Affordable5G’s first year scientific publications output

Affordable5G’s consortium had the chance to publish a significant number of publications in international scientific conferences as well as highly-indexed journals, combining novel ideas and contributions from academic and industrial partners.


Affordable5G’s first year publications can be generally classified into three major categories:

  • Presentation of the architectural approach and innovations of Affordable5G
  • AI/ML algorithms for network optimization (mainly regarding the RAN part)
  • Performance evaluation of massive MIMO configurations in order to support the vision of ultra-dense networking.

In the context of the first category, our main goal was to present the novel approach of the architectural design of Affordable5G, that combines the following key aspects, crucial for a successful and cost-effective deployment of private 5G networks: dis-aggregated multi-vendor 5G RAN with open interfaces; full-fledged 5G Standalone core; data analytics and AI/ML-based network optimization; network slicing; RAN sharing and neutral hosting; resource and service orchestration over heterogeneous infrastructures; network monitoring and telemetry that spans all architectural layers and efficient 5G transport network with support for TSN. A similar approach combining all the aforementioned key enablers for 5G private networks did not occur in international literature, thus motivating us to the aforementioned design and implementation.

In the context of the second category, we have been motivated by the deployment of AI/ML algorithms in 5G and B5G networks, in an effort to proactively manage all network resources and provide high data rate services with improved bandwidth efficiency and minimum latency. Towards this direction, AI/ML techniques are of utmost importance for the deployment of an intelligent, cognitive, and flexible 5G system, primarily due to the memorization and accurate decision-making capabilities. To this end, we were mainly interested in the Deep Reinforcement Learning (DRL) approach, since it has the potential to high-dimensional decision making problems.

In the third category, the promising research field of massive MIMO technology is considered, deployed either in a centralized or in a distributed way, operating in the millimeter wave (mmWave) frequency band. The small size of mmWave antenna configurations provides the unique opportunity to have a multitude of flexible BS deployments for the support of mutable traffic and overall throughput improvement.

The entire list and links to the publications has been updated and it’s available to the public on the dedicated section of the project’s website: be our guest and catch up with the project’s work so far!