A QoS Based MAC Protocol For Wireless Ad-hoc Network

International Journal of Network Security & Its Applications (IJNSA), Vol.3, No.2, March 2011
DOI : 10.5121/ijnsa.2011.3215 196
A QoS Based MAC Protocol For Wireless Ad-hoc
Network
Vandana khare Dr. D. SrinivasRao Dr. Y. Madhavee Latha
Associate professor, ECE Professor- ECE Principal
CMEC,Hyd(A.P) JNTUH,Hyd (A.P) MRECW, Hyd(A.P)
Email:vandna369@gmail.com Email: dsraoece@yahoo.co.uk Email: madhuvsk2003@yahoo.co.in
Abstract—Multimedia communication over wireless Ad-hoc networks has become the driving
technology for many of the important applications, experiencing dramatic market growth and promising
revolutionary experiences in personal communication, gaming, entertainment, military, security,
environment monitoring, and more. The advances in wireless communications and growth of real-time
applications have necessitated the development of wireless networks that can support high Quality of
Service (QoS) and power control. A node in an ad hoc network is normally battery operated which poses
a huge constraint on the power consumption of such a node. Hence, designing a power efficient MAC
protocol for ad hoc wireless networks is a major challenge. In this paper, we propose a CDMA based
power controlled medium access protocol for mobile and ad hoc network (MNA). The protocol conserves
power and provides QoS guarantees for multimedia traffics. In that network one of the fundamental
challenger in MANETs is how to increase the overall network throughputs well as reading the delay
while maintaining how energy consumption for packet processing in communication. simulation results
shows that the performance of the protocol with increase in traffic while QoS is better in terms of Energy
consumption, throughput, & communication delay than existing protocol.
Keywords: QoS, MANETS, Multimedia, CDMA throughput, delay.
I. INTRODUCTION
In recent years wireless communication networks have become increasingly popular. Many
types of wireless services have become available, including cellular systems, satellite
communication networks, and wireless local area networks (WLANs) [1, 2]. The increasing
popularity of WLANs and wireless devices has led to greater interest in wireless ad hoc
networks. An ad hoc network [3] is formed by wireless, potentially mobile hosts, without
requiring the use of any fixed infrastructure, and can be set up in the environment where the
wiring of a conventional network is difficult or not economically feasible. Wireless ad hoc
networks face challenges that are not present in wired networks. In wired networks,
transmission errors typically occur at a low rate and interference among different
communication flows is minimal. Collision detection is usually fast and easy in wired
networks. Wireless communication, however, requires a shared transmission medium that is
highly error-prone. Hence, in wireless communication, there is a much higher chance for
collisions to occur. It is also more difficult to detect a collision in a wireless network. Often the
lack of a reply message is the only way for a node to detect a collision. Therefore, compared to
a wired network, a wireless network requires a different and more complicated medium access
control (MAC) layer. This paper focuses on the issues on MAC layer for wireless networks.
The rest of this paper is organized as follows: Section II discusses the basics of video
compression and the 802.11 protocol stack respectively. Section III outlines the video QoS
enhancing algorithm. Section IV discusses the simulation set up. Section V enumerates and
analyses the simulation results. Finally, we conclude the paper in Section VI.

197
II. RELATED WORK
Several CDMA-based MAC protocols for MANETs have been proposed in past.
These protocols, in general, are based on random channel access, whereby a terminal with a
packet to transmit can proceed immediately with its transmission, irrespective of the state of the
channel. We refer to such schemes as random access CDMA (RA-CDMA). Under appropriate
code assignment and spreading-code schemes, RA-CDMA protocols are guaranteed to be free
of primary collisions. However, the nonzero cross-correlations between different CDMA codes
can induce multi-access interference (MAI), resulting in secondary collisions at a receiver
(collisions between two or more transmissions that use different CDMA codes). This problem
is known as the near-far problem. The near-far problem can cause a significant reduction in
network throughput, and is to be overcome for designing CDMA-based MAC protocols for
MANETs.
For the integration of CDMA with ad hoc architecture various proposal were made in
past. In [4] the addresses part of the packet are spread using the common code, while the rest of
the packet is spread using the transmitter-based approach. A receiver notes the address of the
source terminal and uses this address to switch to the corresponding code. In [5] the authors
proposed the coded tone sense protocol, in which K busy tones are associated with K spreading
codes. During packet reception on a certain code, the receiving station broadcasts the
corresponding busy tone. In [6] all terminals send the RTS-CTS packets on a common code,
while the data packets are sent using a transmitter- or a receiver-based approach. Somewhat
similar approaches were proposed in [7] and [8]. In all the above protocols, the authors assume
perfect orthogonality between spreading codes, i.e., they ignore the near-far problem.
A reservation-based scheme was proposed in [9], whereby small control packets are
used to request slot assignments for data packets. The authors investigated the use of FHSS to
avoid MAI. Their approach, however, cannot be used for DSSS, which is the method of choice
in recent wireless standards (e.g. IS-95). In [10] and [11] the authors proposed distributed
channel assignment algorithms for SS Multihop networks. Those protocols, however, do not
allow for any MAI, and hence cannot support concurrent transmissions of signals with different
codes. Clustering as proposed in [12] is another interesting approach for power control in
CDMA networks. It simplifies the forwarding function for most terminals, but at the expense of
reducing network utilization (since all communications have to go through the cluster heads).
This can also lead to the creation of bottlenecks.
In [13] the authors proposed the use of a multi-user detection circuit at the receiver to mitigate
the near-far problem in MANETs. The proposed scheme also requires the use of GPS receivers
to provide accurate position and timing information. Such a scheme relies heavily on physical
layer techniques to mitigate MAI, and makes no effort to account for MAI at the MAC layer.
Moreover, although it is feasible to deploy multi-user GPS receivers at the base station,
presently it is impractical (and expensive) to implement such receivers within the mobile
terminal. Recently, an interesting approach for joint scheduling and power control in ad hoc
networks was proposed [14]. This approach, however, requires a central controller for
executing the scheduling algorithm, i.e., it is not a truly distributed solution. Furthermore, it
assumes the existence of a separate feedback channel that enables receivers to send their SNR
measurements to their respective transmitters in a contention free manner. In [15] and [16] the
authors analyzed RA-CDMA protocols for MANETs in the presence of MAI. They assumed
that transmissions of all neighbors produce the same noise effect, and therefore, the SNR

198
threshold can be converted into a threshold on the number of transmissions (n) in the receiver’s
neighborhood. A packet is correctly received when that number is less than the predetermined
threshold n. hence; the protocol was called CDMA/n. Although such an approximation may not
be accurate in topologies where nodes are not equally spaced, it shows that MAI can
significantly degrade network performance in such approaches
III. PROPOSED PROTOCOL CONCEPT
The users are the sources which send the message. This message will be coded and the signal
will be modulated by the carrier signal. The codes will be generated with the linear feedback
shift register (LFSR) mechanism. The generated code is known as pseudo random noise
sequence. For example when the message is of 8 bit length, The generated code will be of 32
bit length. This code is generated for single user. So similarly we can generate different codes
for different users and modulate these signals by using modulator. And these signals are
transmitted by transmitting antenna. The generation of the code plays an important role in the
CDMA communication system. For generating these codes we use the LFSR mechanism. The
band width present for the user can be utilized properly in CDMA communication system. As
the CDMA communication system sends the data for ‘n’ number of users at a time. So we can
transmit the number of data of different users at the same time. All the messages coded will be
kept in the transmitting section to transmit the data. So the transmitted data contains number of
messages produced by the different sources. The receiving antenna receives the signals and
demodulates the signal in the block of demodulator. The message received will be decoded
after demodulation. The decoded message is send to the matched filter. The matched filter
matches the data and sends to de-spreader to make the message to reach destination. Hence the
messages generated by the different users will be transmitted and will be received by the
different users at different places which makes the global requirement comes true by using
CDMA communication system. The source generates the message. The CDMA code is
allocated to the message generated by the source. The network creation takes place for the
generated code by the CDMA communication system. The source/destination routes the
message in the desired manner. Finally selects an optimal path for transferring the message.
Request to send (RTS) and clear to send (CTS) mechanisms are verified and the message is
send i.e. performance of the controlled mechanism is performed using the RTS and CTS
mechanisms.
IV. DESIGN APPROACH
Several CDMA-based MAC protocols for MANETs have been proposed in the literature
[1, 2, 3, 4]. These protocols, in general, are based on random channel access, whereby a
terminal with a packet to transmit can proceed immediately with its transmission (starting,
possibly, with an RTS/CTS exchange), irrespective of the state of the channel. We refer to such
schemes as random access CDMA (RA-CDMA). Under appropriate code assignment and
spreading-code schemes, RA-CDMA protocols are guaranteed to be free of primary collisions.
However, as explained the nonzero cross-correlations between different CDMA codes can
induce multi-access interference (MAI), resulting in secondary collisions at a receiver
(collisions between two or more transmissions that use different CDMA codes). In the
literature, this problem is known as the near-far problem. As shown in Section 2, the near-far
problem can cause a significant reduction in network throughput, and hence cannot be
overlooked when designing CDMA-based MAC protocols for MANETs. Accordingly, the

199
main goal of this paper is to provide a CDMA-based MAC solution for MANETs that
addresses the near-far problem.
In our protocol, the transmission powers are dynamically adjusted such that the MAI at
any receiver is not strong enough to cause a secondary collision. This results in a significant
improvement in network throughput at no additional cost in energy consumption. In fact, the
proposed protocol is shown to achieve some energy saving compared to the 802.11 schemes.
To the best of our knowledge, this is the first attempt to address the near-far problem in the
design of MAC protocols for MANETs.
4.1. THE NEAR-FAR PROBLEM IN RA-CDMA
The roots of the near-far problem lies in the fact that unlike FDMA and TDMA
channels which can be completely orthogonal, CDMA codes suffer from nonzero cross-
correlation between codes. When a CDMA receiver de-spreads a signal, it effectively computes
the cross-correlation between the signal and a locally generated PN sequence. If this PN
sequence is identical to the one used to spread the signal at the transmitter (i.e., the message is
intended to this receiver), cross-correlation computations restore the original information data.
Otherwise, such computations result in either a zero or a nonzero value, depending on whether
the system is synchronous or asychnronous. A system is called time-synchronous if all signals
originate from the same transmitter, as in the case of the downlink of a cellular CDMA
network. In here, synchrony is manifested in two ways. First, different transmissions that are
intended for different receivers will have a common time reference.
Second, from the viewpoint of a given mobile terminal, all signals (intended or not) propagate
through the same paths, and thus suffer the same time delays. In synchronous systems, it is
possible to design completely orthogonal spreading codes. In fact, in the IS-95 standard for
cellular CDMA networks [19], each user of the channel is assigned a Hadamard (or Walsh)
code. These codes are orthogonal and are used to “channelize” the available bandwidth.
On the other hand, a system is called time-asynchronous if signals originate from
multiple transmitters, as in the case of the uplink of cellular networks and also in MANETs.
The reasons behind the naming are twofold. First, since signals originate from different
transmitters, it is generally not feasible to have a common time reference for all the
transmissions that arrive at a receiver. Second, these transmissions propagate through different
paths; thus, they suffer different time delays [20]. In an asynchronous system, it is not possible
to design spreading codes that are orthogonal for all time offsets [19]. In this case, the cross-
correlation between codes cannot be neglected. In fact, codes that is orthogonal in synchronous
systems (e.g., Hadamard codes) exhibit high cross-correlation when not perfectly synchronized.
Instead, PN codes that are designed specifically to have low cross-correlation are used. While
the code design problem is crucial in determining the system performance, of greater
importance is the problem of nonzero cross-correlation of the PN codes [21]. Unintended
transmissions add nonzero MAI during the dispreading at a receiver. The near-far problem is a
severe consequence of MAI, whereby a receiver who is trying to detect the signal of the ith
transmitter may be much closer in distance to, say, the jth
transmitter than the ith transmitter.
When all transmission powers are equal, the signal from the jth
transmitter will arrive at the
receiver in question with a sufficiently larger power than that of the ith
transmitter, causing
incorrect decoding of the ith
transmission (i.e., a secondary collision).

4.2 ALLOCATION SCHEME
In our design, we use two frequency channels, one for data and one for control (i.e.,
FDMA-like partitioning). A common spreading code is used by all nodes over the control
channel, while several terminal
architecture is shown in Figure 4.1. Note that the different codes used over the data channel are
not perfectly orthogonal. However, because of the frequency separation, a signal over the
control channel is completely orthogonal to any signal (or code) over the data channel. The
splitting of the available bandwidth into two non
needed to allow a terminal to transmit and receive simultaneously over the control and data
channels, irrespective of the signal power. As explained, this protocol utilizes the fact to allow
interference-limited transmissions that use (quasiorthogonal) data channel codes to proceed
concurrently.
Figure 4.1: Data and contro
4.3 Channel Model
In designing the protocol,
(1) The channel gain is stationary for the duration of the control and the ensuing data
packet transmission periods;
(2) The gain between two terminals is the same in both directions.
(3) Data and control packets between a pair of terminals observe similar channel gains.
4.3.1 NOISE IN COMMUNICATI
Noise plays a crucial role in communication systems. In theory, it determines the
theoretical capacity of the channel. In practice it determines the number of errors occurring in a
digital communication.
Noise is a random signal whose value cannot be predicted. It can only be
assumed of taking a particular value, or range of values randomly. The probability density
function (pdf) of a random signal, or random variable x, is defined to be the probability
that the random variable x takes a value between
ALLOCATION SCHEME
like partitioning). A common spreading code is used by all nodes over the control
channel, while several terminal-specific codes can be used over the data channel. This
y orthogonal to any signal (or code) over the data channel. The
splitting of the available bandwidth into two non-overlapping frequency bands is fundamentally
els, irrespective of the signal power. As explained, this protocol utilizes the fact to allow
limited transmissions that use (quasiorthogonal) data channel codes to proceed
Figure 4.1: Data and control codes in the proposed protocol.
The channel gain is stationary for the duration of the control and the ensuing data
packet transmission periods;
The gain between two terminals is the same in both directions.
Data and control packets between a pair of terminals observe similar channel gains.
NOISE IN COMMUNICATION SYSTEMS:
random signal whose value cannot be predicted. It can only be
of a random signal, or random variable x, is defined to be the probability
dom variable x takes a value between and . Defined as:
200
like partitioning). A common spreading code is used by all nodes over the control
can be used over the data channel. This
y orthogonal to any signal (or code) over the data channel. The
overlapping frequency bands is fundamentally
els, irrespective of the signal power. As explained, this protocol utilizes the fact to allow
limited transmissions that use (quasiorthogonal) data channel codes to proceed
The channel gain is stationary for the duration of the control and the ensuing data
Data and control packets between a pair of terminals observe similar channel gains.
random signal whose value cannot be predicted. It can only be
of a random signal, or random variable x, is defined to be the probability

The probability that the random variable will take a value lying between
integral of the pdf over the interval
The probability
A density satisfying equation is termed
The probability distribution function
variable, x is less than :
From the rules of integration:
The meaning of the two functions
Continuous distribution. An example of a continuous distribution is the Normal, or Gaussian
distribution:
where m is the mean value of
normalised. This expression is important as many naturally occurring noise sources can be
described by it, e.g. white noise. Further, we can simplify the expression by considering the
source to be a zero mean random variable, i.e. m = 0.
Figure 4.2 A zero mean Gaus
The probability that the random variable will take a value lying between and
integral of the pdf over the interval :
is unity. Thus:
isfying equation is termed normalised.
The probability distribution function is defined to be the probability that a random
The meaning of the two functions and become clearer with the aid of examples.
where m is the mean value of . The constant term ensures that the distribution is
expression is important as many naturally occurring noise sources can be
source to be a zero mean random variable, i.e. m = 0.
A zero mean Gaussian distribution with
201
is then the
is defined to be the probability that a random
become clearer with the aid of examples.
. The constant term ensures that the distribution is
expression is important as many naturally occurring noise sources can be

How would this be used If we want to know the probability of, say, the noise signal,
having the value , we would evaluate:
In general, to evaluate
then from with m = 0 we have:
The distribution function is usually written in terms of the error function
where is given by:
which has the property:
The integral is difficult to evaluate and is approximated by use of tables which exist for various
value of x. Some values of
the value falling outside .
Table 2: Some values for the error function
A more practically useful measure is to consider
How would this be used If we want to know the probability of, say, the noise signal,
, we would evaluate:
, if we use:
is usually written in terms of the error function
are shown in Table. Note the extremely low probability of
Some values for the error function
A more practically useful measure is to consider , which gives us:
202
How would this be used If we want to know the probability of, say, the noise signal, ,
:
are shown in Table. Note the extremely low probability of
, which gives us:

This is useful in that it gives us a feeling of how likely it is that the value of the noise signal will exceed
certain thresholds as given by k. A similar table to that of Table can be evaluated for
Table.
Table1: Some values for the error function,
So we see that it is highly unlikely for our noise signal to take values outside the range
The complementary error function,
Discrete distribution. Probability density functions need not be continuous. If a random
variable can only take discrete values, its pdf takes the forms of lines. An example of a discrete
distribution is the Poisson distrib
where . The most celebrated example of the Poisson distribution is its accurate
description of the number of men in the Austro
each year. We cannot predict the value a random variable may take on a
We can introduce measures that summarise what we expect to happen on average. The two
most important measures are the mean (or expectation) and the standard deviation.
The mean of a random variable x is defined to be:
or, for a discrete distribution:
(If n can take only a limited range of values we adopt the convention that
range).
In the examples above we have assumed that the mean of the Gaussian distribution to be 0, the
mean of the Poisson distribution to is found to be
parlance, the average value. On average
A similar table to that of Table can be evaluated for and is given in
Some values for the error function,
So we see that it is highly unlikely for our noise signal to take values outside the range
function, , is defined as:
distribution is the Poisson distribution:
. The most celebrated example of the Poisson distribution is its accurate
description of the number of men in the Austro-Hungarian army killed by kicks from a horse in
each year. We cannot predict the value a random variable may take on a particular occasion.
most important measures are the mean (or expectation) and the standard deviation.
of a random variable x is defined to be:
can take only a limited range of values we adopt the convention that
mean of the Poisson distribution to is found to be . The mean of a distribution is, in common
parlance, the average value. On average
203
and is given in
So we see that it is highly unlikely for our noise signal to take values outside the range .
. The most celebrated example of the Poisson distribution is its accurate
Hungarian army killed by kicks from a horse in
particular occasion.
outside this
. The mean of a distribution is, in common

then, the Austro-Hungarian army lost
deviation is a measure of the spread
standard deviation means the distribution (and hence occurrences) are close to the mean. A
large value indicates a wide range of possible outcomes. The standard deviation
be:
or, for a discrete distribution:
The square of the standard deviation is called the variance. The Gaussian distribution contains
the standard deviation within its definition. The Poisson distribution has a standard deviation of
.
In many cases the noise present in communication signals can be modeled as a zero
Gaussian random variable. This means that its amplitude at a particular time has a pdf.
statement that noise is zero-mean says that, on average, the noise signal takes the value zero.
The mean power in the noise signal is equal to the variance of the pdf. We have already seen
that the signal-to-noise ratio (SNR) is an important quantity
a communication channel. The noise power referred to in the definition of SNR is the mean
noise power. It can therefore be rewritten as:
SNR=10log s/(σ* σ)
Moreover, if the source of this noise is thermal noise we that
source of noise, the variance is sometimes written in the terms of
that:
Note that is an energy, because
4.4 CONTROLLED ACCESS CDMA (CA
CA-CDMA protocol is contention based and uses a modified RTS
mechanism. RTS and CTS packets are transmitted over the control channel (on the common
code) at a fixed (maximum) power Pmax. These packets are received by all potentially
interfering nodes, as in the IEEE 802.11 scheme.
For the ensuing data packet, the receiver and the transmitter must agree on two
parameters: the spreading code and the transmission power. Code selection can be done
according to any code assignment scheme. As the transmission power increases,
rate at the intended receiver decreases (i.e., link quality improves), but the MAI added to other
ongoing receptions increases. In a CA
army lost men per year as a result from horse kicks. The standard
spread of the probability distribution around the mean. A small
e value indicates a wide range of possible outcomes. The standard deviation is defined to
ithin its definition. The Poisson distribution has a standard deviation of
In many cases the noise present in communication signals can be modeled as a zero
Gaussian random variable. This means that its amplitude at a particular time has a pdf.
mean says that, on average, the noise signal takes the value zero.
noise ratio (SNR) is an important quantity in determining the performance of
noise power. It can therefore be rewritten as:
Moreover, if the source of this noise is thermal noise we that . Irrespective of the
source of noise, the variance is sometimes written in the terms of the power per unit Hz, so
is an energy, because
.
4.4 CONTROLLED ACCESS CDMA (CA-CDMA) PROTOCOL
CDMA protocol is contention based and uses a modified RTS-CTS reservation
ring nodes, as in the IEEE 802.11 scheme.
according to any code assignment scheme. As the transmission power increases, the bit erro
intended receiver decreases (i.e., link quality improves), but the MAI added to other
ongoing receptions increases. In a CA-CDMA protocol, terminals exploit knowledge of the
204
men per year as a result from horse kicks. The standard
of the probability distribution around the mean. A small
is defined to
ithin its definition. The Poisson distribution has a standard deviation of
In many cases the noise present in communication signals can be modeled as a zero-mean,
Gaussian random variable. This means that its amplitude at a particular time has a pdf. The
mean says that, on average, the noise signal takes the value zero.
in determining the performance of
. Irrespective of the
the power per unit Hz, so
CTS reservation
the bit error
intended receiver decreases (i.e., link quality improves), but the MAI added to other
CDMA protocol, terminals exploit knowledge of the

205
power levels of the overheard RTS and CTS messages to determine the power that they can use
without disturbing the ongoing receptions.
4.5 INTERFERENCE MARGIN
An interference margin is needed to allow terminals at some distance from a receiver to start
new transmissions in the future. This margin computation is explained below. Consider an
arbitrary receiver i. Let µ∗ be the Eb/N0 eff ratio that is needed to achieve the target bit error
rate at that receiver. It follows from that to achieve the target error rate, we must have
where Pthermal is the thermal noise power and P(i) MAI is the total MAI at receiver I, so the
minimum required received power is (P(i) 0 )min = µ∗(Pthermal + P(i)*MAI). The interference
margin strongly depends on the network load, which itself can be conveyed in terms of the so-
called noise rise (ξ(i)), defined as follows:
Note that (P(i) 0 )min = ξ(i)µ∗Pthermal is also dependent on the noise rise. While more
capacity can be achieved by increasing the noise rise (i.e., allowing larger P(i) MAI), the
maximum allowable noise rise is constrained by two factors. First, Federal Communications
Commission (FCC) regulations limit the power to some fixed value (e.g., 1 Watt for 802.11
devices). Given this maximum transmission power, as the noise rise is increased, the received
power (P(i) 0 )min must increase (µ∗ and Pthermal are constants) and hence, the maximum
range (or coverage) for reliable communication will decrease. Second, increasing the noise rise
increases the power used to transmit the packet, which in turn increases energy consumption.
Energy is a scarce resource in MANETs, so it is undesirable to trade off energy for throughput.
We set the interference margin used by a transmitter to the maximum planned noise rise
(ξmax), which is obtained by taking into account the above two restrictions on ξ(i).
4.6 ACESSING SCHEME
The admission scheme allows only transmissions that cause neither primary nor secondary
collisions to proceed concurrently. RTS and CTS packets are used to provide three functions.
The format of the RTS packet is similar to that of the IEEE 802.11, except for an additional
two-byte field that contains the P(j) map value. The format of the RTS packet is as shown in
figure 4.2.
Figure4.3: Format ofthe RTS packet in the CA- CDMA protocol.

206
These packets allow nodes to estimate the channel gains between transmitter-receiver
pairs. Second, a receiver i uses the CTS packet to notify its neighbors of the additional noise
power (denoted by P(i) noise) that each of the neighbors can add to terminal i without
impacting i’s current reception. These neighbors constitute the set of potentially interfering
terminals. Finally, each terminal keeps listening to the control channel regardless of the signal
destination in order to keep track of the average number of active users in their neighborhoods.
4.6.1 PACKET TRANSFER:
The process of packet transfer over the network is explained as follow. If a terminal j
has a packet to transmit, it sends a RTS packet over the control channel at Pmax, and includes
in this packet the maximum allowable power level (P(j) map) that terminal j can use that will
not disturb any ongoing reception in j’s neighborhood. Upon receiving the RTS packet, the
intended receiver, say terminal i, uses the predetermined Pmax value and the power of the
received signal P (ji) received to estimate the channel gain Gji = P(ji) received/Pmax between
terminals i and j at that time. Terminal i will be able to correctly decode the data packet if
transmitted at a power P(ji) min given by:
where P(i) MAI-current is the effective current MAI from all already ongoing transmissions.
Note that because of the assumed stationarity in the channel gain over small time intervals, Gji
is approximately constant throughout the transmissions of the control packet and the ensuing
data packet. Now, P(ji) min is the minimum power that terminal j must use for data
transmission in order for terminal i to correctly decode the data packet at the current level of
interference. This P(ji) min, however, does not allow for any interference tolerance at terminal
i, and thus all neighbors of terminal i will have to defer their transmissions during terminal i’s
ongoing reception (i.e., no simultaneous transmissions can take place in the neighborhood of i).
The power that terminal j is allowed to use to send to i is given by:
If P(ji) allowed < P(ji) min, then the MAI in the vicinity of terminal i is greater than the one
allowed by the link budget. In this case, i responds with a negative CTS, informing j that it
cannot proceed with its transmission. This is to prevent transmissions from taking place over
links that provides high MAI. This consequently increases the number of active links in the
network (subject to the available power constraints). On the other hand, if P(ji) allowed > P(ji)
min, then it is possible for terminal i to receive j’s signal but only if P(ji) allowed is less than
P(j) map (included in the RTS). This last condition is necessary so that transmitter j does not
disturb any of the ongoing transmissions in its vicinity. In this case, terminal I calculates the
interference power tolerance P(i) MAI-future that it can endure from future unintended
transmitters. This power is given by

207
Note that the factor 3W/2 comes from the spreading gain. The next step is to equitably
distribute this power tolerance among future potentially interfering users in the vicinity of
i. The rational behind this distribution is to prevent one neighbor from consuming the entire P(i)
MAI-future.
The distribution of this power tolerance is given as;
If terminal i keeps track of the number of simultaneous transmissions in its neighborhood,
donated by K(i) inst. Monitored by the RTS/CTS exchanges over the control channel. In
addition, i keeps an average K(i) avg of K(i) inst over a specified window. Then, K(i) is
calculated as:
where β > 1 is a safety margin.
While communication it is observed that when the within interference is more than the neighbor
interference the level of effect observed is high to reduce this interference effect the neighbor
interference is to be reduced. On the calculation if the average interference level per node the
CTS packets are generated with the available interference margin with the required power
transmission request to the neighboring node as shown in figure 4.3.
Figure 4.4: Format of the CTS packet in the proposed protocol.
This demanded power Derived from the CTS packet is then compared with the available power
limit and transmitted back for acceptance over the control channel to forward the packet. In
case the requested power is more than the limiting power the request is denied.
V. RESULT
Our scheme is tested for different QoS parameters like throughput parameters, energy
consumption, Simultaneous transmission probability, Communication delay.
5.1 Throughput plot-
The throughput of Ca-CDMA is less compare to that of the 802.11 with respect to the offered
load.
Fig 5.1: Throughput for CDMA Scheme

208
5.2 Energy Consumption plot-
The energy consumption of Ca-CDMA is less compare to that of the 802.11 with respect to
the offered load.
Fig 5.2 Energy consumption for CDMA Scheme
5.3 Simultaneous transmission probability-
Simultaneous transmission probability is more in Ca-CDMA communication
compared to 802.11 with respect to the load offered.
Fig5.3 transmission probability for CDMA Scheme
5.4 Communication Delay –
From the above graph we observed that as offered load increases for CA-CDMA with time,
communication delay goes down as compared with 802.11.
Fig5.4 communication delay for CDMA Scheme
VI CONCLUSION
The MAC protocol literature was surveyed to understand the state of the field and the
existing level of research. There are several QoS issues in the MAC layer of Ad hoc networks,
and the existing MAC protocols have largely concentrated on improving and optimizing these
issues in isolation of each other. Here, we introduce a new QoS-aware Mac protocol for Ad hoc
networks that simultaneously addresses several QoS issues. In this work, we proposed a

209
CDMA-based power controlled MAC protocol for wireless ad hoc networks. This protocol,
called CA-CDMA, accounts for the multiple access interference, thereby solving the near-far
problem that undermines the throughput performance in MANETs. CACDMA uses channel-
gain information obtained from overheard RTS and CTS packets over an out-of-band control
channel to dynamically bind the transmission power of mobile terminals in the vicinity of a
receiver. It adjusts the required transmission power for data packets to allow for interference-
limited simultaneous transmissions to take place in the neighborhood of a receiving terminal.
We compared the performance of our protocol with that of the IEEE 802.11 scheme. Our
simulation results showed that CA-CDMA can improve the network throughput by up to 280%
and, at the same time, achieve 50% reduction in the energy consumed to successfully deliver a
packet from the source to the destination. To the best of our knowledge, CA-CDMA is the first
protocol to provide a solution to the near-far problem in CDMA ad hoc systems at the protocol
level.
Our future work will focus on other capacity optimizations such as the use of directional
antennas in CDMA-based protocols. Because of MAI effects, CDMA benefits significantly
from smart antennas. The 802.11 scheme allows nodes to increase their information rate up to
11 Mbps when the power at the receiver is far more than necessary to achieve 2 Mbps. It could
be possible to improve the proposed scheme by increasing the information rate (i.e. decreasing
the processing gain) when the MAI is much less than the planned interference margin when
load is low. This is desirable to allow the developed protocol to adapt to different working
conditions in terms of the load offered by the users in the network.
VII REFERENCES:
[1] S. Chakrabarti and A Mishra, “QoS Issues in Ad Hoc WirelessNetworks”, IEEE
Communications Magazine, Vol. 39, No. 2, February2001.
[2] J. J. Garcia-Luna-Aceves and J. Raju, “Distributed Assignment of Codes for Multihop
Packet-Radio Networks”, IEEE MILCOM,November 1997 Barbara Hughes and Vinny Cahill,
“Towards Real-time Event-based Communication in Mobile Ad Hoc WirelessNetworks.” 2nd
International Workshop on real-time LANs in the Internet Age, 2003.
[3] Y. B. Ko and N. H. Vaidya, “Location Aided Routing (LAR) in Mobile Ad Hoc Networks”,
ACM MOBICOM, 1998.
[4] E. Sousa and J. A. Silvester. Spreading code protocols for distributed spread-spectrum
packet radio networks. IEEE Transactions on Communications, 36(3):272–281, Mar. 1988.
[5] K.-W. Hung and T.-S. Yum. The coded tone sense protocol for multihop spread-spectrum
packet radio networks. In Proceedings of the IEEE GLOBECOM Conference, volume 2, pages
712–716, 1989.
[6] J. Garcia-Luna-Aceves and J. Raju. Distributed assignment of codes for multihop packet-
radio networks. In Proceedings of the IEEE MILCOM Conference, volume 1, pages 450–454,
1997.
[7] M. Joa-Ng and I.-T. Lu. Spread spectrum medium access protocol with collision avoidance
in mobile ad-hoc wireless network. In Proceedings of the IEEE INFOCOM Conference,
volume 2, pages 776–783, 1999.

210
[8] S.-L. Wu, C.-Y. Lin, Y.-C. Tseng, and J.-P. Sheu. A new multi-channel mac protocol with
on-demand channel assignment for multi-hop mobile ad hoc networks. In Proceedings of the
International Symposium on Parallel Architectures, Algorithms and Networks, pages 232–237,
2000.
[9] J. E. Wieselthier, A. Ephremides, and J. A. B. Tarr. A distributed reservation-based CDMA
protocol that does not require feedback information. IEEE Transactions on Communications,
36(8):913–923, Augut 1988.
[10] I. Cidon and M. Sidi. Distributed assignment algorithms for multihop packet radio
networks. IEEE Transactions on Computers, 38(10):1353–1361, Oct. 1989.
[11] A. Ephremides and T. V.Truong. Scheduling broadcasts in multihop radio networks. IEEE
Transactions on Communications, 38(4):456–460, Apr. 1990.
[12] T. J. Kwon and M. Gerla. Clustering with power control. In Proceedings of the IEEE
MILCOM Conference, volume 2, pages 1424–1428, 1999.
[13] V. Rodoplu and T. Meng. Position based CDMA with multiuser detection (P-
CDMA/MUD) for wireless ad hoc networks. In IEEE Sixth International Symposium on
Spread Spectrum Techniques and Applications, volume 1, pages 336–340, 2000.
[14] T. ElBatt and A. Ephremides. Joint scheduling and power control for wireless ad-hoc
networks. In Proceedings of the IEEE INFOCOM Conference, volume 2, pages 976–984, 2002.
[15] M.-S. Chen and R. Boorstyn. Throughput analysis of code division multiple access
(CDMA) multihop packet radio networks in the presence of noise. In Proceedings of the IEEE
INFOCOM Conference, pages 310–316, 1985.
[16] O. deSouza, M.-S. Chen, and R. R. Boorstyn. A comparison of the performance
of protocols in packet radio networks. In Proceedings of the IEEE MILCOM Conference,
volume 2, pages 455–460, Oct 85.

A QoS Based MAC Protocol For Wireless Ad-hoc Network

More Related Content

What's hot (20)

Similar to A QoS Based MAC Protocol For Wireless Ad-hoc Network (20)

Recently uploaded (20)

A QoS Based MAC Protocol For Wireless Ad-hoc Network