arXiv subjects

Sara Ayoubi

Publications and source records attributed to Sara Ayoubi.

5 recordsLinked to original sources

arXiv2026

SCOUT: Semantic scene COverage via Uncertainty-guided Traversal

Robots that operate over extended periods should not merely visit space; they should progressively understand it. Yet most 3D scene graph pipelines treat perception as a post-processing stage over a fixed dataset, decoupling scene representation from the decisions that determine what is observed in the first place. We present SCOUT, an online semantic exploration framework that closes this loop by coupling active traversal with probabilistic scene graph construction. Given a prior 2D occupancy map and posed RGB-D observations, SCOUT incrementally builds an uncertainty-aware 3D scene graph whose nodes maintain fused geometry and posterior beliefs over open-vocabulary object labels, while edges encode structural relations such as on, inside, belong, and next to. These beliefs are fed back to an uncertainty-guided traversal planner, which selects viewpoints by balancing expected semantic certainty gain, geometric coverage gain, and travel cost. In this way, the robot revisits ambiguous objects when additional evidence matters and expands into unseen free space when the scene remains incomplete. The resulting system treats semantic scene completeness as an operational objective rather than a passive by-product of semantic mapping, moving toward autonomous agents that can patrol, update, and reason about evolving indoor environments with minimal human intervention.

cs.RO↗

arXiv2024

The Challenges of Trace-Driven Wi-Fi Emulation

Wi-Fi link is unpredictable and it has never been easy to measure it perfectly; there is always bound to be some bias. As wireless becomes the medium of choice, it is useful to capture Wi-Fi traces in order to evaluate, tune, and adapt the different applications and protocols. Several methods have been used for the purpose of experimenting with different wireless conditions: simulation, experimentation, and trace-driven emulation. In this paper, we argue that trace-driven emulation is the most favorable approach. In the absence of a trace-driven emulation tool for Wi-Fi, we evaluate the state-of-the-art trace-driven emulation tool for Cellular networks and we identify issues for Wi-Fi: interference with concurrent traffic, interference with its own traffic if measurements are done on both uplink and downlink simultaneously, and packet loss. We provide a solid argument as to why this tool falls short of effectively capturing Wi-Fi traces. The outcome of our analysis guides us to propose a number of suggestions on how the existing tool can be tweaked to accurately capture Wi-Fi traces.

cs.NI↗

arXiv2023

LoLa: Low-Latency Realtime Video Conferencing over Multiple Cellular Carriers

LoLa is a novel multi-path system for video conferencing applications over cellular networks. It provides significant gains over single link solutions when the link quality over different cellular networks fluctuate dramatically and independently over time, or when aggregating the throughput across different cellular links improves the perceived video quality. LoLa achieves this by continuously estimating the quality of available cellular links to decide how to strip video packets across them without inducing delays or packet drops. It is also tightly coupled with state-of-the-art video codec to dynamically adapt video frame size to respond quickly to changing network conditions. Using multiple traces collected over 4 different cellular operators in a large metropolitan city, we demonstrate that LoLa provides significant gains in terms of throughput and delays compared to state-of-the-art real-time video conferencing solution.

cs.NI↗

arXiv2021

Traffic Refinery: Cost-Aware Data Representation for Machine Learning on Network Traffic

Network management often relies on machine learning to make predictions about performance and security from network traffic. Often, the representation of the traffic is as important as the choice of the model. The features that the model relies on, and the representation of those features, ultimately determine model accuracy, as well as where and whether the model can be deployed in practice. Thus, the design and evaluation of these models ultimately requires understanding not only model accuracy but also the systems costs associated with deploying the model in an operational network. Towards this goal, this paper develops a new framework and system that enables a joint evaluation of both the conventional notions of machine learning performance (e.g., model accuracy) and the systems-level costs of different representations of network traffic. We highlight these two dimensions for two practical network management tasks, video streaming quality inference and malware detection, to demonstrate the importance of exploring different representations to find the appropriate operating point. We demonstrate the benefit of exploring a range of representations of network traffic and present Traffic Refinery, a proof-of-concept implementation that both monitors network traffic at 10 Gbps and transforms traffic in real time to produce a variety of feature representations for machine learning. Traffic Refinery both highlights this design space and makes it possible to explore different representations for learning, balancing systems costs related to feature extraction and model training against model accuracy.

cs.NI↗

arXiv2019

Inferring Streaming Video Quality from Encrypted Traffic: Practical Models and Deployment Experience

Inferring the quality of streaming video applications is important for Internet service providers, but the fact that most video streams are encrypted makes it difficult to do so. We develop models that infer quality metrics (\ie, startup delay and resolution) for encrypted streaming video services. Our paper builds on previous work, but extends it in several ways. First, the model works in deployment settings where the video sessions and segments must be identified from a mix of traffic and the time precision of the collected traffic statistics is more coarse (\eg, due to aggregation). Second, we develop a single composite model that works for a range of different services (i.e., Netflix, YouTube, Amazon, and Twitch), as opposed to just a single service. Third, unlike many previous models, the model performs predictions at finer granularity (\eg, the precise startup delay instead of just detecting short versus long delays) allowing to draw better conclusions on the ongoing streaming quality. Fourth, we demonstrate the model is practical through a 16-month deployment in 66 homes and provide new insights about the relationships between Internet "speed" and the quality of the corresponding video streams, for a variety of services; we find that higher speeds provide only minimal improvements to startup delay and resolution.

cs.NI↗