THE IEEE 802.15.4 LOW-RATE WPAN STANDARD
As noted in Chapter 1, the scope of the IEEE 802.15.4 task group, as
defined in its original Project Authorization Request, is to "define the PHY and
MAC specifications for low data rate wireless connectivity with fixed, portable
and moving devices with no battery or very limited battery consumption
requirements typically operating in the Personal Operating Space (POS) of 10
meters." Further, the purpose of the project is "[t]o provide a standard for
ultra low complexity, ultra low cost, ultra low power consumption and low data
rate wireless connectivity among inexpensive devices. The raw data rate will be
high enough (maximum of 200 kbps) to satisfy a set of simple needs such as
interactive toys, but scaleable down to the needs of sensor and automation needs
(10 kbps or below) for wireless communications."[2] The maximum and
minimum raw data rates were later raised to 250 and 20 kb/s,
respectively.[3]
This diverse set of goals requires the IEEE 802.15.4 standard to
be extremely flexible. Unlike file transfer protocols such as IEEE 802.11 that
are designed for a single application, the IEEE 802.15.4 standard supports a
nearly infinite variety of possible applications in the POS. These applications
vary from those requiring high data throughput and relatively low message
latency, such as wireless keyboards, mice, and joysticks, to those requiring
very low throughput and able to tolerate significant message latency, such as
intelligent agriculture and environmental sensing applications. The IEEE
802.15.4 standard supports both star and peer-to-peer connections, and is
therefore able to support a wide variety of network topologies and routing
algorithms. When security is used, the AES-128 security suite[4] is required.
The standard employs beacons, although their use is optional. The
beacon period is variable in binary multiples of 15.36 ms, up to a maximum of
15.36 ms × 214 = 4 minutes, 11.65824 seconds, so that the optimum
trade-off can be made between message latency and network node power consumption
for each application. Beacons may be omitted for applications that have duty
cycle limitations, as can happen on networks in the 868 MHz band (which has
regulatory limits on network node duty cycle), or applications that require
network nodes with constant reception. Channel access is contention based, via a
carrier sense multiple access mechanism with collision avoidance (CSMA-CA); the
beacon is followed by a contention access period (CAP) for devices attempting to
gain access to the channel. The length of the CAP is adjustable as a fraction of
the period between beacons. A "battery life extension" mode is also available
that limits the CAP to a fixed time of approximately 2 ms. To address the needs
of applications requiring low message latency, the standard supports the use of
optional guaranteed time slots (GTSs), which reserve channel time for individual
devices without the need to follow the CSMA-CA access mechanism.
The standard has a 16-bit address field, meaning that up to
(28 − 2) × (28 − 2) = 64,516 devices may be assigned logical addresses
(two values in each byte are reserved); however, the standard also includes the
ability to send messages with 64-bit extended addresses, allowing an almost
unlimited number of devices to be placed in a single network. Message
transmission can be fully acknowledged; each transmitted frame (with the
exception of beacons and the acknowledgments themselves) may receive an explicit
acknowledgment. This produces a reliable protocol; the overhead associated with
explicit acknowledgments is acceptable given the low data throughput typical of
wireless sensor networks. The use of acknowledgments is optional with each
transmitted frame, however, to support the use of passive acknowledgment
techniques. These techniques are used in some ad hoc routing schemes, for example, the gradient routing
(GRAd) algorithm discussed in Chapter 5.
The IEEE 802.15.4 standard incorporates many features designed to
minimize power consumption of the network nodes. In addition to the use of long
beacon periods and the battery life extension mode, the active period of a
beaconing node can be reduced (again by powers of two), allowing the node to
sleep between beacons.
Coexistence with other services using the unlicensed bands with
IEEE 802.15.4 devices was also a major factor in the protocol design, and is
evident in many of its features. For example, dynamic channel selection is
required; should interference from other services appear on a channel being used
by an IEEE 802.15.4 network, the network node in control of the network (the
personal area network [PAN] coordinator) scans other available channels to find
a more suitable channel. In this scan, it obtains a measure of the peak energy
present in each alternative channel and then uses this information to select a
suitable channel. This type of scan can also be used prior to the establishment
of a new network. Prior to each frame transmission (other than beacon or
acknowledgment frames), each IEEE 802.15.4 network node must perform two clear
channel assessments (CCAs) as part of the CSMA-CA mechanism to ensure the
channel is unoccupied prior to transmission.
A link quality indication (LQI) byte is attached to each received
frame by the physical layer before it is sent to the medium access control
layer. The receiving node expects to used this information for a number of
purposes, at the discretion of the network designer:
-
It can be used as an indication of channel impairment,
perhaps leading to the need to perform the dynamic channel selection process and
move to another channel.
-
It can be used for power control of its own transmitter,
under the assumption of a symmetrical channel.
-
It may be used as part of a relative location determinatin
algorithm, to estimate the location of each network node relative to its
peers.
-
It may be used as part of a network routing algorithm to
establish packet routes based only on link quality between network
nodes.
The LQI may be generated from a signal level determination, a
signal-to-noise determination, or a combination of the two, at the discretion of
the network node implementer. This enables both received signal strength
indication (RSSI) and correlation-based signal quality estimators to be used.
Although a byte (8 bits) is reserved for the LQI, to ease the burden on
implementers that do not desire to make use of it (an appropriate decision for
some applications), the standard specifies that at least eight unique values shall be used in the LQI,
including 0 × 00 and 0 × FF are to be associated with the lowest and highest
quality IEEE 802.15.4 signals detectable by the receiver, respectively.
To maximize the utility of the standard, the IEEE 802.15.4 task
group had to balance the desire to enable small, low-cost, and low-power network
nodes with the desire to produce a standard that met a wide variety of market
applications. The resulting standard includes three types of network node
functionality:
-
PAN coordinator. The PAN coordinator
is the node (strictly speaking, the coordinator node) that initiates the network
and is the primary controller of the network. The PAN coordinator may transmit
beacons and can communicate directly with any device in range. Depending on the
network design, it may have memory sufficient to store information on all
devices in the network, and must have memory sufficient to store routing
informatin as required by the algorithm employed by the network.
-
Coordinator. The coordinator may
transmit beacons and can communicate directly with any device in range. A
coordinator may become a PAN coordinator, should it start a new network.
-
Device. A network device does not
beacon and can directly communicate only with a coordinator or PAN
coordinator.
These three functions are to be embodied into two physically
different device types:
-
Full function device (FFD)). An FFD
can operate in any of the three network roles (PAN coordinator, coordinator, or
device). It must have memory sufficient to store routing information as required
by teh algorithm employed by the network.
-
Reduced function device (RFD). An RFD
is a very low cost device, with minimal memory requirements. It can only
function as a network device.
The archetypical RFD is the wireless light switch. It must be as
inexpensive to produce as possible, is likely to be battery-powered, and has
very limited functional requirements, needing to communicate only with a light.
The light itself, however, may be the archetypical FFD because it can be
(slightly) more expensive, has access to mains power, and can have additional
network functions as a more permanent feature of the building.
As noted earlier, the IEEE 802.15.4 standard supports multiple
network topologies. In the standard, two general types are discussed — star
networks and peer-to-peer networks. In the star network, the master device is
the PAN coordinator (an FFD), and the other network nodes may either FFDs or
RFDs. In the peer-to-peer network, FFDs are used, one of which is the PAN coordinator. RFDs may be used in a
peer-to-peer network, but they can only communicate with a single FFD belonging
to the network, and so do not have true "peer-to-peer" communication.
Similar to all IEEE 802 wireless standards, the IEEE 802.15.4
standard standardizes only the physical and medium access control (MAC) layers.
The IEEE 802.15.4 standard, in fact, incorporates two physical layers:
-
The lower band: 868.0–868.6 MHz (for
Europe), plus the 902–928MHz (for much of the Americas and the Pacific Rim)
-
The upper band: 2.400–2.485 GHz
(substantially worldwide)
The channel numbers and their center frequencies are defined as
follows:
where k is the channel number.
Both lower and upper bands employ a form of direct sequence
spread spectrum (DSSS). In the lower band, binary phase shift keying (BPSK) with
raised-cosine pulse shaping is employed. In the 868-MHz band, a data rate of 20
kb/s and a chip rate of 300 kc/s are used, while in the 902–928-MHz band, a data
rate of 40 kb/s and a chip rate of 600 kc/s are used. In the upper band, a
modified version of the scheme described in Chapter 3 is used;[5] offset quadrature phase shift
keying (O-QPSK) with half-sine pulse shaping is employed at a chip rate of 2
Mc/s, along with a 16-ary orthogonal symbol scheme sent at 62.5 ksymbols/s,
resulting in a data rate of 250 kb/s. The PN sequences of each of the orthogonal
symbols are related to each other through cyclic shifts and/or conjugation
(inversion of chips transmitted on the Q-channel). Because all of the possible
received symbols may be easily derived from a single PN sequence, this scheme
simplifies the design of the receiver in comparison with other 16-ary DSSS
methods that would require the storage of sixteen PN sequences.
Sections
