Congestion Control and Fairness for ManytoOne Routing in Sensor Networks

1
Congestion Control and Fairnessfor Many-to-One
Routingin Sensor Networks

• Oct. 27, 2005
• Howon Lee
• Communication and Information Systems Lab.
• SenSys04

2
Outline

• Introduction
• Motivation
• Congestion control
• Fairness
• Simulation Implementation Results
• Conclusions

3
Introduction

• Sensor motes send data to the BS
• Many-to-One
• Intermediate motes relay messages from far away
  motes to the BS
• Routing of messages form a tree rooted at the
  BS

BS
4
Motivation

• Distributed system no global knowledge
• Motes near the base has to transmit more packets
  than far away motes bottleneck
• Packets generated by far away motes may be
  dropped (unfairness)
• Loss of packet due to buffer overflow and
  collision (congestion)

5
Congestion Control

• Design Network Stack Model

Congestion
Application
Transport
Network
Data-Link
Routing
Physical
MAC
6
Congestion Control

• Types of Congestion
• Type A - Simultaneous transmission interference
  packet loss reduction in throughput
• Type B - Data generation rate is faster than the
  transmission rate - Buffer overflow

7
Congestion Control

• The idea
• Type A congestion can be reduced by phase
  shifting (ex. Jittering)
• Type B congestion can be reduced by buffer
  monitoring

8
Congestion Control

• The idea
• The two different rates we need
• data generation rate rate at which data is
  passed from application
• transmission rate rate at which packets, both
  locally produced and from downstream motes, are
  transmitted upstream
• transmission rate of parent mote should ideally
  be greater than or equal to data generation rate
  of all motes downstream

9
Congestion Control

• The idea

Application
Transport
Transmission rate
Network
Data Generation rate
Data-Link
Physical
10
Congestion Control

• The idea

Transmission rate of parent should be gt
5pkts/sec
Each mote generates data at the rate 1pkt/sec
11
Congestion Control

• The Algorithm
• Repeatedly run at each mote (localized algorithm)
• determines local maximum transmission rate
• divide transmission rate by total number of
  children motes to give data generation rate
• disseminate min(own_rate, parent_rate) downstream

Queue overflow adjustment incorporated
12
Congestion Control

• The Algorithm
• Adjustment
• Let the data generation rate for a child
  determined by the current node is denoted by rc
• Let r denote data generation rate for the current
  when the queues are overflowing or about to
  overflow
• disseminate min(parent_rate, r, rc) downstream

Current nodes data gen rate sent by its parent
13
Congestion Control

• The Algorithm

Transmission rate 50 pkts/sec
Each childs data generation rate should be lt
10pkts/sec
14
Congestion Control

• Lets look into the methods for
• determining effective transmission rate
• determining number of downstream motes
• disseminating data generation rate downstream
• adjusting data generation rate if buffers are
  overflowing

15
Congestion Control

• Calculating effective Transmission Rate

ack pkt
data pkt
data pkt
no ack, timeout
data pkt
no ack, timeout
Total time taken
16
Congestion Control

• Counting number of Downstream Motes
• each mote sums childrens count, adds 1 (for
  itself), then transmits count to parent (data
  aggregation)

5
2
1
1
1
17
Congestion Control

• Disseminating data generation rate
• downstream
• via control packets
• piggy-backed on data packets

18
Congestion Control

• Adjusting data generation rate if buffers are
• overflowing

19
Congestion Control
Taking mean data gen. rate

• The Solution Explained
• The packet generation rate assignment is fair
• Why?
• Type B congestion is minimized
• How?
• Type A congestion is reduced
• How?

Queue overflow check
20
Congestion Control
A
Rate updates takes time to propagate, causing
phase shift
D
B
C
E
E
21
Fairness

• Requirement
• The base station should receive the same number
  of packets from each mote
• The idea
• Transmit number of packets from each subtree with
  equal probability
• ? Probabilistic Selection (PS)
• Within each period of time, transmit number of
  packets from each subtree equal to size of that
  subtree
• ? Epoch-Based Proportional Selection (EPS)

22
Fairness

• Mechanisms
• Per-child packet queues
• Maintenance of per-child tree size
• Obtained as before

• FIFO Queues
• small in size, independent of subtree size

23
Fairness

• Algorithm
• Probabilistic Selection

24
Fairness

• Algorithm
• Epoch based proportional selection
• We define an epoch to have units of packets and
  is an integer multiple of the total number of
  nodes in this subtree and a positive integer n.
• With each epoch we transmit from each queue
  exactly n times the number of nodes serviced by
  that queue.
• Queues are FIFO
• No work conservation is implemented

25
Fairness
In one time period transmit 2 from A, 1 from B, 1
from C and 1 from this parent node

• The idea

C
A
1
B
2
1
26
Fairness

• Algorithm
• FIFO Queue

• In node A, packets from B are received according
  to Bs epoch
• No order is imposed on childs transmission
• Within each epoch one packet per child

27
Fairness

• Algorithm
• Selecting the next packet to transmit

• In node A, packets from B are received according
  to Bs epoch
• No order is imposed on childs transmission

28
Fairness

• Algorithm
• Effect of non-work conservation
• Congestion in any branch of the network will
  cause rates to decrease throughout all other
  parts of the network

29
Fairness

• Algorithm
• Necessity of ARQ
• Implementing EPS without ARQ
• Suppose link between parent A and a child Q is
  very lossy
• A may miss time intervals to transmit as it waits
  for Q to send data
• Entire performance deteriorates due to one single
  link !
• Best solution is to implement hop-by-hop ARQ

30
Results

• Simulation
• Packet level simulation
• Bit rate, control packet size, data packet size
• No interference from motes more than one hop away
• Uses MACA as MAC protocol
• Round-robin servicing of queues to show effect
  of loss of packets further away from the BS

31
Results

• Simulation

• The (Congestion Control EPS) graph
• is a constant at 5000 packets
• The (Congestion Control PS) graph fluctuates
  slightly around 5000 packets (probabilistic)

32
Results

• Implementation on mica2dot motes
• MACA with transport layer ACKs
• 10 motes deployed indoors, within 15 feet of one
  another
• motes arbitrarily construct routing tree
• compare with round-robin servicing of queues

33
Results

• Implementation

• BS receives more packets from nodes close by, in
  round-robin.

34
Conclusion

• Simple and scalable (? implementation)
• We showed that by measuring the effective
  available bandwidth and obtaining the size of the
  sub-trees, we can divide that bandwidth equally
  amongst all downstream nodes.
• ? Congestion control
• Using, Probabilistic Selection or Epoch-based
  Proportional Selection with ARQ, we can enable
  equal numbers of packets from all nodes to reach
  the base station.
• ? Fairness
• Since we solve congestion control and fairness in
  the transport layer, our solution is independent
  of the routing in the network layer, as well as
  the MAC protocol in the data-link layer

35
Thank you
Questions?