KO LAYOUT 4/5/06 12:37 PM Page 56

WIRELESS MESH NETWORKING

EMERGING STANDARDS FOR

WIRELESS MESH TECHNOLOGY
MYUNG J. LEE AND JIANLIANG ZHENG, CITY UNIVERSITY OF NEW YORK
YOUNG-BAE KO AND DEEPESH MAN SHRESTHA, AJOU UNIVERSITY, KOREA

ABSTRACT the PSTN, wireless mesh technology allows

designers to build electronic networks without
Wireless mesh networking is a promising ripping apart buildings or tearing up streets to
technology for numerous applications which wire miles of copper or fiber cables. Yet it has a
appeals especially to those applications that can- flexible coverage and can seep where it is likely
not be directly supported by other wireless tech- to be beyond the reach of other wired and wire-
nologies. The commercial success of Wi-Fi and less technologies. Besides, mesh connectivity sig-
the advances in many wireless technologies have nificantly enhances network performance,
in part stimulated the development of wireless including fault tolerance, load balancing, and
mesh networks. The persistent driving force, throughput. In addition, the self-configuring and
however, comes from the envisioned advantages self-healing features of wireless mesh networks
MP of wireless mesh techniques themselves, includ- not only enable them to be deployed on the fly
ing extended coverage, robustness, self-configu- and on the cheap, but also enhance system
ration, easy maintenance, and low cost. This resilience and reliability. Moreover, with mini-
article presents an overview of this emerging mal up-front investment and being easily
technology, focusing on the technical merits and adjustable and expandable, wireless mesh net-
related standards activities of IEEE and ZigBee works cater to the requirements of various con-
that are actively applying the concept of multi- sumers, large or small. With all these advantages,
hop mesh techniques in the field of wireless net- wireless mesh technology will open a world of
The commercial working technologies, ranging from personal possibilities and develop a burgeoning market in
success of Wi-Fi and area networks (PANs) to metropolitan area net-
works (MANs).
the foreseeable future, even though considerable
research efforts are still needed.
the advances in As the World Wide Web has revolutionized

many wireless INTRODUCTION the way people acquire information and suc-
ceeded in creating a huge market, engineers are
A disruptive technology, wireless mesh network- now trying to weave another web, but one with-
technologies have in ing, is knocking on the doors of the communica- out threads. Although wireless mesh technology
tion industry. While this technology is not new is still in its infancy, its potential to likely trans-
part stimulated the conceptually, it was not until recently that form our world appears enormous. Some of the
development of researchers and enterprises, as well as con-
sumers, have begun to think of this technology
scenarios and applications where wireless mesh
technology is likely to provide a more versatile
wireless mesh as something over the next hill rather than pie in or affordable solution than other wired or wire-
the sky. Wireless networks have been making less technologies include, but are not limited to,
networks. The inroads into private residences, office buildings, the following:
universities, and other industrial and commercial • Extensive coverage areas, for example, offices,
driving force, venues around the globe in the past several campus networking, stadiums, or spanning a
years. This commercial success together with sprawling facility
however, comes some emerging wireless technologies, especially • Areas that are unwired, under-wired, or hard-
from the envisioned radio technologies, suggest it is now time for
wireless mesh networking to come into play.
to-wire, such as highways, conduits, golf cours-
es, or farmlands
advantages of Seeing its potential to reshape the landscape of • Emergency situations such as fire fighting, dis-
communications, major consumer-electronics aster recovery, and military operations.
wireless mesh companies as well as small startups are staking While a few companies have been rolling out
out this emerging technology and preparing for proprietary wireless mesh products for some
techniques the market to take off. time, the involvement of international standard
Both the wired Internet and the public groups, the major driving force behind various
themselves. switched telephone network (PSTN) are essen- technologies, has signaled the arrival of the wire-
tially mesh networks that have long been pre- less mesh era. As can be seen from Table 1,
sent. In stark contrast to the wired Internet and IEEE has been playing a key role in the devel-

56 1536-1284/06/$20.00 © 2006 IEEE IEEE Wireless Communications • April 2006

KO LAYOUT 4/5/06 12:37 PM Page 57

opment of wireless mesh standards with network

Types of mesh networks and Corresponding standards used as a
coverage ranging from PAN to MAN. The
related marketing alliances basis
remainder of this article presents an overview of
wireless mesh standards activities of IEEE that
WMAN mesh (WiMAX) IEEE 802.16a
includes IEEE 802.16, 802.11s, 802.15.5, and
those of ZigBee.
WLAN mesh (Wi-Fi) IEEE 802.11s
STANDARDS FOR WIRELESS MAN MESH LR-WPAN mesh (ZigBee) IEEE 802.15.5
The IEEE 802.16 working group (WG) defines
the physical (PHY) layer and the medium access
n Table 1. Wireless mesh standards.
control (MAC) sublayer standards targeted for
wireless networking in metropolitan area net-
works (MANs). The IEEE 802.16 WirelessMAN
standard [1], associated with Worldwide Interop-
erability for Microwave Access (WiMAX) and BS Mesh BS
published in April 2002, was designed to operate
in the licensed 10–66 GHz frequency range that
requires line-of-sight (LOS) towers covering up
to 5 km, similar to base stations (BS) in cellular
communications. It builds a backhaul network
for broadband wireless access at much lower
cost, as compared to existing wired counterparts
such as DSL and cable. This standard was initial-
ly created for point-to-multipoint (PMP) broad-
SSs
band applications, with an aim to provide higher (a) (b)
data rate (up to 75 Mb/s) for each subscriber
station (SS). Following this standard, IEEE
802.16 WG has set up several task groups (TGs)
n Figure 1. Illustration for the IEEE 802.16a standards: a) point-to-multipoint
mode; b) mesh mode.
from ‘a’ to ‘g’ to address several extensions in
fixed and portable/mobile broadband wireless
access (BWA) in metropolitan areas as follows: for any new node to synchronize with the exist-
• a — Addition of a mesh mode ing network, which contains network configura-
• b — Providing quality-of-service (QoS) feature tion information such as the mesh BS
• c — Supporting interoperability identification number and base channel in use.
• d — Extensions of PHY layer The new node (also called a candidate node)
• e — Supporting mobility actively scans for the existing network. Hearing
• f — Supporting multihop functionality in IEEE the MSH-NCFG message, it establishes synchro-
802.16e nization and initiates a network entry process. If
• g — Providing efficient handover and QoS there are many nodes sending MSH-NCFG mes-
To focus our attention on mesh-related activi- sages, it chooses one potential sponsoring node
ties, in the following subsection we describe the and sends a MSH-NENT (mesh network entry)
enhancement in the basic IEEE 802.16 standard message with a request to join the mesh net-
(i.e., the 802.16a mesh mode) and the new study work. After authorization, it receives a 16-bit
group named mobile multihop relay (MMR). node identifier (i.e., node ID) upon request to
the mesh BS, which uniquely identifies the node
IEEE 802.16 MESH within a mesh during a normal operation and is
The IEEE 802.16a standard [2] incorporates carried by all unicast and broadcast frames.
what is known as the “mesh mode” in addition Two types of TDMA-based packet scheduling
to the PMP mode defined in IEEE 802.16. This mechanisms for channel access are available in
new standard operates in the licensed and unli- the mesh mode: centralized scheduling and dis-
censed lower frequencies of 2–11 GHz that tributed scheduling. In the centralized approach,
allows non-line-of-sight (NLOS) communica- a mesh BS allocates the radio resources for all
tions, spanning up to a 50 km range. SSs within a certain hop range. In distributed
A key difference between the PMP mode and scheduling, all nodes coordinate with each other
the mesh mode is the ability of latter to enable for accessing the channel, including the mesh
multihop communications. While the PMP mode BS. Medium-access coordination between sta-
requires each SS to be connected to a BS, neigh- tions is performed by broadcasting the individual
bors can directly communicate with each other schedules to all neighboring nodes.
in the mesh mode (see Fig. 1). Hence, an SS in The IEEE 802.16a mesh mode has two limi-
the mesh mode serves as a router relaying traffic tations. First, its operation targets only fixed
between the SSs, until it arrives at the target BS broadband applications. Second, it is not com-
(also called the “mesh BS”) that connects the patible with the existing PMP mode. To address
mesh to backhaul link and other external net- these limitations, another study group (SG)
works. called the “Mobile Multihop Relay (MMR)” was
A new node joining the mesh network needs established in July 2005 under the 802.16 WG.
to go through the network entry and self-config- The main purpose of this SG is to study the pos-
uration process. Active nodes that are the part sibility for extending the PMP mode for an SS
of mesh network periodically advertise MSH- outside the coverage of a BS and to support
NCFG (mesh network configuration) messages mobile stations (MS) by using multihop relaying

IEEE Wireless Communications • April 2006 57

KO LAYOUT 4/5/06 12:37 PM Page 58

Mesh” was formed in May 2004 under the 802.11

WG to address the need for wireless mesh in
802.11s mesh links WLANs. In the following subsections, we pre-
sent a brief overview of the ongoing IEEE
MPP
802.11s activities followed by the description of
STAs its proposed network architecture and the primi-
tive services provided by the medium access co-
ordination function (MCF) sublayer.

IEEE 802.11S: GENERAL OVERVIEW

The IEEE 802.11s extended service set (ESS)
mesh aims at applying multihop mesh techniques
MP to specify a WDS that can be used to build a
wireless infrastructure for small-to-large-scale
MP MP WLANs. Hence, the ESS or WLAN mesh can
be considered as an IEEE 802.11-based WDS, a
MAP subset of the DS that consists of a set of devices
interconnected with each other via wireless links,
resulting in a ‘mesh of connectivity.’
Legacy 802.11s links The activities of 802.11s TG comprise the
n Figure 2. IEEE 802.11s network architecture. specification of a new protocol suite for the
installation, configuration, and operation of
WLAN mesh. Its implementation shall be atop
the existing PHY layer of IEEE 802.11a/b/g/n
techniques using relay stations (RSs). An RS operating in the unlicensed spectrum of 2.4-
relays information between an SS/MS and a BS and 5-GHz frequency bands. The specification
or between other RSs or between an RS and a shall include the extensions in topology forma-
BS [3]. Hence, unlike the mesh mode, dedicated tion to make the WLAN mesh self-configure as
RSs forms a treelike topology for relaying the soon as the devices are powered up. A path
traffic to the BS. The MMR SG submitted a selection protocol will be specified in the MAC
project authorization request (PAR) for approval layer instead of network layer for routing data
to create a TG named TGj at the March 2006 in the multihop mesh topology. This standard is
meeting. expected to support MAC-layer broadcast/mul-
ticast in addition to the unicast transmissions.
STANDARDS FOR WIRELESS LAN MESH This standard shall also accommodate devices
The IEEE 802.11 family of standards is currently that are able to support multichannel opera-
the most successful wireless networking stan- tions, or are equipped with multiple radios,
dards for wireless LANs. The initial specification with an aim to boost the capacity of the overall
was completed in 1999 (IEEE 802.11a/b) [4] and network. The specification is expected to be
later extended in 2003 (IEEE 802.11g). It defines adopted as part of the working group standard
the PHY and MAC sublayer for the devices used by March 2008.
in WLAN networking. It includes a dedicated
device called an access point (AP), to which end- PROPOSED IEEE 802.11S
user devices or stations (STAs) with 802.11-com-
pliant network interface cards connect for NETWORK ARCHITECTURE
accessing network services. Such a set of STAs The proposed WLAN Mesh architecture is
managed by a single AP is called a basic service depicted in Fig. 2. Any IEEE 802.11-based entity
set (BSS). An independent BSS (IBSS), also (either AP or STA) that partially or fully sup-
popularly referred to as an “ad hoc network,” is ports a mesh-relay function is defined as a mesh
a self-contained network in which STAs can point (MP). The minimal MP operations include
directly communicate with each other without neighbor discovery, channel selection, and form-
APs. A set of one or more BSS interconnected ing an association with neighbors. Besides, MPs
by a distribution system (DS) forms an extended can directly communicate with their neighbors
service set (ESS), and can be regarded as a sin- and forward traffic on behalf of other MPs via
gle 802 network segment. A DS contains either a bidirectional wireless mesh links. A set of MPs
single AP in a standalone network or a wired and the mesh links form a WDS, which distin-
(cables)/wireless system that connects APs. guishes itself from the BSS as defined in the
IEEE 802.11 standards continue to advance legacy IEEE 802.11[4].
with various amendments, for example, 802.11e The proposed WLAN mesh also defines a
for QoS and 802.11n for data rates in excess of mesh access point (MAP), which is a specific
100 Mb/s. These standards are still limited MP but acts as an AP as well. The MAP may
because of their dependency upon the wired net- operate as a part of the WLAN mesh or in one
work and unspecified wireless distribution sys- of the legacy 802.11 modes. A mesh portal
tem (WDS). Furthermore, 802.11 standards (MPP) is yet another type of MP through which
primarily aim at fulfilling one-hop communica- multiple WLAN meshes can be interconnected
tion needs and hence are affected by the prob- to construct networks of mesh networks. An
lems of throughput degradation and unfairness MPP can also co-locate with an IEEE 802.11
when applied to multihop networks. portal and function as a bridge/gateway between
Motivated by the foregoing issues and limita- the WLAN mesh and other networks in the DS.
tions, a separate TG called “IEEE 802.11s ESS To uniquely identify a WLAN mesh, a common

58 IEEE Wireless Communications • April 2006

KO LAYOUT 4/5/06 12:37 PM Page 59

mesh ID is assigned to each MP, similar to the

use of service set identifier (SSID) to represent
an ESS in legacy 802.11 networks.

THE MEDIUM ACCESS COORDINATION FUNCTION Mesh interworking with other 802 networks
The major components of the proposed 802.11s
Mesh
MCF are shown in Fig. 3. Built on top of the measurement
legacy physical layer specification, 802.11s shall Mesh topology Medium access
learning, routing and coordination
explicitly provide the WLAN mesh services. forwarding
These include topology learning, routing and
forwarding, medium access coordination, mesh
configuration and management, topology discov- Discovery and
ery and association, mesh measurement, inter- Mesh security association
working, and security functions [5].
For interworking of the WLAN mesh with 802.11 service
other networks, the IEEE 802.1D standard shall Mesh configuration and management
integration
be incorporated in the mesh portals (MPPs),
which define the interworking framework and
service access interface across all 802 standards. PHYs
Similarly, security architecture shall be based on
the IEEE 802.11i standard, which specifies secu-
rity features for all WLAN networks. In the fol- n Figure 3. Architecture for the 802.11s MCF sublayer.
lowing subsections, we describe only the major
services and functions germane to the mesh
capability. the 802.11-1999 specification [4], with two new
extensions for quality of service (QoS) and mesh
Mesh Topology Learning, Routing, and Forwarding — control. Upon receiving such frames, the MP
This service set focuses on the peer-to-peer dis- checks for its authenticity and its destination
covery of MPs. It enables automatic topology MAC address before forwarding. On arrival of
learning, establishes links, and eventually forms the 3-address frame in MAP from the associated
a dynamic data delivery path across the WLAN STA, the frame shall be converted to the 4-
mesh. address format and forwarded towards the desti-
Topology discovery and formation: A new nation. Similarly, forwarding of multicast and
node (a candidate MP) initially gathers informa- broadcast traffic is also supported if they are 4-
tion from neighboring MPs that belong to the address data frames from the known source. The
active WLAN mesh, by either the active scan- time-to-live subfield present in the data frame is
ning (i.e., sending probe messages) or the pas- decremented by every forwarding MP for con-
sive listening (i.e., receiving periodic beacons) trolling the broadcast traffic in the WLAN mesh.
mechanism. Based on factors such as peer capa-
bility, power saving capability, security informa- Medium Access Coordination — The medium access
tion, and link quality, two peers associate with coordination of the proposal [6, 7] is based on
each other, forming a partial or full mesh topol- the enhanced distributed channel access
ogy. (EDCA) mechanism used in 802.11e [8]. The
Path selection protocol: The MCF architec- proposed MAC mechanisms facilitate congestion
ture provides an extensible framework such that control, power saving, synchronization, and bea-
the enhanced routing protocols and metrics tai- con collision avoidance. Using the proposed
lored for particular applications can be imple- mechanisms, it shall be possible to enable multi-
mented and used as required. A layer-2 path ple channel operations in multiradio or single
selection protocol is supposed to handle unicast radio, as well as mixed, environments. The mul-
and broadcast/multicast data delivery in the tichannel MAC proposal is based on the com-
WLAN mesh. Since the network might have mon channel framework (CCF) [9], which is
both nonmobile and mobile MPs, a hybrid rout- compliant to the legacy channel access mecha-
ing protocol that includes both proactive and on- nisms. It is primarily designed for efficient use of
demand schemes is expected to be more suitable. link capacity by enabling multichannel opera-
Therefore, a hybrid scheme of using the ad hoc tions in nodes that are equipped with a single
on-demand distance vector (AODV) and the radio interface.
optimized link state routing (OLSR) protocol is
proposed to support wide range of application Mesh Configuration and Management — Self-configur-
scenarios [6,7]. In addition, radio-aware metrics ing paths and links offer one of the main advan-
that reflects actual link condition are proposed, tages of mesh networks. Since their deployment
thereby making the routing protocols more can be unmanaged, autonomic management
robust against link failures. For example, an air- modules are required for their continuous oper-
time metrics [6] reflects the cost of channel, ation. Protocols for association between MPs
path, and packet error rate. Similarly, another and the nodes outside the WLAN mesh can
metric named weighted radio and load aware minimize the burden of manual configuration
(WRALA) [7] is based on the protocol overhead for the service provider.
at the MAC and PHY layers, size of the frame, Mesh management ensures the smooth oper-
bit rate, link load, and error rate. ation of the network. Since any available MP
Forwarding scheme: WLAN mesh traffic con- can route packets, a failure of a particular
sists of 4-address data frames similar to that of device is not likely to affect the network as a

IEEE Wireless Communications • April 2006 59

KO LAYOUT 4/5/06 12:37 PM Page 60

• During the initialization phase, nodes join the

network and an ART tree is formed. After
A [8][6]
initialization, the network enters operation
phase, in which normal communications start.
• During the operation phase, new nodes are
B J still allowed to join the network, and if there
[5][1] [5] is a substantial change of either the number of
nodes or the network topology, the network
may need to be reconfigured.
• If the tree is broken, then the recovery phase
C H K
[1][2][1] [3][1] is triggered. Notice that recovery phase is dif-
[1] ferent from the other two phases in that only
the affected part of the tree needs to enter
the recovery phase (other unaffected parts are
still in operation phase).
D E G I L O An ART tree is formed during the initializa-
[1][1]
[0] [1] [0] [0] [0] tion phase. The ART tree formation is function-
ally divided into two stages: association and
address assigning. During the association stage,
beginning from the root, nodes gradually join the
F M N network and a tree is formed. But this tree is not
[0] [0] [0] an ART tree yet, since no node has been assigned
a logic address. After the tree reaches its bottom,
n Figure 4. Calculation of number of nodes along each branch. a down-top procedure is used to calculate the
number of nodes along each branch, as shown in
Fig. 4. The numbers in brackets indicate the
numbers of nodes within branches below a cer-
whole. However, the system should still be able tain node. When the numbers of nodes are
to report the malfunctioning of devices. Fur- reported from bottom to top, each node can also
thermore, the interfaces need to support indicate a desirable number of addresses. The
802.11h so as to enable compliance with dynam- ultimate result of address assignment is that each
ic frequency selection (DFS) requirements. node has an ARTT built. During the operation
Support for radio frequency auto-configuration phase, link failures or routing node failures will
is expected to be provided for efficient multi- trigger the recovery phase. The ARTT is con-
hop transmissions, power saving, and improving structed in such a way that tree repair and recov-
the total capacity. ery can be accomplished without changing any
assigned address. The details of routing and tree
STANDARDS FOR WIRELESS PAN MESH repair/recovery can be found in [11].

IEEE 802.15.5 Meshed Adaptive Robust Tree — A meshed ART

(MART) can be formed on top of an ART. In
IEEE 802.15.4 [10] specifies the PHY layer and Fig. 5, where the original ART is connected
MAC sublayer functions of LR-WPANs. Cur- using black lines, additional magenta lines are
rently, the ZigBee Alliance, which uses 802.15.4 added so that the network now looks more like a
as a basis, has been working on the specifica- mesh than a tree. But from each individual
tions for network layer, application layer, and node’s point of view, the network is still a tree.
security of LR-WPANs. The IEEE 802.15.5 TG, Any two nodes connected through a magenta
kicked off in May 2004, is currently working to line treat each other as a child and adds an
provide an architectural framework for interop- ARTT entry for each other. For example, node
erable, stable, and scalable wireless mesh K treats node H as a child, and vice versa. Note
topologies for WPAN devices. Here we briefly that ancestors and descendents, no matter
describe only the low-rate mesh architecture in whether they are one level or multiple levels
the baseline document [11]. The proposal is away from the node, are not meshed (i.e., not
based on the meshed tree approach; it addresses connected to the node through the magenta
meshed tree routing, multicasting, and key pre- lines). By forming a MART, it is possible to
distribution. route a packet through a shorter path. Another
advantage of MART is that some SPFs are
Adaptive Robust Tree — The tree defined in the removed. For instance, if the link between nodes
proposal is called adaptive robust tree (ART) J and K is broken, packets from node K to node
based on the facts that logical addresses are H or I can still be routed.
adaptively assigned during the tree formation
procedure to reflect the actual network topology ZIGBEE
and that the tree is free of single point of fail- In this subsection, we briefly describe the stan-
ures (SPFs). In ART, each node keeps an ART dards activities of low-rate wireless personal area
table (ARTT) to track its branches. Each branch networks (LR-WPANs) in the ZigBee Alliance.
is assigned one or more blocks of consecutive
addresses. Functionally, three phases are defined Stack Architecture — The ZigBee stack architecture
in ART: the initialization (or configuration) is depicted in Fig. 6. As mentioned above, the
phase, the operation phase, and the recovery IEEE 802.15.4 [10] standard defines the PHY
phase. layer and the MAC sublayer for LR-WPANs.

60 IEEE Wireless Communications • April 2006

KO LAYOUT 4/5/06 12:37 PM Page 61

The ZigBee Alliance has been working on the

application (APL) layer, the network (NWK)
A
layer, and the security services. To date, the Zig-
Bee standard is the only market-ready wireless
mesh standard.
There are three logical device types defined
B J
in a ZigBee network, namely, the ZigBee coor-
dinator, the ZigBee router, and the ZigBee end
device. A ZigBee coordinator is the 802.15.4
PAN coordinator and must be a full-function
device (FFD). A ZigBee router is a FFD partici-
C H K
pating in a ZigBee network. A ZigBee router is
not the ZigBee coordinator but may act as an
802.15.4 coordinator within its personal operat-
ing space. A ZigBee end-device is either a FFD
or a reduced-function device (RFD), which is
neither a ZigBee coordinator nor a ZigBee D E G I L O
router. The capabilities and functionalities of
these three types of devices are different in the
network.

Network Layer — The network (NWK) layer pro- F M N

vides the data service and the management ser-
vice to the APL layer. Here we focus on one of
the most important functions of NWK layer — n Figure 5. Meshed ART.
mesh routing.
ZigBee routing combines tree routing with on-
demand nontree routing. Tree routes along a sin- router-capable children or to its parent. This
gle tree branch are generally optimal if the tree routing scheme is referred to as cluster-tree
has been optimized with respect to the routing routing.
cost(s) under consideration (e.g., hop count, link With the cluster-tree, a device can immedi-
quality, and power). Therefore, optimal on- ately begin to transmit data packets to other
demand nontree routes are expected to be orthog- devices once it joins the network, without going
onal with a high probability of tree routes in the through the route discovery procedure. Howev-
sense that they mainly connect different tree er, most cluster-tree routes are not optimal in
branches. As a result, tree routes and nontree terms of hop count. Cluster-tree routing also
routes interconnect all nodes and form a mesh. results in uneven traffic distribution. That is, a
Through the primitive association supported node at a smaller tree level normally needs to
by IEEE 802.15.4 [10], a logical tree, referred to handle more traffic than a node at a larger tree
as a cluster-tree [12], can be formed along with level. As such, a node at a smaller tree level dies
the setup of an LR-WPAN. The first node in the more quickly than the other nodes due to its
network will designate itself as the ZigBee coor- quick battery depletion. Without additional aid-
dinator and begin to accept association requests ing mechanisms, single point of failure (SPF)
from other nodes. Any node already in the net- and network partition could easily happen.
work can determine whether to allow other Therefore, in ZigBee networks, cluster-tree rout-
nodes to join it, that is, whether to act as a Zig- ing is combined with another on-demand table-
Bee router, depending on the availability of its driven routing, which is currently based on the
resources such as memory and energy. In the ad hoc on-demand distance vector junior (AOD-
cluster-tree, a node can have a maximum num- Vjr) [14]. AODVjr is a simplified version of the
ber of C m children and a node can be at most AODV routing protocol [15], which is capable of
Lm levels (i.e., hops) away from the root of the finding optimal or near-optimal routes, and thus
tree (Cm and Lm are two networkwide constants helps to reduce the message delivery latency.
determined by the ZigBee coordinator). By Nevertheless, it requires more memory com-
knowing C m, L m, and its own address and tree pared with cluster-tree routing to store routing
level, a node can calculate the address blocks to entries and also incurs higher control overhead.
be assigned to its children. Based on ZigBee As most routes are formed on demand, the ini-
Network Specification V 1.0 [13], a slightly dif- tial latency caused by route discovery is high. In
ferent version of cluster-tree algorithm is used. general, AODVjr is suitable for devices with suf-
Compared with the original cluster-tree [12], the ficient memories, and favors long communica-
new version distinguishes two types of devices tion sessions.
when assigning logic addresses, that is, routers ZigBee routing combines cluster-tree routing
and end devices. A router is still assigned an and AODVjr routing, and makes tradeoffs
address block, which can be further assigned to between them according to network conditions
its children, but an end device only gets a single and application requirements.
address and thus cannot have any children.
Based on the above address-assignment logic, a CONCLUSION
node can easily determine how to forward a data
packet by looking at the destination address, Many forces are drawing researchers as well as
that is, whether to forward the data packet to manufacturers to wireless technologies. The
one of its end-device children or to one of its explosive growth of wireless communications has

IEEE Wireless Communications • April 2006 61

KO LAYOUT 4/5/06 12:37 PM Page 62

Application (APL) layer

Application framework

ZDO public
interfaces
ZigBee device object
Application Application (ZDO)
object 240 object 1

Endpoint 240 Endpoint 1 Endpoint 0

APSDE-SAP APSDE-SAP APSDE-SAP

ZDO management
APSME-
Application support sublayer (APS)

SAP
APS security APS message Reflector
management broker management

plane
Security
service
provider NLDE-SAP
Network (NWK) layer

NLME-
SAP
NWK security NWK message Routing Network
management broker management management

MLDE-SAP MLME-SAP
Medium access control (MAC) layer

PD-SAP PLME-SAP
Physical (PHY) layer
2.4 GHz radio 868/915 MHz radio

IEEE 802.15.4 ZigBee alliance End manfacturer Layer Layer

defined defined defined function interface

SAP: service access point

PD: PHY data PLME: PHY layer management entity
MCPS: MAC common part sublayer MLME: MAC sublayer management entity
NLDE: NWK layer data entity NLME: NWK layer management entity
APSDE: APS data entity APSME: APS management entity

n Figure 6. ZigBee stack architecture.

driven the cost of radio devices down and the gram supervised by the Institute of Information
quality up. With the ability to reduce installation Technology Assessment (IITA), the Ministry of
costs, add flexibility, and ease deployment and Information and Communication (MIC) in
maintenance hassles, the attractiveness of wire- Korea, and the Samsung Advanced Institute of
less technologies needs little reinforcement. Technology.
Recently there has been a whole slew of research
dedicated to wireless mesh networking. Behind REFERENCES
this are some of the unique features of wireless [1] IEEE Std. 802.16-2001, “IEEE Standard for Local and
mesh networks such as cost effectiveness, extend- Metropolitan Area Networks,” 2001.
ed coverage, fault tolerance, load balancing, self- [2] IEEE Std. 802.16-2004, “IEEE Standard for Local and
configuring, self-healing, and relatively smaller Metropolitan Area Networks,” 2004.
[3] Peterson et al., “IEEE 802.16mmr-06/007, Definition of
upfront investment. Terminology Used in Mobile Multihop Relay,” 2006.
Large companies and industry alliances are [4] IEEE Std 802.11, “IEEE Standard for Local and
now actively involved in research on wireless Metropolitan Area Networks,” 1999.
mesh networks, and several IEEE standards task [5] IEEE 802.11-04-0968r11, “Issues for Mesh Media Access
Coordination Component in 11s,” Jan. 2005.
groups have also been established to work on [6] Aoki et al., “802.11 TGs Simple Efficient Extensible
new standards for wireless mesh networks. With Mesh (SEE-Mesh) Proposal,” IEEE 802 11-05/0562r01,
all these efforts and other advances in wireless 2005.
technologies, we are ushering in a new era — [7] Sheu et al., “802.11 TGs MAC Enhancement Proposal,”
IEEE 802 11-05/0575r4, 2005.
the wireless mesh era. [8] IEEE P802.11e/D13.0, “Amendment: Medium Access
Control (MAC) Quality of Service (QoS) Enhancements,”
ACKNOWLEDGMENTS Jan. 2005.
We thank the referees for their helpful com- [9] S.-W. Lee et al., “802.11 TGs MAC Enhancement Pro-
posal,” IEEE 802 11-05/0589r0, June 2005.
ments and suggestions. We would also like to [10] IEEE P802.15.4/D18, “Draft Standard: Low Rate Wire-
thank Rakesh Taori and Ji-Hoon Lee at the less Personal Area Networks,” Feb. 2003.
Samsung Advanced Institute of Technology, who [11] See http://www.ieee802.org/15/pub/05/15-05-0260-00-
reviewed the earlier drafts of this article. This 0005-802-15-5-mesh-networks.pdf.
[12] L. Hester et al., “neuRFon Netform: A Self-Organizing
work is supported by the Ubiquitous Autonomic Wireless Sensor Network,” Proc. 11th IEEE ICCCN Conf.,
Computing and Network project (The 21st Cen- Miami, FL, Oct. 2002.
tury Frontier R&D Program), the ITRC pro- [13] ZigBee Specification, v. 1.0, June 2005.

62 IEEE Wireless Communications • April 2006

KO LAYOUT 4/5/06 12:37 PM Page 63

[14] I. Chakeres and L. Klein-Berndt, “AODVjr, AODV Sim- module as well as the ZigBee network ns2 simulation mod-
plified,” ACM SIGMOBILE Mobile Comp. and Commun. ule. His research interests include wireless sensor networks
Rev., July 2002, pp. 100–101. and wireless mobile ad hoc networks, medium access con-
[15] C. Perkins, E. Belding-Royer, and S. Das, “Ad Hoc on trol, wireless mesh routing, cross-layer design, embedded
Demand Distance Vector (AODV) Routing,” IETF RFC system, network security, and IPv6 technologies.
3561, July 2003.
Y OUNG -B AE K O is currently an assistant professor in the
BIOGRAPHIES School of Information and Computer Engineering at Ajou
M YUNG J ONG L EE [SM] ([email protected]) received B.S. University, Korea, leading the Ubiquitous Networked Sys-
and M.S. degrees from Seoul National University, Korea, tems (UbiNeS) Laboratory. Prior to joining Ajou University
and a Ph.D. degree in electrical engineering from Columbia in 2002, he was with the IBM T. J. Watson Research Cen-
University, New York, in 1990. He is currently a professor ter, Hawthorne, New York, as a research staff member in
with the Department of Electrical Engineering of City Uni- the Department of Ubiquitous Networking and Security. He
versity of New York and the director of the SAIT–CUNY received his Ph.D. degree in computer science from Texas
Joint Laboratory. He has been a visiting professor to Telcor- A&M University, and B.S and M.B.A. degrees from Ajou
dia and Samsung AIT. His recent research interests are on University. His research interests are in the areas of mobile
various issues of wireless sensor/ad hoc networks. He has computing and wireless networking. In particular, he is
published more than 100 journal and conference papers, actively working on mobile ad hoc networks, wireless
and holds 20 U.S. and international patents (pending mesh/sensor networks, and various ubiquitous networked
included). He is the Chair of IEEE 802.15.5 WPAN Mesh system issues. He was the recipient of a Best Paper award
Task Group and Vice Chair of ZigBee NWG. He received the from ACM Mobicom 1998. He has served on the program
CUNY Excellence Performance Award in 1999 and the Best committees of several conferences and workshops. He also
Paper Award at IEEE CCNC 2005. serves on the editorial board of ACM Mobile Computing
and Communications Review. See http://uns.ajou.ac.kr/
JIANLIANG ZHENG received his B.S. degree in applied physics ~youngko for further details.
from Harbin University of Science and Technology in China,
his M.S. degree in physics from Shanghai Jiao Tong Univer- DEEPESH M. SHRESTHA ([email protected]) received his B.E
sity in China, and his Ph.D. degree in electrical engineering in computer engineering from Kathmandu University,
from the City University of New York. He has published six Nepal, and his M.E in computer science from BITS, Pilani,
journal papers and seven conference papers. He also holds India. He is currently a Ph.D. student in the Ubiquitous
five U.S. and international patents (pending). He has been Networked System Laboratory at Ajou University, Korea.
actively participating in international standardizations, such His research interests include routing and medium access
as IEEE 802.15.5 Mesh WPAN TG and ZigBee Alliance. He control protocol design for wireless mesh and sensor net-
contributed the IEEE 802.15.4 LR-WPAN ns2 simulation works.

IEEE Wireless Communications • April 2006 63