PSG COLLEGE OF TECHNOLOGY

Department of Instrumentation and Control Systems Engineering

18UC02-SMART SENSORS AND NETWORKS

DESIGN PRINCIPLES FOR WSNS

Presented by:
G.R.POOVARAGHAVAN.
S.RAHUL.
R.S.VISHNU PRASATH.
 PRESENTATION OVERVIEW
• Introduction
• Distributed organization
• In-network processing
• Data centricity
• Adaptive fidelity and accuracy
• Exploit location information
• Exploit activity patterns
• Exploit heterogeneity
• Component-based protocol stacks and cross-layer optimization
 INTRODUCTION

• Energy efficiency and scalability are important parameter for

wireless sensor networks.
• But these goals themselves do not provide many hints on how
to structure a network.
• A few basic principles have emerged, which can be useful
when designing networking protocols.
• Following them would result in inferior solutions.
 DISTRIBUTED ORGANIZATION
• Both the scalability and the robustness optimization goal and some
other goals goal are involved to organize the network in a distributed
fashion.
• That means that there should be no centralized entity in charge.
• Self-organization is a commonly used term for this principle.
• When organizing a network in a distributed fashion, it is necessary to
be aware of shortcomings of this approach.
• A centralized approach can produce solutions that perform better or
require less resources (in particular, energy).
 CONT..
• To combine the advantages one possibility is to use
centralized principles in a localized fashion.
• It can be done by dynamically electing out of the set of equal
nodes, specific nodes that assume the responsibilities of a
centralized agent.
• Such elections result in a hierarchy, which has to be dynamic.
• The election process should be repeated continuously or else
the resources of the elected nodes be overtaxed.
 IN-NETWORK PROCESSING
• When organizing a network in a distributed fashion the nodes in the
network are not only executing application programs.
• They are also actively involved in taking decisions about how to
operate the network.
• This is a specific form of information processing that happens in the
network but is limited to information about the network itself.
• It is possible to extend this concept by also taking the concrete data.
• The concrete data that is to be transported by the network into
account in this information processing, making in-network
processing a first-rank design principle.
 CONT…

• There Several techniques for in-network processing exist.

Aggregation.
Distributed source coding and distributed compression.
Distributed and collaborative signal processing.
Mobile code/Agent-based networking.
 AGGREGATION
• The simplest in-network processing technique is aggregation.
• Suppose a sink is interested in obtaining periodic measurements
from sensors, but it is only relevant to check whether difference
between minimum and maximum value is too big.
• It is evidently not necessary to transport are readings from all
sensors to the sink, it suffices to send the average or the minimum
and maximum value.
• Transmitting data is considerably more expensive than even
complex computation is the great energy-efficiency benefits of
this approach.
• This function should be compassable.
 DISTRIBUTED SOURCE CODING AND
DISTRIBUTED COMPRESSION
• Is it possible to reduce the number of transmitted bits.
• but still obtain the full information about all sensor readings at the
sink. It is related to the coding and compression problems known
from conventional networks.
• where a lot of effort is invested to encode for example a video
sequence, to reduce the required bandwidth.
• Traditional coding schemes tend to put effort into the
encoding which might be too computationally complex for
simple sensor nodes.
 DISTRIBUTED AND COLLABORATIVE
SIGNAL PROCESSING
• The in-networking processing approaches discussed so far have not
really used the ability for processing in the sensor nodes.
• have only used in networking processing for trivial operations like
averaging or finding the maximum.
• When complex computations on a certain amount of data is to be
done.
• it can still be more energy efficient to compute these functions on
the sensor nodes despite their limited processing power.
MOBILE CODE/AGENT-BASED NETWORKING
• The possibility of executing programs in the network, other
programming paradigms or computational models are feasible.
• One such model is the idea of mobile code or agent-based
networking.
• The idea is to have a small, compact representation of program code
that is small enough to be sent from node to node.
• This code is then executed locally, for example, collecting
measurements and then decides where to be sent next.
• This idea has been used in various environments.
 DATA CENTRICITY
• In traditional communication networks, the focus of a Communication
relationship is usually the pair of Communicating peers – the Sender and
the Receiver of data.
• In a WSN, the interest of an application is not in the identity of a
particular sensor node, it is much rather in the actual information
reported about the physical environment.
• The data is at the center of attention (ignoring the identity of nodes) is
called Data-Centric Networking.
• In a data-centric application, all the application would have to do is state
its desire to be informed about events of a certain type – presence of
elephant – and the nodes in the network that possess elephant detectors
are indirectly informed about this request.
 CONT…
• Data-centric networking allows very different networking architectures
compared to traditional, identity-centric networks. Used in in-network
processing techniques like data fusion and aggregation .
• Data Centric networking and addressing is also claimed to improve
performance and especially energy efficiency of a WSN.
• The reason for this is data-centric solutions scale better by being
implementable using purely local information about direct neighbors and
easier integration of a notion of adaptive accuracy into a data-centric
framework.
 IMPLEMENTATION OPTIONS FOR DATA-
CENTRIC NETWORKING
• Overlay networks and distributed hash tables

• Publish/Subscribe Approach

• Databases
 OVERLAY NETWORKS AND DISTRIBUTED
HASH TABLES

• In peer-to-peer networking, the solution for an efficient

lookup of retrieval of data from an unknown source is usually
to form an overlay network, implementing a Distributed Hash
Table (DHT) .

• The desired data can be identified via a given key (a hash)

and the DHT will provide one (or possibly several) sources
for the data associated with this key
 PUBLISH/SUBSCRIBE
• The provision of the information is not well matched with the
synchronous characteristics of a request/reply protocol.
• A means is necessary to express the need for certain data and the
delivery of the data, where the data as such is specified and not the
involved entities. This is Publish/Subscribe approach.
• Any node interested in a given kind of data can subscribe to it, and any
node can publish data, along with information about its kind as well.
• Example : Youtube, Chrome WebPages etc,.
 DATABASES
• A different view on WSN is to consider them as (dynamic)
databases
• Being interested in certain aspects of the physical
environment that is surveyed by a WSN is equivalent to
formulating queries for a database.
• SQL-based querying of a WSN can be extended to an easy-
to-grasp interface to wireless sensor networks, being capable
of expressing most salient interaction patterns with a WSN.
 ADAPTIVE FIDELITY AND ACCURACY

• The amount of energy should be contingent for making the

fidelity computation of results.
• This node should be extended should be extended from a
single node to a entire network
• In function approximation ,when more sensors are used ,the
approximation is better.
• More energy has to be invested for function approximation
application
 CONTD…
• Similar examples are detection and tracking applications in
WSN.
• The application should be able :
To adapt to its current status of network.
How many nodes have already failed.
How much energy could be scavenged.
 EXPLOIT LOCATION INFORMATION

• To exploit local information when such information is

present.
• The location of an event is very crucial to determine the
location of sensor nodes.
• One such information can simplify the design .
• The operation of communication protocols improve energy
efficiency.
 EXPLOIT ACTIVITY PATTERNS
• Activity pattern is a wireless sensor network and it is different
from traditional networks.
• The data rate averaged over a long time and it is small and this
change dramatically when something happen.
• Once an event is happened it has been detected by large number
is sensors, cause a well known even shower effect.
• It has switch between modes of quiescence and high activity
 EXPLOIT HETEROGENEITY
• Sensor nodes can be heterogeneous by constructions.
• Some modes have larger batteries, farther reaching
communication devices or more processing power.
• All nodes can be started from an equal state, because each nodes
has to perform more tasks.
• Heterogeneity of the network is burden or an opportunity.
• It is asymmetric assignment of tasks.
• It should invest their energy for long distance communication.
• It has to be balanced against the potential benefits.
 COMPONENT BASED PROTOCOL
• Component based protocol is a layering based protocol and it is a
default collection of components and it forms sensor nodes.
• WSN requires even if only simple.
 Form physical layer.
 MAC(medium access control)
 Linked layer protocols.
• Wireless sensor network requires.
 Routing
 Transport layer functionalities.
 CROSS LAYERED OPTIMIZATION
• Cross layer information exchange holds great promise for
protocol optimization, and without danger.
• The imprudent use of cross layer design leads to.
Feedback loops.
Endangering both functionalities.
Performance of the entire system.
• The above reasons should not be easily disregarded
• Care has to be taken to avoid such unexpected feedback loops.
REFERENCES
• Protocols and Architectures for Wireless Sensor Networks by Holger
Karl and Andres Wilig
THANK YOU