Natural emergency and disaster scenarios are unfortunate occurrences with far reaching after effects. In addition to human casualties, natural calamities can destroy the power grid, telephone networks, and mobile phone towers. The result being, there is no alternative for disseminating critical updates such as disease and safety alerts, locations, and directions to survivors. The above problem is an outcome of the lack of network systems that are designed to survive and serve after natural disasters. In order to remedy this problem, this project is designing, implementing, and deploying a solar powered self-sustainable emergency mesh to provide critical communication infrastructure to survivors and rescue personnel.
This work employs a mesh architecture that is energy-efficient and self-sustainable. It relies on renewable energy sources such as solar to provide near-perpetual lifetime to mesh nodes while serving critical updates to survivors. Since renewable energy scavenging is notoriously unpredictable, this project uses a clean-slate low power hardware and software systems design for the nodes. Additionally, in the event of node failures due to variability inherent in renewable energy scavenging and extreme environmental conditions during the aftermath of a natural calamity, the mesh automatically redistributes the data on the failed nodes to maintain sufficient redundancy. Finally, the mesh design uses common wireless technology such as Wi-Fi and light weight web-based services for compatibility with off-the-shelf laptops, mobile phones, and PDAs carried by disaster survivors. The culminating goal of this project is to save human lives. With the mesh infrastructure in place, human casualties during post-disaster times can be minimized. Additionally, during non-emergency scenarios, the system can be used to disseminate police alerts" and disease alerts" (e.g. swine-flu) that keep individuals well informed and cautious. In addition to a broad societal impact, this project, through educational courses provides hands-on experience to undergraduates and graduates in building mobile, embedded, and geospatial systems.
Project Summary: Disconnected environments can occur during major power outages, malicious attacks on communications infrastructure, natural disasters, or in remote operating environments. A primary challenge in such disconnected environments is access to information for end users since centralized webservices such as Google maps are unavailable. To mitigate this problem, in this project we have designed, implemented, and evaluated a solar panel powered self-sustainable emergency mesh network that can serve map data to users in a disconnected environment on common handhelds such as laptops and mobile phones. Intellectual Merit: Designing a robust fallback data dissemination mesh operational after a disaster requires addressing fundamental problems in embedded, mobile, and geospatial systems. First, our mesh node must be highly available and must operate near perpetually from energy harvested using medium sized solar panels. To address this problem, our node architecture uses a hierarchical design that combines a low power processor with a high power processor. The system maintains high availability by keeping the lower power processor on most of the time, and duty-cycling the higher power processor. Through intelligent task sharing the wakeup and prediction algorithms the system can maintain always-on operation at minimal energy consumption. Second, our system must provide useful map data to survivors of natural disasters. To this end, our system uses a custom software map stack that disseminates map data to users on their common handhelds like laptops and smartphones. These maps provide directions, generated in-situ by the map stack, for survivors from their present location to safe locations. Third, we have developed a solar panel emulation platform that uses a high gain PNP Darlington transistor to emulate solar panel behavior in the laboratory. The emulator allows in lab evaluation of the mesh node obviating the challenges of outdoor deployment and evaluation of renewable energy-driven systems. Broader Impacts: The system developed as part of the project is going to be used for two other applications--- disseminate "police alerts" and "disease alerts" that keep individuals well informed and cautious. We have applied the hierarchical system design to the design of low power wearable sensors for gesture recognition in patients with limited mobility. Additionally, we have incorporated the lessons learned as part of the project in undergraduate and graduate courses at University of Arkansas and University of Maryland. We have also created a small workforce of undergraduate and graduate students trained in various aspects of hardware systems design, GIS system development, and embedded software design through this project.
