Image Processing Reference
In-Depth Information
share the same channel, and frequency agility is minimal. There is no frequency hopping, and the
only option is to scan for a channel with the least amount of interference at startup.
There are two classes of network devices in ZigBee: full-function devices (FFD) and reduced-
function devices (RFD). The former can route messages in mesh networks and act as the network
coordinator, whereas the latter can only communicate with one FFD in a star network setup.
ZigBee can operate in both beaconed and nonbeaconed mode. In the beaconed mode, the nodes
are to some extent synchronized and the superframe is divided into slots. The slots in the
frame are generally contention-based, using carrier-sense multiple access with collision avoidance
(CSMA/CA). There is an option to dedicate up to seven of these slots to specific nodes in order
to increase determinism, so-called guaranteed time slot (GTS). However, support for this is not
mandatory and use of this feature might break interoperability.
.... Security
In the version of the specification, security is not mandatory. However, support for authen-
tication, integrity, and encryption for both network and application layer is present. MAC layer
security available through .. is not explicitly addressed in the ZigBee standard, and its use
might break interoperability between different vendors' products. Replay attacks are protected against
using sequential numbering. ZigBee makes use of the security mechanisms in ..; Counter with
CBC-MAC
∗
(CCM) with AES- encryption, but with the option to employ encryption-only or
integrity-only.
Three key types are used: master key, link key, and network key. he master key is necessary to be
abletojointhenetwork.helinkkeyisusedforend-to-endencryptionandwouldbythatprovide
the highest level of security at the price of higher storage requirements. The network key is shared
between all devices, and thus presenting a lower level of security, though with the benefit of reduced
storage requirements in the devices. All keys can be set at the factory, or be handed out from the trust
center (residing in the network coordinator), either over the air or through a physical interface. For
commercial grade application, the trust center can control the joining of new devices and periodically
refresh the network key.
27.5.2.2 WirelessHART
.... Architecture
WirelessHART was designed based on a set of fundamental requirements: it must be simple (e.g.,
easy to use and deploy), self-organizing and -healing, flexible (e.g., support different applications),
scalable (i.e., fit both small and large plants), reliable, secure, and support existing HART technology
(e.g., HART commands, configuration tools, etc.).
Figure . shows that the architecture of WirelessHART mimics the OSI layer design. Wire-
lessHART is based on the PHY layer specified in the IEEE ..- standard [],
†
but specifies
a new data-link (including MAC), network, transport, and application layers. he figure also shows
the backward compatibility of WirelessHART because it shares the Transport and Application layer
with HART.
.... Basic Functionality
WirelessHART is a TDMA-based network. All devices are time synchronized and communicate in
prescheduled fixed length time-slots. TDMA minimizes collisions and reduces the power consump-
tion of the devices.
∗
Cipher Block Chaining-Message Authentication Code.
†
WirelessHART only supports the globally available . GHz ISM frequency band.
