Wireless Sensor Networks

C.P. Gupta
Asso. Prof. & Head
Computer Engg.
Rajasthan Technical University
Kota
What are wireless sensor networks
(WSNs)?

network of (possibly low-size and low-

complexity) devices denoted as nodes that can
sense the environment and communicate the
information gathered from the monitored field
through wireless links; the data is forwarded,
possibly via multiple hops relaying, to a sink that
can use it locally, or is connected to other
networks (e.g., the Internet) through a gateway.
What are wireless sensor networks
(WSNs)?
• Networks of typically
small, battery-
powered, wireless
devices.
– On-board processing,
– Communication, and
– Sensing capabilities.
• Low-power processor. Sensors P
– Limited processing.
O
• Memory. W
Storage Processor
– Limited storage. E
• Radio. R

– Low-power. Radio
– Low data rate.
– Limited range. WSN device schematics
• Sensors.
– Scalar sensors: temperature, light, etc.
– Cameras, microphones.
• Power.
WSN Applications
• Monitoring.
– Scientific, ecological applications.
• Non-intrusiveness.
• Real-time, high spatial-temporal resolution.
• Remote, hard-to-access areas.
• Surveillance and tracking.
– Reconnaissance.
– Perimeter control.
• “Smart” Environments.
– Agriculture.
– Manufacturing/industrial processes.

1-5
Sensing Tasks
• Different types of sensors: seismic, low sampling rate
magnetic, thermal, visual, infrared, acoustic and radar,
able to monitor:
– temperature,
– humidity,
– vehicular movement,
– lightning condition,
– pressure,
– soil makeup,
– noise levels,
– the presence or absence of certain kinds of objects,
– mechanical stress levels on attached objects, and
– the current characteristics such as speed, direction, and size of
an object.

6
Military Applications
• Desirable characteristics of sensor networks
– rapid deployment,
– self-organization
– fault tolerance
• Example applications
– Monitoring friendly forces, equipment and ammunition
– Battlefield surveillance
– Reconnaissance of opposing forces and terrain
– Targeting
– Battle damage assessment
– Nuclear, biological and chemical attack detection and
reconnaissance
7
Environmental Applications
• Desirable characteristics of sensor networks
– untethered sensors
– No interruption to the environment
– Redundancy
• Example applications
– Forest fire detection: Strategically, randomly, and densely
deployed sensor nodes can relay the exact origin of the fire.
– Biocomplexity mapping of the environment: integrating
information across temporal and spatial scales.
– Flood detection: rainfall, water level and weather sensors supply
information to the centralized database system.
– Precision Agriculture: the pesticides level in the drinking water,
soil erosion, and air pollution.

8
Home applications
• Home automation: smart sensor nodes and actuators can be buried
in appliances, such as vacuum cleaners, micro-wave ovens,
refrigerators, and VCRs. They are connected to external networks.

• Smart environment: Furniture and appliances (and servers) learn to

provide the needed service.

• Smart kindergarten: to provide parents and teachers with the

abilities to comprehensively investigate students’ learning
processes; to collect, manage, and fuse the information of the
sensors

Many more applications

9
Factors Influencing Sensor Network Design

• fault tolerance;
• scalability;
• production costs;
• operating environment;
• sensor network topology;
• hardware constraints;
• transmission media; and
• power consumption.
10
Fault tolerance
• Failures:
– lack of power,
– physical damage in harsh environment
– Interference by other objects (e.g. radios) and other sensors.
• Fault tolerance: the ability to sustain sensor network
functionalities without any interruption due to failures
– reliability Rk(t) or fault tolerance of a sensor node (using the
Poisson distribution), i.e., the probability of not having a failure
within the time interval (0,t):
Rk(t) = exp(-λkt)
λk : the failure rate of sensor node k
• The environment is important to the fault tolerance of
algorithms and protocols
11
Scalability
• # of sensors: hundreds, thousands, to millions,
depending on the type of applications.
• Density can be expressed as:
μ(R) = (N π R2) / A
where N is the number of scattered sensor nodes in
region A; and R, the radio transmission range.
Basically, μ(R) gives the number of nodes within the
transmission radius of each node in region A.
Note: often we consider 2-dimensional space.
• Density also depends on the applications.
12
Costs
• Per node cost is important for large sensor
networks. It has to be kept low.
• Bluetooth radio system: $5 now, but still
too expensive for sensors. PicoNode:
targeted to be < 50c.
• More challenging, with large amount of
functionalities

13
Hardware constraints
• All components must be contained in a “matchbox”
• Limited energy
• Low speed (MHz) and small OS kernel (KBs)
• Small memory (KBs)
• Transceiver (kbps, short range, feet-meters)

14
Sensor network topology
• topology maintenance a challenging task due to
– # of nodes, failures, dynamics etc
• Pre-deployment and deployment phase: no careful planning.
considerations:
– the installation cost,
– no need for any pre-organization and pre-planning,
– the flexibility of arrangement, and
– better self-organization and fault tolerance.
• Post-deployment phase
– topology changes are due to change in position, reachability (due to jamming,
noise, moving obstacles, etc.), available energy, malfunctioning, etc
– How to maintain the topology change?
• Re-deployment of additional nodes phase
– Adding new sensors

15
Transmission media
• Wireless communication, formed by radio,
infrared or optical media

16
 Design Challenges

Why are WSNs challenging/unique from a

research point of view?
• Typically, severely energy constrained.
– Limited energy sources (e.g., batteries).
– Trade-off between performance and lifetime.
• Self-organizing and self-healing.
– Remote deployments.
• Scalable.
– Arbitrarily large number of nodes.
Design Challenges (Cont’d)
• Heterogeneity.
– Devices with varied capabilities.
– Different sensor modalities.
– Hierarchical deployments.
• Adaptability.
– Adjust to operating conditions and changes in
application requirements.
• Security and privacy.
– Potentially sensitive information.
– Hostile environments.

1-18
Think Back -Sensor Network
Server

Sensor field

Gateway

Internet

19
Energy/Power Considerations
• Terms
– Cell, Battery
– Energy (Joule)
– Power (J/s or Watt)
– Ampere-hour (AH)
– Deep-cycle
– MCU
– Sleep Modes
– ADC
– BPS

20
Where Does The Power Go?

21
Microcontroller Unit (MCU)
• Intel’s StrongARM, Atmel AVR (PIC?)
• Low power modes
– Active, Idle, nap, shutdown, sleep modes
– For some MCUs, in deep sleep modes, the
power consumption can almost be negligible
– Takes longer to wake from a deep sleep than
just a nap
– Wakeup time also takes power
– Wakeup impact processing

22
Radio
• Radio typically contain embedded controller that
provides many functions
– Uses RSSI to adjust transmit power
– Error detection and correction in hardware
• Several modes
– Receive only, transmit + receive, idle, etc.
• Transmit in general requires most power
• Careful consider radio spec and modes
• Mode change can consume a lot of power
– May be better to shutdown completely rather than go into
idle mode

23
Bandwidth vs. Power Consumption

• Higher bandwidth (BPS) generally requires

– Better S/N
• => more signal power
– More sophisticated modulation
encoding/decoding algorithms
• => more powerful CPU, more power
• Receive power normally much lower than
transmit power

24
Conventional vs WSN Power
Management
• Conventional
– Well developed techniques
– Objective is to minimize power consumption of individual
device: sleep modes, low voltage, low clock speed, etc.
• WSNs
– Objective is to maximize lifetime of individual node, but
more importantly
– The network as a whole
• For example
– It may be better for a node to deplete its power source
for the greater good of the network

25
Sensors
Passive & low power (~mW and smaller)
– Soil moisture, temperature, light, humidity
• Active & high power
– Anemometers, disdrometers, cameras
• Many sensors are inherently analog, but some sensors
have digital interfaces (provided by embedded
controllers)
• Conditioning/wakeup times need to be considered
• Analog-Digital Converters (ADC)
– Can be a major power consumer
– More bits and high conversion rate requires more power
– Don’t over specify

26
Battery
• Uses chemical reaction to provide electrical
energy
– Temperature depended
• Batteries are often the most bulky part of a
mote
• Capacity measured in Ampere-hours or Ah.
Note that the capacity does consider voltage…
– The capacity is the nominal number of hours it can
supply a given current

27
DC/DC Converter
• Why?
– Battery does not provide constant voltage
– Battery does not provide convenient voltage (e.g.,
MCU needs 2.7 V)
– Linear regulator wastes energy
• How does it work?
– Chops DC, make AC, use inductive element
(analogy transformer) to change voltage levels with
minimal loss of energy
– Provides voltage regulation
• Efficiencies 80% and up
28
Concept: Deep Cycle
• Normal batteries
– Can’t deplete the battery more that say 10%
without damage to battery
– E.g., for a 100 Ah battery one can only use 10
Ah, and then should recharge
• Deep cycle batteries
– Can tolerate much deeper discharge (say 50-
60%) without damage

29
Energy Budget
3V -> 5V DC/DC Converter
90% Effective • Radio (5V)
3V, 1.25 Ah battery – Power down 25 A
– Receive 50 mA
– Transmit 150 mA

• Sensors (5V)
WSN Node – Average ~ 1 mA

WSN node is programmed to turn on radio every 5 seconds, listen on the network for
0.5 seconds, and then power down the radio.

It is estimated that it will be required to make a 10-second transmission once every 5

minutes.

Estimate how long the node can operate. Assume that 70% of the battery capacity is
available.

30
 Energy Advantage of WSNs
• Model for received power:

Pt
Pr  
R
Pt  R  Pr

• Where  in range 2-4

• N transmissions each over a distance R, is more efficient than one
transmission over a distance NR by a factor:

( NR )  Pr  1
 
 N
N ( R Pr )
31
 Energy Advantage of WSNs
• Derive, explain, comment on, this equation:

( NR )  Pr  1
 
N
N ( R Pr )
• Thus for  = 3, and
10 ,10 hops,the advantage ofW SN com m unication is

  10 31  100
Tim em.ore energy efficient.

32
Protocol stack

33
Transport layer: current and
future
• The need for transport layer? Maybe not.
– How to address?
– Do we need reliable service at transport layer
– Congestion control
• No attempt thus far to propose a scheme or to discuss
the issues related to the transport layer
• Open research issues
– Challenging problem due to the influencing factors, especially
the hardware constraints such as the limited power and
memory.
• no buffer like TCP
• acknowledgements too expensive.
– “new schemes that split the end-to-end communication probably
at the sinks may be needed where UDP type protocols are used
in the sensor network and traditional TCP/UDP protocols in the
Internet or Satellite network.”
34
Network layer: current and
future
• All about routing at this layer (unstructured network)
– Those ad hoc routing techniques, Like DSR? Do not usually fit
the requirements:
• Power efficiency
• Sensor networks are mostly data centric.
• Data aggregation is useful
• Both attribute-based addressing and location awareness.
• How to route, i.e., choice of routing metrics
– Maximum available power (PA) route
– Minimum energy (ME) route
– Minimum hop (MH) route
– Maximum minimum PA node route

35
Network layer: current and
future
• Flooding: each node receiving a packet repeats it by
broadcasting (simple flooding versus TTL-based)
• Gossiping: no broadcast, but send the incoming packets
to a randomly selected neighbor.
• Sensor protocols for information via negotiation (SPIN)
• Low-energy adaptive clustering hierarchy (LEACH):
clustering-based protocol
• …
• “need to be improved or new protocols need to be
developed to address higher topology changes and
higher scalability. Also, new internetworking schemes
should be developed to allow easy communication
between the sensor networks and external networks”
36
Data link layer: current and
future
• multiplexing of data streams, data frame detection,
medium access and error control, reliable point-to-point
and point-to-multipoint connections
• Medium access control (MAC): establish links, fairly and
efficiently share communication resources, must have
– built-in power conservation,
– mobility management, and
– failure recovery strategies.
• Fixed allocation (like TDMA) and random access (like
CSMA)

37
Data link layer: current and
future
• Open research issues
– MAC for mobile sensor networks
– Energy requirement for sensor network self-organization.
– Error control coding schemes, FEC, erasure code etc
– Power saving modes of operation

38
The active sub-areas
• Routing
• Topology control
• Data management, aggregation and query
• MAC protocols
• Target tracking, resource discovery
• Monitoring and maintenance

39