Transcript ad hoc wireless networks

AD HOC WIRELESS
NETWORKS
BY: Radwa Bayoumy
OUTLINE
Introduction
Application
Design Principles and Challenges
Protocol Layers
Cross-Layer Design
Network Capacity Limits
Energy-Constrained Networks
Introduction:
An ad-hoc (or "spontaneous") network
is a local area wireless or temporary plugin connections.
In Latin, ad hoc literally means "for
this," further meaning "for this purpose
only,"
Applications
An ad-hoc network has been applied to
future office or home networks in which
new devices can be quickly added, using,
for example, Bluetooth technology in
which devices communicate with the
computer and perhaps other devices using
wireless transmission.
Examples of applications
Data Networks
Home Networks
Device Networks
Sensor Networks
Distributed Control Systems
Data Networks
These types of networks support data
exchange between laptops, palmtops,
personal digital assistance (PDA), and
other information devices. According to the
area being covered, the category of the
network is being identified weather it’s
LANs, MANs, and WANs.
Although the wireless LANs has a good
performance at low cost, the ad hoc have some
advantages over it. Only one access point is
needed for communication with an already wired
network and in certain cases no access point is
needed.
It’s inefficient if a node had to go through an
access point or a base station to exchange
information for example: between two PDAs
right next to each other.
In wireless MANs the multihop routing
is a necessary as they cover a large area.
For the high mobility users (example:
military programs) it’ complicated to
communicate with the lake of centralized
network control. The ad hoc network had
offered the solution for such case with a
limited success.
In Wireless WANs it’s used where
network infra structure cannot be
developed and network that must be build
up and torn out quickly as in military
application.
Home Networks
All the house electronics equipment
can be linked together by the help of the
ad hoc network. Such a network could
enable a smart room from the light
adjustment to the fire alarm that is
connected to the fire department. These
types of application are mostly used by the
people with certain disabilities.
Device Networks
Short range wireless connection
between devices is supported by device
Networks. Usually such connection is
wired. Devices as cell phones, modems,
headsets, PDAs, computers, printers,
projectors, network access point, the main
technology drivers for a network between
them can be one at low cost, low power
radio with network capabilities such as
Bluetooth, Zigbee, and UWB.
Sensor Networks
These networks consist of small nodes
with sensing, computation, and wireless
network. Sensor array can be deployed
and used for remote sensing in nuclear
power plants, mines, and military
applications.
Distributed control systems
Distributed Control applications, with
remote plants, sensors, and actuator
linked together by wireless communication
channel are enable by ad hoc wireless
networks. The ad hoc wireless networks
can be used to support coordinated control
of multiple vehicles in an automated
highway system.
Design Principles and
Challenges
The most design principles and challenges of an
ad hoc network come from the fact of its lack of
infrastructure. For wireless networks there is no
peer to peer communication.
On the other side ad hoc wireless network has
peer to peer communication, networking and
control functions that are distributed among all
nodes, and routing that can exploit intermediate
nodes are relays.
Ad hoc wireless networks may create
structure to improve network performance,
although it’s not a fundamental design
requirement of the network. Ad hoc wireless
networks can form a backbone infrastructure
from a subset of nodes in a network or some
nodes may be chosen to perform as base
stations for neighboring nodes. If a node in this
backbone subset leaves the network, the
backbone can be reconfigured.
One of the biggest challenges in ad hoc
wireless networks design is energy
constrain. These constrains arise in
wireless network nodes power by batteries
that cannot be recharged. Therefore
energy consumption must be optimized
over all aspects of the network design.
Protocol Layer
A protocol is a set of rules that govern the
operation of functional units to achieve
communication. There is an international
standard called OSI (open system
interconnection) model that was developed as a
framework for protocol layer in data networks.
From this standard model, different models were
derived as the TCO (transport control protocol)
and IP (internet protocol). The TCP and IP
protocol consist of five layer model.
Application
Transport
Network
Access
Physical
Physical Layer Design
The Physical Layer which is also
referred to as the link layer, deals primarily
with transmitting bits over a point-to-point
wireless link
In ad hoc networks bits are pocketsize
for transmission. The link packets error
rate (PER) is determined by physical layer
along with the channel and interference
conditions.
A multiple antennas can be used with ad hoc
wireless network. In fact it can increase the data
rate on the link, by providing diversity to fading
so that average BER is reduced leading to a
fewer retransmission. It also can provide
directionality to reduce fading and the
interference of signal causes to anther.
furthermore multiplexing will increase the link
rate, which reduces overcrowding and delay in
the link and benefits all multiple routes using that
link.
Considering the transmitted power of all nodes in the
network, it must be optimized with respect to all layers
that it impacts. As increasing transmit power at the
physical layer reduce PER and that decrease the
retransmission required at the access layer, but a high
transmit power from one node of the network can cause
significant interference to the other nodes. Therefore
SINR drives the performance in an ad hoc wireless.
The transmit power coupled with adaptive
modulation and coding for a given node defines it’s “local
neighborhood”- the collection of nodes that it can reach
in a single hop - and defines the context in which access,
routing, and other higher layer protocols operate.
Access Layer Design
Access layer is the layer that controls
how different users share the available
spectrum and ensures successful
reception of packets transmitted over this
shared spectrum.
There are two types of access either
multiple access or random access.
Multiple access divides the signaling
dimensions into dedicated channel.
The common used methods of multiplexing
are TDMA, FDMA, and CDMA.
In random access, channels are assigned to
achieve user dynamically, and in multihop
networks these protocols must content with
hidden and expose terminals.
The power control across the network
is also a part of the access layer functions.
The main role of power control is to insure
that SINR targets can be met on all links in
the network.
Assume an ad hoc wireless network with K
nodes and N links between different transmitterreceiver pairs of these nodes. The SINR link K is
given by
 ij 
Gij Pi
 j  k i Gkj Pk
 T
The access layer is also responsible for retransmitting of
packet received in error over the wireless link.
The data packets have an error detection code that is used by
the receiver to determine if one or more bits in the packets were
corrupted and cannot be corrected. For such packet:
the receiver will discard the corrupted packets and inform the
transmitter by a feedback that the packet must be retransmitted.
or
The access layer can save it and use a form of diversity to
combine the corrupted packet with the retransmission packets for a
higher probability of correct packet reception.
or
An alternative retransmitting the original packets in its entirety
the transmitter can just send some additional coded bits to provide a
stronger error correction capability for the original packets to correct
for its corrupted bits.
Network Layer Function
Neighbor discovery
Routing
Dynamic Resource Allocation
How does neighbor discover
occurs?
A group of neighboring nodes with some
initial transmit power
If not discovered
The transmit power is increased
Until all nodes connections established or
the maximum power is reached
Routing
Flooding Routing
Proactive Routing
Reactive Routing
Flooding
A packet is broadcast to all
nodes within receiving range
Nodes also broadcast the packet
Forward continues until the packet
reaches it destination
Advantages of Flooding
Highly flexible to changing network
topologies
Requires little routing overhead
Disadvantages of Flooding
Wasting Bandwidth
Battery power of transmitting nodes
Centralization
approach information about channel conditions and
network topology
forwarded to a centralized location that
computes the routing tables for all nodes in the
network
Advantages
minimum average delay
minimum number of hops
minimum network congestion.
Disadvantages
cannot adapt to fast change in the channel
condition and network topology
requires much over head for periodically
collecting local node information
Reactive routing
source node initiates a route-discovery process when
it has data to send
It will determine if one or more route are available to
the destination.
Advantages
can be obtained with relatively little overhead
Disadvantages
significant initial delay
Advantages
can be obtained with relatively little overhead
Disadvantages
significant initial delay
Resource Allocation and flow
Control
Identify the
route a packet
should follow
from a source
to its
destination
How are Routing, resource
allocation & flow control related?
Routing based on minimum delay
Delay is a function of the link data rate or
capacity
The higher the capacity, the more data
that can flow with min. delay
The link capacity depends on the
resources allocated to the link
Fij
Dij =
Cij- Fij
•Fij traffic flow
•Dij delay on a link
•Cij capacity
Link Utilization Formula
Fij
Dij
=
Cij
Transport layer
Transport Layer Functions:(end to end functions)
Error recovery
Retransmission
Reordering
Flow control
Application Layer Functions:
Generates the data to be sent over
the network
Processes the corresponding data
received over the network
Provides compression of the
application data along with error
correction and cover up
MDC
Multiple description coding
A form of compression
Multiple descriptions of the data are
generated
The original data can be reconstructed
from any of these descriptions with
some loss
Cross Layer Design
The layering approach to wireless network
design, where each layer of the protocol stack is
unaware to the design and operation of other
layers, has not worked well in general
Cross-layer design clearly requires:
information exchange between layers,
adaptively to this information at each
layer,
diversity built into each layer

interaction between layers
Ease of modifying the functionality of one of the
layers
May need to generate a new protocol stack each
time a small
change is modified
Long term survivability of such architectures?
Network Capacity Limits
Capacity is the set of maximum data rates
possible between all nodes.
The capacity region has dimension of
K(K − 1)
For a large K,


the per-node rate is 1/√K logK ,
and the throughput is √K/logK
Energy-Constrained
Networks
Batteries in nodes and devices:
Nodes are powered by batteries with a
limited life time
Devices with rechargeable batteries
must conserve energy to max. time
between recharges
Some devices can not be
recharges
Some operate solely form the
environment
Energy constraints associated with
node operation:
Hardware operation
Transmit power
Signal Processing
Design consideration with energy
constrained nodes:
Modulation & Coding
MIMO and Cooperative MIMO
Access, Routing, & Sleeping
Cross Layer Design under Energy
Constraints
Capacity Per Unit Energy
Modulation & coding are based on:
Required transmit power
Data rate
BER
Complexity
Modulation
The design choice should be based on
the total energy consumption
Circuit energy consumption
increases with transmition time
Decrease transmition time &
putting nodes to sleep
Example between M-ary Modulation and binary
modulation
Coding
Reduces the required transmit
energy per bit for a given BER target
Some coding schemes encode bits
into a codeword that is longer than
the original bit sequence:
Such as:
block codes
Convolution codes
continue
The total transmit energy required
for the codeword to be sent , a
longer transmission time
consumes more circuit energy,
and band width expansion
MQAM is more efficient
then MFSK
MIMO
Multiple Inputs Multiple outputs
Functions:
Multiplexing gain:
Provides a higher data rate
Diversity gain:
Provides a lower BER in
fading
Continue
Provide energy savings over a single
antenna system for most transmition
distances
Why?
The reason is that MIMO systems can
support a higher data rate for a given
energy per bit. So, it transmits the bits
quicker and then it shuts down
Cooperative MIMO
Small nodes that can not support
multiple antennas they are grouped
together to form a transmitter while
others form a receiver
Distance between the nodes is small
so the energy associated is small
Cooperative MINO
Access
How to be more energy efficient?
Minimizing collision
Optimizing transmit power
Routing
Routing is affected by energy
consumption distributed across all
node
Routing optimization to minimize
end to end consumption
How
By applying the Standard
optimization procedure
Sleep
Nodes consume power even in stand by mode
How to solve?
By scheduling sleep periods for nodes
Each node to only listen during a certain period of
time