Adhoc and Sensor Networks - Chapter 07

Copyright © 2006, Dr. Carlos Cordeiro and Prof. Dharma P. Agrawal, All rights reserved. 1
Introduction
TCP Protocol Overview
 Designed and Fine-Tuned to Wired Networks
 TCP Basics
 TCP Header Format
 Congestion Control
 Round-Trip Time Estimation
TCP and MANETs
 Effects of Partitions on TCP
 Impact of Lower Layers on TCP
Solutions for TCP over Ad Hoc
 Mobility-Related
 Fairness-Related
Conclusions and Future Directions
Table of
Contents

Introduction
 TCP most widely used transport protocol
 Ad hoc networks composed exclusively of wireless
links
 All nodes can move freely and unpredictably
 TCP needs to distinguish the nature of errors

Designed and Fine Tuned
to Wired Network
 Design heavily influenced by “end-to-end argument”
 Excessive intelligence in physical and link layers to handle error
control, encryption or flow control
 Performance often dependant on flow control and congestion control
 Time outs and retransmission handle error control
 It is important to incorporate link layer acknowledgement and error
detection/correction functionality
 High error rates, longer delays and mobility makes MANET
environments extremely challenging

TCP Basics
 Byte Stream Delivery: TCP interfaces between the application
layer above and the network layer below and TCP decides
whether to segment or delineate the byte stream in order to
transmit data in manageable pieces to the receiver, hence called
“byte stream delivery service”
 Connection-Oriented: Two communicating TCP entities (the
sender and the receiver) must first agree upon the willingness to
communicate
 Full-Duplex: TCP almost always operates in full-duplex mode,
and TCP exhibit asymmetric behavior only during connection
start and close sequences (i.e., data transfer in the forward
direction but not in the reverse, or vice versa)

A number of mechanisms help provide the guarantees:
 Checksums: All TCP segments carry a checksum,used by the receiver
to detect errors with either the TCP header or data
 Duplicate data detection:TCP keeps track of bytes received in order to
discard duplicate copies of data that has already been received
 Retransmissions: TCP must implement retransmission schemes for
data that may be lost or damaged and the lack of positive
acknowledgements, coupled with a timeout period calls for a
retransmission
 Sequencing: It is TCP's job to properly sequence segments it receives
so that it can deliver the byte stream data to an application in order
 Timers: TCP maintains various static and dynamic timers on data sent
and if the timer expires before receiving an acknowledgement, the
sender can retransmit the segment
Reliable TCP guarantees

TCP Header Format

 Source Port: This is a 16-bit number identifying the application
where the TCP segment originated from within the sending host
 Destination Port: A 16-bit number identifying the application the
TCP segment is destined for on a receiving host
 Sequence Number: A 32-bit number, identifying the current position
of the first data byte in the segment and after reaching 232
-1, this
number will wrap around to 0
 Acknowledgement Number: A 32-bit number identifying the next
data byte the sender expects from the receiver and is one greater than
the most recently received data byte
 Header Length: A 4-bit field that specifies the total TCP header
length in 32-bit words (or in multiples of 4 bytes), with the largest
TCP header of 60 bytes
 Reserved: A 6-bit field currently unused and reserved for future use
TCP Frame Details

Control Bits
 Urgent Pointer (URG) – If this bit field is set, the receiving TCP should
interpret the urgent pointer field
 Acknowledgement (ACK) – If this bit is set, the acknowledgment field is valid
 Push Function (PSH) – If this bit is set, the receiver should deliver this
segment to the receiving application as soon as possible
 Reset Connection (RST) – If this bit is present, it signals the receiver that the
sender is aborting the connection and all the associated queued data and
allocated buffers can be freely relinquished
 Synchronize (SYN) – When present, this bit field signifies that the sender is
attempting to “synchronize” sequence numbers
 No More Data from Sender (FIN) – If set, this bit field tells the receiver that
the sender has reached the end of its byte stream for the current TCP
connection

 Window: This is a 16-bit integer used by TCP for flow control in the form
of a data transmission window size
 Checksum: A sender computes the checksum value of 16-bits, based on
the contents of the TCP header and data fields and is compared with the
value the receiver generates using the same computation
 Urgent Pointer: This 16-bit field tells the receiver when the last byte of
urgent data in the segment ends
 Options: Depending on the option(s) used, the length of this field varies in
size, but it cannot be larger than 40 bytes due to the maximum size of the
header length field (4 bits)
 Padding:It may be necessary to “pad” the TCP header with zeroes so that
the segment ends on a 32-bit word boundary as defined by the standard
 Data: This variable length field carries the application data from TCP
sender to receiver and this field coupled with the TCP header fields
constitutes a TCP segment
TCP Details

Congestion Control
 Slow Start: (A mechanism to control the transmission rate) Whenever a
TCP connection starts, the Slow Start algorithm at the sender initializes a
congestion window (CWND) to one segment and the congestion window
increases by one segment for each acknowledgement returned
 Congestion Avoidance: When Slow Start forces a network to drop one or
more packets due to overload or congestion, Congestion Avoidance is used
to reduce the transmission rate
 Fast Retransmit: When a duplicate ACK is received, the sender does not
know if this is because a TCP segment was lost or because a segment was
delayed and received out of order at the receiver and if more than two
duplicate ACKs are received by the sender, it does not even wait for the
Retransmission Timeout to expire and retransmits the segment (as
indicated by the position of the duplicate ACK in the byte stream)
 Fast Recovery: The sender has implicit knowledge that there is data still
flowing to the receiver since duplicate ACKs can only be generated when a
segment is received and the sender only enters Congestion Avoidance
mode

 Round-Trip Time Estimation: When a host transmits a TCP packet to
its peer, and the reply does not come within the expected period, the
packet is assumed to have been lost and the data is retransmitted
 Over an Ethernet, no more than a few microseconds should be needed
for a reply
 This process called Round-Trip Time (RTT) estimation
 If the RTT estimate is too low, packets are retransmitted
unnecessarily; if too high, the connection can sit idle while the host
waits to timeout
Time Estimation

TCP and Manets
Challenges
 As the topology changes, the path is interrupted and TCP goes into
repeated, exponentially increasing time-outs, with severe performance
impact
 TCP performance in ad hoc multi-hop environment depends critically
on the congestion window in use
Significant TCP unfairness
 TCP injects packets at an increasing rate into the network until a
packet loss is detected and then, the sender shrinks its CWND,
retransmits the lost packet and resumes transmission at a lower
increasing rate
 If the losses persist at every retransmission, the sender doubles its wait
timer (i.e., the RTO) so that it can wait longer for the ACK of the
current packet being transmitted and is known as the exponential back
off strategy

Drawback of TCP Exponential Back Off

Effects of Partitions on TCP
Node 5 moves away from node 3 (short-term partition)

Node 5 moves away from node 3 (long-term partition)
Long Term Partition

1
2
3
4
5
6
7
8
9
The routing protocol reestablishes the path through node 6
Reestablishing Path

No communication between the partitions
Long Term Network Partition

Impact of Lower Layers on TCP
MAC Layer Impact:
 It is intended for providing an efficient shared broadcast channel
through which the involved mobile nodes can communicate
 In IEEE 802.11, RTS/CTS handshake is only employed when the
DATA packet size exceeds some predefined threshold
 Each of these frames carries the remaining duration of time for the
transmission completion, so that other nodes in the vicinity can hear
it and postpone their transmissions
 The nodes must await an IFS interval and then contend for the
medium again
 The contention is carried out by means of a binary exponential
backoff mechanism which imposes a further random interval
 At every unsuccessful attempt, this random interval tends to become
higher

Issues at the MAC Layer
 Consider a linear topology in which each node can only communicate with
its adjacent neighbors
 In addition, consider that in Figures 7.7(a) and 7.7(b) there exist a single
TCP connection running between nodes 1 and 5

TCP Throughput
TCP throughput is inversely proportional to the
number of hops
Larger the number of nodes a TCP connection needs to span, lower is
the end-to-end throughput, as there will be more medium contention
taking place in several regions of the network

Capture Conditions
 In Figure 7.7(c) where there are two independent connections,
(connection 2-3) (connection 4-5)
 Assuming that connection 2-3 experiences collision due to the
hidden node problem caused by the active connection 4-5 , node
2 will back off and retransmit the lost frame
 At every retransmission, the binary exponential backoff
mechanism imposes an increasingly backoff interval, and
implicitly, this is actually decreasing the possibility of success for
the connection 2-3 to send a packet as connection 4-5 will
“dominate” the medium access once it has lower backoff value
 In consequence, the connection 2-3 will hardly obtain access to
the medium while connection 4-5 will capture it

Network Layer Impact
 Routing strategies play a key role on TCP performance
 There have been a lot of proposed routing schemes and,
typically, each of them have different effects on the TCP
performance

DSR
 DSR protocol operates on an on-demand basis in which a node
wishing to find a new route broadcasts a RREQ packet
 The problem with this approach concerns the high probability
of stale routes in environments where high mobility as well as
medium constraints may be normally present
 The problem is exacerbated by the fact that other nodes can
overhear the invalid route reply and populate their buffers with
stale route information
 It can be mitigated by either manipulating TCP to tolerate such
a delay or by making the delay shorter so that the TCP can deal
with them smoothly

TORA
 TORA has been designed to be highly dynamic by establishing
routes quickly and concentrating control messages within a
small set of nodes close to the place where the topological change
has occurred
 TORA makes use of directed acyclic graphs, where every node
has a path to a given destination and established initially
 This protocol can also suffer from stale route problem similar to
the DSR protocol
 The problem occurs mainly because TORA does not prioritize
shorter paths, which can yield considerable amount of out-of-
sequence packets for the TCP receiver, triggering
retransmission of packets

Path Asymmetry Impact
 In ad hoc networks, asymmetry can occur by different reasons
including lower layer strategies
 Loss Rate Asymmetry: It takes place when the backward path is
significantly more error prone than the forward path
 Bandwidth Asymmetry: Here, forward and backward data follow
distinct paths with different speeds and this can happen in ad hoc
networks as well, since all nodes need not have the same interface
speed
 Media Access Asymmetry: This type of asymmetry may occur due to
characteristics of the wireless shared medium as TCP ACKs may have
to contend for the medium along with TCP data, which may cause
excessive delay as well as drops of TCP ACK packets

Route Assymetry
 Route asymmetry implies in distinct paths in both directions
 Route asymmetry is associated with the possibility of different
transmission ranges for the nodes
 The inconvenience with different transmission ranges is that it can
lead to conditions in which the forward data follow a considerably
shorter path than the backward data (TCP ACK) due to lack of power
in one (or more) of the nodes in the backward path
 However, multi-hop paths are prone to have low throughput and TCP
ACKs may face considerable disruptions

Solutions for TCP over Ad Hoc
 Mobility-Related
 TCP-Feedback
 TCP sender can effectively distinguish between route failure and
network congestion by receiving Route Failure Notification (RFN)
messages from intermediate nodes
 Upon receipt of a Route Re-establishment Notification (RRN)
message from the routing protocol, the sender leaves the frozen
state and resumes transmission using the same variables values
prior to the interruption
 A route failure timer is employed to prevent infinite wait for RRN
messages, is started whenever a RFN is received and upon
expiration of this timer, the frozen timers of TCP are reset hence
allowing the TCP congestion control to be invoked normally

 Explicit Link Failure Notification (ELFN) is a cross-layer proposal
in which TCP also interacts with the routing protocol in order to
detect route failure and take appropriate actions
 ELFN messages are sent back to the TCP sender from the node
detecting the failure
 ELFN messages contain sender and receiver addresses and ports, as
well as the TCP sequence number
 Whenever the TCP sender receives an ELFN message, it enters a
“stand-by” mode in which its timers are disabled and probe packets
are sent regularly towards the destination in order to detect route
restoration
 Upon receiving an ACK packet, the sender leaves the “stand-by”
mode and resumes transmission using its previous timer values and
state variables
The ELFN Approach

 Relies on the idea that routing error recovery should be
accomplished in a fast fashion by the routing algorithm
 It disables such a mechanism whenever two successive
retransmissions due to timeout occur, assuming that it
actually indicates route failure
 TCP sender doubles the RTO once and if the missing packet
does not arrive before the second RTO expires, the packet is
retransmitted again and again, but the RTO is no longer
increased
Fixed Retransmission Timeout
(RTO)

 The Ad hoc TCP (ATCP) protocol does not impose
changes to the standard TCP itself and instead, it
implements an intermediate layer between the network
and the transport layers in order to provide an enhanced
performance to TCP
 ATCP relies on the ICMP protocol and on the Explicit
Congestion Notification (ECN) scheme to detect /
distinguish network partition and congestion, respectively
 The intermediate layer keeps track of the packets to and
from the transport layer so that the TCP congestion
control is not invoked when it is not really needed
The ATCP Protocol

 Mobility in MANETs is extremely frequent and the packet
usually arrive out-of-order (OOO) at the destination
 The TCP-DOOR (Detection of Out-Of-Order and Response)
protocol focuses on the idea that OOO delivery of packets
can happen frequently in MANETs as a result of nodes
mobility
 TCP-DOOR implements a detection of such deliveries at
both entities: TCP sender and TCP receiver
TCP-DOOR (Detection
Out-Of-Order and
Response )

 The approaches that rely on feedback information from
inside the network (TCP-F, ELFN-based, ATCP) may fail in
situations where TCP sender is unable to receive data from
the next hop node
 The usage of explicit notification by the intermediate nodes,
such as ECN, raises many security concerns
 The assumption in TCP-DOOR that OOO packets are
exclusive results of route disturbance may not be true in a
quite a few scenarios
 The main concern addressed by the approaches presented so
far is how to avoid the TCP exponential backoff mechanism
when losses take place by factors other than congestion
Main Drawbacks

COPAS
 A protocol called COPAS (COntention-based PAth
Selection) has been proposed to address TCP performance
drop due to the capture problem and resulting unfairness
 COPAS implements two novel routing techniques in order
to contention-balance the network, namely, the use of
disjoint forward (for TCP data) and reverse (for TCP
ACK) paths to reduce the conflicts between TCP packets
traveling in opposite directions
Fairness-
Related

Route Establishment in COPAS
C
D
E
F
S
B H
I
J
4
4 2
5
3 1
8 7
G
D
E
C
F
S
B
G
H
I
J
4
4 2 5
3
1
8
7
TCP ACK
TCP Data(a) – Network contention
perceived at node D
(b) – Routes selected by node D

Average Aggregate Throughput
Simulation results of COPAS applied to scenario of 50 and 100 nodes
50 Nodes
100 Nodes

 Two unique features of ad hoc wireless networks are the key to
understand unfair TCP behaviors: Spatial reuse constraint and the
location dependency
 View a node and its interfering neighbors to form a neighborhood (the
neighborhood of a node X is formed by all nodes within communication
range of X)
 Flows get different feedback in terms of packet loss rate and packet delay
when congestion happens
 The main achievement of NRED is the ability to detect early congestion
and drop packets proportionally to a flow’s channel bandwidth
utilization
Neighborhood RED
(Randomly Early
Detection)

Node A’s
Neighborhood and
Distributed Queue
 Keep estimating the size of
neighborhood queue
 Once queue size exceeds certain
threshold, a drop probability is
computed
 This is propogated to provide
cooperative packet drop

Conclusions and Future Directions
 Concerning the error-detection strategies used in each approach, they
may be classified as network detection and end node detection
 Each approach has its advantages and disadvantages, and ideally, it is
better to combine the advantages of each one
 The interactions between TCP and MAC protocols could be improved
by using either using smaller values for the maximum TCP window size
or larger MAC IFS intervals, respectively
 It might be useful to investigate the possibility of increasing the
maximum number of possible retransmissions at the MAC layer as an
attempt to increase the probability of success of the local retransmission
scheme
 With regards to multipath routing strategies, further evaluation
towards improvements with respect to TCP support is needed
 Power management is a very important topic within MANETs, as they
are supposed to be composed mostly of battery powered devices
 Thus, power aware approaches offer increasing interest while little has
been done with regards to TCP, as is not power aware
 Interoperation between wireless mobile ad hoc networks and wired
networks is another subject that has not been adequately addressed
from TCP perspective
 Security considerations have become nowadays a hot issue in wireless
environments as wireless mediums are much more susceptible to
malicious users than the wired ones

Adhoc and Sensor Networks - Chapter 07

More Related Content

What's hot (20)

Similar to Adhoc and Sensor Networks - Chapter 07 (20)

More from Ali Habeeb (20)

Recently uploaded (20)

Adhoc and Sensor Networks - Chapter 07