In 2022, 3 years after the initial 5G rollout, through a cross-country US driving trip (from Los Angeles to Boston), the authors of [28] conducted an in-depth measurement study of user-perceived experience (network coverage, performance, and QoE of a set of major 5G ''killer'' apps) over all three major US carriers. The study revealed disappointingly low 5G coverage and suboptimal network performance -- falling short of the expectations needed to support the new generation of 5G ''killer apps. Now, five years into the 5G era, widely considered its midlife, 5G networks are expected to deliver stable and mature performance. In this work, we replicate the 2022 study along the same coast-to-coast route, evaluating the current state of cellular coverage and network and application performance across all three major US operators. While we observe a substantial increase in 5G coverage and a corresponding boost in network performance, two out of three operators still exhibit less than 50% 5G coverage along the driving route even five years after the initial 5G rollout. We expand the scope of the previous work by analyzing key lower-layer KPIs that directly influence the network performance. Finally, we introduce a head-to-head comparison with Starlink's LEO satellite network to assess whether emerging non-terrestrial networks (NTNs) can complement the terrestrial cellular infrastructure in the next generation of wireless connectivity.

Despite 5G’s promise of enhanced capacity and lower latency, recent measurement studies have shown that users experience significant cellular performance degradation under vehicular mobility. In this paper, we go beyond prior work that merely characterizes 5G performance, by uncovering the underlying mechanisms responsible for throughput degradation episodes during vehicular mobility. Through extensive measurements across 3,687 km of driving routes in the U.S., we collect granular performance and network KPI data from three major carriers across diverse radio access technologies. We introduce a novel KPI-driven clustering methodology that not only quantifies the frequency and duration of performance degradations, but also critically identifies their root causes by analyzing KPIs and their interactions. Our analysis reveals previously unidentified patterns: persistent higher degradation rates in uplink versus downlink, technology-specific vulnerability signatures in LTE and 5G deployments, and the predominance of compound degradation mechanisms where multiple factors interact to create severe throughput reductions. These findings provide essential information for developing mobility-aware resource management strategies to address the unique challenges of vehicular connectivity.

With rapid evolution of mobile core network (MCN) architectures, large-scale control-plane traffic (CPT) traces are critical to studying MCN design and performance optimization by the R&D community. The prior-art control-plane traffic generator SMM heavily relies on domain knowledge which requires re-design as the domain evolves. In this work, we study the feasibility of developing a high-fidelity MCN control plane traffic generator by leveraging generative ML models. We identify key challenges in synthesizing high-fidelity CPT including generic (to data-plane) requirements such as multimodality feature relationships and unique requirements such as stateful semantics and long-term (time-of-day) data variations. We show state-of-the-art, generative adversarial network (GAN)-based approaches shown to work well for data-plane traffic cannot meet these fidelity requirements of CPT, and develop a transformer-based model, CPT-GPT, that accurately captures complex dependencies among the samples in each traffic stream (control events by the same UE) without the need for GAN. Our evaluation of CPT-GPT on a large-scale control-plane traffic trace shows that (1) it does not rely on domain knowledge yet synthesizes control-plane traffic with comparable fidelity as SMM; (2) compared to the prior-art GAN-based approach, it reduces the fraction of streams that violate stateful semantics by two orders of magnitude, the max y-distance of sojourn time distributions of streams by 16.0%, and the transfer learning time in deriving new hourly models by 3.36×.

With faster wireless networks and server GPUs, offloading high-accuracy but compute-intensive AR tasks implemented in Deep Neural Networks (DNNs) to edge servers offers a promising way to support high-QoE Augmented/Mixed Reality (AR/MR) applications. A cost-effective way for AR app vendors to deploy such edge-assisted AR apps to support a large user base is to use commercial Machine-Learning-as-a-Service (MLaaS) deployed at the edge cloud. To maximize cost-effectiveness, such an MLaaS provider faces a key design challenge, i.e., how to maximize the number of clients concurrently served by each GPU server in its cluster while meeting per-client AR task accuracy SLAs. The above AR offloading inference serving problem differs from generic inference serving or video analytics serving in one fundamental way: due to the use of local tracking which reuses the last server-returned inference result to derive results for the current frame, the offloading frequency and end-to-end latency of each AR client directly affect its AR task accuracy (for all the frames).

After 4 years of rapid deployment in the US, 5G is expected to have significantly improved the performance and overall user experience of mobile networks. However, recent measurement studies have focused either on static performance or a single aspect (e.g., handovers) under driving conditions of 5G, and do not provide a complete picture of cellular network performance today under driving conditions - a major use case of mobile networks. Through a cross-continental US driving trip (from LA to Boston, 5700km+), we conduct an in-depth measurement study of user-perceived experience (network coverage/performance and QoE of a set of major latency-critical 5G "killer'' apps) To understand the root cause of the observed network performance, while collecting low-level 5G statistics and signaling messages. Our study shows disappointingly low coverage of 5G networks today under driving and highly fragmented coverage by cellular technologies. More importantly, network and application performance are often poor under driving even in areas with full 5G coverage. We also examine the correlation of technology-wise coverage and performance with geo-location and the vehicle's speed and analyze the impact of a number of lower layer KPIs on network performance.

For its stringent QoE requirement, augmented reality (AR) has been widely hailed as a representative of ultra-high bandwidth and ultra-low latency apps that will be enabled by 5G networks/edge clouds. Such a portrait of AR by the telco and cloud industry raises an important research question - can 5G enable latency-critical applications such as (edge-assisted) AR? In this paper, we conduct to our knowledge the first in-depth measurement study of whether 5G mmWave in combination with in-network edge cloud can support the baseline edge-assisted object detection. After we discover 5G mmWave is unlikely to achieve the level of uplink network performance needed to support a baseline edge-assisted object detection implementation in the near future, we quantify the performance benefits in retrofitting app-level optimizations developed in the pre-5G era on top of baseline edge-assisted object detection, as well as the performance benefits from hardware upgrade on the edge. We find that these optimizations can significantly boost object detection performance over both LTE and 5G mmWave; however, the improvement with 5G mmWave over LTE is marginal, and 5G mmWave still fails to provide satisfactory performance in all scenarios under consideration. Overall, we conclude that today's 5G mmWave deployment is not a deciding factor in enabling edge-assisted object detection.

Immersive applications such as Augmented Reality (AR) and Mixed Reality (MR) often need to perform multiple latency-critical tasks on every frame captured by the camera, which all require results to be available within the current frame interval. While such tasks are increasingly supported by Deep Neural Networks (DNNs) offloaded to edge servers due to their high accuracy but heavy computation, prior work has largely focused on offloading one task at a time. Compared to offloading a single task, where more frequent offloading directly translates into higher task accuracy, offloading of multiple tasks competes for shared edge server resources, and hence faces the additional challenge of balancing the offloading frequencies of different tasks to maximize the overall accuracy and hence app QoE.

5G networks are being deployed rapidly across the world and have opened door to many uplink-oriented, bandwidth-intensive applications such as Augmented Reality and Connected Autonomous Vehicles. However, the roll-out is still in the early phase and the nature of deployment also varies across different geographic regions of the world. In this demo, we present NextG-UP, an Android-based tool designed to help understand the performance and evolution of 5G networks around the world. The crowd-sourcing mobile app collects various cellular network metrics and runs a short uplink throughput/latency test.

5G networks are being deployed rapidly across the world and have opened door to many uplink-oriented, bandwidth-intensive applications such as Augmented Reality and Connected Autonomous Vehicles (CAV). However, the roll-out is still in the early phase and the nature of deployment also varies across different geographic regions of the world. In this demo, we present NextG-UP, an open-source Android-based tool designed to help understand the performance and evolution of 5G networks around the world. The crowd-sourcing mobile app collects various cellular network metrics and runs a short uplink throughput/latency test.