Low Earth Orbit (LEO) satellite networks, such as Starlink, are increasingly used for Internet access, emergency response, and backup connectivity. However, tracking when and where Starlink usage changes remains difficult because direct sources of evidence, such as operator telemetry and subscriber records, are not publicly available. We present a cross-dataset tool for detecting high-confidence shifts in Starlink usage from public measurements. The tool combines three complementary vantage points: APNIC estimates of visible users, M-Lab speed-test activity, and Cloudflare Radar traffic trends. It first identifies changes in user trends independently in each time series, then reports candidate events when multiple vantage points observe same-direction changes within a narrow temporal window. We evaluate the approach through country-level case studies of Jamaica, Ukraine, and Germany. In Jamaica, the tool identifies a sharp increase in Starlink usage around the Hurricane Melissa response. In Ukraine, it detects usage shifts consistent with reported changes in Starlink deployment and availability. In Germany, it avoids treating a prominent APNIC-only rise-and-fall pattern as a supported event. These results show that cross-source public measurements can provide the community with a practical and high-confidence way to track event-driven shifts in how LEO networks are used and relied upon over time.

Modern servers employ multiple resource management algorithms, including fast core schedulers and overload controllers to balance application performance and resource utilization. Individual algorithms and their amalgamation are required to meet these tight performance requirements. In this paper, we demonstrate that state-of-the-art core schedulers and overload controllers produce poor performance when deployed simultaneously. Fundamentally, the design assumptions of each controller are violated by the other controller. An overload controller assumes that all resources are dedicated to an application, while a core scheduler assumes that all incoming load will be admitted. To overcome this fundamental limitation, we present CoreSync, a server-driven credit-based protocol for joint core scheduling and overload control. CoreSync relies on the basic idea that the admitted load should be proportional to the allocated resources. However, strict proportionality can lead to low utilization when admitted load does not materialize at the server (e.g., when demand drops). Thus, CoreSync uses partial proportionality to balance latency, throughput, and utilization. Our evaluation across synthetic and real-world workloads shows that CoreSync outperforms state-of-the-art schedulers and overload controllers. In particular, in overload scenarios, CoreSync improves throughput by up to 6%. At low loads, CoreSync reduces the 99th percentile latency by up to 1.7× and improves CPU utilization by up to 1.4×.

In this paper, we study the viability of LEO networks as a failover network. We contextualize our analysis by framing the capacity of satellite networks relative to lost capacity due to submarine cable failure. Specifically, we focus on scenarios where LEO networks act as failovers for submarine cables, providing a concrete target capacity to be fulfilled by the satellite network. We introduce a new model and simulator that help us estimate the failover capacity. We identify key factors determining the actual capacity available on the satellite network: the total area of the country, the terminal distribution policy used by the government, the spectrum allocation and traffic engineering policies used by the LEO network operator. Based on our findings, we make policy recommendations to governments that can result in an increase of up to 1.8× in the failover capacity without requiring additional infrastructure. However, we find after implementing all our recommendations, with 200k terminals deployed and no competing traffic in the network, a satellite network can only satisfy 0.9-14.7% of the capacity lost due to submarine cable failure in four out of six case studies.

Modern datacenter operators aim to maximize the utilization of limited and expensive resources, especially CPU cores. Achieving such an objective requires fast and accurate core scheduling policies. Meanwhile, operating at high utilization requires the employment of overload controllers that shed excess load beyond the allocated capacity. Currently, no analytical techniques exist to study the interactions between these controllers. In this paper, we use performance verification to establish bounds on the throughput and latency achieved by a server that employs state-of-the-art fine-grained core allocation and overload control mechanisms. Our model enables system performance analysis under a wide range of workload and system configurations (e.g., RTT, load, and burstiness). We show that worst-case throughput and latency degrade by $1.8 \times$ and $2.3 \times$, respectively, under high load and burstiness due to the interactions between overload control and fast core allocation. We validate our findings using simulations that demonstrate the plausibility of the identified worst-case behavior under realistic conditions.

Architecting a modern data center network is increasingly complicated. Seeking the highest performance and support for emerging workloads, network architects planning a buildout must choose from a large selection of switching components, NICs, network stacks, congestion control algorithms, routing schemes, measurement systems, virtualization software, centralized bandwidth allocators and security mechanisms, all from various vendors. Today, manual planning by human experts is time-consuming at best, and can easily result in overlooked design choices or missed complex inter-dependencies.

Modern data center applications require servers to respond to requests from thousands of clients while maintaining micro-second-scale SLOs. Efficient operation of the data center infrastructure requires assigning the exact amount of resources needed by the application, no more and no less. A burst of incoming traffic that exceeds allocated capacity can cause long queues, requiring efficient and responsive overload control schemes. A sudden drop in demand requires re-allocation of resources to other applications. Within a single host, several mechanisms have been proposed for overload control [1, 3, 8, 9] and dynamic core allocation [2, 4, 6, 7]. The state of the art in both control loops is designed to react to microsecond-level changes in load. We present a simulation-based study of the interaction between Overload Controllers and Core Allocators at microsecond timescales, examining their macroscopic implications at larger timescales.

This paper studies the problem of allocating tasks from different customers to vehicles in mobility platforms, which are used for applications like food and package delivery, ridesharing, and mobile sensing. A mobility platform should allocate tasks to vehicles and schedule them in order to optimize both throughput and fairness across customers. However, existing approaches to scheduling tasks in mobility platforms ignore fairness.

The diversity of paths on the Internet makes it difficult for designers and operators to confidently deploy new congestion control algorithms (CCAs) without extensive real-world experiments, but such capabilities are not available to most of the networking community. And even when they are available, understanding why a CCA underperforms by trawling through massive amounts of statistical data from network connections is challenging. The history of congestion control is replete with many examples of surprising and unanticipated behaviors unseen in simulation but observed on real-world paths. In this paper, we propose initial steps toward modeling and improving our confidence in a CCA's behavior. We have developed CCAC, a tool that uses formal verification to establish certain properties of CCAs. It is able to prove hypotheses about CCAs or generate counterexamples for invalid hypotheses. With CCAC, a designer can not only gain greater confidence prior to deployment to avoid unpleasant surprises, but can also use the counterexamples to iteratively improvetheir algorithm. We have modeled additive-increase/multiplicative-decrease (AIMD), Copa, and BBR with CCAC, and describe some surprising results from the exercise.

Cloud services are deployed in datacenters connected though high-bandwidth Wide Area Networks (WANs). We find that WAN traffic negatively impacts the performance of datacenter traffic, increasing tail latency by 2.5x, despite its small bandwidth demand. This behavior is caused by the long round-trip time (RTT) for WAN traffic, combined with limited buffering in datacenter switches. The long WAN RTT forces datacenter traffic to take the full burden of reacting to congestion. Furthermore, datacenter traffic changes on a faster time-scale than the WAN RTT, making it difficult for WAN congestion control to estimate available bandwidth accurately.

Modern end-host network stacks have to handle traffic from tens of thousands of flows and hundreds of virtual machines per single host, to keep up with the scale of modern clouds. This can cause congestion for traffic egressing from the end host. The effects of this congestion have received little attention. Currently, an overflowing queue, like a kernel queuing discipline, will drop incoming packets. Packet drops lead to worse network and CPU performance by inflating the time to transmit the packet as well as spending extra effort on retansmissions. In this paper, we show that current end-host mechanisms can lead to high CPU utilization, high tail latency, and low throughput in cases of congestion of egress traffic within the end host. We present zD, a framework for applying backpressure from a congested queue to traffic sources at end hosts that can scale to thousands of flows. We implement zD to apply backpressure in two settings: i) between TCP sources and kernel queuing discipline, and ii) between VMs as traffic sources and kernel queuing discipline in the hypervisor. zD improves throughput by up to 60%, and improves tail RTT by at least 10x at high loads, compared to standard kernel implementation.

BGP was initially created assuming by default that all ASes are equal. Its policies and protocols, namely BGP, evolved to accommodate a hierarchical Internet, allowing an autonomous system more control over outgoing traffic than incoming traffic. However, the modern Internet is flat, making BGP asymmetrical. In particular, routing decisions are mostly in the hands of traffic sources (i.e., content providers). This leads to suboptimal routing decisions as traffic sources can only estimate route capacity at the destination (i.e., ISP). In this paper, we present the design of Unison, a system that allows an ISP to jointly optimize its intra-domain routes and inter-domain routes, in collaboration with content providers. Unison provides the ISP operator and the neighbors of the ISP with an abstraction ISP network in the form of a virtual switch. This abstraction allows the content providers to program the virtual switch with their requirements. It also allows the ISP to use that information to optimize the overall performance of its network. We show through extensive simulations that Unison can improve ISP throughput by up to 30% through cooperation with content providers. We also show that cooperation of content providers only improves performance, even for non-cooperating content providers (e.g., a single cooperating neighbour can improve ISP throughput by up to 6%).

Low-cost mini-drones with advanced sensing and maneuverability enable a new class of intelligent sensing systems. To achieve the full potential of such drones, it is necessary to develop new enhanced formulations of both common and emerging sensing scenarios. Namely, several fundamental challenges in visual sensing are yet to be solved including (1) fitting sizable targets in camera frames; (2) positioning cameras at effective viewpoints matching target poses; and (3) accounting for occlusion by elements in the environment, including other targets. In this article, we introduce Argus, an autonomous system that utilizes drones to collect target information incrementally through a two-tier architecture. To tackle the stated challenges, Argus employs a novel geometric model that captures both target shapes and coverage constraints. Recognizing drones as the scarcest resource, Argus aims to minimize the number of drones required to cover a set of targets. We prove this problem is NP-hard, and even hard to approximate, before deriving a best-possible approximation algorithm along with a competitive sampling heuristic which runs up to 100× faster according to large-scale simulations. To test Argus in action, we demonstrate and analyze its performance on a prototype implementation. Finally, we present a number of extensions to accommodate more application requirements and highlight some open problems.

The basic MAC mechanisms in IEEE 802.11 (WiFi) have remained largely unchanged for over 20 years. In this paper, we argue that the prevalence of WiFi makes it almost impossible to improve its performance through changes that require modifying hardware, firmware, or drivers. New applications, however, continue to exert novel performance demands. We suggest that changes should be developed as augmentation-only solutions through above-driver, kernel-level software modifications. An augmentation-only solution needs to maintain inter-operability and afford transparency in performance to existing WiFi devices, as well as enable minimum overhead upgradability. Our goal is to demonstrate the feasibility of MAC augmentation according to these principles. To this end, we leverage soft scheduling, where nodes are asked for a best-effort attempt to adhere to a given schedule. We allow the soft scheduler to coexist with and work at a different time scale from WiFi's Distributed Coordination Function (DCF); allowing it to reduce the time nodes spend contending for the medium while allowing DCF to handle only missed schedule slots and schedule divergence. We present a new Soft Token Passing Protocol (STPP) as an instance of this family of Soft Scheduling Protocols. We then show how STPP can be made part of a MAC protocol with specific performance improvement goals by developing the Wireless Low-Latency Local Links (WL4) system. We evaluate WL4 on a five node microbenchmark and quantify the system's overhead on network throughput and latency. We show that soft scheduling, via STPP, enables WL4 to adhere to our augmentation principles while improving the latency within the system.

White spaces are portions of the TV spectrum that are allocated but not used locally. Ifaccurately detected, white spaces offer a valuable new opportunity for highspeed wireless communications. We propose a new method for white space detection that allows a node to actlocally, based on a centrally constructed model, and at low cost, whiledetecting more spectrum opportunities than best known approaches. Weleverage two ideas. First, we demonstrate that low-cost spectrum monitoringhardware can offer "good enough" detection capabilities. Second, we develop amodel that combines locally-measured signal features and location to more efficiently detect white space availability. We incorporate these ideas into the design,implementation, and evaluation of a complete system we call Waldo. We deployWaldo on a laptop in the Atlanta metropolitan area in the US covering 700 km2. Our results show that usingsignal features, in addition to location, can improve detection accuracy by up to10x for some channels. We also deploy Waldo on an Android smartphone,demonstrating the feasibility of real-time white space detection with efficientuse of smartphone resources.

Low-cost mini-drones with advanced sensing and maneuverability enable a new class of intelligent visual sensing systems. This potential motivated several research efforts to employ drones as standalone surveillance systems or to assist legacy deployments. However, several fundamental challenges remain unsolved including: 1) Adequate coverage of sizable targets; 2) Target orientation that render coverage effective only from certain directions; 3) Occlusion by elements in the environment, including other targets.

Traffic shaping, including pacing and rate limiting, is fundamental to the correct and efficient operation of both datacenter and wide area networks. Sample use cases include policy-based bandwidth allocation to flow aggregates, rate-based congestion control algorithms, and packet pacing to avoid bursty transmissions that can overwhelm router buffers. Driven by the need to scale to millions of flows and to apply complex policies, traffic shaping is moving from network switches into the end hosts, typically implemented in software in the kernel networking stack.

Wireless sensor and actuator networks have been extensively deployed for enhancing industrial control processes and supply-chains, and many forms of surveillance and environmental monitoring. The availability of low-cost mobile robots equipped with a variety of sensors in addition to communication and computational capabilities makes them particularly promising in target coverage tasks for ad hoc surveillance, where quick, low-cost or non-lasting visual sensing solutions are required, e.g. in border protection and disaster recovery. In this paper, we consider the problem of low complexity placement and orientation of mobile cameras to cover arbitrary targets. We tackle this problem by clustering proximal targets, while calculating/estimating the camera location/direction for each cluster separately through our cover-set coverage method. Our proposed solutions provide extremely computationally efficient heuristics with only a small increase in number of cameras used, and a small decrease in number of covered targets.

Cognitive radios emerged as a solution for utilizing the spectrum which is considered a limited resource. Multi-hop routing in cognitive radio networks (CRNs) has been gaining increasing attention as it enables future large-scale CRNs. However, many existing protocols flood the network with control packets in route discovery phase, which leads to wasting bandwidth. In this paper, we introduce a location-aware probabilistic route discovery technique for CRNs that leverages gossiping with dynamic probabilities to reduce the flooding overhead without affecting the quality of the discovered routes. The proposed technique can be used on top of any routing protocol regardless of whether it relies on a common control channel or not. Evaluation of our technique through ns2 simulations for improving different classes of routing protocols shows a significant reduction in the number of control packets by up to 75% and an increase in throughput by up to 400%.

Cyber-Physical Systems (CPS) have the promise of presenting the next evolution in computing with potential applications that include aerospace, transportation, and various automation systems. These applications motivate advances in the different sub-fields of CPS such as mobile computing, context awareness, and computer vision. However, deploying and testing complete CPSs is known to be a complex and expensive task. In this paper, we present the design, implementation, and evaluation of Up and Away (UnA): a testbed for Cyber-Physical Systems that use Unmanned Aerial Vehicles (UAVs) as their main physical component. UnA aims to abstract the control of physical system components to reduce the complexity of UAV oriented CPS experiments. UnA provides APIs to allow for converting CPS algorithm implementations, developed typically for simulations, into physical experiments using a few simple steps. We present two scenarios of using UnA's API to bring mobile-camera-based surveillance algorithms to life, thus exhibiting the ease of use and flexibility of UnA.

Cognitive radio networks (CRNs) provide a solution to increase the utilization of the scarce radio frequency spectrum. Building testbeds for CRNs is one of the main challenges that can affect the wide deployability of such networks. In this paper, we present the design, implementation, and evaluation of CogFrame: a framework that facilitates the development of cost-efficient large-scale CRNs routing protocols testbeds. The framework allows the designers to focus on the CRNs routing protocols by abstracting the PHY and MAC layers while providing the necessary cross layer functionalities. CogFrame works with standard computers and WiFi cards to reduce the cost while allowing integration with other special hardware for more flexibility. In addition, CogFrame provides different modules for implementing and emulating complex scenarios such as regulatory authority policies, mobility management, and topology management. We benchmark the performance of CogFrame and compare it to standard ns-2 simulations and USRP2 implementations. In addition, we case study a location-aided routing protocol for CRNs using both CogFrame and ns-2 simulations. Our results highlight the ease of implementation, low-cost, and realistic replication of the CRN environment, showing the promise of CogFrame as a testbed for future CRNs implementations.

The widespread usage of WLANs and mobile devices has fostered the interest in localization systems for wireless environments. The majority of research in the context of wireless-based localization systems has focused on device-based active localization, in which devices are attached to tracked entities. Recently, device-free passive localization (DfP) has been proposed where the tracked entity is neither required to carry devices nor to participate actively in the localization process. Previous studies have focused on small areas and/or controlled environments. In this paper, we present the design, implementation, and analysis of Nuzzer, a large-scale DfP localization system, which tracks entities in real environments, rich in multipath. We first present probabilistic techniques for DfP localization of a single entity and evaluate their performance both analytically and in typical office buildings. Our results show that Nuzzer gives location estimates with less than 2-meters median distance error. We then give an algorithm for estimating the number of entities in an area of interest and localizing them into coarse-grained zones to enhance the scalability of the system. This indicates the suitability of Nuzzer to a large number of application domains.

WLAN Device-free passive (DfP) indoor localization is an emerging technology enabling the localization of entities that do not carry any devices nor participate actively in the localization process using the already installed wireless infrastructure. This technology is useful for a variety of applications such as intrusion detection, smart homes and border protection. We present the design, implementation and evaluation of RASID, a DfP system for human motion detection. RASID combines different modules for statistical anomaly detection while adapting to changes in the environment to provide accurate, robust, and low-overhead detection of human activities using standard WiFi hardware. Evaluation of the system in two different testbeds shows that it can achieve an accurate detection capability in both environments with an F-measure of at least 0.93. In addition, the high accuracy and low overhead performance are robust to changes in the environment as compared to the current state of the art DfP detection systems.

WLAN Device-free Passive (DfP) localization is an emerging technology that uses the widely deployed WiFi networks for detecting and localizing human presence within indoor environments. This paper presents an accurate and low-overhead technique for detecting human presence based on non-parametric statistical anomaly detection. This technique constructs profiles capturing the signal strength characteristics when no human is present within the area of interest and uses these profiles to identify any anomalies in the signal strength due to human motion activity. To adapt to changes in the environment, the constructed profiles are regularly updated by signal strength readings with low anomaly probability. Exponential smoothing is then used to reduce the effect of noisy readings in order to enhance the detection accuracy. Our work proved to be more robust and accurate than other DfP detection techniques, achieving a high detection accuracy of 4.7% miss detection rate and 3.8% false alarm rate, while requiring minimal deployment overhead.