Search for: All records

Crowdsourced navigation platforms infer congestion from GPS-derived telemetry, enabling real-time traffic estimation and control at city scale. However, because these systems rely on device-reported location traces, they are vulnerable to Sybil attacks, where an adversary injects fabricated “vehicles” and synthetic slowdowns using emulators, scripted clients, or device farms. Such attacks have been publicly demonstrated to create “ghost” cars and phantom congestion that distort the displayed traffic layer and influence routing decisions. We propose using street cameras as a real-time validation layer for GPS-based traffic inference. Street cameras add an independent signal grounded in physical observation (“eyes on the road”), rather than device-reported telemetry alone. They also sit on a different attack surface than mobile devices: manipulating camera evidence targets physical infrastructure rather than the mobile client, and is often more expensive to scale than client-side Sybil injection. By cross-checking camera-observed vehicle activity against GPS-based mobile traces, camera feeds provide a conservative verification layer, and can be used to raise integrity alerts. In this paper, we implement an end-to-end GPS-camera pipeline that (i) matches cameras to nearby road links, (ii) aligns and stabilizes the streams, (iii) scores mismatch using a residual-based detector to produce alerts, and (iv) extends coverage to blind spots using nearby intersections. We evaluate the system using New York City DOT street cameras (200+ active in Manhattan) together with GPS-based traffic levels collected via the TomTom API. The pipeline achieves 91.5% detection under a moderate injected attack (δ = +0.05) and reaches ≈99% detection for stronger attacks. We further address blind spots: using only neighbors, we retain a high-load consensus signal in 96.68% of Manhattan tested locations. 

We propose a data synthesis pipeline for generating realistic traffic scenes and safety-critical rare events under natural language instructions while providing agent relation annotations. The pipeline structurally comprises the scene planner, agent generator, waypoint filter, event reasoner, and trajectory refiner, while incorporating a language model backend for controlled inference. By decomposing high-level semantic reasoning and low-level scene execution, our framework is able to produce physically grounded agent trajectories that satisfy the social relation specifications. The pipeline is used to generate a dataset of 370 traffic scenes based on an urban traffic intersection, featuring agent relations such as collision and yielding, which are safety-critical but challenging to specify in real world traffic data. We evaluate the quality of the synthesized agent trajectories by the simulation-to-reality gaps, where the pipeline achieves an 84% of instruction satisfaction rate equipped with the Claude3.5-Sonnet backend. We further showcase the usage of the synthesized dataset by testing traffic scene perception and precognition using a simple agentic pipeline, both outperforming non-LLM baselines by a noticeable margin. 

We present a measurement-based statistical model for the backscatter power ratio of monostatic RF sensing in urban canyons with moving clutter, suitable for large-scale system level performance evaluation of RF sensing in 6G networks. A narrowband (CW) 140 GHz sounder used a monostatic radar arrangement with an omnidirectional transmit antenna illuminating streets and a spinning horn 2o receive antenna offset vertically (less than 1 m away) collecting backscattered power as a function of azimuth and time below building height in Manhattan and Valparaiso, Chile. A concise outdoor deterministic model of average backscattered power dependent on distance to nearest building-wall reproduces observations with 3.3 dB RMS error or better. Distribution of power variation in azimuth around this average is reproduced within 0.5 dB by a random azimuth spectrum with a lognormal distribution. Temporal fluctuations for various antenna aims and locations were found to be well modeled by a Rician distribution, with lognormally distributed K-factor, with 0.47- 0.73 correlation coefficient to backscatter power deviation from mean. The statistical model does not require a detailed environmental description, aiming to reproduce backscatter clutter statistics (as opposed to a deterministic response) faithfully and efficiently, essential for large-scale system-level performance evaluation. 

We develop novel RF canceler tuning algorithms for the phase and amplitude (P&A) based full-duplex (FD) radios in the open-access NSF PAWR COSMOS testbed. The RF canceler is a critical component in an FD radio, supporting FD operation by performing self-interference cancellation (SIC). To enable algorithm development, we use a mobile P&A-based FD radio to collect a dataset of canceler performance across different environments within the COSMOS testbed. This dataset is used to train neural networks (NNs) to predict the optimal configuration of the RF canceler using input features derived from an estimate of the self-interference channel. The best model achieves within 1 dB of the optimal cancellation with a one-shot prediction, while also achieving a 99% speedup compared to a baseline gradient descent algorithm. To the best of our knowledge, this is the first practical implementation of a NN-based tuning algorithm for a custom RF SI canceler suitable for small form factor devices, utilizing a software-defined radio for evaluation. 

We propose a framework for adaptive data collection aimed at robust learning in multi-distribution scenarios under a fixed data collection budget. In each round, the algorithm selects a distribution source to sample from for data collection and updates the model parameters accordingly. The objective is to find the model parameters that minimize the expected loss across all the data sources. Our approach integrates upper-confidence-bound (UCB) sampling with online gradient descent (OGD) to dynamically collect and annotate data from multiple sources. By bridging online optimization and multi-armed bandits, we provide theoretical guarantees for our UCB-OGD approach, demonstrating that it achieves a minimax regret of O(T 1 2 (K ln T) 1 2 ) over K data sources after T rounds. We further provide a lower bound showing that the result is optimal up to a ln T factor. Extensive evaluations on standard datasets and a real-world testbed for object detection in smartcity intersections validate the consistent performance improvements of our method compared to baselines such as random sampling and various active learning methods. 

—We consider a decentralized wireless network with several source-destination pairs sharing a limited number of orthogonal frequency bands. Sources learn to adapt their transmissions (specifically, their band selection strategy) over time, in a decentralized manner, without sharing information with each other. Sources can only observe the outcome of their own transmissions (i.e., success or collision), having no prior knowledge of the network size or of the transmission strategy of other sources. The goal of each source is to maximize their own throughput while striving for network-wide fairness. We propose a novel fully decentralized Reinforcement Learning (RL)-based solution that achieves fairness without coordination. The proposed Fair Share RL(FSRL)solution combines: (i) state augmentation with a semiadaptive time reference; (ii) an architecture that leverages risk control and time difference likelihood; and (iii) a fairness-driven reward structure. We evaluate FSRL in more than 50 network settings with different number of agents, different amounts of available spectrum, in the presence of jammers, and in an ad-hoc setting. Simulation results suggest that, when we compare FSRL with a common baseline RL algorithm from the literature, FSRL can be up to 89.0% fairer (as measured by Jain’s fairness index) in stringent settings with several sources and a single frequency band, and 48.1% fairer on average.