Improved Algorithm for MAC Layer Spectrum Sensing in Cognitive Radio Networks from Dynamic Spectrum...

8/10/2019 Improved Algorithm for MAC Layer Spectrum Sensing in Cognitive Radio Networks from Dynamic Spectrum Mana

1/16

www.tjprc.org [email protected]

International Journal of Computer Networking,

Wireless and Mobile Communications (IJCNWMC)

ISSN(P): 2250-1568; ISSN(E): 2278-9448

Vol. 4, Issue 6, Dec 2014, 75-90

TJPRC Pvt. Ltd.

IMPROVED ALGORITHM FOR MAC LAYER SPECTRUM SENSING IN COGNITIVE

RADIO NETWORKS FROM DYNAMIC SPECTRUM MANAGEMENT PERSPECTIVE

DIPAK P. PATIL1& VIJAY M. WADHAI2

1Research Scholar, SGBAU Amravati, Maharashtra, India

2Principal, SIT, Kondhwa, Pune, Maharashtra, India

ABSTRACT

Spectrum scarcity and effective utilization of the available spectrum is still a challenge in wireless

communication. Due to static spectrum allocation (DSA) policy most of the spectrum remains under-utilized. Hence

Dynamic Spectrum Allocation technique is proposed to use the available spectrum opportunistically. Cognitive Radio is a

promising technique for efficient utilization of idle authorized spectrum since it is able to sense the spectrum and reuse the

frequency when the licensed user is absent. In this paper, a new sensing mechanism which can be realized as two layer

mechanisms, i.e. Physical Layer and MAC Layer is presented. The issue of finding the spectrum opportunities by

sensing-period adaptation and minimizing sensing period from channel utilization perspective is addressed. A new

algorithm is developed for channel selection and sensing period minimization. Results show an increase in channel sensing

and reduction in sensing time compared to non-optimal schemes.

KEYWORDS:Cognitive radio, Channel Usage, Dynamic Spectrum Management, MAC Layer, Spectrum Sensing

1. INTRODUCTION

We are well aware with the importance of the spectrum resources in the wireless communications. Today's wireless

networks have resulted in spectrum inefficiency because of the static spectrum allocation policy. In static spectrum

allocation policy, the radio frequency bands are licensed to the authorized users. This static policy based spectrum

management framework can guarantee that the radio frequency spectrum will be exclusively licensed to an authorized user

on a long term basis for the specified region. However this can cause inefficient spectrum usage. Where a large portion of

the assigned spectrum is used sporadically, leading to an underutilization of the allocated spectrum. And this underutilization

is due to the fact that an authorized user may not fully utilize the spectrum at all times in all locations. Hence to meet the

increasing spectrum demands for wireless applications/services needs of flexible spectrum management technique are arises

[1].

Dynamic spectrum access (DSA) techniques are proposed to solve the spectrum inefficiency problem, for

improving the flexibility of the spectrum usage by considering all dimensions and issues of spectrum usage. To deal with the

conflicts between spectrum congestion and spectrum under-utilization, Cognitive Radio (CR) has been recently proposed as

the solution to current low usage of licensed spectrum problem. The CR can dynamically adjust the operating parameters

over a wide range depending on availability of the spectrum. The CR provides the capability to share the wireless channels

in an opportunistic manner. However, the basic requirement is to ensure that the existing licensed users are not affected by

such transmissions. CR networks are also used to provide high bandwidth to the mobile users through heterogeneous

wireless architectures and dynamic spectrum access techniques.


2/16

76 Dipak P. Patil & Vijay M. Wadhai

Impact Factor (JCC): 5.3963 Index Copernicus Value (ICV): 3.0

The CR networks, however, impose unique challenges due to the high fluctuation in the available spectrum, as well

as the diverse quality of service (QoS) requirements of various applications. For addressing these challenges, each CR user

in the CR network must be able to determine the availability of the spectrum and select the best available channel. Also it

has to coordinate access to this channel with licensed users and vacate the channel when a licensed user is detected [2].

Sensing the availability of a channel is commonly recognized as one of the most fundamental issue of a CR due to

its crucial role of sensing spectrum opportunities and detecting the existence of primary users [3-5].

A CR allows a cognitive user to access an available spectrum unoccupied by a primary user(PU) and improve the

spectrum utilization in the spectrum. However, CRs are adapted as lower priority to a PU. This fundamental requirement is

to avoid interference to the PUs in their vicinity. To implement without interference to the primary signal, the CR needs to

sense the spectrum opportunities in wireless environments before accessing the channel. Hence spectrum sensing is the most

important issue of the CR technique. Spectrum sensing can be realized as a two-layer mechanism. The Physical (PHY) layer

sensing focuses on efficiently detecting PU signals to identify opportunities by changing its modulation/encoding schemesand parameters. Several well-known PHY-layer detection methods such as energy detection, matched filter, and

cyclostationary detection have been proposed as candidates for the PHY layer sensing. On the other hand, the MAC-layer

sensing determines when secondary users (SU) have to sense which channels are free. This type of sensing, despite its

importance, has received far less attention than other related topics. Nowadays Medium Access Control (MAC) layer

sensing is the major area of research from channel allocation and SUs coordination. [6-8]. In this paper new sensing

mechanism, which can be realized as two layer mechanism by sensing period adaptation and minimizing sensing period

from channel utilization perspective is proposed.

The rest of the paper is organized as follow. Section II describes proposed system for dynamic spectrum

management. Section III defines spectrum sensing function and presents a cross -layer approach for its implementation.

The network/channel models and our assumptions are discussed in Section IV. Algorithm for sensing period optimization

and channel selection is presented in Section V. Results is shown in Section VI, followed by conclusion in Section VII.

2. PROPOSED SYSTEM FOR DYNAMIC SPECTRUM MANAGEMENT USING COGNITIVE

RADIO

The proposed system presents design and architecture of dynamic spectrum management (DSM) using CR and is

shown in Figure 1.

Figure 1: Communication Model for DSM System


3/16

Improved Algorithm for MAC Layer Spectrum Sensing in Cognitive 77Radio Networks from Dynamic Spectrum Management Perspective


The high level communication model of the system for DSM consist of major components like a reconfigurable

radio with configurable parameters like center frequency, power, bandwidth, frame length, modulation, spatial transmission

pattern etc[9]. A sensing engine will accept inputs from the external environment such as the radio frequency (RF),

but possibly other sources such as data sources on the internet or other networked nodes. Learning and reasoning engine willaccept inputs from the sensing engine and policy database where user policy and base station policies are already defined.

The learning and reasoning engine will also determine an appropriate configuration for the radio components. The reasoning

engine may be capable of learning based on experience.

A configuration database is required to maintain the current configuration of the radio components. A simple CR

system might have single reconfigurable radio component with a reasoning engine accepting sensing information from local

node but not from external data sources. Finally, a policy database may exist that determines what behavior is acceptable

under what circumstances. This database may be dynamically configurable allowing policies to be changed when required.

3. SPECTRUM SENSING

A Cognitive Radio is a radio capable to sense the spectral environment over a wide frequency band and exploit this

information for providing wireless links to meet the user communication requirements opportunistically [3]. While many

other characteristics have also been discussed, and considered. PHY and MAC functions that are linked to spectrum sensing

as illustrated in Figure 2.

Since CR is considered as lower priority for SUs of spectrum allocated to a PU, a fundamental requirement is to

avoid interference to PUs in their vicinity. On the other hand, PU networks have no requirement to change their

infrastructure for spectrum sharing with cognitive radio networks (CRN). Therefore, CRs should be able to independently

detect PU presence through continuous spectrum sensing. Different classes of PUs would require different sensitivity and

rate of sensing for the detection.

Figure 2: Cross Layer Functionalists Related to Spectrum Sensing

In general, CR sensitivity should outperform PU receiver by a large margin in order to prevent what is essentially a

hidden terminal problem. This is the key issue that makes spectrum sensing very challenging research problem. Meeting the

sensitivity requirement of each primary receiver with a wideband radio would be difficult enough, but the problem becomes

even more challenging if the sensitivity requirement is raised by additional 30-40 dB. This margin is required because CR

does not have a direct measurement of a channel between PU receiver and transmitter and must base its decision on its local

channel measurement to a PU transmitter. This type of detection is referred to as local spectrum sensing, in severe multipath

fading, or inside buildings with high penetration loss while in a close neighborhood there is a PU who is at the marginal


4/16



reception, due to its more favorable channel conditions. Even though the probability of this scenario is low, CR should not

cause interference to such PU. The implementation of the spectrum sensing function also requires a high degree of flexibility

since the radio environment is highly variable, both because of different types of primary user systems, propagation losses,

and interference. The main design challenge is to define RF and analog architecture with right trade-offs between linearity,sampling rate, accuracy and power, so that digital signal processing techniques can be utilized for spectrum sensing,

cognition, and adaptation. This also motivates research of signal processing techniques that can relax challenging

requirements for analog, specifically wideband amplification, mixing and A/D conversion of over a GHz or more of

bandwidth, and enhance overall radio sensitivity.

4. PREREQUISTES

4.1 Network Topology

A group of SUs is assumed to form a single-hop wireless sensor network (WSN) within the transmission range of

which there are no other sensor network (SN) interfering or cooperating with that SN. In a practical (CRN), however, the

interference among adjacent SNs should be dealt within the context of internetwork coordination of channel sensing and

allocation. Although the coordination issue is not the main focus of this paper, our proposed scheme can coexist with any

other scheme by dynamically adapting the pool of available channels for an SN in such a way that those channels are not

used simultaneously by other SNs.

Every SU in the SN is assumed to be equipped with a single identical antenna that can be tuned to any

combination of N consecutive licensed channels. This can be done by the Orthogonal Frequency Division Multiplexing

(OFDM) technique with adaptive and selective allocation of OFDM subcarriers to utilize any subset of N licensed channels

at the same time [10-12]. Note that equipping each SU with more than one antenna might cause severe interference among

its antennas, thus degrading the SU's performance [13]. We therefore focus on SUs, each equipped with a single antenna.

Each SU works as a transceiver, as well as a sensor in its SN. An important role of sensing is incumbent detection that is,

determining the presence/absence of PUs on a channel. Energy and feature detections are two prominent PHY sensing

schemes for incumbent detection. Energy detection, however, cannot differentiate PU signals from SU signals since it only

measures the energy of a signal. Although feature detection can be used to overcome this difficulty, it may be harder to

detect PU signals if SU signals interfere with them during sensing. Hence, IEEE 802.22 introduced the concept of a quiet

period during which all SUs should suspend their transmission so that any sensor monitoring the channel may observe the

presence/absence of PU signals without interference [14-15]

It is assumed that a channel will be sensed within a quiet period whose schedule should be negotiated/reserved

among SUs. It is also assumed that all SUs in an SN should participate in sensing a channel at the same time for each

scheduled measurement period to enhance the detection of PU signals even in a fading/shadowing environment.

Fading/shadowing is known to become a serious problem in achieving the desirable sensing quality in terms of incumbent

detection and false alarm probabilities [16 to18]. To overcome these difficulties, collaborative sensing has been proposed,

which requires multiple sensors to cooperate. Hence, we will use a basic collaboration policy of allowing all SUs

participate in simultaneously sensing a channel.

4.2 Channel Usage Model

Spectrum sensing is used to detect a channel's availability. Depending on availability of a channel, it is modeled as

occupied and unoccupied periods. The state of the channel can be described as and . Such a model captures the time


5/16



period in which the channel can be utilized by SUs without causing any interference to PUs. Once an period is

discovered, SUs can utilize any portion of the remaining period for their own transmission.

For channel ( = 1,2, . . . . ,N) we model the sojourn time of an ON period as a random variable with the

probability density function (pdf) Similarly, the probability density function of the sojourn time in an

period is given as . periods are assumed to be independent and identically distributed (iid) and so are

periods. We also assume that and periods are independent of each other. Let ) denote the state

( / )of channel at time. Then becomes a semi-Markov process in that whenever the process enters

/ state, the time until the next state transition is governed by pdf Since there are only two

possible states, the behavior of this process can be analyzed by using the theory of alternating renewal processes [20, 21].

Figure 3 shows the state transition model of this semi-Markov process.

Sensing can be cast as a sampling procedure of the given channel process to discover its state at each

sensing instant. Let period correspond to the value 1/0.Then, sensing produces a binary random sequence for each

channel. Figure 4 illustrates periodic sensing with sensing period and sensing time. Here, is a predefined amount of time

for a single measurement in order to achieve the desirable level of detection quality by PHY-layer sensing. For example, it

has been proposed in the IEEE 802.22standard [16] that less than 1 ms should be spent for fast sensing with energy

detection. We assume that is predetermined by PHY layer sensing, and it is small relative to and On the other

hand, channel utilization , which is defined as the fraction of time in which channel is in ON state, is given as

Figure 3: State Transition Diagram of Semi Markov Process

4.3 Oppurtunity Usage Model

Opportunity represents spectrum availability ( period) in a licensed channel. An opportunity in a channel can

only be discovered by sensing the channel. As discussed earlier, it is necessary to perform collaborative sensing to overcome

uncertainties in a wireless spectrum such as fading/shadowing. In collaborative sensing, the sample of a channel collected by

an SU must be shared/ synchronized with other SUs so that each SU can decide channel's availability. Since the cooperation

among SUs is not a focus of this paper, we assume that the sensing time includes both PHY-layer detection time

(for example, 1 ms) and data synchronization time in collaborative sensing. Whenever sensing is performed on a channel

and an opportunity on the channel is discovered, the channel is merged into a pool of available channels where the pool is

called a logical channel. Therefore, a logical channel can include 0 N licensed channels depending on their availability at

that instant. The logical channel is treated as if it were a single channel whose capacity is equal to the sum of all licensed


6/16



channels merged into it. This can be done by using the OFDM technique with selective allocation of subcarriers to the

channels to be utilized [10-12]. In this way, more than one channel in the logical channel can be simultaneously utilized by a

single SU. The term home channel will be used to represent a licensed channel which is merged into the logical channel and

being utilized by SUs. In contrast, a channel that does not belong to the logical channel is called a foreign channel. For SUsto share the logical channel we assume the following medium-access model:

SUs with packets to transmit compete with each other to gain exclusive access to the logical channel

while an SU is transmitting, other SUs keep silent

The SU who has gained exclusive access to the channel should listen to the medium before transmission to detect

returning PUs [10].

Figure 4: The Illustration of Sensing of an ON-OFF Alternating Channel

For minimizing interference, the return of PUs on a home channel should be detected promptly. This can be done

by the Listen-before-Talk policy where every SU has to listen to the medium before commencing any packet transmission.

Hence, we can assume that returning PUs can be detected within a reasonably small amount of time so that the channel can

be vacated by SUs promptly. To vacate the channel due to returning PUs, OFDM should reconfigure subcarriers to exclude

the channel band from usage. As a special case, if the home channel to be vacated is the only member of the logical channel,

there will be no more channels to utilize. We call this situation as channel switching in that the SUs should switch from the

current channel to a new idle channel. It is important to find the new idle channel as soon as possible so that SUs can resume

their data transmission with the least interruption.

4.4 Sensing Period Optimization

When proactive sensing is employed by an SN and each channel is sensed periodically with its own sensing period,

we would like to optimize the set of N sensing periods to maximize the discovery of opportunities. Since sensing is

nothing but a sampling process, it is not possible to exactly identify each state transition between and periods.

Hence, the time portion of a discovered period between the start time and the discovery time of the period cannot

be utilized. In addition, some periods may remain undiscovered at all if sensing is infrequent. However, blindly

increasing the sensing frequency is not desirable, as it will increase the sensing overhead, which is proportional to the sum of

( ). Note that the sensing overhead is the time overhead during which all data traffic among SUs must be suspended to

measure a channel's availability. This trade-off must be captured in the construction of an equation to find the optimal

sensing frequencies/periods. Therefore, for each channel , we define two mathematical terms, (Unexplored

Opportunity) and (Sensing Overhead),where = .) is defined as the average fraction of


7/16



time during which channel 's opportunities are not discovered in case channel is being periodically sensed with its sensing

period .On the other hand, is defined as the average fraction of time during which channel 's discovered

opportunities are interrupted and not utilized due to sensing of one of channels. An already discovered opportunity within

a channel will be interrupted by sensing because we assumed that 1) an SU is equipped with a single antenna and 2) all SUs

in the SN must participate in sensing a channel. That is, the SUs must suspend the use of a discovered channel when it

senses other channels since data transmission and sensing cannot take place at the same time with one antenna [21].

This situation is depicted in Figure 5.

Figure 5: Concept of SOHi: Channel 1's Discovered Opportunity

cannot be Utilized during Sensing of Channel 2

(1)

(2)

Where is a vector of optimal sensing periods. As a boundary condition of

, shouldbe satisfied, providing a lower bound of .

4.4 Analysis of UOPi (TPi)

Define as the average of opportunitieson channel during provided a sample is collected at

time . In case the state transition occurs at , is used to denote the same metric. Possible

scenarios of those four functions are shown in Figure 6.

Figure 6: Illustration of or each Function, Two Possible Cases 1 and 2 are Shown.

Denotes the Remaining Time in the Current Period Starting from in Case the State TransitionOccurs at, x is Used Instead of


8/16



Define as the average of opportunities on channel during provided a sample is collected at

time . In case the state transition occurs at , is used to denote the same metric. Possible

scenarios of those four functions are shown in Figure 6.

According to the renewal theory, for an alternating renewal process that has been started a long time ago, the

remaining time in the current state from the sampling time has its pdf

of [20,21].where and is cdf of the period. This is shown in Figure

7. Where is a random variable of the remaining time in the period.

Figure 7: The Density Functions of the Remaining time in the Current Off Period

Similarly, the pdf of the remaining time in the state from is given as Using the above facts, we

can derive the following equations.

(3)

(4)

(5)

(6)

By taking Laplace transform,, one can obtain

(7)

(8)

(9)

(10)

Hence it leads to,


9/16



(11)

(12)

Now, we find an expression of in terms of and A new term is defined as the

average fraction of time during which usable opportunities are not discovered between two consecutive samples in case the

first sample is . Then,

In case is collected at time, opportunities existing in cannot be discovered since there is no

more sensing between two sampling times and . Since the amount of opportunities in and is given

as

We get,

(13)

In case is collected at time, the opportunity discovered at startsto be utilized until PU's return. If the

period lasts more than after , there will not be any unexplored portion of opportunities in ( ).On the contrary, if

PU's emerge at ( < < ), any opportunities in( , , ) could not be explored since the next sampling time is

.Hence,

(14)

Two examples of are introduced here. In case channel is ON/OFF periods are Erlang-distributed, we

have

On the other hand, for exponentially distributed periods, we have

,


10/16



These results are reasonable in the sense that no opportunity is discovered

since no sensing will be performed. Therefore, becomes .

Analysis of

With reference to the earlier definition, is the average fraction of time during which channel s discovered

opportunities cannot be utilized due to sensing N channels. To express mathematically, we introduce a concept of

observed channel usage pattern. Since a channels usage pattern is partially observed by SU's via sensing at

discrete time points, the exact renewal times (that is, state transition times such as cannot be

observed by SU's. Instead, we use an observed pattern of channel i to derive In the observed

model, a channels period starts when the period is discovered. Once an period is discovered, however, the

next state transitionto the following period is assumed to be recognized via the Listen-before-Talk policy.

Figure 8, illustrates the concept of the new model. This model of channel utilization is called modified channel utilization,

denoted by iwhich is given as = + . Using the new model, is derived as

Figure 7: The Observed Channel-usage Pattern Model

Here (1- ) implies the time fraction in which channel s opportunities are discovered. The reason for using

instead of is that is only concerned with the discovered portion of periods by its definition. The second term

, means the cumulative sensing overhead due to sensing on channels.

5. PROPOSED ALGORITHM

Start

Step 1:Get Original Signals and SU Signals.

Step 2:Get Discovered Signals, SOH Signals.

Step 3:Initialize the respective parameters.


11/16



Step 3.1:Discovery Signals -1 SU send in the channels, else 0.

Step 3.2:SOH Signals -1 SU stop current sending because any other SU or same SU need sensing, otherwise 0.

Step 3.3:Initialize the required variables with some default value (mostly zero).

Step 4:Start Process of the data.

Step 4.1:Start process for each channel.

Step 4.2:Check Last Sensing State and change the counter states.

Step 4.3:Along with that check for the Channel is free or not and check its counter state.

Step 4.4:Change Channel Delay counter if SU in Delay time.

Step 4.5:Change the Channel Tpcounter value.

Step 5:Get new SU Packet from each channel.

Step 5.1:There is no packet to be send/received.

Step 5.2: Sense the channels in an ascending order according to their utilization.

Step 5.3:At sensing delay SU Ti second to stop any SU.

Step 5.4:Idle Channels enough.

Step 5.5:Sleep for Sleep in Time second.

Step 6:Transmit/Receive.

Step 6.1:Add SU Delay and SU Send Packet.

Step 6.2:Remove Packet from beginning of Queue.

Step 6.3:Remove Idle Channel from ending of Queue.

Step 6.4:Sense the channels periodically.

Step 6.5:Check TpEnding for any Channel.

Step 6.6:Reset Channel Tpcounter.

Step 6.7:Estimate channels parameters.

Step 6.8:Adapt sensing period for each channel.

Step 7:Calculate.

Step 7.1:No error PU OFF.

Step 7.2:Error UOP1 opportunities existing cannot discovered.

Step 7.3:Error UOP0 PUs emerge in opportunities.

Step 7.4:No error PU On.


12/16



End.

6. SIMULATION RESULTS

For verification of effectiveness of the proposed scheme, we have applied the performance metric as Achieved

Opportunity Ratio (AOR).AOR is a used to represent the efficiency of the proposed sensing period optimization in terms of

the percentage of total opportunities it can discover. Mathematically it is represented as,

In simulation we have shown results in comparison with actual AORmax, as practically is not possible to achieve

one hundred percent AORmax. We also studied how close the estimation results would be to the actual channel parameters

and how well estimates track time-varying channels. The proposed scheme is comparatively evaluated against other

schemes. For the AOR test, our scheme with sensing period optimization is compared to the reference scheme without

sensing-period optimization.

Figure 9: The Observed Channel-usage Pattern Model

Figure 9 shows the performance of the proposed system in terms of AOR with respect to the reference schemes.

For owing effective utilization of the available, we considered a reference of one hundred percent of which is

practically un achievable due to the sensing overhead (SOH

i

) and the missed portion of opportunities (UOP

i

). Thus, it isconsidered the analytical maximum of utilizable opportunities as (AORmax). The sensing-period optimization offers more

than 90 percent of the analytical maximum of discovered spectrum availability regardless of the tested conditions

Following are the factors responsible for the maximum utilization of the available opportunities

Time-varying parameters,

Adaptation time for sensing periods.

Perturbation of estimated parameters from the actual parameters.

About 25 percent more discoveries observed than the reference scheme. As the initial Tip is chosen smaller than0.05 seconds or larger than 1.0 seconds. In fact, no reference scheme can outperform the proposed scheme. As the initial Tip


13/16



is chosen farther away from the optimal one the performance of the reference scheme degrades greatly for two reasons.

Figure 10: Channel Utilization Estimation of ui

: N=4

Figure 11: ESimation of OFFN=4

Figure 10 and Figure 11 show the accuracy of our channel-usage pattern estimation. Each point in the Figure

indicates the estimate produced within an estimation cycle, Testimation. Dashed lines represent the actual target channel

parameters. The plot of, as well as ,follows the actual channel parameters very closely even when they are time

varying.

7. CONCLUSIONS

The cross layer issue of spectrum sensing is addressed by optimizing the sensing-time for MAC layer spectrum

sensing. The proposed algorithm helps to optimize the sensing period. The new model for spectrum sensing is prepared for

improving the channel Utilization, by using the available spectrum opportunities more efficiently, by considering the

underlying ON-OFF channel usage patterns. This is done by developing an improved optimization algorithm utilization of

the maximum spectrum opportunities for secondary users and improves spectrum efficiency from dynamic spectrum

management perspective. It is believed that the proposed methodology can be beneficial for utilizing maximum spectrum

opportunities for secondary users and improves spectrum efficiency. Results show comparatively better performance in

channel utilization and sensing period adaption. In future interconnection of PHY and MAC layer spectrum sensing can be

possible to address cross layer issue of spectrum sensing.


14/16



8. REFERENCES

1. FCC, Spectrum Policy Task Force Report, ET Docket, No. 02-155, November 02 (2002).

2.

I. F.Akyildiz et al, NeXt Generation/Dynamic Spectrum Access/Cognitive Radio Wireless Networks: A Survey,

Comp. Networks J, vol. 50, 2127-2159, September (2006).

3. Vision RFC, http://www.ir.bbn.com/projects/xmac/rfc/rfcvision. pdf, (2007).

4.

Arch. RFC, http://www.ir.bbn.com/projects/xmac/rfc/rfcaf. pdf, (2007).

5. FCC Notice of Proposed Rulemaking and Order, ET Docket, No. 03-108, December (2003).

6.

D.Cabric, S. M. Mishra, and R. W. Brodersen, Implementation Issues in Spectrum Sensing for Cognitive

Radios, Proc. Asilomar Conf. Signals, Systems, and Computers, pp. 772-776, November (2004).

7. H. Tang, Some Physical Layer Issues of Wide-Band Cognitive Radio Systems, Proc. IEEE International Symp.

Dynamic Spectrum Access Networks (DySPAN 05), pp. 151- 159, November (2005).

8.

S. Shankar, C. Cordeiro, and K. Challapali, Spectrum Agile Radios: Utilization and Sensing Architectures, Proc.

IEEE International Symp. Dynamic Spectrum Access Networks (DySPAN 05), pp. 160-169, November (2005).

9. D. P.Patil, V. M.Wadhai, Dynamic Spectrum Management Techniques Using Cognitive Radio: Issues and

Proposed System, Journal of Computing, VOLUME 3, ISSUE 10, October (2011).

10.

C.-T. Chou, Adaptive Quality-of-Service Provisioning in Wireless/ Mobile Networks, PhD dissertation, Univ.

of Michigan, (2004).

11. T. Keller, L. Hanzo, Adaptive Multicarrier Modulation: A Convenient Framework for Time-Frequency

Processing in Wireless Communications, Proc. IEEE, vol. 88, no. 5, pp.611-640, May (2000).

12.

R. Rajbanshi, Q. Chen, A. M. Wyglinski, G. J. Minden, and J. B. Evans, Quantitative Comparison of Agile

Modulation Techniques for Cognitive Radio Transceivers, Proc. Fourth IEEE Consumer Comm. and Networking

Conf. Workshop Cognitive Radio Networks, pp. 1144-1148, Jan. (2007).

13. H. Kim, C. Cordeiro, K. Challapali, and K.G. Shin, An experimental Approach to Spectrum Sensing in Cognitive

Radio Networks with O-the-Shelf IEEE 802.11 Devices, Proc. Fourth IEEE Consumer Comm. and Networking

Conf. Workshop Cognitive Radio Networks, pp. 1154-1158, Jan. (2007).

14.

IEEE 802.22 Working Group on Wireless Regional Area Networks, http://www.ieee802.org/22/, (2008).

15. C. Cordeiro, K. Challapali, and M. Ghosh, Cognitive PHY and MAC Layers for Dynamic Spectrum Access and

Sharing of TV Bands, Proc. First International Workshop Technology and Policy for accessing Spectrum

(TAPAS 06), Aug. (2006).

16.

A. Ghasemi and E.S. Sousa, Collaborative Spectrum Sensing for Opportunistic Access in Fading Environments,

Proc. IEEE International Symp. Dynamic Spectrum Access Networks (DySPAN 05), pp. 131-136, Nov. (2005).

17. Ganesan and Y. Li, Cooperative Spectrum Sensing in Cognitive Radio Networks, Proc. IEEE International

Symp. Dynamic Spectrum Access Networks (DySPAN 05), pp.137-143, Nov. (2005).


15/16



18. E. Visotsky, S. Kuner, and R. Peterson, On Collaborative Detection of TV Transmissions in Support of

Dynamic Spectrum Sharing, Proc, IEEE International Symp. Dynamic Spectrum Access Networks (DySPAN

05), pp. 338-345, Nov. (2005).

19.

D. R. Cox, Renewal Theory, Butler and Tanner (1967).

20. S. M. Ross, Stochastic Processes, John Wiley & Sons (1983)

21.

H. Kim, K. G. Shin, Efficient Discovery of Spectrum Opportunities withMAC-LayerSensing in Cognitive Radio

Networks, Proc. IEEE Transactions on Mobile Computing,VOL. 7, NO. 5, MAY (2005).


16/16

Date post:	02-Jun-2018
Category:	Documents
Upload:	tjprc-publications
View:	215 times
Download:	0 times

Improved Algorithm for MAC Layer Spectrum Sensing in Cognitive Radio Networks from Dynamic Spectrum...

Documents