Dr Angela Doufexi, CCR

1
Dr Angela Doufexi, CCR

• Wireless Sensor Networks
• - wireless perspective

2
Introduction - WSN

• Network of large number of nodes scattered
  through an area of interest
• Usually, each node has finite battery lifetime
  and thus limiting computing and communication
  capabilities
• Nodes can cooperate to perform advanced signal
  processing tasks
• Wide range of applications
• Military
• Health
• Home
• Emergency and Disaster Services
• Monitoring of physical phenomena
• Video sensor networks (increased data rate)

3
Requirements

• The Physical and MAC layers of WSN must be energy
  efficient
• Energy consumption must be minimised while
  satisfying throughput and delay requirements
• PHY layer energy efficient transmission schemes
  (current systems IEEE 802.15.4 Zigbee, 250 kbps
  at 2.4 GHz (DSSS))
• Impulse Radio UWB (Low transmit power, Excellent
  localization)
• Energy consumption includes both the transmission
  energy and the circuit energy consumption
• In the MAC layer, energy efficiency is achieved
  by putting a node in a power saving mode (sleep
  and listen period, example S-MAC)

4
MIMO in WSN

• MIMO systems can support higher data rates under
  the same transmit power budget or for the same
  throughput requirement MIMO systems require less
  transmission energy than SISO systems
• However in WSN networks energy consumption (from
  all circuit blocks) is comparable with
  transmission energy
• Circuit complexity in MIMO structures is higher
  than SISO. Multiple antennas in a small node may
  not be realistic. If the constellation is
  optimally chosen energy efficiency in MIMO
  systems increases.
• Cooperation can reduce energy consumption
  (virtual MIMO)
• MIMO techniques include STBC and Spatial
  Multiplexing

5
Source and Channel Coding in WSN

• A system with FEC can provide an objective
  reliability using less power than a system
  without FEC (we can also consider HARQ)
• However additional parity bits and
  encoding/decoding complexity require additional
  energy consumption. Codes include LDPC, IRA.
• WSN are made up of multiple nodes generating
  correlated information signals to be compressed
  and transmitted in wireless systems.
• We can employ lightweight signal processing at
  the individual sensor nodes and transfer the
  complexity at the receiver.
• Wyner-Ziv coding, Stepian Wolf information
  theoretic results
• Distributed video coding with joint decoding in
  high data rate wireless video sensor networks.

6
Further Research

• Realistic channel models Propagation modelling
  for specific applications.
• Computation of parameter estimates based on
  distributed observations collected at individual
  sensors. Distributed estimation algorithms
  (communication with neighbours)
• Packet Combining in multi hop WSN
• Cooperative communications for low cost sensors
• Joint source channel coding
• Cross Layer Design