FOAM 2020 Summer Updates
The Foamspace team has been very busy this summer and we are excited to share some of our updates. Our previous blog post described the radio hardware, custom firmware, general architecture of a FPGA powered software defined radio(SDR). This blog post will cover the first FOAM Zone for Proof of Location which has been installed across the Brooklyn Navy Yard, a large milestone in our development, allowing field testing, experiments and localization to be conducted in a real world environment. Like many we have had to navigate hurdles brought on by the pandemic, however we are fortunate enough to continue development and leverage the New Lab product realization facilities for agile design iterations, prototypes and tests for our hardware stack.
Overview of the Installation
Through the development process of FOAM Location, verified location services independent of GPS, extensive testing was done in ideal settings to avoid any interference or obstruction, for example a field or beachfront. Having successfully validated our technology and localization protocol the next step was to test in a dense, urban environment with as many obstacles as possible. For this, we have collaborated with the Brooklyn Navy Yard(BNY) to permanently install our radio nodes for ongoing testing. As the first operational test network, the BNY Zone will serve as an ongoing sandbox environment.
The FOAM Location protocol requires at least four Zone Anchor radios to form a zone. This installation hosts four rooftop radio nodes with good visibility between themselves. Previously, we shared photos of the first Zone Anchor enclosure design; however, this prototype was optimized for indoor use, and so was ill-equipped for an indefinite outdoor installation. Further iterations produced a new enclosure design for outdoor use which addressed new requirements for mounting, power management, and antenna placement.
Unlike tests done in an ideal environment such as an open field, the BNY installation has the radio nodes at differing altitudes, which needs to be taken into consideration when running the protocol and further enables these tests to mirror a real world production environment as closely as possible. The altitude of the nodes range from 10m to 60m and the largest distance between them is 500 m. This installation currently exists across three building rooftops and four locations in the BNY, with additional rooftop locations to be added as the locations are secured.
Components of an Experiment
Given the overview of the installation, let’s unpack what it means to run tests and experiments. From our developer’s point of view, a zone is a collection of remotely-controlled radios, which are coordinated to transceive LoRa packets to and from one another and additionally configured to log those packets to a remote data store, such as a blockchain. Specifically, radios are provisioned through a remote task execution and configuration management system developed at FOAM. This allows us to easily explore different combinations of experiment parameters, from new firmware builds to varying LoRa spreading factors.
When transmissions are coordinated across multiple radio nodes, one can perform localization of each of the transmitting gateways by using pseudoranges, which are estimations of the distance between a transmitter and a receiver. The configuration described above allows for a set of gateways to receive and transmit LoRa packets and generate pseudoranges, performing localization requires further analysis such as multilateration.
Installation Results
The BNY Zone has been running successfully for the past 10 weeks and we are able to run experiments each day. The first thing we look for in the results of an experiment is the Zone Quality. This is measured by the visibility each node has to the other three and is plotted on a matrix graph. A full matrix indicates 100% visibility and a healthy Zone and that all packets were received. Even if all packets are received, some may be faulty and unreadable. Adding additional nodes can enable localization without full visibility. The radio nodes in this installation are able to transmit, receive and decode packets from all of the others. Further, by the process described above each Zone Anchor can localize each other accurately and plot their locations on a map dashboard.
With the results of the Zone itself clear, we additionally run tests daily that incorporate a dynamic and moving node, which we do with a person walking with a radio node as well as on bicycles and vehicles. A very positive sign for this real world environment is that all Zone Anchors on the roofs at different altitudes are able to receive packets from the a dynamic node and decode them successfully, despite the dynamic node being lower on the street level with obstacles such as buildings. We have been able to localize a dynamic node at a consistent accuracy comparable to GPS. Currently we are building animated dashboards for richer, real-time visualizations and asset tracking.
Zone Maintenance
Running experiments and making updates each day takes a lot of time and energy from our team. This includes visiting the rooftops to swap out components as needed, adjust antenna placements, re-assessing power sources and acting as a dynamic node for tests. Currently we are working on establishing a permanent power source as well as exploring Solar power in. Zone Anchor assembly is underway for more Zones!
Next Steps
The work done on the BNY Zone has allowed up to stress test the protocol, implement robust monitoring and testing procedures while improving our localization algorithms. Work done on a dashboard for the whole network and to display localization in real-time is nearing completion. Next we will upgrade this Zone to be a testnet for presence claims, verified and fraud proof location data stored on the blockchain.
The full vision for FOAM Location for Zone interconnectivity is reliant on the maturity of underlying L1 and L2 infrastructure. In recent months there has been impressive progress on components such as IBC, Peggy and Ethermint. That said, the work to include a blockchain and web3 functionality has already been completed by our team while working on the Kepler Cosmos SDK. Additionally, our team continues to make contributions to the core Cosmos protocol. Once our sdk is implemented we will follow up with a demo of a user obtaining a Presence Claim from the BNY Zone and stored on the local Tendermint chain. From there we will be seeking to engage with community for support in hosting new Zones. Please reach out if you would like to participate and have access to locations for testing.
Conclusion
In conclusion, circling back to the beginning of this post, we have had to navigate hurdles brought on by the pandemic, however we are fortunate enough to continue development and leverage the New Lab product realization facilities for agile design iterations, prototypes and tests for our hardware stack. The installation of the BNY Zone was delayed by mandatory work from home orders. However, both the New Lab and the Brooklyn Navy Yard remained open, essential and became the center of New York City’s relief efforts.
Facilities pivoted to making urgently needed face shields and hand sanitizer; Brooklyn Navy Yard Businesses Mobilize To Help Fight Coronavirus Pandemic. At the beginning of the crisis one of the most crucial needs were Ventilators. “New York Needed Ventilators. So They Developed One in a Month”, an emergency project was established and led by the New Lab with a team of experts to build open source ventilators that are low cost. The Foamspace team is proud to have contributed to those efforts and provide our programming expertise and want to share in full the work we have been pursuing. Thank you for reading and we look forward to sharing our next update.
