Marcin Hoffmann

Mr. Marcin Hoffmann

orcid

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
28 November 2023 in IEEE Journal on Selected Areas in Communications
Verified
Journal Article
Open access
MHMarcin Hoffmann
Marcin Hoffmann
SJSalim Janji
Salim Janji
ASAdam Samorzewski
Adam Samorzewski
ŁKŁukasz Kułacz
Łukasz Kułacz
CACezary Adamczyk
Cezary Adamczyk
HBHanna Bogucka
Hanna Bogucka
The concept of open radio access networks (RAN) creates numerous opportunities for developing new technology and economy branches. At the same time, a flexible and modular approach in the disaggregated RAN entails the need for careful design of the overall RAN architecture and the implementation and deployment process of new applications. It is assumed that dedicated and specialized software companies may deliver the latter. A joint effort must be guaranteed among different sectors (industry, academia, and standardization bodies) to make the whole process efficient, safe, and reliable. Here, one of the critical driving forces origins from the open-source community that often stimulates the development of a specific technology. In this paper, we address the challenges that have to be faced by third-party application developers in the context of Open RAN. Based on many implemented applications (called xApps or rApps), we compare various available solutions. We pose the most critical issues that must be tackled in the near future to stimulate the progress in open RAN development further. In particular, we compare available open platforms for xApp development and testing. We present the details of implementing four selected applications describing the problems encountered. The paper is split into two logical parts - first, we identify the key ambiguities related to the development of new xApps, which address more complicated use cases like beam management. In the second part, we present the challenges associated with detailed software implementation in existing open platforms. In the first case, we show that dedicated beam mobility management xApp can reduce beam switches and keep beam failures low. However, it requires access to detailed localization information. Similarly, the signaling storm detection xApp provides expected performance under the assumption that there is access to detailed information on, e.g., time advance resolution parameter. We conclude here that several aspects still need to be well-defined to allow smooth software implementation; these include the rules for data reporting in time, parameters available in service models, and localization features. Concerning the second logical part, related to low-level implementation, we compare the numerical results of the traffic steering and quality-of-service-based resource allocation xApps and draw conclusions related to implementation and testing. In particular, we point out problems associated with the simulator, the software, and conflicts inside. Finally, we identify the key challenges which should be treated as incentives for joint academia-industry cooperation in the field of Open RAN. Thus, the paper presents the lesson learned during the first years of xApp development.

1 March 2021 in Computer Communications
Verified
Journal Article
Closed access
MHMarcin Hoffmann
Marcin Hoffmann
PKPawel Kryszkiewicz
Pawel Kryszkiewicz
AKAdrian Kliks
Adrian Kliks

16 September 2021 in IEEE Access
Verified
Journal Article
Open access
MHMarcin Hoffmann
Marcin Hoffmann
PKPawel Kryszkiewicz
Pawel Kryszkiewicz
To provide users with high throughputs, the fifth generation (5G) and beyond networks are expected to utilize the Massive Multiple-Input Multiple-Output technology (MMIMO), i.e., large antenna arrays. However, additional antennas require the installation of dedicated hardware. As a result the power consumption of a 5G MMIMO network grows. This implies, e.g., higher operator costs. From this angle, the improvement of Energy Efficiency (EE) is identified as one of the key challenges for the 5G and beyond networks. EE can be improved through intelligent antenna switching, i.e., disabling some of the antennas installed at a 5G MMIMO Base Station (BS) when there are few User Equipments (UEs) within the cell area. To improve EE in this scenario we propose to utilize a sub-class of Machine Learning techniques named Reinforcement Learning (RL). Because 5G and beyond networks are expected to come with accurate UE localization, the proposed RL algorithm is based on UE location information stored in an intelligent database named a Radio Environment Map (REM). Two approaches are proposed: first EE is maximized independently for every set of UEs’ positions. After that the process of learning is accelerated by exploiting similarities between data in REM, i.e., REM-Empowered Action Selection Algorithm (REASA) is proposed. The proposed RL algorithms are evaluated with the use of a realistic simulator of the 5G MMIMO network utilizing an accurate 3D-Ray-Tracing radio channel model. The utilization of RL provides about 18.5% EE gains over algorithms based on standard optimization methods. Moreover, when REASA is used the process of learning can be accomplished approximately two times faster.

26 August 2024 in IEEE Access
Verified
Journal Article
Open access
MHMarcin Hoffmann
Marcin Hoffmann
MDMarcin Dryjanski
Marcin Dryjanski
Recently, energy efficiency (EE) has been pointed out as one of the key requirements within mobile networks. The development of intelligent algorithms providing Radio Access Networks (RAN) with EE features is possible when having access to the Key Performance Indicators (KPIs), and proper control actions, e.g., cell on/off switching, or Radio Frequency (RF) channel reconfiguration. These features together with a Machine Learning (ML) framework are available in the O-RAN architecture. This paper provides an overview of the EE framework according to the use cases specified by the O-RAN ALLIANCE. It is followed by the implementation of the Energy Saving rApp (ES-rApp). The rApp utilizes Deep Q-Learning (DQL) to increase EE through intelligent RF channel reconfiguration. Simulation results show up to a 24.8% EE gain over the static RF Channel Configuration (RCC).

31 January 2020 in EURASIP Journal on Wireless Communications and Networking
Verified
Journal Article
Open access
PDF file
MHMarcin Hoffmann
Marcin Hoffmann
PKPawel Kryszkiewicz
Pawel Kryszkiewicz
GKGeorgios P. Koudouridis
Georgios P. Koudouridis
In 5G networks information about localization of a user equipment (UE) can be used not only for emergency calls or location-based services, but also for the network optimization applications, e.g., network management or dynamic spectrum access by using Radio Environment Maps (REM). However, some of these applications require much better localization accuracy than currently available in 4G systems. One promising localization method is Global Navigation Satellite System (GNSS)-based Real-Time Kinematics (RTK). While the signal received from satellites is the same as in traditional GNSS, a new reception method utilizing real-time data from a nearby reference station (e.g., 5G base station) results in cm-level positioning accuracy. The aim of this paper is to obtain a model of the RTK localization error for smartphone-grade GNSS antenna under open-sky conditions, that can be used in 5G network simulators. First, a tutorial-style overview of RTK positioning, and satellite orbits prediction is provided. Next, an RTK localization simulator is implemented utilizing GNSS satellites constellations. Results are investigated statistically to provide a simple, yet accurate RTK localization error framework, which is based on two Gauss-Markov process generators parametrized by visible satellites geometry, UE motion, and UE-satellite distance error variance.

28 May 2023
Verified
Conference Paper
Closed access
MHMarcin Hoffmann
Marcin Hoffmann
PKPawel Kryszkiewicz
Pawel Kryszkiewicz
The Massive Multiple-Input Multiple-Output (M-MIMO) is considered as one of the key technologies in 5G, and future 6G networks. From the perspective of, e.g., channel estimation, especially for high-speed users it is easier to implement an M-MIMO network exploiting a static set of beams, i.e., Grid of Beams (GoB). While considering GoB it is important to properly assign users to the beams, i.e., to perform Beam Management (BM). BM can be enhanced by taking into account historical knowledge about the radio environment, e.g., to avoid radio link failures. The aim of this paper is to propose such a BM algorithm, that utilizes location-dependent data stored in a Radio Environment Map (REM). It utilizes received power maps, and user mobility patterns to optimize the BM process in terms of Reinforcement Learning (RL) by using the Policy Iteration method under different goal functions, e.g., maximization of received power or minimization of beam reselections while avoiding radio link failures. The proposed solution is compliant with the Open Radio Access Network (O-RAN) architecture, enabling its practical implementation. Simulation studies have shown that the proposed BM algorithm can significantly reduce the number of beam reselections or radio link failures compared to the baseline algorithm.