communications.

Francesc Wilhelmi, Sergio Barrachina Muñoz, Cristina Cano, Ioannis Selinis, Boris Bellalta; "Spatial Reuse in IEEE 802.11ax WLANs". arXiv:1907.04141. 2019. Submitted to IEEE Communications Surveys and Tutorials.


Journal Papers

Álvaro López Raventós, Boris Bellalta; "Concurrent Decentralized Channel Allocation and Access Point Selection using Multi-Armed Bandits in multi BSS WLANs", Computer Networks, 2020.

Barrachina-Muñoz, Sergio, Francesc Wilhelmi, and Boris Bellalta. "Online Primary Channel Selection for Dynamic Channel Bonding in High-Density WLANs."  IEEE Wireless Communications Letters, 2020.

Nurchis, Maddalena, and Boris Bellalta. "Target Wake Time: Scheduled Access in IEEE 802.11 axWLANs." IEEE Wireless Communications (2019).

Barrachina-Muñoz, Sergio, Francesc Wilhelmi, and Boris Bellalta. "To overlap or not to overlap: Enabling channel bonding in high-density WLANs." Computer Networks 152 (2019): 40-53.

Bellalta, Boris, and Katarzyna Kosek-Szott. "AP-initiated multi-user transmissions in IEEE 802.11 ax WLANs." Ad Hoc Networks 85 (2019): 145-159.

Wilhelmi, Francesc, Sergio Barrachina-Muñoz, Boris Bellalta, Cristina Cano, Anders Jonsson, and Gergely Neu. "Potential and pitfalls of multi-armed bandits for decentralized spatial reuse in WLANs." Journal of Network and Computer Applications 127 (2019): 26-42.

Barrachina-Munoz, Sergio, Francesc Wilhelmi, and Boris Bellalta. "Dynamic Channel Bonding in Spatially Distributed High-Density WLANs." IEEE Transactions on Mobile Computing. 2019.
Wilhelmi, Francesc, Cristina Cano, Gergely Neu, Boris Bellalta, Anders Jonsson, and Sergio Barrachina-Muñoz. "Collaborative spatial reuse in wireless networks via selfish multi-armed bandits." Ad Hoc Networks 88 (2019): 129-141.
Conference Papers

S. Barrachina, B. Bellalta, E. Knightly ; "Wi-Fi All-Channel Analyzer". ACM WinTech 2020 (a Mobicom 2020 workshop).

Wilhelmi, Francesc, Sergio Barrachina-Muñoz, and Boris Bellalta. "On the Performance of the Spatial Reuse Operation in IEEE 802.11ax WLANs." IEEE Conference on Standards for Communications and Networking (CSCN) 2019.

López-Raventós, A., Wilhelmi, F., Barrachina-Muñoz, S., & Bellalta, B. "Machine learning and software defined networks for high-density WLANs." IEEE WiMob 2019 workshops.

Carrascosa, Marc, and Boris Bellalta. "Decentralized AP selection using Multi-Armed Bandits: Opportunistic E-Greedy with Stickiness." IEEE Symposium on Computers and Communications (ISCC) (2019).

Barrachina-Muñoz, Sergio, Francesc Wilhelmi, Ioannis Selinis, and Boris Bellalta. "Komondor: a Wireless Network Simulator for Next-Generation High-Density WLANs." In 2019 Wireless Days (WD), pp. 1-8. IEEE, 2019.

Adame, Toni, Marc Carrascosa, and Boris Bellalta. "The TMB path loss model for 5 GHz indoor WiFi scenarios: On the empirical relationship between RSSI, MCS, and spatial streams." In 2019 Wireless Days (WD), pp. 1-8. IEEE, 2019.

Dovelos, Konstantinos, and Boris Bellalta. "Breaking the Interference Barrier in Dense Wireless Networks with Interference Alignment." 2018 IEEE International Conference on Communications (ICC). IEEE, 2018.

Cañizares, Guillem, and Boris Bellalta. "Improving User's Experience through Simultaneous Multi-WLAN Connections." EWSN 2018.

Invited Talks

Invited talk in the IEEE 5G Summit, Thessaloniki, Greece: "WiFifor IoT: Deterministic Access in IEEE 802.11ax WLANs", by Boris Bellalta (October 2018).

Invited talk in the Workshop on Next-Generation Wireless with NS-3, Florence, Italy: "Towards the implementation of 11ax features in Komondor", by Francesc Wilhelmi and Sergio Barrachina (June 2019)

Two talks in the ITU-T Focus Group on Machine Learning for Future Networks including 5G (ML5G): "Decentralized learning implications in the performance of dense WLANs", by Francesc Wilhelmi [ML5G-I-011] (2018); "A proposal to reuse Komondor, an IEEE 802.11ax-oriented simulator for future wireless networks in FG ML5G", by Francesc Wilhelmi [ML5G-I-125] (2019).

PhD Thesis

Part of the PhD thesis of Sergio Barrachina (2020), Francesc Wilhelmi (2020), Konstantinos Dovelos (2021) and Alvaro Lopez (2022) have been developed under this project. In brackets, the expected defense date.

Master Thesis

Three master thesis have been carried out in the framework of this project:

    • Marc Carrascosa: Predicting user satisfaction to optimize AP selection in WLANs using Random Forests. 2019. Master in Intelligent Systems.
    • Abdullah Zeidan: Experimental Analysis of MU-MIMO performance in IEEE802.11ac Wave2. 2019. Master in Wireless Communications.
    • Guillem Cañizares: 2019. Enhancing Wireless Communications Performance using Multiple Interfaces in IEEE 802.11n/ac WiFi StandardMaster in Wireless Communications.

Bachelor Thesis

One bachelor thesis have been carried out in the framework of this project:

    • Marc Carrascosa: Improving user association in high-density WLANs using machine learning. 2018. Telecomunications degree, UPF.