John Bellardo

Jigsaw: solving the puzzle of enterprise 802.11 analysis

The combination of unlicensed spectrum, cheap wireless interfaces and the inherent convenience of untethered computing have made 802.11 based networks ubiquitous in the enterprise. Modern universities, corporate campuses and government offices routinely de-ploy scores of access points to blanket their sites with wireless Internet access. However, while the fine-grained behavior of the 802.11 protocol itself has been well studied, our understanding of how large 802.11 networks behave in their full empirical complex-ity is surprisingly limited. In this paper, we present a system called Jigsaw that uses multiple monitors to provide a single unified view of all physical, link, network and transport-layer activity on an 802.11 network. To drive this analysis, we have deployed an infrastructure of over 150 radio monitors that simultaneously capture all 802.11b and 802.11g activity in a large university building (1M+ cubic feet). We describe the challenges posed by both the scale and ambiguity inherent in such an architecture, and explain the algorithms and inference techniques we developed to address them. Finally, using a 24-hour distributed trace containing more than 1.5 billion events, we use Jigsaws global cross-layer viewpoint to isolate performance artifacts, both explicit, such as management inefficiencies, and implicit, such as co-channel interference. We believe this is the first analysis combining this scale and level of detail for a production 802.11 network.

Measuring packet reordering

The Internet architecture provides an unsequenced datagram delivery service. Nevertheless, many higher-layer protocols, such as TCP, assume that packets are usually delivered in sequence, and consequently suffer significant degradation when packets are reordered in flight. While there have been several recent proposals to create protocols that adapt to reordering, evaluating their effectiveness requires understanding the dynamics of the reordering processes prevalent in the Internet. Unfortunately, Internet packet sequencing is a poorly characterized and understudied behavior. This failing can be largely attributed to the lack of accurate and universally applicable methods for measuring packet reordering. In this paper, we describe a new set of active measurement techniques that can reliably estimate one-way end-to-end reordering rates to and from arbitrary TCP-based servers. We validate these tools in a controlled setting and show how they can be used to measure the time-domain distribution of the reordering process along a given path.

INSPIRE: Interplanetary NanoSpacecraft Pathfinder in Relevant Environment

The INSPIRE project would demonstrate the revolutionary capability of deep space CubeSats by placing two nanospacecraft in Earth-escape orbit. Prior to any inclusion on larger planetary missions, CubeSats must demonstrate that they can operate, communicate, and be navigated far from Earth – these are the primary objectives of INSPIRE. Spacecraft components, such as a JPL X-band radio and a robust watchdog system, would provide the basis for future high-capability, lower-cost-risk missions beyond Earth. These components should enable future supplemental science and educational opportunities at many destinations. The nominal INSPIRE mission would last for three months and achieve an expected Earth-probe distance of 1.5x10 km (dependent upon escape velocity as neither spacecraft will have propulsion capability). The project would monitor onboard telemetry; operate, communicate, and navigate with both spacecraft; demonstrate cross-link communications; and demonstrate science utility with an onboard magnetometer and imager. Lessons learned from this pathfinder mission should help to inform future interplanetary NanoSpacecraft and larger missions that might use NanoSpacecraft components.

PolySat's Next Generation Avionics Design

The CubeSat platform provides a unique challenge for flight software design due to the incredible size and power constraints. A number of tradeoffs must be made to balance effectiveness, fault tolerance, and cost. These basic requirements have been combined with the lessons learned from Cal Polys past 8-bit avionics system to design a significant revision based around a 32-bit microprocessor running Linux. This work analyzes both generations of avionics design, including a discussion of major design principles that are relevant to other CubeSat missions.

Onboard Autonomy on the Intelligent Payload EXperiment CubeSat Mission

The Intelligent Payload Experiment (IPEX) is a CubeSat that flew from December 2013 through January 2015 and validated autonomous operations for onboard instrument processing and product generation for the Intelligent Payload Module of the Hyperspectral Infrared Imager (HyspIRI) mission concept. IPEX used several artificial intelligence technologies. First, IPEX used machine learning and computer vision in its onboard processing. IPEX used machine-learned random decision forests to classify images onboard (to downlink classification maps) and computer vision visual salience software to extract interesting regions for downlink in acquired imagery. Second, IPEX flew the Continuous Activity Scheduler Planner Execution and Re-planner AI planner/scheduler onboard to enable IPEX operations to replan to best use spacecraft resources such as file storage, CPU, power, and downlink bandwidth. First, the ground and flight operations concept for proposed HyspIRI IPM operations is described, followed by a description ...

Real-Time Orbital Image Analysis Using Decision Forests, with a Deployment Onboard the IPEX Spacecraft

Automatic cloud recognition promises significant improvements in Earth science remote sensing. At any time, more than half of Earths surface is covered by clouds, obscuring images and atmospheric measurements. This is particularly problematic for CubeSats, a new generation of small, low-orbiting spacecraft with very limited communications bandwidth. Such spacecraft can use image analysis to autonomously select clear scenes for prioritized downlink. More agile spacecraft can also benefit from cloud screening by retargeting observations to cloud-free areas. This could significantly improve the science yield of instruments such as the Orbiting Carbon Observatory 3 mission. However, most existing cloud detection algorithms are not suitable for these applications, because they require calibrated and georectified spectral data, which is not typically available onboard. Here, we describe a statistical machine-learning method for real-time autonomous scene interpretation using a visible camera with no radiometric calibration. A random forest classifies cloud and clear pixels based on local patterns of image texture. We report on experimental evaluation of images from the International Space Station ISS and present results from a deployment onboard the IPEX spacecraft. This demonstrates actual execution in flight and provides some preliminary lessons learned about operational use. It is a rare example of a machine-learning system deployed to an autonomous spacecraft. To our knowledge, it is also the first instance of significant artificial intelligence deployed on board a CubeSat and the first ever deployment of visible image-based cloud screening onboard any operational spacecraft.

Autonomy for remote sensing — Experiences from the IPEX CubeSat

The Intelligent Payload Experiment (IPEX) is a CubeSat mission to flight validate technologies for onboard instrument processing and autonomous operations for NASAs Earth Science Technologies Office (ESTO). Specifically IPEX is to demonstrate onboard instrument processing and product generation technologies for the Intelligent Payload Module (IPM) of the proposed Hyperspectral Infra-red Imager (HyspIRI) mission concept. Many proposed future missions, including HyspIRI, are slated to produce enormous volumes of data requiring either significant communication advancements or data reduction techniques. IPEX demonstrates several technologies for onboard data reduction, such as computer vision, image analysis, image processing and in general demonstrates general operations autonomy. We conclude this paper with a number of lessons learned through operations of this technology demonstration mission on a novel platform for NASA.

Reactive Encapsulation Mappings in HIDRA

Scalability analysis of the Internet has resulted in two main concerns: rapid growth of the forwarding table and BGP’s poor convergence properties when distributing hundreds of thousands of routes. HIDRA [5], a backward-compatible architecture designed with feasibility-of-implementation in mind, has been proposed as one solution to reduce the size of the default-free zone (DFZ) forwarding table. This work extends HIDRA, greatly reducing the number of routes maintained by BGP, yet preserves a practical, incremental deployment strategy. The proposed protocol also provides end networks direct, finer-grained control over the distribution of packets flowing into their network and provides for efficient mobility support. The new mapping protocol is prototyped on a small network testbed and shown to work in all tested circumstances, including normal network operation, link failures, and transitional routing environments. Additionally, IP Mobility is discussed and shown to work in this environment without triangle routing and only minimal additional overhead.

HeL1oNano: The first CubeSat to L1?

HeL1o Nano: The mission concept Mission summary – A technology demonstration and Heliophysics science mission – Based on 6U cubesat of 14 kg placed on a Lissajous orbit about Sun Earth Lagrange point 1 (SEL1) – Twin spacecraft option to improve reliability

Onboard machine learning classification of images by a cubesat in Earth orbit

10 x 10 cm) that was launched into Earth orbit on December 5, 2013. It carries a random forest classifier that performs onboard analysis of images collected by its five 3-megapixel cameras (Altinok et al., submitted). The classifier was trained on the ground prior to launch using test imagery from a high-altitude balloon flight. The classifier architecture was adapted from the TextureCam instrument, which integrates image acquisition, processing, and classification into a single system (Wagstaff et al., 2013; Bekker et al., 2014). Each time IPEX acquires a new image, it classifies each pixel as cloudy, clear, the planetary limb, or outer space. This figure shows AI MATTERS, VOLUME 1, ISSUE 4! JUNE 2015