@inproceedings{Kappelt2025:MachineLearning,

title = {Machine Learning for In-Network Spatial Localization within Wireless Mesh Networks},

author = {Peter {Kappelt} and Christian Leonard {Vielhaus} and Justus {Rischke} and Marek {Sobe} and Frank H. P. {Fitzek}},

year  = {2025},

date = {2025-06-08},

urldate = {2025-06-01},

booktitle = {IEEE International Conference on Communications (ICC)},

pages = {5.97},

address = {Montreal, Canada},

abstract = {Wireless localization is a crucial technology for determining the spatial 

position of nodes within a wireless network, with applications ranging from 

indoor navigation to asset tracking and robotic movement. However, 

traditional localization methods often rely on specialized hardware, while 

other localization methods suitable for commodity hardware tend to have 

limited accuracy. To improve localization accuracy, we propose models that 

integrate machine learning (ML) techniques for distance and position 

estimation. We observe improved localization performance by incorporating 

not only signal strength indicators from the physical layer, but also 

features specific to the mesh link layer of wireless mesh networks as model 

inputs. Furthermore, a proposed end-to-end machine learning localization 

model demonstrates promising performance by leveraging diverse network 

features for the direct inference of node position estimates. 

We evaluate the models using experimental data from a measurement campaign 

we conducted in an office environment hosting a wireless mesh testbed. Our 

method achieves an absolute position error of less than two meters. Using 

both information specific to mesh networks and novel machine learning 

methods yields improved performance, making our approach a promising 

solution for industrial applications.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{Zimmermann2501:COSMIC,

title = {COSMIC - Coding for Optimized Sending in Multimedia InterPlanetary Communication},

author = {Sandra {Zimmermann} and Paul {Schwenteck} and Christian Leonard {Vielhaus} and Juan A. {Cabrera} and Frank H. P. {Fitzek}},

year  = {2025},

date = {2025-01-14},

urldate = {2025-01-01},

booktitle = {International Conference on Information Networking (ICOIN)},

pages = {6},

address = {Chiang Mai, Thailand},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{Latzko24:Offloading,

title = {Accelerator-Aware Computation Offloading under Timing Constraints},

author = {Vincent {Latzko} and Christian Leonard {Vielhaus} and Mahshid {Mehrabi} and Frank H. P. {Fitzek}},

doi = {10.1109/ICNC59896.2024.10556064},

year  = {2024},

date = {2024-02-19},

urldate = {2024-02-19},

booktitle = {International Conference on Computing, Networking and Communications 

(ICNC)},

pages = {706-710},

address = {Big Island, USA},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{Latzko2402:Experiment_as_Code,

title = {Experiment as Code - An Automated Testbed for Cloud Scaling and Operations},

author = {Vincent {Latzko} and Zhenzhen {Jia} and Christian Leonard {Vielhaus} and Mahshid {Mehrabi} and Frank H. P. {Fitzek} },

doi = {10.1109/ACDSA59508.2024.10468018},

year  = {2024},

date = {2024-02-01},

urldate = {2024-02-01},

booktitle = {International Conference on Artificial Intelligence, Computer, Data Sciences and Applications (ACDSA)},

pages = {5},

address = {Mahe, Seychelles},

abstract = {Cloud Computing has been widely adopted in industry, administrations, as well as research. At the same time, development and testing of new technologies is troubled by inconsistent environments. Exchanging modules of a system during research becomes a major obstacle. As a result, baselines for comparisons are lacking, and evaluating new developments may lead to heavy time cost. Big industrial players typically circumvent this problem by relying on the scale of their datacenters, experiments, and problems, which makes them inherently interesting. 

We propose a softwarised testbed that abstracts away inevitable hardware differences. Its design closely follows the established philosophy to separate workloads and declare expressive interfaces between them. The result is a coherent testbed, consisting of interchangeable, but state of the art, components. It establishes Experiment-as-Code, where an experiment is declared to load generators programmatically. In sum, the testbed yields reproducible results. We verify networking, computational and interacting components by reproducing known results. We illustrate one use case of scaling under uncertainty the applicability for research. The whole system is hosted on-premise, with minimal hardware requirements, which enables especially academia to contribute. The low-effort in setup also allows handover to students and staff with the goal to ensure knowledge transfer.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{Zimm2306:Information,

title = {Information Flow Graph for Distributed Caching without Newcomers over a Broadcast Medium},

author = {Sandra {Zimmermann} and Paul {Schwenteck} and Willi {Mei\ssner} and Christian Leonard {Vielhaus} and Juan A. {Cabrera} and Frank H. P. {Fitzek} and Martin {Reisslein}},

doi = { 10.1109/WoWMoM57956.2023.00017},

year  = {2023},

date = {2023-08-02},

urldate = {2023-06-01},

booktitle = {International Symposium on a World of Wireless, Mobile and 

Multimedia Networks (WoWMoM)},

publisher = {IEEE},

address = {Boston, MA, USA},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{Stei2306:Demo,

title = {Demo: The Future of Dog Walking - Four-Legged Robots and Augmented Reality},

author = {Jannek {Steinke} and Justus {Rischke} and Peter {Sossalla} and Johannes {Hofer} and Christian Leonard {Vielhaus} and Nico {vom Hofe} and Frank H. P. {Fitzek}},

doi = {10.1109/WoWMoM57956.2023.00060},

year  = {2023},

date = {2023-08-02},

urldate = {2023-08-02},

booktitle = {International Symposium on a World of Wireless, Mobile and 

Multimedia Networks (WoWMoM) Demo Session (WoWMoM 2023 Demo Session)},

publisher = {IEEE},

address = {Boston, MA, USA},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{vomH2306:Demo,

title = {Demo: Robotics meets Augmented Reality: Real-Time Mapping with Boston Dynamics Spot and Microsoft HoloLens 2},

author = {Nico {vom Hofe} and Peter {Sossalla} and Johannes {Hofer} and Christian Leonard {Vielhaus} and Justus {Rischke} and Jannek {Steinke} and Frank H. P. {Fitzek}},

doi = {10.1109/WoWMoM57956.2023.00062},

year  = {2023},

date = {2023-08-02},

urldate = {2023-08-02},

booktitle = {International Symposium on a World of Wireless, Mobile and 

Multimedia Networks (WoWMoM) Demo Session (WoWMoM 2023 Demo Session)},

publisher = {IEEE},

address = {Boston, MA, USA},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{Latz2306:Analysing,

title = {Analysing and Learning Low-Latency Network Coding Schemes},

author = {Vincent {Latzko} and Christian Leonard {Vielhaus} and Mahshid {Mehrabi} and Frank H. P. {Fitzek}},

doi = {10.1109/WiMob58348.2023.10187820},

year  = {2023},

date = {2023-07-26},

urldate = {2023-06-20},

booktitle = {International Conference on Wireless and Mobile Computing, 

Networking and Communications (WiMob)},

address = {Montreal, Canada},

abstract = {Forward Error Correction (FEC) has become an integral part of communication technology to address expected losses during transfer of data. Among 

various layer three block codes, Random Linear Network Coding (RLNC) has emerged as an adaptable, powerful approach. However, its most straight-forward implementation, Full Vector Coding (FVC), introduces too much delay for widespread adoption. Handcrafted schemes were introduced to optimise coding delay, while keeping resilience reasonably high. These works have resulted in tailRLNC and PACE. For the first time we analyse their respective behaviours in a fair and comparable manner, as non-recovered packets were statistically ignored previously. 

We then introduce an environment that uses a consistent, unbiased simulator and interface it with a Deep Reinforcement Learning (DRL) agent. This is the first time RLNC is joined with DRL. Our deep Q-network (DQN) based agent effectively uses an optimisation loop and utilises a customisable, expressive and extendable parametric loss function to learn a protocol. We demonstrate our agent recovers hand-tailored schemes and achieves state of the art.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}