Arman Maghsoudnia – Blog

Swiss Atlas Is Live

2026-10-10T00:00:00+00:00

Swiss Atlas</a> is live: an interactive map of Switzerland at 100 m resolution that shows where people live and where the mobile antenna sites are.

It combines the Federal Statistical Office’s STATPOP 2024 population grid with OFCOM’s antenna sites on swisstopo maps. You can color the map by age, nationality or any of the 77 published attributes, summarize a radius or a commune, compare antenna sites with the population, and share or export any view. It runs in the browser, in English, German, French and Italian.

The code is on GitHub</a>.

Paper Accepted to SIGCOMM 2026

2026-08-11T00:00:00+00:00

Our paper SAGE: A Real-Time AI System for Reducing Latency in NextG Cellular Networks</a> has been accepted to SIGCOMM 2026</a>. 

The paper was published in the ACM Digital Library on August 11, 2026.

LateFrame Is Now Public

2026-05-08T00:00:00+00:00

We have made LateFrame</a> public. 
LateFrame is a Linux UDP traffic generator for experiments where packet inter-arrival time matters. It is meant for cases where tools such as ping and fping are useful references, but not precise packet schedulers.
The tool supports synthetic traffic generation with controlled constant, Poisson, or Gaussian timing, as well as PCAP replay through UDP encapsulation while preserving packet timing.
This makes LateFrame especially useful for networking research, where repeatable traffic patterns, accurate packet timing, and reproducible workload replay are often needed to evaluate systems under realistic and controlled conditions.
The repository includes the source code, build instructions, comparison scripts, and measurement data for reproducing the current timing and replay results.

Paper Available on arXiv

2025-11-26T00:00:00+00:00

We have posted our paper LatencyScope: A System-Level Mathematical Framework for 5G RAN Latency</a> on arXiv. 
Abstract</h2>
This paper presents LatencyScope, a mathematical framework for accurately computing one-way latency (for uplink and downlink) in the 5G RAN across diverse system configurations. LatencyScope models latency sources at every layer of the Radio Access Network (RAN), pinpointing system-level bottlenecks—such as radio interfaces, scheduling policies, and hardware/software constraints—while capturing their intricate dependencies and stochastic nature. The framework also includes a configuration optimizer that uses these mathematical models to search hundreds of billions of configurations and identify settings that meet latency–reliability targets under user constraints. We validate LatencyScope on two open-source 5G RAN testbeds (srsRAN and OAI), demonstrating close agreement with empirical latency distributions and significant improvements over prior analytical models and widely used simulators (MATLAB 5G Toolbox and 5G-LENA). Finally, LatencyScope identifies system configurations that meet Ultra-Reliable Low-Latency Communications (URLLC) targets, enabling network operators to efficiently determine optimal system setups.

Workshop Paper Presented at SIGCOMM 2025

2025-09-08T00:00:00+00:00

We presented our paper Towards URLLC with Open-Source 5G Software</a> at the OpenRIT6G Workshop</a> at SIGCOMM 2025</a> in Coimbra, Portugal. 
Abstract</h2>
Ultra-Reliable Low-Latency Communication is a key feature of 5G, yet achieving its strict one-way latency target remains challenging in real-world deployments. While previous work proposes latency-reduction techniques, most are theoretical or simulation-based and overlook practical bottlenecks in actual systems. In this paper, we analyze and optimize latency with open-source 5G RAN software. We characterize latency sources arising from 5G specifications and implementation-level factors, along with their complex interplays. Guided by this analysis, we introduce improvements reducing one-way latency by 39.28% in the downlink and 55.38% in the uplink. Our results show the importance of system-level experimentation and provide a blueprint for advancing toward URLLC targets in both 5G and future cellular networks.
</a></figure>

Best Demo Award at MobiCom 2024

2024-11-23T00:00:00+00:00

We presented our demo SliceGuard: Secure and Dynamic 5G RAN Slicing with WebAssembly</a> at ACM MobiCom 2024 and won the Best Demo Award! 
</a></figure>
</a></figure>

Paper Presented at HotNets 2024

2024-11-18T00:00:00+00:00

We presented our paper Ultra-Reliable Low-Latency in 5G: A Close Reality or a Distant Goal?</a> at HotNets 2024 in Irvine, California, USA. 
Abstract</h2>
Ultra-Reliable Low-Latency Communication (URLLC) was introduced in 5G to meet the stringent requirements of latencies as low as 0.5 ms and reliability levels of up to 99.999% for specific applications. Despite more than a decade of discussion, achieving these targets in real-world deployments remains a significant challenge. We argue that it is still unclear whether—and how—URLLC can be realized in practice, and that a holistic, system-level perspective addressing all inherent bottlenecks is required. Inspired by a real-world 5G testbed, we present such a comprehensive view and demonstrate how latency requirements can be met by outlining the necessary design choices across all system layers, including both processing and radio units.
</a></figure>

Paper Accepted to HotNets 2024

2024-09-13T00:00:00+00:00

Our paper Ultra-Reliable Low-Latency in 5G: A Close Reality or a Distant Goal? has been accepted to HotNets 2024</a>. 
We presented the paper on November 18, 2024, in Irvine, California, USA.

Passed My Candidacy Exam

2024-08-26T00:00:00+00:00

I have passed my candidacy exam and am now officially a PhD candidate. 

I started my PhD!

2023-09-01T00:00:00+00:00

I am pleased to announce that I have officially begun my PhD at EPFL. 
</a></figure>
</a></figure>
</a></figure>