Agriculture Reference
In-Depth Information
for low-power wireless sensor networks. This group of developers has now grown
into an international consortium, the TinyOS Alliance, which makes TinyOS the
de facto
standard OS for WSNs. The latest released version, TinyOS 2.1.1, supports
most commercial WSN hardware platforms (TinyOS, 2012). TinyOS programs are
very robust and efficient and supports low-power operations including multi-hop net-
working, network-wide submillisecond time synchronization, data collection to a
designated root or gateway, reliable data dissemination to every node in a network,
and installing new codes over the wireless network.
The scheduling mechanism used in TinyOS is specifically designed to fit with
WSN platforms with limited system resources, low-power consumption, and high
concurrency.
TinyDB (2010) was developed to handle the data and extract useful information
running. It provides a simple, SQL-like interface to allow users to easily inquire, fil-
ter, aggregate, and route the data through power-efficient in-networking processing
algorithms without writing NesC code. Table 13.2 summarizes the operation systems
developed for WSN applications.
13.4.3 P
OWER
S
OURCE
FOR
W
IRELESS
S
ENSOR
N
ETWORK
Most WSN platforms are powered by batteries, such as alkaline, cell, and lithium
batteries, which can be replaced when needed. Although many manufacturers claim
long battery life for their products, users often face much shorter battery life of
the power sources, especially when external sensors, high sampling rate, and high
data transmitting rate are used and extreme environment conditions are encountered.
To develop long-lasting and truly autonomous wireless WSNs, consistent and stable
power sources are crucial. Now the ultralow power consumption design on WSN
hardware and software lessen the energy requirement for operations. Microscale
energy harvesting (at micro- or milliwatt levels) from ambient environment became
a feasible approach to be incorporated with WSN applications. In the 2010-2011
Energy Harvesting Report, IDTechEx, a consulting firm in printed electronics,
RFID, and energy harvesting, forecasts a market of more than $2 billion in 2016
just for the harvesting elements, excluding power storage and electronic interfaces,
and nearly 10 billion energy harvesting devices are projected to be sold in 2020
(IDTechEx, 2010). Harvesting energy from ambient environment offers an excit-
ing future for long-term WSN deployments, especially in agricultural applications.
Currently, research and development and field testing on energy harvesting and stor-
age technologies are still going on.
The success of WSNs is determined by the quality of data generated, which need
to be processed, filtered, interpreted, stored, and displayed to end users. On one
hand, WSNs provide users detailed knowledge of the target environment, and on
the other hand, they create a huge load for data processing and handling. Limited
resource on WSN components also challenges traditional methods on the data man-
agement. In WSN applications, the data are commonly used in two ways: (1) queries
on current data and (2) queries on historical data (Diao et al., 2005). The current data
are often used for decision-making to determine control operations. For example, if a
soil moisture level is below a predefined threshold, a water pump should be powered
