TeamPhone: Networking Smartphones for Disaster Recovery

Zongqing Lu, Guohong Cao, Thomas La Porta

Introduction

In the immediate aftermath of a major disaster, such as an earthquake or hurricane, the failure of centralized cellular infrastructure often creates a "communication vacuum." This lack of connectivity severely hampers rescue efforts and leaves survivors isolated. While specialized emergency communication tools exist, they are rarely available to the general public at the moment of crisis. TeamPhone addresses this challenge by leveraging the devices people already carry: smartphones. By integrating ad-hoc networking, opportunistic routing, and energy-efficient protocols, TeamPhone transforms standard mobile devices into a resilient lifeline for disaster recovery.

TeamPhone network architecture in a disaster scenario, illustrating the interaction between rescue workers, survivors, and command centers through various communication links.

The Challenges of Emergency Networking

The authors identify three critical hurdles to using smartphones in disaster scenarios:

  1. Infrastructure-less Communication: The system must function without reliance on cellular towers or fixed internet access.
  2. Battery Life: Continuous wireless broadcasting (WiFi/Bluetooth) for peer discovery can drain a smartphone in hours, whereas rescue operations may last for days.
  3. Localization: GPS frequently fails in indoor environments or beneath debris, making it difficult for survivors to report their exact locations to rescuers.

System Architecture and Methods

TeamPhone is an Android-based system consisting of two primary subsystems: the Messaging System and the Self-Rescue System.

The Messaging System

Designed primarily for rescue workers, the Messaging System employs a hybrid communication strategy. It prioritizes standard cellular connections when available. In the absence of infrastructure, it switches to Ad-hoc networking (AODV) to facilitate multi-hop communication between devices. If the network becomes partitioned, the system utilizes Opportunistic networking (DTN2). This "store, carry, and forward" approach treats mobile users as "data mules," using their physical movement to bridge gaps in the network and ensure eventual message delivery.

The Self-Rescue System

The Self-Rescue System focuses on extreme energy efficiency for survivors. Rather than requiring every device to remain active, TeamPhone uses graph theory to organize nearby phones into "cliques." The system then implements a Distributed Wake-up Schedule based on the concept of "Maximal Independent Sets."

Technical illustration of the one-hop network cliques and the corresponding distributed wakeup schedules designed to minimize energy consumption.

As illustrated in the figure above, this scheduling ensures that at any given time, at least one node within a clique is awake to listen for signals from rescue teams, while the remaining nodes stay in a low-power sleep state. This coordination provides nearly the same discovery coverage as an always-on network but at a fraction of the energy cost.

Relative Positioning

To address the limitations of GPS, TeamPhone builds a relative coordinate system. By analyzing WiFi signal strength (RSSI) between devices in a clique, the system uses triangulation to map the positions of survivors relative to one another. This provides rescue teams with a local map of survivor clusters, which is often more actionable than inaccurate GPS coordinates in a disaster zone.

Key Results and Evaluation

The system was evaluated using a physical testbed of four Samsung Galaxy S3 smartphones and extensive topological simulations.

  • Energy Efficiency: The wake-up scheduling algorithm achieved approximately 70% energy savings compared to constant radio activity. This can extend a device's operational life from roughly 40 hours to over 120 hours, a critical factor in multi-day rescue operations.
  • Network Performance: In ad-hoc mode, the system achieved a maximum throughput of 20 Mbps at close range. Even at a distance of 100 meters, it maintained a throughput of 2 Mbps, sufficient for transmitting text, voice, and low-resolution images.
  • Latency Reduction: By integrating opportunistic routing, the system reduced round-trip delays from 60ms to 20ms in specific multi-hop scenarios by identifying more efficient paths through moving nodes.

Limitations and Future Work

Despite its potential, TeamPhone faces several hurdles for mass adoption. Currently, the system requires "root" privileges to modify WiFi drivers for ad-hoc mode, necessitating cooperation from OS developers or manufacturers. Additionally, future work will focus on balancing the "awake" duty based on individual battery levels and improving positioning accuracy by incorporating more precise ranging technologies, such as Ultra-Wideband (UWB), to mitigate the effects of signal noise in complex environments.

Summary updated

BibTeX
@article{lu2017teamphone,
  author = {Zongqing Lu and Guohong Cao and Thomas La Porta},
  title = {{TeamPhone: Networking Smartphones for Disaster Recovery}},
  journal = {IEEE Trans. Mobile Computing},
  year = {2017},
  doi = {10.1109/tmc.2017.2695452},
}
Figure