A SURVEY ON DYNAMIC SPECTRUM ACCESS TECHNIQUES IN COGNITIVE RADIO

International Journal of Next-Generation Networks (IJNGN) Vol.4, No.4, December 2012
DOI : 10.5121/ijngn.2012.4403 27
A SURVEY ON DYNAMIC SPECTRUM ACCESS
TECHNIQUES IN COGNITIVE RADIO
Pinki Yadav1
, Subhajit Chatterjee2
and Partha Pratim Bhattacharya3
1, 3
Department of Electronics and Communication Engineering
Faculty of Engineering and Technology,
Mody Institute of Technology & Science (Deemed University)
Lakshmangarh, Dist. Sikar, Rajasthan, Pin-332311, India.
2
Department of Electronics and Communication Engineering
Regent Education and Research Foundation
Barackpore, Kolkata, India
1
pinki.yadav15@yahoo.in
2
chsubhajitch@gmail.com
3
hereispartha@gmail.com
Abstract
Wireless networks are characterized by fixed spectrum policy. With increasing demands for wireless
communication efficiently using the spectrum resources has become an essential issue. Cognitive radio is a
form of wireless communication which is used to sense the spectrum and find the free spectrum. It is used
by unlicensed users without causing interference to the licensed user. Cognitive radio with the dynamic
spectrum access is key technology which provides the best solution by allowing a group of Secondary users
to share the radio spectrum originally allocated to the primary users. Dynamically accessing the unused
spectrum is known as dynamic spectrum access (DSA) which becomes a promising approach to increase
the efficiency of spectrum usage. In this paper, DSA models are discussed along with different methods
such as game theory based method, a measurement-based model, network coded cognitive control channel,
Markovian Queuing model, the Delay performance of threshold policies, fuzzy logic based method and
spatio-temporal spectrum management model.
Keywords
Wireless communication system, cognitive radio, dynamic spectrum access, fuzzy logic, threshold policies,
spectrum management.
1. INTRODUCTION TO COGINITIVE RADIO
The ever-increasing demand of the wireless communication applications and services affirms the
importance of the effective usage of the limited radio spectrum. Each user is assigned a license to
operate in certain frequency bands. Most of the time spectrum remains unused and it is very
difficult to find the unused spectrum. The allocated spectrum has not been utilized properly and it
varies with time, frequency and geographical locations. For the removal of the spectrum scarcity
and the unutilized spectrum band, Cognitive radio and Dynamic spectrum access technology has
been introduced.

28
Cognitive radio is a figure of wireless communication in which a transceiver can perceive, which
communication channels are in use and which are not, and accordingly switch into empty
channels while avoiding busy ones. This optimizes the use of available radio-frequency spectrum
while minimizing interference to other users. It is a hybrid technology involving software
defined radio as applied to spread spectrum communication [1].
The concept of cognitive radio was first officially presented by “Joseph Mitola” III, at the Royal
Institute of Technology in 1998 and published later in an article by Mitola and Gerald Q.
Maguire, Jr in 1999 [2]:
As defined by Haykins, the cognitive radio is an “intelligent wireless communication system that
is aware of its surrounding environment (i.e., outside world) and uses the methodology of
understanding by-building to learn from the environment and adapt its internal states to statistical
variations in the incoming RF spur by making corresponding changes in certain operating
parameters (e.g., transmit-power, carrier-frequency, and modulation strategy) in real-time, with
two primary objectives in mind, highly reliable communication whenever and wherever needed;
efficient utilization of the radio spectrum” [3]. The FCC (Federal communication commission)
ruled in November 2008, in which unused part of RF spectrum (known as white spaces) made
available for public use. White Space Device Technology prevents interferences, such as
spectrum sensing and geolocations capabilities [1]. The parameter can altered on the basis
observations of multiple factors from external and internal environment of cognitive radio
environment, like radio frequency spectrum, user behaviour, and network state. Spectrum in 400-
1000MHz range is shown in Figure1 [4].
Figure 1 Spectrum utilization
Efficient utilization of spectrum improves by allowing a secondary user (SU) to utilize a licensed
band when the primary user (PU) is absent. So the detection of spectrum hole is important as
shown in Figure 2 [5].

29
Figure 2 Spectrum hole concept
Spectrum holes is basically a concept which correspond to the latent opportunities for safe use of
spectrum i.e. non-interfering and considered as multidimensional regions within frequency, time,
and space. The main challenge for secondary radio systems is to be able to robustly sense when
they are within such a spectrum hole. To agree to a amalgamated discussion of the core issues in
spectrum sensing, the “Weighted Probability of Area Recovered (WPAR)” metric is introduced
where the performance of a sensing strategy and the “Fear of Harmful Interference” are
evaluated [6]. Cognitive radio is such a inimitable radio technology where user has to decide
which part of the spectrum is available. When a user function in a approved (licensed) band (i.e.
spectrum sensing), the best available channel is selected (i.e. spectrum management), other users
access this channel (i.e. spectrum sharing) and the channel is vacated when a licensed user is
detected (i.e. spectrum mobility). Regulatory bodies in various countries (including the Federal
Communications Commission (FCC) in the United States, and Ofcom in the United Kingdom)
found that most of the radio frequency spectrum was inefficiently utilized. The idea of spectrum
sharing is shown in Figure 3 below [7].
Figure 3 Spectrum Access by Cognitive Radio

30
There are two types of cognitive radio, full cognitive radio and spectrum sensing cognitive radio.
Full cognitive radio takes into account all the parameters that a wireless node or network can be
aware of. Full cognitive radio is also known as “Mitola Radio”. Spectrum sensing cognitive radio
is used to find channels in the radio frequency spectrum [1].
The radio spectrum animatedly changes the functions in accordance with the surrounding i.e. he
radio frequency, transmission power, modulation scheme, communication protocol can be
changed by cognitive radio without any alteration of the hardware environment.
The main functions of cognitive radios are: [8]
1) Spectrum Sensing: - Spectrum sensing supports to detect the spectrum holes (Underutilized
bands of the spectrum) providing high spectral resolution capability. Spectrum sensing technique
classifies into direct and indirect method. Direct method is also known as frequency domain
approach in which estimation is carried out directly from signal. Indirect method is known as time
domain approach in which estimation is performed using autocorrelation of the signal [9].
Spectrum sensing techniques are classified into three categories:
• Transmitter detection: cognitive radios should be capable of determining a signal (if
any) from a primary transmitter which is locally present in a certain spectrum and there
are several approaches to perform this and those are namely :
Matched filter detection
Energy detection
Cyclostationary feature detection
• Cooperative detection: Cooperative detection refers to spectrum sensing methods where
information from multiple Cognitive radio users is incorporated for primary user
detection.
• Interference based detection.
2) Spectrum management: - Spectrum management capturers the best available spectrum to meet
user communication requirements. Cognitive Radio decides the best spectrum band to meet the
quality of service requirements over all available spectrum bands. Therefore spectrum
management functions are important for cognitive radios. Spectrum management functions can be
classified into: i) Spectrum analysis. ii) Spectrum decision.
3) Spectrum mobility: - Spectrum mobility is defined as the process when a cognitive radios users
exchange its frequency of operation. Cognitive radio network target is to use the spectrum in a
dynamic manner by allowing the radio terminals to operate in the best available frequency band
and also maintaining seamless communication requirements during the transition to better
spectrum.
4) Spectrum sharing: - Spectrum sharing is function of cognitive radio which provides the fair
spectrum scheduling method. Spectrum sharing has one of the major challenges open spectrum
usage. It can be regarded to be similar to generic media access control (MAC) problems in
existing systems.

31
Opportunistic use of the radio-frequency (RF) spectrum is enabled by cognitive radio for dynamic
spectrum access, where unlicensed users are allowed to utilize licensed bands under the
circumstance that interference is as little as possible with the licensed bands. A cognitive radio
operates or transmits on bands detected as being free, leaving them whenever a primary user is
sensed. [10].
2. DYNAMIC SPECTRUM ACCESS
Dynamic spectrum access(DSA) is a technique by which a radio system dynamically adapts to
available spectrum holes with limited spectrum use rights, in response changing circumstance and
objectives: interference created changes the radio’s state, changes in environmental constraints
[11]. The main objective of DSA is to overcome two types of interference concern: harmful
interference caused by malfunctioning device and harmful interference caused by malicious user.
Dynamic spectrum management is also referred to as dynamic spectrum access. DSA that was
first demonstrated in 2006 by the Defense Advanced Research Project Agency (DARPA) and
Shared Spectrum Company (SSC) of Vienna, VA [12]. In [13], DSA is such an advanced
approach to spectrum management which is closely correlated to other management techniques
such as flexible spectrum management and spectrum trading. This allows users to access a
particular piece of spectrum for a defined time period or defined area which they cannot exceed
without re-applying for the resources. A DSA procedure would follow the following steps:-
• Monitor spectrum to see which frequencies have no other radio activity (i.e. they are not
being used by anyone).
• Agree with other dynamic spectrum access devices in the network which frequencies will
be used, via same previously agreed common channel.
• Being communicating on the agreed frequency band.
• Continue to monitor the spectrum for attempts other user to access this spectrum.
• Change frequency bands and adjust power as necessary.
The monitoring and managing of the radio resources are done by a single device in the network
(centrally managed network) or by each of the devices individually and cooperatively
(autonomous network).
In this paper, dynamic spectrum access models are discussed in section 2 which helps to improve
performance of a communication network as a whole. In section 3, different methods of dynamic
spectrum access are discussed.
3. MODELS OF DYNAMIC SPECTRUM ACCESS:-
Dynamic spectrum access is a new concept which has overcome the limitations of fixed
frequency spectrum assignment and management. Dynamic spectrum access models for cognitive
radio can be categorized as dynamic exclusive-use, open sharing model, and hierarchical access
model as shown in Figure 4 [14].

32
Figure 4 Dynamic Spectrum Access Models
a) Dynamic exclusive use model: - Dynamic exclusive use model maintains the basic
structure of the current spectrum regulation policy and Spectrum bands are licensed to services
for exclusive use. The main objectives of this model are to improve the flexibility and spectrum
efficiency. It manages spectrum in finer scales of time, space and frequency and use dimensions.
There are two types of dynamic exclusive use model they are:
i) Spectrum property rights: In this approach licenses are allowed to sell and trade spectrum and
to freely choose the technology. The economy and market an important role to play in lashing
toward the most gainful use of this limited resource.
ii) Dynamic spectrum allocation: Spectrum efficiency is improved by Dynamic spectrum
allocation through dynamic spectrum assignment by using the spatial and chronological traffic
statistics of different services. It can be said in other way i.e in a given area and at a given
instance, spectrum is allocated to services for elite use. This allocation varies in a much sooner
scale than the current policy. Based on an ddedicated-use model, elimination of white space is
not possible in spectrum resulting from the burst nature of wireless traffic.
b) Open sharing model: - Open sharing model is also called spectrum commons model. This
model uses open sharing among peer-users as the foundation for managing a spectral region. This
model is supported from the phenomenal success of wireless services operating in the unlicensed
industrial, scientific and medical (ISM) radio band. It has three types [7]: Uncontrolled-
commons, Managed-commons, Private-commons.
i) Uncontrolled-commons:-This is also referred as open spectrum access. When a spectrum band
is managed no entity has exclusive licensed to the spectrum band. It is maximum transmit power
constraint.
ii) Managed-commons:-This represents an effort to avoid the tragedy of commons by imposing a
limited form of order or structure of spectrum access.
iii) Private-commons:-Spectrum owner specifies technology and protocol for the CR user access.
CR user may receive a command from spectrum owner (transmission parameter). CR user may
sense and access the spectrum.

33
c) Hierarchical access model: - Hierarchical access model [15], in which radio spectrum can be
simultaneously shared between primary (licensed) user and secondary (unlicensed) user. CR users
can opportunistically access the radio spectrum if it is not occupied or fully utilized by primary
users. There are also two types of model; Spectrum overlay and Spectrum underlay.
i) Spectrum overlay:-The spectrum overlay model actively explored in a going DARAP XG
program and advocated by Mitola targets for aggressive opportunistic exploitation of white-space
or spectrum “gaps” in spatial-temporal domain [7]. Cognitive Radio will have to identify the idle
spectrum band, which are not used by licensed users at a certain time and location and use those
idle spectrum bands dynamically to unlicensed users. This model is shown in Figure 5 (a). This
model allows primary and secondary transmission. Secondary users can use part of their power
for secondary communication and remain part of the power to relay primary transmission, these
enabling premises for an overlay system model [16]. Example of spectrum overlay is TDMA,
FDMA, and OFDMA system [17].
ii) Spectrum underlay: - Some constraints are imposed by spectrum underlay approach on the
transmission power of secondary users so that they are operated well below the noise floor of
primary users. By scattering transmitted signals over a wide frequency band (UWB), secondary
users can latently achieve a short-range high data rate with very low transmission power shown in
Figure 5 (b). In the worst case, it is assumed that the primary users transmit all the time; this
approach does not depend on detection and utilization of spectrum white space [14].
Figure 5 Spectrum overlay and underlay approaches
In spectrum overlay approach; secondary CR user can transmit with a high transmission power to
increase their rates for giving spectrum opportunities however they have to identify the idle
frequency bands which are not used by primary users. Similarly, in spectrum underlay approach,
the secondary users do not need to identify the spectrum opportunities and can transmit
simultaneously coexisting with primary users however they are not allowed to transmit with high
transmission power even if the entire RF band is idle (i.e., entire RF is not used by primary users)
[17]. Therefore overlay is also known as interference model and underlay is known as
interference avoidance model [16].
There are many methods of DSA which work based on these models and discussed below.

34
4. METHODS OF DYNAMIC SPECTRUM ACCESS
We have discussed different methods of dynamic spectrum access in this section along with their
feature.
i) Game theory for cognitive radio networks
A Game theory is a mathematical tool that analyzed and planned the interaction among multiple
decision makers. There are three major components in a strategic-form game model [18]:
1) A finite set of players denoted by N;
2) A set of action, denoted by Ai, for each player i; and
3) Payoff/utility function, denoted by ui: A R, which measure the outcome for player i
determined by the actions of all player, A =Xi NAi .
In above definition and notations, a strategic game is denoted by ‹N, (Ai) ;( ui) ›.
Game theoretic spectrum sharing schemes are classified into four parts:
1) Non-cooperative spectrum sharing games: - Nash equilibrium is a solution to understand non-
cooperative game theory. Nash equilibrium often suffers from extreme competition among selfish
players in a non-cooperative game and the outcome of the game is inefficient. In non-cooperative
spectrum sharing game with rational network users, each user only cares about his/her own
benefit and chooses the optimal strategy that can maximize his/her pay off function and such
outcomes of the non-cooperative game is termed as Nash equilibrium.
2) Economic games, auction games and mechanism design: - It can be applied to the economic
world to deal with how people interact with each market. The key concept of the game theory is
rationality and equilibrium. Often players are the seller and buyer in the market (such as firms,
individuals and so on). Payoff functions are defined as the utility or revenue that players want to
maximize and equilibrium strategies are considerable interested.
Auction theory is an applied branch of game theory which analyzes the interaction in the auction
market and researches. An auction conduct by an auctioneer is a process of buying and selling
product by eliciting bid from possible buyers and deciding the auction outcome based on the bids
and auction rules. The auction rules or auction mechanism is obtaining from whom the goods are
allocated to and how much price they have to pay.
The structure of a game is to design the mechanism and this game structure is “designed” by a
game designer called a “principal” who want to choose a mechanism for his/her own interest and
resource constraints and incentive constraints are Co equally considered in an allocation problem
with private information.
3) Cooperative games: - Cooperative spectrum sharing is a game in which network users have an
agreement on how to utilize and distribute the spectrum resources. Two types of cooperative
games are i) Bargaining games ii) Coalitional games.

35
4) Stochastic games: - A stochastic games is an extension of Markov decision process by
considering the interactive competition among different agent. A stochastic spectrum sharing
game is such a game in which network users adapt their strategies according the changing
environment and other user’s strategies. There are many other research challenges such as,
Defining a proper payoff function, Efficiency of equilibrium, Issues in mechanism design, Issues
in stochastic games, security.
ii) A Measurement-based model for dynamic spectrum access in WLAN Channel
A Measurement-based model is proposed [19], in which Continuous-time Semi-Markov model is
used that captures the WLAN behavior so good enough to be used for deriving optimal control
strategies within a decision theoretic frame work. A measurement-based model is based on actual
measurements in the 2.4GHZ ISM band using a vector signal analyzer to collect complex base
band data. Continuous-time Semi-Markov model is able to capture the data with good accuracy.
A measurement setup shown in figure 6. Different from existing publication that use a
commercial WLAN adapter card to obtain packet trace but here they use a vector signal analyzer
(VSA) to capture the raw complex baseband data. These data have to find busy and idle periods
of the channel. In Measurement setup, both WLAN and VSA are provided as shown in Figure 6
respectively.
Figure 6 Measurement setup
The WLAN setup consists of a net gear WGT624 wireless router and three computers with
wireless adapter card (two net gears WG311T and WG511T). The router operates in a 22MHZ
frequency band around 2.462 GHZ (channel 11). The router as well as work stations locates in the
same room resulting high signal to noise ratio between nodes and no hidden terminals. Traffic is
generated using Distributed Internet Traffic Generator (D-ITG), which allows to statically
characterizing parameters such as inter-departure time and packet length. This measurement is
based on two traffic scenario. One of the high rate UDP traffic from workstations to the router
(the other computers are turned off) and to verify the compatibility of the setup. After capturing
the transmission of WLAN, employ Agilent 89640A vector signal analyzer to collect the complex
data base band sample. The device internally down converts 2.462GHZ to an internal IF
frequency at a sample rate of 44MHZ. Continuous-time Semi-Markov process in which allows an
arbitrary specification of sojourn time distribution in each state. A Semi-Markov process is a
stochastic process whose transition behavior characterized in two steps. First step is the transition
between states follow a markov chain and specified by transition matrix in Equation 1, where pij
denoted the probability that transition from state i to state j occur.

36
P= (1)
Secondly, given that the system is in state i and will transition to state j, the sojourn time t in state
i is distributed according to cumulative distribution function Qij (t). The estimator in Equation 2
(2)
In which the transition count nij is the number of transitions i→j occurring in our observation
Sequence. Similarly
(3)
In Equation 3 is the number of times the system resides in the state. High SNR transmission
between nodes and no hidden terminals the sequence of states DATA→SIFS→ACK is essentially
deterministic (the corresponding transition probabilities are very close to one). Hence, it is
possible to simplify the model by lumping, these states together. While this model inhibited the
occurrence of collisions, retain good accuracy since collisions are infrequent. The Markov model
is as shown in Figure 7.
Figure 7 (a) Markov model. (b) The lumped model (with deterministic DATA SIFS ACK
transitions is shown on the right)
In Figure 7 ‘transmit’ state (a lumped version of DATA, SIFS, and ACK states with deterministic
transitions), and an idle state is shown. The transition probabilities for this simplified semi-
Markov model is now trivial, since every transmit state must be followed by an idle period
Continuous time Semi-Markov model captures the idle periods remaining between the burst
transmissions of the wireless LAN. This model strikes a good compromise between accuracy and
computation complexity.
iii) Dynamic Spectrum Access using a network coded cognitive control channel
Dynamic Spectrum Access using network coded cognitive control channel [20], allows the users
to opportunistically and efficiently access the available channel for communication. Important
aspects of opportunistic spectrum access are i) Implementation of the control channel. ii) Multi
channel medium access control. iii) Primary user detection. iv) Secondary users reuse the unused
spectrum of primary uses. In this method all secondary users visit all channels in a pseudo
random fashion and exchange control information whenever they happen to meet in any channel.

37
The control information exchanged by secondary users consist of all the information which is
needed to select switch patterns as well as a resource allocation for data communication,
according to a pre-defined deterministic algorithm. This method has important aspects such as,
primary user performed over all channels in each allocation period, to track the varying pattern of
primary user’s activity. The detection information gathered by each during an allocation period is
to be disseminated to all users using the control channel. Cooperative detection is carried out by
each user using the same deterministic algorithm. The resource allocation algorithm, which is run
independently by each user, assigns transmission opportunities only on free channels. This
method is completely distributed and it does not need dedicated spectrum resources for control
purposes but rather leverages of the virtual control channel which is implementing network
coding techniques and exploit a cooperative detection strategy to identify unused spectrum. This
method works without the establishment of a control channel, this is achieved by having each
secondary user detect the presence of primary activity and decide whether to access the spectrum
independently from another user. The main issue is that channel switch pattern needs to cover all
channels; so that primary user detection can be performed effectively but the control information
cannot be transmitted by secondary users on those channels in which primary users are active.
This causes a degradation of the dissemination performance of network coded cognitive control
channel-DSA with respect to NC4-MAC. The performance evaluation of NC4-DSA with respect
to primary user detection, interference performance, the spectrum reuse efficiency of the
secondary users and good put in dynamic spectrum access as shown in Figure 8, Figure 9, Figure
10, and Figure11. In Figure 8, the value of the signal to noise ratio of the primary user is
The normal probability scale is used for both axes; cooperative detection strategy
allows to achieve significant improvements in the achievable tradeoff between primary user
detection and false alarm probability. The detection improves with increasing Ns and S.
Figure 8 Primary user detection
In Figure 9 and Figure 10, Pinter f and Preuse is plotted as a function of the mean activity duration
and inactivity duration 1/ of the primary user respectively for c=10,
.

38
Figure 9 Interference to the primary user Figure 10Efficiency of secondary user reuse
caused by secondary users of spectrum unused by primary user
In Figure 9 Pinter f and Preuse are decreases when Increases, since the longer activation period of
the primary user allows secondary user to detect it and avoid interfering with it. In the case where
Ns =10 and , probability of a single detection attempt very low due to low SNR of the
primary user and benefits of cooperation are limited due to the small number of users. In Figure
10 the probability of reusing unused spectrum depends on the mean duration of the inactive
period of the primary user.
Figure 11Goodput for secondary spectrum access
The Good put the performance of NC4-DSA for both the original “cooperative detection only”
version and the “cooperative + LBT” variant. The resulting performance is shown in Figure 11,
function of the primary user activity For C = 10, dB, = 5 dB, = 35, =
600, and =3 dB when also LBT is used. Effect on the detection of the primary users and the
reuse efficiency of unused spectrum on the overall good put is limited, since detection works
almost perfectly with the value of S that is practical for the dissemination of control information
in the primary / secondary scenario; this effect does not increase in π1.

39
iv) Markovian Queuing Model for Dynamic Spectrum Allocation in centralized
Architecture
Markovian Queuing model for dynamic spectrum allocation in which centralized architecture is
used. In centralized network the master/controller of this ad-hoc network coordinates spectrum
allocation with the surrounding CR in the network. This CR ad-hoc network assumes to coexist
with the network of licensed users where the controller of licensed user is updated with the CR
coordinating engine. A centralized network eliminates hidden terminal problem and provide
better coverage and efficiency spectrum handover technique. Each SU is consisting of two
transrecievers, one is dedicated to control and second software defined radio based. The SDR
based transrecievers scan the availability of spectra in its vicinity and forward the information of
these spectrum holes to the master/controller. In this case, SU form an infrastructure less network
and the BS are infrastructure based network. The equivalent model of network queues is shown in
Figure12 [21].
Figure 12 Queuing Model for DSA in Cognitive Radio
These queues are in the special case of stochastic processes, characterized by the arrival process
of service requests, waiting list of requests to be processed. The queue stacking all the entries of
SUs is referred to secondary user queue and the entire request entering this queue are served on
the first come first serve (FCFS) basis. At any time when bandwidth needs to be allocated to the
SU, the head considers both the request from the SU and the PU, who need it licensed channel.
Therefore while distributing a number of frequencies for PU and SU; the arrival rates of both the
users are summed to access the frequencies with the Head. This queue so formed is referred to as
bandwidth allocation queue (BAQ). Markov process is used to analyze the Queuing model. The
blocking probability PB for the bandwidth request made by CR that finds all the channels with
Head as occupied is given by Erlarg-B formula as given below in Equation 4:
= (4)
The Blocking probability is shown in Figure13. Variation in PB with respect to change in number
of available channels in the system as 2,5,7,10,13, and 15. Blocking probability increases with
increase in SU traffic in the network.

40
Figure 13 Blocking probabilities ( ) Against SU utilization in the system ( ). The Variation in
is depicted with different numbers of channels (S), available with the system.
v) Delay performance of threshold policies for dynamic spectrum access
The delay performance of threshold policies for DSA is proposed [22]. A simple time-threshold
policy for the SU to minimize the average delay while satisfying the collision probability
constraint of the PU (primary user). Such policies perform closely to an optimized policy found
by a Markov chain formulation, while facilitating analytical analysis of the delay and collision
probability. The time-threshold policies have important aspects which are as follow:
1. The threshold policy performs closely to the optimal policy found by Markov decision process
(MDP) formulating, but simpler in structure and provide better insights and being
computationally more efficient than the MDP policy.
2. The threshold policies facilitate analytical characterization of the delay and collision
probability under general PU is busy and idle distribution.
The time threshold policy with the threshold Γ is defined such that the SU will transmit only
when the following three condition: i) the channel is sense idle. ii) t < Γ (note that Γ is in general
different from the Γ∗ that maximizes capacity), and iii) the SU queue length M is greater than the
zero. The time threshold Γ should be adjusted to satisfy the PU collision probability constraint. In
Figure14, examine the effect of busy distribution on ܹ and pc . In which four different busy
distributions with the same mean: exponential, uniform, Weibull, and the fixed busy time are
considered respectively. Figure 14(a) shows that the busy distribution affects ܹ significantly.
The exponential busy distribution and the fixed busy distribution induce the largest delay and the
smallest delay, respectively. On the other hand, for the same Γ, Figure 14(b) shows that ‫݌‬c
changes only slightly with busy distributions.

41
Figure 14 Performance of threshold policies as a function of Γ. Delay and collision probability
Comparison of various busy distributions. Assume a uniform idle distribution and p=0.11.
Figure 15: Delay performance of threshold policies as a function of .
Figure 15 shows the delay of the threshold policy as a function p. For each p, choose Γ of the
threshold policy such that ߟ=0.001 a uniform busy distribution and various idle distributions. For
each idle distribution, there exists some ‫∗݌‬ such that when ‫݌‬ < ‫,∗݌‬ the threshold policy becomes
the greedy policy. Find that ‫∗݌‬ = 0.06, 0.075, 0.095 for the exponential, uniform, and Weibull
distribution, respectively. Figure15 shows that for these idle distributions, the delay of the
threshold policy is similar in the region ‫݌‬ < ‫݌‬ ∗ where the greedy policy is optimal. For the
exponential distribution, ‫∗݌‬ = 0.06 is the highest arrival rate such that ‫ܿ݌‬ does not exceed ߟ. In
comparison the capacity of the uniform distribution is higher. Therefore, when ‫݌‬ ∈ [0.075,
0.114], a threshold policy can be found to ensure that ‫݌‬ܿ = ߟ at the rate of increased delay. Thus
that the delay of the threshold policy increases rapidly as ‫݌‬ approaches ‫ܥ‬ = 0.114. The Weibull
distribution has the highest capacity. When ‫݌‬ ∈ [0.095, 0.2], the delay of the threshold policy
increases with ‫,݌‬ but at a slower rate than that of the uniform distribution. The accuracy of the
above graph has been confirmed for several commonly used busy and idle distributions. This
method is shown that the busy time distribution significant impact of delay of the SU, while the
idle distribution largely determines the transmission threshold and collision probability.

42
vi) Fuzzy logic Based System
Fuzzy logic based spectrum access method is proposed in [23]. Fuzzy logic is a multi-valued
logic. It uses many input parameters to take the decision. Here distance, signal strength, velocity
and spectrum efficiency are considered as input parameters. If the signal strength of the intended
channel is high ratio will decide to change of the channel. When the node velocity is more, more
will be the chance for a mobile node to change position. The ratio of the required spectrum by the
secondary user to the total available spectrum (spectrum efficiency) is determining parameter in
this parameter dynamic spectrum access policy radio will use unused vacant spectrum. The
distance between the primary licensed and secondary unlicensed user has been one of the
determining parameters in which the secondary user at a closer distance to the licensed primary
user, give priority to access spectrum. Simulation results show that [23], the chance of taking
decision increases if the signal strength of the channel offered by primary user is high and
distance between primary and secondary users is low as shown in Figure16. As the velocity
increases the chance of the spectrum accessing is more if the distance is small as shown in Figure
17.
Figure 16 Opportunistically spectrum access decision possibility (velocity=50 Km/hr and ratio of
the required spectrum to available spectrum =0.5)
Figure 17 Opportunistically spectrum access decision possibility (Distance between the
primary and secondary user =50 meters and user velocity =50 Km/hr)

43
vii) Spatio-Temporal Spectrum Management Model
Spatio-Temporal Spectrum Management Model is used for coordinating DSA networks [24]. This
Model is reducing complexity of the problem into temporal dynamic spectrum allocation problem
and spatial dynamic spectrum allocation problem. In this model regional spectrum broker (RSB)
coordinated the temporal dynamic spectrum allocation for a given region within the assumed
spatial distribution of the spectrum demand is homogeneous. Spectrum Broker Coordinator
(SBC) is a centralized entity, which stores the spectrum demands of the regions, and spectrum
management at the border of the regions is realized based on this information. Architecture of the
model is shown in Figure 18.
Figure 18 Spatio-Temporal DSA architecture
In TDSA method the service providers of the region send their demands for spectrum to RSB.
The RSB allocates continuous spectrum blocks to the requesters separated by guard bands. The
size of the blocks may vary in time. Moreover demanding another spectrum block, service
providers are returning spectrum blocks that they do not need. The requests are batch-processed
at given time-intervals.
The Spectrum Dynamic Spectrum Allocation (SDSA) handles spectrum demands arising at the
same time in different regions. The main objective of the SDSA is attuning the different demands
within different regions on the way, where the least interference arises in the overlapping regions.
In order to realize this, the RSB needs to have information about the actual spectrum allocation of
the neighboring regions. To collect this information, a time snapshot of the spectrum usages
inside a region is sent by the RSBs to the Spectrum Broker Coordinator (SBC) Based on this
information, the RSBs can manage the problem of interference in the overlapping regions
accordingly. The hierarchy of the RSBs and SBC is shown in Figure 18. Simulation results are
shown in Figure 19 and Figure 20. In region 1 the demand of NSP1 is much larger than that of the
region NSP2, in region 2, it was just opposite way. The numbers of carriers used by the providers
in the both regions as a function of time .The excess spectrum required to fulfill the demands in
the case of overhearing is denoted by a darker tone shown in Figure.

44
Figure 19 Allocated spectrum sizes for both providers in region 1.
Figure 20 Allocated spectrum sizes for both providers in region 2
5. CONCLUSIONS
Cognitive radio is an emerging technology, which can provide faster and more reliable wireless
service by utilizing the existing spectrum band more efficient. Here discussed different dynamic
spectrum access models. Game theory is an important tool in studying, modeling and analyzing
the Cognitive interaction process. A Measurement based model in which the continuous time
semi Markov model is used. Continuous-time semi Markov model captures the idle periods
remaining between the bursty transmissions of a wireless LAN. This model strikes a good
compromise between accuracy and computation complexity. DSA using a NC4 method
opportunistically and efficiently accesses the available channel for communication. In the
Markovian queuing model a centralized architecture is proposed for allocation of
bandwidth and to find blocking probability. The delay performance of threshold policies
is to minimize the delay of SU subject to a PU collision probability constraint. A fuzzy
logic based spectrum management technique help to take wise decision with respect to
spectrum sharing in cognitive radio networks. Spatio-Temporal spectrum management model
simplify the spectrum allocation problem and described architecture that splits the complex
problem into temporal and spatial parts.

45
References:
[1] http://searchnetworking.techtarget.com/resource definition from WhatIs_com.htm, “What is cognitive
radio (CR)?”
[2] www.wikipedia.org.
[3] K. –C. Chen, Y. –J. Peng, N. Prasad and Y. –C. Liang, S. Sun, “ Cognitive Radio Network
Architecture: Part I – General Structure”, Proceeding of the ACM International Conference on
Ubiquitous Information Management and Communication, Seoul, 2008
[4] Przemyslaw Pawelczak “Cognitive Radio: From Utopia to Reality” Freeband Ambient
Communication Event, Enscheda, 4 July 2006
[5] Anita Garhwal1 and Partha Pratim Bhattacharya2 “A Survey on Dynamic spectrum Access
techniques for cognitive radio” International Journal of Next-Generation Networks (IJNGN) Vol.3,
No.4, December 2011 pp.15-32.
[6] Rahul Tandra, Shridhar Mubaraq Mishra and Anant Sahai “What is a spectrum hole and what does it
take to recognize one?” IEEE Journal and Magazine, Vol.97, issue 5, May 2009, pp 1-22.
[7] Anita Garhwal, Partha Pratim Bhattacharya, “Dynamic Spectrum Access in Cognitive Radio: a brief
review”, International Journal of Computer Application in Engineering Sciences, Special Issue on
Computer Networks & Security, Vol.1 December 2011, pp. 149-153.
[8] J. Mitola III and G. Maquire (1999), “Cognitive Radio Making Software Radios More
Personal”,IEEE Pres. Comm.,6(4), 1999, pp 13-18.
[9] Mansi Subhedar1 and Gajanan Birajdar2 “Spectrum Sensing Technique in Cognitive Radio Network
Spectrum Access Networks: A Survey” International Journal of Next-Generation Networks (IJNGN)
Vol.3, No.2, June 2011 pp. 37-51.
[10] Alice Crohas “Practical Implementation of a Cognitive Radio System for Dynamic Spectrum Access”
July2008pp 1-46.
[11] Przemyslaw s Pawelczak, Gerard J.M Janssen and R.Venkatesha Prasad “Performance Measures of
Dynamic Spectrum Access Networks”IEEE Globecom, 28 November 2006 pp.1-22.
[12] http://www.mwrf.com/
[13] Brief note “Dynamic spectrum access” commission for communication regulation document no.7/22,
13 April 2007 pp.1-20.
[14] Qings Zhao and Brian M. Sadler “A Survey of Dynamic Spectrum Access” IEEE Signal Processing
Magazine, May 2007 pp.79-89.
[15] Xavier Gelabert “Cognitive Radio Enabling Opportunistic Spectrum Access (OSA): Challenges and
Modeling Approaches” UPF-DTIC Weekly Research seminar 28 October 2010.
[16] Sudhir Srinivasa and Syed Ali Jafar “The Throughput Potential of Cognitive Radio: A Theoretical
Perspective” IEEE Communications Magazine, May 2007 pp.73-79.
[17] Danda B. Rawat and Gongjun Yan “Spectrum Sensing Methods and Dynamic Spectrum Sharing in
Cognitive Radio Networks: A Survey” International Journal of Research and Reviews in Wireless
Sensor Networks Vol. 1, No. 1, March 2011.
[18] Beibei Wang, Yongle Wu, K.J. Ray Liu “Game theory for cognitive radio networks: An overview”
Computer Networks 4, 13 April 2010 pp. 2537–2561.
[19] Stefan Geirhofer, Lang Tong and Brian M. sadler “Measurement-Based Model for Dynamic
Spectrum Access in WLAN Channel” communication and network consortium 19 July 2011.
[20] Nicola Baldo, Alfred Asterjadhi and Michele Zorzi “Dynamic Spectrum Access Using a Network
Coded Cognitive Control Channel” IEEE Transaction On Wireless Communication, Vol. 9, NO. 8,
August 2010 pp. 2575 -2587.

46
[21] Prabhjot Kaur, Arun Khosla and Moin Uddin “Markovian Queuing Model for Dynamic Spectrum
Allocation in Centralized Architecture for Cognitive Radios” IACSIT International Journal of
Engineering and Technology, Vol.3, No.1, February 2011 ISSN: 1793-8236.
[22] Rong-Rong Chen and Xin Liu “Delay Performance of Threshold Policies for Dynamic Spectrum
Access” IEEE Transactions on Wireless Communication, VOL. 10, NO. 7, July 2011, pp.2283-2293.
[23] Partha Pratim Bhattacharya, Ronak Khandelwal, Rishita Gera, Anjali Agarwal, “Smart Radio
Spectrum Management for Cognitive Radio”, International Journal of Distributed and Parallel
Systems, Vol. 2, no. 4, July 2011
[24] Lászl ó Kovács and Attila Vidács “SpatioTemporal Spectrum Management Model for Dynamic
Spectrum Access Networks” Published by ACM August 2006, Proceedings of the first international
workshop on Technology and policy for accessing spectrum (TAPAS).
Authors
Pinki Yadav was born in India on February 2, 1987. She received her B.E. Degree in
Electronics and Communication Engineering from Mandsaur Institute of Technology,
Mandsaur (University RGPV Bhopal), M.P., India in 2010 and currently is an M. Tech
(Signal Processing) student in Mody Institute of Technology and Science (Deemed
University), Rajasthan, India.
Subhajit Chatterjee was born in India. He has more than 12 years experience in
teaching starting and 3 years in industry . He has served some reputed educational
institutes in different positions from Lecturer to Teacher In Charge of department.
Presently he is working as Assistant Professor, Department of Electronics &
Communication Engineering, Regent Education and Research Foundation, Barackpore.
He has number of publications and his area of interests are Analog & Digital
Communications, Solid State Devices, Telecommunication, Microprocessors & Microcontrollers. He has
co authored two books on Microprocessors & Microcontrollers and Telecommunication & Switching
respectively. He has reviewed chapters of books published by reputed publishers.Presently he is doing his
research on Spectrum Access in Cognitive Radio.
Dr. Partha Pratim Bhattacharya was born in India on January 3, 1971. He has 15 years of
experience in teaching and research. He served many reputed educational Institutes In India
in various positions starting from Lecturer to Professor and Principal. At present he is
working as Professor in Department of Electronics and Communication Engineering in the
Faculty of Engineering and Technology, Mody Institute of Technology and Science
(Deemed University), Rajasthan, India. He worked on Microwave devices and systems and
mobile cellular communication systems. He has published a good number of papers in
refereed journals and conferences. His present research interest includes mobile communication and
cognitive radio. Dr. Bhattacharya is a member of The Institution of Electronics and Telecommunication
Engineers, India and The Institution of Engineers, India. He is the recipient of Young Scientist Award from
International Union of Radio Science in 2005. He is working as the chief editor, editorial board member
and reviewer in many reputed journals.

A SURVEY ON DYNAMIC SPECTRUM ACCESS TECHNIQUES IN COGNITIVE RADIO

More Related Content

What's hot (19)

Similar to A SURVEY ON DYNAMIC SPECTRUM ACCESS TECHNIQUES IN COGNITIVE RADIO (20)

More from ijngnjournal (20)

Recently uploaded (20)

A SURVEY ON DYNAMIC SPECTRUM ACCESS TECHNIQUES IN COGNITIVE RADIO