Outline of the report

Introduction
QoS associated with Real-time applications
Real-time Traffic
Real-time for the Internet
References

Introduction
There are different kinds from which we can express the real-time constraints on
communication channels, but in general we have the two expressions Real-time" and
communication and around those expressions we can define the constraints and the
channel characteristics. There are ways to provide services according to their type and
requirements. Traditional networks protocols introduce the services with Best Effort
performance. Best Effort means that the services are delivered with the best
performance on average. There is also a concern with reliability, and mechanisms are used
to make sure that no data are lost, corrupted, or misordered during transit. It does not
matter if the packets or the message are delivered on time or not. Other services those
are more concerned with timing issues. In most cases, there is a requirement that data be
delivered at a constant rate equal tothe sending rate. In other cases, a deadline is
associated with each block of data, such that the data are not usable after the deadline
has expired and these services are called real-time services. In real-time applications
or services the connection is established only if there are performance guarantees to fulfill
the requirements of the application (i.e.: digital video, digital audio, streaming). These
requirements are different from one application to another as in this case we deal with
different types of traffic. There is also a need to quantify the requirements or make the
quality of the system measureable or handle different types of traffic that meets the
service needswhich is called Quality of Service QoS.
We can find that there are three players we have to deal with: QoS, kind of traffic, and
Network type; wealso should employ them for real time applications. If we consider the
packet transfer from the source to the destination, many issues raised that cause low
performance. A low performance is known if one or more of the following problems result:

Low throughput
Due to varying load from disparate users sharing the same network resources, the
bit rate (the maximum throughput) that can be provided to a certain data stream
may be too low for real-time multimedia services if all data streams get the same
scheduling priority.

Dropped packets
The routers might fail to deliver (drop) some packets if their data is corrupted or
they arrive when their buffers are already full. The receiving application may ask for
this information to be retransmitted, possibly causing severe delays in the overall
transmission.

Errors
Sometimes packets are corrupted due to bit errors caused by noise and
interference, especially in wireless communications and long copper wires. The
receiver has to detect this and, just as if the packet was dropped, may ask for this
information to be retransmitted.

Latency
It might take a long time for each packet to reach its destination, because it gets
held up in long queues, or takes a less direct route to avoid congestion. This is
different from throughput, as the delay can build up over time, even if the
throughput is almost normal.

Jitter

Packets from the source will reach the destination with different delays. A packet's
delay varies with its position in the queues of the routers along the path between
source and destination and this position can vary unpredictably. This variation in
delay is known as jitter and can seriously affect the quality of streaming audio
and/or video.

Out-of-order delivery
When a collection of related packets is routed through a network, different packets
may take different routes, each resulting in a different delay. The result is that the
packets arrive out of order. This problem requires special additional protocols
responsible for rearranging out-of order packets.

QoS associated with Real-time applications

In general, the real-time communications constraints are often around latency ,delay
jitter and loss rate. They can be divided into two types: Soft RT and Hard RT
communications and the constraints should be made on each type of them. Soft RT
communications have bounds on latency ,delay jitter and there is a maximum tolerable
packet loss rate for guaranteed service while Hard RT communications have no tolerance
for the packet loss. They have also maximum latency limit and delay jitter limit.
There are some issues that affect these constraints. For example, latency or packet
delay or the end-to-end packet delay results from stages that encounter the packet
through its journey from the source to the destination. Propagation delay ,
Packetization delay ,Switching delay and Queuing delay are the components that
form the latency. All can be considered fixed or can be assumed, but queuing delay is the
most important and should be studied.
Regarding the delay jitter, It can be eliminated by buffering at the receiver. Buffering is
just a mean for providing some delay so that more packets can be received and then
playbacked properly without pauses or interruptions. The amount of space needed for
buffering (i.e. Buffer size) can be calculated in terms of the peak rate of transmission and
the delay jitter. For instance, consider a video source that emits video frames with a rate
of 20 frames per second and each frame has 4 Mbytes of data and a jitter of 2 seconds
influences the transmission ,so the buffer size can be obtained by assembling data for 2
seconds before playback. The amount of data that should be assembled within 2 seconds
will be 20 (frame / second ) x 4 (Mbytes / frame ) x 2 second of jitter = 160 Mbytes which
is the buffer size. Bounding the delay jitter and reduce it can significantly affect the size of
buffer.
A study should also be made for the jitter and the buffering.

Real-time Traffic
Each of the Soft and Hard RT traffic has to be modeled to be studied in order to estimating
the bounds to jitter ,latency and loss rate for both or in general, introduce a guaranteed
quality of the service. The source of traffic is often a sensor that samples the data into a
digital form and then transmits this data through network.
The categories of traffic include:

CBR (Constant Bit Rate) Traffic

The source of data generates a fixed-sized data in a well known interval.
The opposite figure visualize the CBR.

VBR (Variable Bit Rate) Traffic

The source of data generates two intervals of data, the first is a fixed-sized data
and the other interval is idle (On/off sources). It also can generate a periodic data
packets with variable size.
The opposite figure visualize the forms of VBR.
P

t
SBR (Sporadic Bit Rate) Traffic
The source generates burst data having no pattern or order in time with variable
size.
It is a special type of VBR.
The opposite figure visualize the forms of SBR.
P

All we should do is to look to our service and then approximates its traffic into one of
categories mentioned above.
There are some models that can be used to express the traffic with some given
parameters:
t
(X min , S max) Model: bounds a traffic source with a peak rate.
This model can describe a CBR traffic that have an inter arrival time less than

X min

and a packet with size

calculated by

S max

as a maximum bound. The peak rate is

S max
X min .
T

( r ,T ) Model :This model divides the time axis into intervals of length

representing a frame. The maximum number of bit that can occur in the interval

is

r . The model provides tight bound for CBR traffic.

(X min , X avg , S max , I )

X avg

inter arrival time,

interval ,

S max

Model :In this model

X min

represents the minimum

represents the average inter arrival time over an I

is the maximum packet size and

is an interval that the

observation is made.
This model can describe a VBR traffic. The peak rate in this model is

S max
the average rate is X avg

( , )

S max
X min

and

Model: In this model, is the maximum burst size and is the average

rate of source traffic for a long time of observation.

packets
long time
a
large duration of observation

The number of
The average traffic can be determined by

The model applies as follows, the maximum number of bits that are generated from
a traffic source is

+ t where t

is any time interval. This model can describe a

bursty traffic source with an acceptable performance. In another words, during any
interval I can handle the maximum burst size plus the average size that is
measured within a long time. we Note that the utilization of the system here is low
since there will be time interval with no burst, but that is the cost we pay in order to
deal with bursts of data successfully.

Multiple rate bounding Model: That is a more

burst traffic, it employs the
rate

,each

(r , T )
over

accurate approximation to the

Model. There will be multiple bounding average

different

{ ( r 1,T 1 )( r 2,T 2 )( r 3, T 3 ) . . } withT 1<T 2<T 3

maximum of bits that can be generated is

average

Over

riTi

if

any

time

interval

intervals

I ,the

Ti1< I < Ti .

As the averaging interval gets longer, the traffic will be bounded by a rate lower
than its peak rate and closer to its long term average rate.

Real-time for the internet

After discussing the concept above, let's see how they are employed in practice. The
internet protocols RTP (Real time Transport Protocol) and RTCP (RTP control
protocol) are discussed in brief. RTP is a standard protocol for the transport of real-time
data, including audio and video. RTP consists of a data part and a control part which is
RTCP. These protocols may provide controlled delivery of multimedia traffic over the
Internet. They provide feedback mechanisms for the quality of service (QoS) monitoring,
such as delay, jitter and packet loss calculations. RTP can be considered as the data part
while RTCP is considered the control part.
RTP can provide the following :
Payload type identification : describes the specific media encoding so that it can
be changed if it has to adapt to a variation in bandwidth.
Source identification: which identifies the originator of the frame.
Sequence numbering:determines the packet order.
Time stamping: detects different delay jitter within a single stream and
compensate for it.
The main functions of the RTCP are:
QoS monitoring and congestion control
Identification
Session size estimation and scaling
The RTCP packets contain direct information for quality-of-service monitoring. The sender
reports (SR) and receiver reports (RR) exchange information on packet losses, delay
and delay jitter.
The sender reports (SR) format can be shown in the figure below:

The Receiver reports (RR) format can be shown in the figure below:

The contents of NTP timestamp (LSB) and NTP timestamp (MSB) from SR and the
contents of the inter arrival jitter field, Last Sender Report timestamp (LSR) field and
Delay since Last Sender Report (DLSR) fields from RR can be used to calculate the
parameters we need (i.e. Round trip delay m inter arrival jitter, packet loss)
The RTP timestamp indicates the relative sending time of this packet.
Round trip delay
The sender can the sender can directly calculate the round-trip delay
according the formula :

D= ALSRDLSR

where

is the time instant when the receiver report was received by the sender.

The middle 32-bits of NTP timestamp are copied by the receiver to LSR field
and the delay since last particular sender's report is stored until a
corresponding receiver report is sent.
Inter-arrival jitter
An estimate for inter-arrival jitter is calculated as follows:
Firstly, the difference

in packet spacing at the receiver compared to the packet

spacing at the sender is calculated according to the formula

D=(RjRi )(SjSi)
where

is the time of arrival and

is the RTP timestamp for a certain packet.

This delay variation value is calculated after every RTP packet. To avoid temporary
fluctuations the final value for inter-arrival jitter estimate is smoothed according to the
following equation:

Ji=(15/16) Ji1+(1/16)D
which gives only a small weight to the most recent observation.
Packet loss
The receiver reports also contain information about the lost packets. The fraction of lost
packets is defined to be the number of packets lost divided by the number of packets
expected, which are calculated based on actually received packets and the highest
sequence number received in RTP packets.

References
[1] Rajib Mall, "Real-Time Systems: Theory and Practice," Pearson, 2008.
[2] C.Aras , J.Kurose , D.Reeves " Real-time communication in packet-switched networks"
Proceedings of the IEEE , vol.83 pp.122-139 ,January 1994.
[3] Data and Computer Communications (8th, Eighth Edition) - By William Stallings.
[4] en.wikipedia.org/wiki/Quality_of_service
[5] http://www.ciscopress.com/articles/article.asp?p=357102
[6] http://technet.microsoft.com/en-us/library/bb742481.aspx
[7] KOISTINEN, T. "Protocol overview: RTP and RTCP" Nokia Telecommunications; 2000