| Abstract: |
Wireless control systems consist of sensing and actuating devices that are commonly driven by a central controller. Wireless communication protocols for Cyber-Physical Systems (CPS) match
this design by employing a ”sense -> collect -> process -> disseminate -> actuate” flow, where typically different protocols are employed for collecting sensor data and
disseminating actuation signals. In this paper, we depart from this traditional design and introduce CaptureCom. By relying on capture effects and in-network processing, it omits the need for a
central controller and for distinct collection and dissemination phases. In CaptureCom, each node transmits its current data (e.g., temperature reading). Upon receiving, nodes integrate (e.g.,
aggregate) the received data with previously received data and concurrently forward the result. Due to capture effects, neighboring nodes correctly receive one of the concurrently sent packets
with high probability. Repeating this process, the network converges to a stable state where all nodes have received the same data (consensus). The impact of our approach is threefold: 1.
CaptureCom closes the loop in CPS: data are processed within the network. Upon completion, all nodes have received with high probability the same data as the base for actuation. Thus, it departs
from the widespread architecture of collecting information at a central controller for processing and then disseminating the results. 2. CaptureCom exploits spatial diversity in low-power
wireless networks: Consecutive, concurrent transmissions spread out across the network allow for data distribution at very low delays and high energy efficiency. 3. Relying solely on concurrent
forwarding and capture effects for communication, CaptureCom simplifies the networking stack by obviating the need for link estimation, neighbor discovery, and routing protocols. |
