Zagazig University

Faculty of engineering

The enhancements of dynamic spectrum access

techniques in cognitive radio ad hoc networks
By

Eng. Abeer Abd Al Rahman Abd Al Hameid Al Attal

A THESIS SUBMITTED FOR THE DEGREE OF

MASTER OF ENGINEERING

Under Supervision of:

Supervisor: Prof. Dr. Mohammed Fouad

Co-Supervisor: Dr. Saleh Ibrahim Saied

© Abeer Al Attal, 2019

 ‫بسم هللا الرمحن الرحمي‬

‫وح ۖ قُ ِل الرُّ و ُح ِم ْن أَ ْم ِر َربِّي‬ ‫ِ‬ ‫الرُّ‬ ‫ن‬

‫ِ‬ ‫ع‬
‫َ‬ ‫ك‬
‫َ‬ ‫َ‬ ‫ن‬‫و‬‫ُ‬ ‫ل‬‫َويَسْأ َ‬
‫َو َما أُوتِيتُم ِّم َن ْال ِع ْل ِم إِاَّل قَلِياًل‬

‫صدق هللا العظمي‬

‫سورة االسراء‬
‫االية ‪85‬‬
  Acknowledgments:

The success of this thesis is a result of the valuable

contributions of many people who merit to be mentioned. I wish
to thank the Almighty Allah for His grace and sustenance upon
my life for the successful completion of this thesis.
I would like to thank Prof. Dr. Mohammed Fouad, my
supervisor, for all his support and guidance. I am lucky to work
with him.
Also, I would like to thank Dr. Saleh Ibrahim Saied. He was
very friendly and co-operative during the whole time. We had
very good technical discussions which give the birth opportunity
to this thesis.
 i. Abstract:
A cognitive radio (CR) is a radio that can be configured dynamically to
use the best wireless channels. It is used for making max channel
utilization and best efficiency without interference by dynamic spectrum
management technique.
Medium access control (MAC) protocol design is defiance in the
performance of cognitive radio CR networks. The existing MAC
protocols for dynamic spectrum access can be classified according to
accessing channel into (i) random access protocols like Dynamic Open
Spectrum Sharing MAC Protocol, (ii) time slotted protocols like
Opportunistic MAC Protocol, and (iii) hybrid access protocols like
Synchronized MAC Protocol.
The categories of spectrum access protocols and performance study of
those protocols will be illustrated. Moreover some enhancements to
DOSS mac protocol are proposed. DOSS enhancement once by using
back off algorithm and also by selective control channel mechanism.

Keywords: Dynamic spectrum access technique, CR Ad-Hoc networks,

CR MAC protocols, Dynamic Open Spectrum Sharing mac protocol,
Back-off algorithm.
Contents
List of Figures

Fig. No. Page No. Fig. Name

Fig.1

Fig.2

Fig.3

Fig.4

Fig.5

Fig.6

Fig.7

Fig.8

Fig.9

Fig.10

Fig.11

Fig.12

Fig.13

Fig.14

Fig.15

Fig.16

Fig.17
List of Tables

Fig. No. Page No. Fig. Name

Table.1

Table.2

Table.3

Table.4

Table.5

Table.6

Table.7

Table.8

Table.9

Table.10
Abbreviations

Abbreviation illustration
CR Cognitive Radio

CRNs Cognitive Radio Networks

PUs Primary Users

SUs Secondary Users

MAC Media Access Control

DOSS Dynamic Open Spectrum Sharing

DSS Dynamic Spectrum Sharing

DSA Dynamic Spectrum Algorithm

Carrier-Sense Multiple Access / Collision
CSMA/CA
Avoidance
SDR Software Defined Radio

SYNC- MAC Synchronized Media Access Control

QoS Quality of Service

BOA Back-Off Algorithm

SMC-MAC Self-scheduled Multi-Channel MAC

FCA Fixed Channel Allocation Scheme

DCA Dynamic Channel Allocation Scheme

HCA Hybrid Channel Allocation Scheme

Introduction
Contents

1. Problem statement.................................................................................1
2. Goals …………………………………………………………………1
3. Thesis organization...............................................................................1

1. Problem statement
The term radio spectrum refers to a limited frequency range, which
divided into different frequency bands, each having specific applications.
Therefore, there is a finite resource with competing demand which means
spectrum scarcity problem.
2. Goals
The Cognitive radio (CR) [1][2] is a futuristic technology formed to solve
the spectrum scarcity problem. A cognitive radio (CR) is the radio that is
able to be programmed and configured dynamically to use the best
wireless communication channels in its neighbor's channels to keep away
from user interference and congestion.
3. Thesis organization
Following this introduction, the thesis has five chapters reminder. These
chapters can be organized as follows:
 Chapter 1: It is an overview includes cognitive radio networks
definitions, characteristics, Network architectures, and cognitive
radio cycle.
 Chapter 2: This chapter shows different CR-MAC protocols
 Chapter 3: This chapter presents the back-off algorithm as an
enhancing technique.
 Chapter 4: This chapter presents the channel allocation schemes.
 Chapter 5: It illustrates the wishing Cognitive Radio and its
Application for Next Generations.
 Chapter 6: Conclusions of the simulation analysis of all proposed
techniques in last chapters and future works.
 The thesis ends with a bibliography of all references cited in the
thesis.
 Chapter 1:
Cognitive Radio Networks: an overview

Contents

1.1 Cognitive radio networks definition and characteristics ……….........1

1.2 Network architectures … … … … … … … … … … … … … … ... 1
1.3 cognitive radio cycle … … … … … … … … … … … … … … … 1

1.1 Cognitive radio networks definition and characteristics

Cognitive radio networks (CRNs) are networks used to solve the scarcity
problem of spectrum limitations and applications growth by increasing
spectrum efficiency and utilization.

In this kind of networks there are two types of users:

• PU: primary users

• SU: cognitive or secondary users

An important consideration is avoiding performance deviance of

the licensed users with the best possible performance of unlicensed users
or cognitive radio (CR) users. It occurs by providing efficient means of
sensing and sharing to keep tolerable interference levels for the PUs that
have a priority usage. The DSA [1] can be illustrated as shown in figure
1.

Dynamic Spectrum Algorithm (DSA) [2] allows CR users to share

holes dynamically whenever PU is absent. The SUs senses the spectrum
bands and determine the available bands. Then, the SUs determine the
best available bands according to its quality of service (QoS)
requirements using DSA for data exchanging between SUs. In case of any
PU appearance in these holes, the SUs must vacate the band immediately
and search for another hole to avoid interference
 Fig. 1. Dynamic Spectrum Sharing.
1.2 Network architectures
There are two types of cognitive radio networks based on the
network architecture [6].

Fig. 2. Infrastructure of cognitive radio networks

A centralized cognitive radio networks, and decentralized cognitive
radio network which called cognitive radio ad-hoc networks. The
centralized cognitive radio networks contain a centralized entity controls
the spectrum access procedures as shown in Figure 2. Each SU provides
the sensing information to the central control object and the control unit
builds the spectrum allocation map or free allocation and usage allocation
channel lists.
 Fig. 2. Infrastructure of cognitive radio ad-hoc networks
In decentralized or ad-hoc cognitive radio network spectrum
sharing occurs in distributed method as shown in figure 3. Each SU
processes the sensing information and spectrum access procedure.
1.3 cognitive radio cycle.
The cycle of cognitive radio cycle [2] shown in figure 4 consists of
four main functions:

Fig. 4. Cognitive radio cycle

A. Spectrum sensing:
The SU can use only free portion of the spectrum (holes). So,
spectrum sensing provides information on spectrum availability [5]. The
most effective method to sense spectrum holes is to detect PU [6] [7].
However, the PU may be not in the transmission range of the SU or there
are some barriers between PUs and SUs leads to some cooperation.
The cooperative sensing between SUs is vital to share information
about the PUs appearance. If the detection time is long the transmission
time will be short and the throughput will below. Therefore, it is
important to control the sensing time.
B. Spectrum decision:
Once the available holes are allocated, it is necessary to select the
most suitable one for SUs according to their quality of service (QoS)
requirements [8]. Both spectrum selection and routing formation are
included in spectrum decision. That is done in the beginning of the
transmission period. A Dynamic Decision Protocol For Adhoc Network
can be illustrated in [9].
Spectrum decision consists of three steps: First step, each spectrum
band is analyzed based on the local observations of SUs and the statistical
information of networks, this is called spectrum analysis. Second step,
based on this analysis, the best spectrum band can be selected and this
step is called spectrum selection function. Third step, the SUs need to
reconfigure communication hardware to the selected spectrum band, this
function is called reconfiguration.
C. Spectrum sharing:
It is responsible for avoiding collisions in overlapping parts of the
spectrum during data transmission.
D. Spectrum mobility:
If the PU is detected in the specific portion of the spectrum in use,
SUs should leave the spectrum immediately and continue their
transmission using another hole.
 Chapter 2:
Cognitive Radio Mac Protocols

Contents

2.1 CR MAC protocols classification.........................................................1

2.2 Random access protocols …...………………………………………1
2.3 Time slotted protocols ….…...………………………………………1
2.4 Hybrid protocols .................................................................................1

2.1 CR MAC protocols classification

Spectrum access protocols [4] can be classified based on the nature of
channel access to as shown in the following figure:

Fig. 5. Classification of cognitive radio MAC protocols

i. Random access CR MAC protocols [4]: this category need
no time synchronization, and based on the CSMA/CA principle.
ii. Time slotted MAC protocols: These MAC protocols
necessitate synchronization, because the time is divided into many
time slots for both the control channel and the data transmission.
iii. Hybrid protocols: These protocols are partially time slotted
in control signaling occurs over synchronized time slots. But, the
data transmission occurs in a random access manner, without any
time synchronization.
2.2 Random access control protocols.
Random access protocols for CR ad-hoc networks may be classified
based on the required number of transceivers per CR user to [5] protocols
have multiple radio transceivers, and protocols have only a single radio.
These protocols are based on (CSMA/CA) principle [6]. There are many
protocols defined in this category like dynamic open spectrum sharing
(DOSS) MAC protocol. Random access control protocols depend on
IEEE 802.11 standardization [7] that is called also request to send and
clear to send (RTS-CTS) hand shaking.

2.2.1 Dynamic Open Spectrum Sharing (DOSS) MAC protocol

algorithm [8]:

This algorithm passes through five sequential steps as shown in the

following figure:

Fig.6. Steps of DOSS MAC protocol

1- Investigation of primary user's presence: is a detection
step urgent to know the free channels in which primary user is idle.
2- Set up three frequency bands: we need a frequency band
for data transmission, control channel, and a busy tone band.
 a. Data transmission band: Is the range of frequencies
particularized for actual data exchanging among cognitive users.
b. Control band: In DOSS algorithm control channel is used to
hold on the flow of control signals or beacons.
c. Busy tone band: A band is reserved only for busy tones
which have a direct one to- one mapping with the data transmission
bands.
3- Spectrum mapping: The mapping of the busy tones with
the data transmission bands are done as follows. The busy tone
band has a frequency range [fl:fh], and the data band is contained in
[Fl: Fh], as shown in Figure 7. The data spectrum is too larger, and
this linear mapping allows neighboring nodes to realize that the
spectrum is actually used by another CR user using the observation
of the corresponding busy tone.

Figure 7 one to one mapping.

4- Spectrum negotiation: The transmitter sends a request
(REQ) packet with the available spectrum bands. Then receiver
replies with a request acknowledgment (REQ_ACK) packet
including the information of the chosen spectrum band. With
respect to the spectrum mapping, the corresponding busy tone tells
its neighbors not to transmit on the chosen data band.
5- Data transfer: If a packet is received correctly, the receiver
replies with data acknowledgment (DATA_ACK) and turns off the
busy tone.

The five steps of the proposed DOSS Mac protocol can be detailed in the
following flow chart.
 Fig. 8. Flow chart for the DOSS algorithm
2.3 Time slotted CR MAC protocols [9]
In this category the control channel is divided into time slots fixed in
length. Each time slot represents one CR node, and nodes can only
communicate in their respective time slots. Each time slot has two
periods:
 a) Listening period
b) Transceiving period.

Synchronization between nodes is a must [10]. This category has some

algorithms like opportunistic MAC protocol which will be illustrated
later, and the Structure of those protocols can be illustrated as shown in
the following figure.

Figure 9: the structure of Time slotted CR MAC protocols

2.3.1 Opportunistic MAC protocol [11]

Each cognitive radio user is equipped with two sets of transceivers.

The first transceiver is specified to operating over the dedicated control
channel to collect the information of available channels, and to negotiate
with the others. The second transceiver consists of a software defined
radio (SDR) [12] that can tune to any n licensed channels to sense,
receive, and transmit packets.

Figure 10: Opportunistic MAC protocol structure

 The control channel divided into periodical time slots. Each time slot is
divided into two phases:
a) The reporting phase: that can be divided into n sub-channels.
b) Negotiation phase: that consists of periods of RTS-CTS
negotiation.

Steps of this protocol:

1. At the beginning of negotiation phase for each time slot, SUs scan
the spectrum band, and then send beacons in the mini slots that
describe the channel state. The first transceiver listens to control
channel upon receiving a beacon at ith mini time slot, upgrade free
channel list and current usage lists.
2. If SDR sense an idle channel, inform control transceiver and the
interested node send RTS over control channel during negotiation
phase.
3. Upon receiving CTS, negotiation is success.
4. Then SDR tunes to selected channel and start data transmission
5. Finally, reply with DATA-ACK.

2.4 Hybrid cognitive Radio MAC protocols :

Hybrid protocols [13] are those protocols which are partially time
slotted and partially random access. They use equally time slots for
control signals and other slots for data but the access within data
transmission slots is randomly. There are many algorithms in this
category like synchronized MAC (SYNC- MAC) protocol that will be
illustrated later.

2.4.1 SYNC-MAC protocol :

The SYN-MAC [14] protocol has a dedicated radio one for listening to
the channel for control messages, the second transceiver is used for data
transmission. In this protocol Time is divided into time slots and each slot
represents a particular data channel. As shown in the following figure.
 Figure 11: the structure of SYNC- MAC protocol.
1. The control signal exchange occurs in the channels
represented by the corresponding slots. At the beginning of each
time slot, the CR users tune their dedicated control radios to the
channel specified by it.
2. Users that wish to start a data transmission must send out a
beacon at this time.
3. Interested neighbors respond with their own list of available
channels
4. Then communication is done in one of those selected
channels.
5. After finishing the channel, returns it to be free.

 Simulation analysis:

Using simulation we can investigate the following parameters:

1- Succession ratio: is the ratio of the numbers of data packets
received successfully to the total transmitting packet. That
calculated after each DATA-ACK.
2- Dropping ratio: is the ratio of the numbers of data packets
dropped at PUs appearance to the total transmitting packet.
Dropping occurs when PU appear to the frequently used channel
by a SU.
3- Blocking ratio: is the ratio of the numbers of data packets
blocked by PUs presence to the total transmitting packet. Blocking
occurs when SU can't locate appropriate channel to use.
1.1.Simulation assumption:
i. In simulation experiments a network of 5 nodes and 4 links
is used.
ii. There are ten sub channels per link.
iii. All licensed channels are homogeneously.
iv. The arrival rate of PUs and SUs is assumed to be Poisson
distribution with rate
λpu and λsu.
v. The service rate of the PUs and SUs are assumed to be with
rate μpu and μsu.
vi. Simulation time is 500,000 time unit.
vii. Dropping occurs when a PU interrupts the transmission of
SUs.
viii. Blocking occurs when: All channels are busy or CCC is used
by other SU.

 Simulation results and discussion:

a) Analyzing the effect of λpu:

In this experiment we use the following parameters: λsu= 0.1

SU/time unit, μpu= 0.2 packet unit/time unit and μsu= 0.25 packet
unit/time unit, the following results can be obtained.

Figure 12 blocking ratio of SU vs. PU arrival rate

 We can observe that in the CR network the SU blocking ratio
increases linearly with PU arrival rate for all three considered MAC
protocols. This linear increment is expected as when PU arrival rate
equals to zero, the SU dropping ratio will be zero because all channels
will be occupied by SUs only. And when PU arrival rate equals to one,
the SU dropping ratio will be maximum because all channels will be
occupied by PUs. But each protocol access channels with different
way so network efficiency exchanges and this explain the different
slops.

Figure 13 dropping ratio of SU vs. PU arrival rate

Notice that the SU dropping ratio increases with PU arrival rate.
This linear increment occurs because; the dropping is done if PU
appears in any spectrum band which is used by SUs. When the PU
arrival rate increases, the PU activity in the network will increase.
Then, the dropping ratio will increase.
 Figure 14 succession ratio of SU vs. PU arrival rate
Succession is number of success packets per total transmitted
packets This means each of succession ratios and throughput depends
on success packets. The succession process of CR user transmission
depends on the number of available channels that SUs can access it at
a time. When the PU arrival rate increases, number of free channel
that CR user can obtain is decrease so number of success packets
decrease.

 Conclusion.

CR networks are investigated to solve the problem of spectrum

scarcity by making efficient and opportunistic use of frequencies
reserved for the use of licensed users. To realize the goals of truly
usage of spectrum in aware communication, the CR devices need to
merge the spectrum sensing, spectrum decision, spectrum sharing, and
spectrum mobility functionalities. The main challenge in CRAHNs is
to help the CR users to communicate reliably in a distributed manner,
over a spectrum environment, without any infrastructure support and
without interfering with PU.
In this research, an overview of medium access control protocols in
cognitive radio networks is presented. Moreover a classification of CR
MAC protocols based on the channel access methods is investigated. .
This research focused on the comparative of CR MAC protocols in
this classification by a simulation performance study depending on
measuring blocking, dropping, and succession ratios of each proposed
protocol against PU arrival rate
 Chapter 3:
DOSS mac protocol with back off algorithm

Contents

3.1 Back-Off algorithm …..........................................................................1

3.2 DOSS protocol with back-off algorithm (BOA-DOSS) ……..………1
3.3 Simulation assumptions and results …................................................1

3.1 Back-Off algorithm

Back-off algorithm (BOA) [17] is a collision resolution mechanism
which is generally used in the Ethernet to schedule re-transmissions after
collisions. This algorithm has been implemented during the contention
interval to enhance the number of successful data packets of users and
hence increase the throughput of cognitive radio network. The
competition and collision among the cognitive users has seriously
minimized by reserving the unused licensed channels. The back-off
mechanism has been implemented to resolve the contention among the
collided cognitive users to increase the number of successful users. So, if
the collision is detected among cognitive users, the collided cognitive
users attempt to utilize another contention slot in the same cycle-time.
That can be occurred using back-off mechanism.
As shown in Figure 15, if a collision takes place between 2 stations or
nodes in the network, they may restart transmission as soon as they can
after the collision. This will always lead to another collision and form an
infinite loop of collisions leading a deadlock. To prevent such scenario
the back-off algorithm is used.
The nodes involved in the collision randomly pick up an integer number
from a set K, where K is called the contention window. If the nodes
collide again because they picked up the same integer, the contention
window size is doubled. The contention window size is calculated by the
relation (2n-1), where n is collision number or collision order. Before the
involved nodes in the collision start the transmission, they listen to the
channel and transmit only if the channel is idle. The node is picked the
smallest integer in the contention window for successful frame
transmission. So, Back-off algorithm defines a waiting time for the nodes
involved in the collision; In other words, the station should wait time
slots before re-transmission.

Fig.15. Back-off algorithm.

In [18], back-off algorithm is discussed and implemented in the self-
scheduled multi-channel MAC (SMC-MAC) protocol for solving the
connection among the SUs and holding the suitable licensed channels for
data transmission.
The SMC-MAC with back-off algorithm [19] is presented with a system
model and its numerical analysis. The spectrum mapping of busy tone in
SMC-MAC protocol occurs in time domain by dividing sensing sharing
period into time slots. The frame structure [19] of SMC-MAC can be
shown in Fig 16. In contrast, mapping of busy tone in DOSS-MAC
protocol occurs in frequency domain using one to one mapping as shown
in Figure 7. The SMC-MAC protocol belongs to the hybrid protocols
which are affected by the synchronization problem.
 Fig.16. SMC-MAC frame
While, the DOSS-MAC protocol can be implemented and analyzed with
simulation for enhancing network performance as driven in section 4. The
main reasons of choosing DOSS-MAC algorithm can be summarized in
the following points:
i. It is a very simple protocol.
ii. This protocol doesn't need synchronization.
iv. The protocol is appropriated for mobile ad-hoc networks.

3.2 DOSS protocol with back-off algorithm (BOA-DOSS)

The proposed DOSS protocol with back-off algorithm (BOA-DOSS) can
be illustrated using pseudo code as shown in Fig. 17. At the beginning of
network initialization, the frequency bands should be set up. Then, a
frequency mapping is formulated as illustrated in DOSS scheme. In the
case of the CR networks, there is PUs which has a priority of
transmission without any delay or interference. Other SUs are allocated
sharing the spectrum holes with PUs. When SU wants to start data
transmission, it sends RTS to idle or free channel according to IEEE
802.11 standardization.
If the channel is immediately free, the receiver replies with CTS signal.
When the source node receives the signal, it starts to transmit data and
turn on the corresponding busy tone. At the case of collision occurrence
by conflicting with another SU, both of the sender and receiver nodes tag
the collision and calculate its order. If the collision order less than max
number of re transmissions allowed, it retransmits data. Else a blocking
occurs and calculated. At the end of successful data transmission, the
receiver replies with data acknowledgment (DATA-ACK). Then, counter
of successful data increases. If a PU is appeared, the corresponding SU
should leave the channel immediately and the dropping is occurred. The
dropping counter will increase. At the end of simulation time block, drop,
and success counters can be obtained.
The ordinary DOSS scheme selects a control channel in a random
method. The common control channel is fixed and tuned only with a
fixed frequency. Hence, high traffic over common control channel and
the probability of blocking occurs without any minimization. So, using
BOA-DOSS algorithm, the probability of blocking can be decreased. In
case of BOA-DOSS algorithm, it has been proposed that each cognitive
user randomly chooses a contention slot which makes it more vulnerable
to collision among the cognitive users. In order to reduce the number of
collisions, the control channel’s contention interval can be modified by
using the back-off algorithm in the contention interval. For example, if
the SU collides with another, the collided cognitive users can select
another contention slot in the same cycle time with the help of back-off
algorithm. And if the selected contention slots are different, both the
cognitive users become successful. Otherwise, if again there is collision,
then the contention window size is increased, and same procedure is
followed.
However, this whole procedure has been presented with the help of
Pseudo code as following.
 Fig. 17. Pseudo code of DOSS protocol with back-off algorithm.

3.3Simulation assumptions and results

ix. Simulation assumptions:

To evaluate the performance of the proposed method and the ordinary

DOSS algorithm, a simulation experiments are carried out using
Matlab™ with the following network assumptions
x. The network consists of five nodes with four connected links;
each link consists of 10 sub channels.
ii. All licensed channels are homogeneously.
xi. The arrival rate of Pus and Sus is assumed to be Poisson
distribution with rate λpu =0.2 and λsu = 0.2, sequentially.
iv. The service rate of the Pus and Sus are assumed to be with rate μpu
= 0.25 and μsu = 0.2, respectively.
xii. Simulation time is 100,000 time unit.
vi. Dropping occurs when a PU interrupts the transmission of Sus.
number of dropped data packets
Dropping ratio=
total number of transmitted packets

vii. Blocking in DOSS protocol occurs when all channels are busy or
common control channel is used by other SU.
number of blocked data packets
Blockingratio=
total number of transmitted packets

Viii. Blocking in DOSS with back-off algorithm occurs when collision

order exceeds the max number of allowable data re transmissions.
xiii. The throughput [19] can be calculated by:
number of data packets received successfully
Throughtput =
total transmission time

3.3.2 Simulation results:

The blocking and dropping ratios can be estimated at 5 retransmission

trials. In these results, the DOSS algorithm will be represented if number
of retransmission n=0; Otherwise, it will be DOSS with back-off
algorithm. Moreover, the effect of varying PU arrival rate λpu from 0 to 1
can be analyzed.
As shown in Fig. 18, the blocking ratio is increased linearly when PU
arrival rate increases. This linear increment occurs due to the decreasing
in free spectrum bands. This linear increment stops at appoint of network
saturation. At this point, the number of blocked data is seemed to be
fixed.

Fig. 18. Blocking ratio versus PU arrival rate

Fig.19. Dropping ratio versus PU arrival rate

Fig. 19 represents the dropping ratio versus PU arrival rate. The dropping
ratio is increased with PU arrival rate. The probability of dropping
increases with increasing PU activity in the network. This increment is
expected because of the increasing of PUs increases the probability of
applying on a used spectrum portion by SUs, which leads to drop the SU
immediately with respect to PU priority.

After using back-off algorithm the blocking ratios decreased due to

increasing number of data re transmissions which increases the
probability of delivering data successfully without blocking while, the
dropping ratio is increased. After a certain number of data retransmission
times, the network congestion is occurred. Therefore, the optimum
number of data retransmission times can be selected.

The trade-off is to select an optimum number of data re transmissions that

decrease blocking and dropping ratio. Drawing the relation between
blocking and dropping ratios versus number of data re transmissions can
be reported as shown in Fig. 20

Fig. 20. Blocking and dropping ratios versus number of data

After 5 trials of data re transmissions blocking ratio will be nearly

saturated as shown in Fig. 20 While, dropping seems to be stabled. Any
more trials than 5 will be worthless and increase network traffic. Hence,
the optimum number of data re transmission for this network scenario is
five trials.

The optimum point of retransmissions will be at n=5. When PU arrival

rate equal to zero, the licensed channels will be occupied by SUs only,
the SU dropping ratio and blocking ratio will be zero in both MAC
protocols. The collision ratio is high if many users transmit in the
beginning of time slot. Thus, the collision ratio in DOSS-MAC protocol
is higher than the collision ratio in the BOT-DOSS MAC protocol.

Fig. 21. Throughput of DOSS and DOSS with BOA versus PU

arrival rate

Therefore, the throughput of DOSS-MAC protocol will be lower than the

throughput of BOT-DOSS MAC protocol as shown in Fig. 21. By
increasing the PU arrival rate, the collision, SU blocking and SU
dropping ratios will be increased. Hence, the throughput will decrease.
Due to the BOT technique, the throughput of the BOT-DOSS MAC
protocol is higher than the throughput of the ordinary DOSS MAC
protocol.

All of these results can be summarized in Table 1 and Table 2.

Numerically, the following two tables illustrate the cases of blocking
ratios, dropping ratios, and successful data transmission ratios in both
case of ordinary DOSS MAC protocol and DOSS MAC protocol with
BOA. It is noticed clearly that succession ratios in case of DOSS MAC
protocol with BOA is enhanced more than 20 % against ordinary DOSS
MAC protocol. It refers to successfully delivered packets. Moreover,
blocking ratios are decreased by nearly 40%.

Table 1: Summary of data obtained from simulation of ordinary DOSS

MAC and DOSS MAC protocol with BOA.

  DOSS MAC protocol DOSS MAC protocol with BOA

Blocking Dropping Successful Blocking Dropping Successful

λpu
ratio ratio data ratio ratio ratio data ratio

0.2 0.0642 0 0.9358 0 0.0146 0.9854

0.4 0.5885 0.0244 0.3871 0 0.1916 0.8084
0.6 0.6159 0.0418 0.3423 0 0.3365 0.6635
0.8 0.6265 0.0665 0.307 0 0.3812 0.6188
1 0.6364 0.1006 0.263 0 0.4212 0.5788

 Conclusion

In this research, an overview of the state of the art for cognitive radio
cycle and accessing protocols is presented. The ordinary DOSS-MAC
protocol for the cognitive radio based wireless networks is explained
briefly. Back-off algorithm is illustrated. A modified DOSS algorithm
based on BOT mechanism is investigated. Moreover, a simulation study
between the proposed technique and the ordinary DOSS-MAC protocol
is illustrated. The accessing channel mechanism of ordinary DOSS-MAC
protocol occurs by selecting a free channel from predefined free usage list
for control signals without any investigation or minimization to blocking
probability. Applying back-off algorithm (BOA) on DOSS-MAC
protocol increases the data transmission with priority of illuminating
blocking ratios.

Using simulation analysis, the proposed DOSS with BOA method has
significantly enhanced the performance. The number of successful data
delivered for cognitive users is increased. Hence, the proposed DOSS
with BOA method has enhanced the throughput compare with the
ordinary DOSS-MAC protocol. The simulated performance analyses of
the network for applying both of strategies prove that using BOA-DOSS
algorithm can decrease blocking ratio and increase throughput without
interfering with PU activities.
 Chapter 4:
Enhancing DOSS MAC protocol with different
Channel allocation schemes

Contents

4.1 Static or fixed channel allocation scheme (FCA) …………..............1

4.2 Dynamic channel allocation scheme (DCA) ..………………………1
4.3 Hybrid channel allocation (HCA) …………………………………1
4.4 DOSS with static channel allocation scheme. ……………………1
4.5 DOSS with dynamic channel allocation scheme…………………1
4.6 DOSS with Back off algorithm and DCA scheme………………1

Channel allocation algorithms research for cellular systems shows the

existing channel allocation schemes can be classified into three classes as
following [17]:
4.1Static or fixed channel allocation scheme (FCA)
The main advantage of FCA schemes is its simplicity. The control
channel in this scheme is assumed to be fixed, not changed during all
network operation time. One of the main disadvantages of FCA scheme is
that it blocks the data transmissions if the number of transmissions
exceeds the number of channels assigned to a network. This problem is
called control channel saturation problem.

4.2 Dynamic channel allocation scheme (DCA)

Unlike FCA, the channels are assigned dynamically in DCA scheme. In
DCA scheme, channels are allocated in a real-time based on the actual
and immediate availability conditions by performing real-time sensing,
while sharing information are saved in each node's list to make decision
about channels allocation. Due to its real-time occupation requirements,
DCA schemes have higher complexity and less efficiency than FCA
scheme. But, it can solve control channel saturation problems in many
cases [18]. There are many additional different management techniques
which illustrate in [19].

4.3 Hybrid channel allocation (HCA)

HCA is another allocation scheme which is considered to be a
combination of FCA and DCA. In HCA scheme some channels are
statically allocated while the other channels are dynamically allocated.
The ratio between statically and dynamically allocated channels is desired
according to the aimed quality of service of the overall network.
In the following section, the performance of DOSS MAC protocol will
be investigated for both static and dynamic control channel allocation
schemes.
4.4 DOSS with static channel allocation scheme.
In this scheme CR user first configure the common control channel
(CCC) which is fixed during all network operation time. Then it collects
all sensing information to allocate the free spectrum holes. Then it can
choose a free channel and transmit a request to send signal according to
IEEE 802.11 standardization over the common control channel. If the
selected channel still free, it replies with a clear to send signal over the
same common control channel. Then the user transceiver tuned to the
selected channel frequency and transmits data packets over it. At the end
of data transmission another control signal is sent over the common
control channel to inform with the channel freeing.
The performance study of DOSS with FCA is explained in chapter (3).

4.5 DOSS with dynamic channel allocation modeling

In this scheme DOSS operates with the same steps of IEEE802.11
standardization. Unlike FCA, the channels are assigned dynamically in
DCA scheme. The control channel is selected in real time of data
transmission request. So DCA can increase network performance by
solving CCC saturation problems and decreasing blocking probability as
shown in the following figures.
Simulation results
Fig. 22 blocking ratio of Static and Dynamic DOSS versus PU arrival
rate

Fig. 23 Dropping ratio of Static and Dynamic DOSS versus PU arrival

rate
Fig. 24 Success ratio of Static and Dynamic DOSS versus PU arrival rate

4.6 DOSS with Back off algorithm and DCA scheme.

In this section DCA scheme of DOSS mac protocol with BOA will be
analyzed and simulated. The blocking and dropping ratios can be
estimated at 5 retransmissions. In these results, the DCA scheme of
DOSS mac protocol will be represented if number of retransmission n=0;
Otherwise, it will be DCA scheme of DOSS mac protocol with back-off
algorithm. Moreover, the effect of varying PU arrival rate λ pu from 0 to 1
can be analyzed.
The blocking ratio is increased linearly when PU arrival rate increases
as shown in Fig. 22. This linear increment occurs due to the decreasing in
free spectrum bands. It stops at appoint of network saturation. At this
point, the number of blocked data is seemed to be fixed. As shown in
figure 25, the blocking can be illuminated after 3 times of retransmission
trails n=3.
 Fig. 25 Blocking ratio versus PU arrival rate
The dropping ratio versus PU arrival rate is presented as shown in Fig.
23. The dropping ratio is increased with PU arrival rate. This increment is
normally occurs due to the increment of increasing PU activity in the
network. As the increasing of PUs increases the probability of applying
on a used spectrum portion by SUs, this leads to drop the SU immediately
with respect to PU priority.
After using back-off algorithm, the blocking ratios decreased when
number of data re transmissions increases. The probability of delivering
data successfully without blocking is increased with the increment of
dropping ratio. After a certain number of data retransmission times, the
network congestion is occurred which leads a saturation in the network.
Therefore, the optimum number of data retransmission times can be
selected.
The trade-off is to select an optimum number of data re transmissions
that decrease blocking and dropping ratio. Fig. 9 shows the relation
between blocking and dropping ratios versus number of data re
transmissions.
Fig. 26 Dropping ratio versus PU arrival rate
That is means, blocking ratio will be nearly saturated after 3 trials of
data re transmissions. The dropping seems to be stabled. Any more trials
than 3will be worthless and increase network traffic. Hence, the optimum
number of data re transmission for this network scenario is three trails.
The optimum point of retransmissions will be at n=3. When PU arrival
rate equal to zero, the licensed channels will be occupied by SUs only,
the SU dropping ratio and blocking ratio will be zero in both MAC
protocols. The collision ratio is high if many users transmit in the
beginning of time slot. Thus, the collision ratio in DOSS-MAC protocol
is higher than the collision ratio in the BOT-DOSS MAC protocol.
 Fig. 27 Success ratio of DCA scheme of DOSS and DCA scheme of
DOSS with BOA versus PU arrival rate
As shown in Fig. 27, the success ratio of DCA scheme of DOSS
-MAC protocol will be lower than the throughput of DCA scheme of
DOSS MAC protocol with BOA. By increasing the PU arrival rate, SU
blocking and SU dropping ratios will be increased. Hence, the throughput
will decrease. Due to the BOT technique, the throughput of the DCA
scheme of BOT-DOSS MAC protocol is higher than the throughput of
the ordinary DCA scheme of DOSS MAC protocol.
Finally, it is clearly observed that the enhancement of DOSS protocol
with BOA and DCA scheme can illuminate blocking occurrence by the
aid of retrying of transmission in BOA and dynamic channel selection in
DCA. It considers handling to blocked packets. This is shown in Fig.27.

 Conclusion

In this research, an overview of cognitive radio network architecture and

accessing protocols is presented. The ordinary DOSS-MAC protocol for
the cognitive radio based wireless networks is explained briefly. A
channel allocation mechanism for DOSS algorithms is investigated with
the ordinary DOSS-MAC protocol.
With respect to DCA scheme DOSS-MAC protocol, accessing
channel mechanism occurs by selecting a free channel from predefined
free usage list for control signals instead of fixing the control channel in
ordinary DOSS algorithm.
Using simulation analyses the proposed DCA-DOSS method has
significantly enhanced the performance of cognitive communication
system by increasing the number of successful cognitive users for the
data transmission. Hence, the proposed has enhanced the throughput as
compare to the FCA- DOSS MAC protocol.
BOA is proposed to enhance the performance of DOSS MAC protocol of
DCA scheme. From simulation BOA proofed enhancing network
performance by increasing opportunities of receiving data packets
successfully.
 Chapter 5:
Cognitive Radio and its Application for Next
Generations
conclusions
References:
Authors' biography:
‫الملخص العربي‬
 ‫ب ِْس ِم اهَّلل ِ َّالرمْح َ ٰ ِن َّالر ِح ِمي‬
‫ان فَضْ ُل هَّللا ِ َعلَ ْي َ‬
‫ك َع ِظي ًما‬ ‫َو َعلَّ َم َ‬
‫ك َما لَ ْم تَ ُك ْن تَ ْعلَ ُم َو َك َ‬

‫صدق هللا العظمي‬

‫سورة النساء‬
‫االية ‪113‬‬
 ‫لكية الهندسة‬
‫جامعة الزقازيق‬

‫حتليل أداء تقنية الوصول ادلينامييك للطيف الرتددي يف حاةل‬

‫الارسال املتعدد لشباكت الارسال املوجه‬

‫رساةل مقدمة من‬

‫عبري عبد الرمحن عبد امحليد العتال‬
‫باكلريوس هندسة ختصص الكرتونيات واتصاالت‬

‫حتت ارشاف‬
‫أ‪.‬د‪ .‬صالح ابراهيم سعيد‬
‫أ‪.‬د‪ .‬محمد فؤاد‬