Increasing the Available White Space SpectrumUsing TV Receiver Information

T oday, TV white spaces are the most promising

spectrum in which cognitive radios (CRs) can

be deployed. The goal of this article is to investi-

gate the potential increase

in available spectrum used by the

CR device. We define two knowl-

edge levels and show that infor-

mation about the TV receivers

can increase the amount of available spectrum for

CR devices in the TV bands by as much as 120 MHz.

In the last few decades, demand for wireless services

has exploded. According to Cisco, mobile data traffic is

expected to have a 26-fold increase from 2010 to 2015 [1].

We know from Shannon’s capacity

formula that increasing the avail-

able bandwidth increases the

capacity of communication sys-

tems. In terms of bandwidth occu-

pancy, the emergence of digital technology over analog

technology has freed up some bandwidth. When TV

broadcasting systems switched from analog to digital

broadcasting, around 100 MHz was freed up, depending

on the country. This part of the spectrum has becomeDigital Object Identifier 10.1109/MVT.2011.2179344

Date of publication: 3 February 2012

© PHOTO F/X2

Hemdan Bezabih, Brage Ellingsæter,

Josef Noll, and Torleiv Maseng

©P

HO

TO

DIS

C

24 ||| 1556-6072/12/$31.00©2012IEEE IEEE VEHICULAR TECHNOLOGY MAGAZINE | MARCH 2012

known as the digital dividend, and specific reallocation of

this spectrum to other services is a current process in

many countries. Because of the high value of spectrum, it

is likely that only the capable operators will be able to buy

the rights to these frequencies.

As the need for spectrum for high-capacity services

became apparent, CR emerged as a hot research topic [2].

The basic idea of a CR is that a licensed band allocated to

a certain wireless service will not always be used. Thus, if

an unlicensed user (secondary user) has the ability to

detect these spectral holes, he or she should be able to

use these without causing harmful interference to the

licensed user (primary user). Over the years, more

research has been done in this area, especially in the areas

of sensing and simultaneous transmission between

primary and secondary users.

One part of the spectrum particularly suited for CR is

the frequencies allocated for DVB-T. This is due to the

reuse factor used in DVB-T systems, which leads a portion

of the allocated spectrum unused at any given location.

The underutilization of these frequencies has been verified

by both the U.S. Federal Communications Commission

(FCC) [3] and the researchers in Switzerland [4]. These

unused frequencies are known as the TV white spaces.

Many researchers have pushed hard to allow the CR

technology to operate in the TV white spaces, and recently,

the FCC has allowed the CRs to operate in the white spaces

given they do not create harmful interference to the TV

receivers. The current criteria for

transmission for a CR are that the

radio must be able to look up its

location in a database and transmit

under the power constraint given

for that frequency in that area [5].

The criterion for transmission is

primarily based on the TV broad-

cast station and CR locations.

Registering the TV broadcast

stations and not the receiver loca-

tions leads to a significant overpro-

tection of certain areas where there

are no TV receivers. We investigate

the potential gain (in amount of

additional available frequencies)

from registration of TV receiver

locations when compared to only

registering TV broadcast station’s

locations. We define white spaces in

the conventional way as a spectrum

that is unused by the TV broadcast

station in the area due to the reuse

factor in DVB-T systems. We intro-

duce gray spaces as an additional

spectrum that is available, given TV

receiver information. To estimate

the amount of gray space, we define the protection area of

a TV receiver based on tolerable interference levels given

by the FCC and Office of Communications (Ofcom) and a

standard propagation model (Okumura–Hata model). We

assume two different types of wireless devices as CRs.

The first is a fixed 4-W radio that is used to illustrate the

IEEE 802.22 base station. The second is a mobile 100-mW

radio, which is thought of as a hand-held device, such as a

mobile phone, laptop, or other small-scale radio device.

We also define two knowledge levels of TV receiver

information that yield different amounts of gray space:

n Knowledge Level 1 (KL1): Only knowing the TV re-

ceiver locations.

n Knowledge Level 2 (KL2): Knowing the TV receiver loca-

tions and the TV channels each receiver subscribes to.

For these information levels, we provide real-value sim-

ulation results based on real location information from

Norway. The main finding is that, by registering the TV

receivers, a significant amount of frequencies can be uti-

lized by CRs in addition to white spaces.

Unlicensed Operation in TV White Spaces

As mentioned earlier, white spaces occur due to the reuse

factor used in DVB-T systems. Or more specifically,

because there is a necessary separation distance between

two TV broadcast stations using the same frequencies to

avoid interference (Figure 1). If a low-power device is

located sufficiently far away from the TV broadcast

TX Method

RX Method

Gray Space

TV Receivers

TV StationGray SpaceWhite SpaceProtection Area

TV Receivers

TV Service Area

White Space

FIGURE 1 The concept of white and gray spaces. The three circles represent the TV coverage

areas where the broadcast stations use the same frequency. By only registering the TV broad-

cast station’s locations, this frequency can only be used by a CR device outside the protection

area of this coverage area (white space). By registering the TV receivers and protecting these

receivers, a CR device can utilize the same frequency as the TV broadcast station within the

coverage area (gray space).

MARCH 2012 | IEEE VEHICULAR TECHNOLOGY MAGAZINE ||| 25

station, it can reuse the frequency without affecting the

reception of the TV receivers.

Accessing White Space

When operating on a secondary basis in licensed bands,

the main considerations are the licensed users. In the TV

band, this is mainly the TV receivers and wireless micro-

phones. In both the United States and United Kingdom, dif-

ferent methods have been considered in accessing and

utilizing white spaces. Research shows that, by using a

frequency-sensing equipment in combination with a geo-

database that gives a list of occupied frequencies, one can

guarantee that the devices operating in a secondary man-

ner meets the required interference levels. The database

is the approach that will be investigated in this article.

The database containing the location of the TV broadcast

stations [referred to as the transmitter (TX) database] regis-

ters TV TXs and their service area, and in addition, other

protected devices such as cable head ends and locations

where wireless microphones are used. This information

ensures that the TV receivers within the service area are

protected; the channels used within the service area are set

as occupied and are not available for a CR device.

Operation Rules

Once a CR device has supplied the TX database with its

location and received the set of available frequencies,

there are still some operational requirements. These are

primarily requirements on the cochannel and adjacent

channel interference levels.

The maximum level of interference a TV receiver can

handle is defined by the carrier-to-interference (C=I) ratio.

The C=I levels given by regulators define how much lower

the CR signal strength must be compared with the TV sig-

nal at the TV receiver. The technical parameters defined

by Ofcom and the FCC are summarized in Table 1.

TV Receiver Registration

When registering TV TX parameters, all possible user loca-

tions within the service area of a given TV TX are protected.

This method is conservative regarding utilizing frequen-

cies, as it protects all locations within an area without con-

sidering whether there is anyone to protect. TV receivers

are not located everywhere within a service area, thus mak-

ing this method ineffective, especially, in rural areas. Some

examples of such scenarios are the following:

n Of the total population, only a portion uses digital

terrestrial television. By using the TX database, those

who are not using the TV service are also protected.

n There may be households within the service area that

are receiving signals from other TV stations using a

different frequency. Using the TX database gives no

option of adjusting the protection to fit certain cases.

n TV coverage areas may overlap, which makes it hard

to make a statement regarding which channels are

used at a location. Using the TX database in locations

that intersect several TV service areas will increase

overprotection.

n Signal strength of the TV TX will vary within a service

area; there are locations that do not have access to

the service or where the probability of service is low

because of terrain and structure. There is no need to

protect those areas within the service area.

The common factor for these issues is that the registra-

tion of TV receiver location together with terrain knowl-

edge would solve these issues to a large extent. One

aspect that has been taken into account with this method

is how the registration of TV receiver location can be

done. One solution can be to provide the location informa-

tion when buying the TV. In some countries (e.g., Norway),

this is already done due to a mandatory license needed to

watch TV. The second, and, perhaps, most appropriate

option, is for the TV user to register the TV receiver loca-

tion online or automatically through an IP connection. The

incentive to the user is, of course, that his signal reception

will be better protected.

One problem is the portable TV devices. If such a

device has an IP connection, it can provide updates on its

location. But as the majority of TV receiver locations have

fixed receivers, we neglect this issue in this article.

TV Receiver Protection

With the knowledge about the TV receivers,

we no longer have to protect the whole

service area of a TV broadcast station from

unlicensed transmission. Instead, we are

able to construct protected areas surround-

ing the TV receiver sites. The spectrum

made available to a CR device by this

method is defined as gray space.

The TV receiver protection area mainly

depends on calculations using TV signal

strength at the TV receiver locations and

known C=I ratios given in Table 1. In addi-

tion, the geographical location of the TV

TABLE 1 Technical parameters defined by Ofcom [5] and the FCC [5].

ParametersOfcom/UnitedKingdom FCC/United States

Bandwidth (MHz) 8 6Min. field strength forservice

50 dBlV/m 41 dBlV/m

Receiver height 10 m 10 mLocation accuracy 100 m 50 mCochannel interference 33 dB 23 dBPower limit As specified by the

databaseFixed device: 4 W

Adjacent channelinterference

�17 dB �26/�28 dB

Power limit 50 mW Portable device:40/100 mW

26 ||| IEEE VEHICULAR TECHNOLOGY MAGAZINE | MARCH 2012

receiver must be estimated. In this article, the TV signal

strength at the TV receiver location is assumed to be 50

dBlV, as this is defined as the minimum field strength

needed to receive TV service by Ofcom (Table 1).

Knowledge Levels

The goal of this article is to investigate the potential increase

in available spectrum that a CR device can utilize due to TV

receiver knowledge. We define two knowledge levels. KL1 is

defined as only knowing the TV receivers’ locations, and KL2

as knowing the TV receivers’ locations and the channels each

receiver subscribes to. As different TV channels are multi-

plexed to increase frequency utilization, KL2 means knowing

which multiplexes (MUXs) a receiver is able to decode.

As this information is stored by the service provider,

KL2 is a plausible assumption. However, this information

is business sensitive and not all subscribers would be will-

ing to share this information. Thus, obtaining this informa-

tion requires strict authorization.

CR Device Parameters

We now estimate the minimum distance between a CR

device and a TV receiver, given the specifications in Table 1.

We calculate the distance for two different CR devices: one

is a fixed 4-W device thought to represent an IEEE 802.22

base station, and the other is a portable 100-mW device.

The required carrier-to-interference ratio at the TV

receiver will limit the maximum signal strength transmit-

ted by the CR device, denoted as ETCR. Denoting the maxi-

mum allowable signal strength from the CR device at the

TV receiver site as ERCR, we have

Lmin½dB� ¼ ETCR

dBlV

m

� �� ERCR

dBlV

m

� �,

where Lmin is the minimum required overall path loss

between the CR device and TV receiver.

For a given CR transmit power, PTCR, the equivalent

term of field strength must be calculated. Field strength is

related to transmit power as follows [7]:

EdBlV

m

� �¼ 10 log10 P ½mW� � 20 log10 d ½m� þ 104:8,

where d is the reference distance for the field strength. In

this article, we set d ¼ 1.

Maximum CR Field Strength at a TV Receiver

In Table 1, the minimum field strength and protection

criteria for the TV receivers were given. We use the values

defined by Ofcom, as they are the most restrictive. The

minimum field strength of the TV signal at the TV receiver

is ERTV ¼ 50 dBlV=m½ �. The requirement for the C=I level

for a 8-MHz band is 33 dB for cochannel and �17 dB for

adjacent channel usage.

The maximum allowed CR field strength at the TV

receiver is

ERCR ¼ ERTV � C=I:

When substituting the values obtained above, we find

the field strength for cochannel usage to be ECORCR ¼

50� 33 ¼ 17 dBlV=m½ � and for adjacent channel usage to

be EadRCR ¼ 50(�17) ¼ 67 dBlV=m½ �.

Propagation Loss Between CR and TV Rx

To calculate the minimum distance between a CR device

and TV receiver, we use the Okumura–Hata propagation

model. By changing the parameters of the model, one

can predict the path loss for three different environ-

ments: urban, suburban, and rural [8]. We choose the

path-loss model for the suburban environment, pri-

marily, because the areas chosen in the simulations fit

this description.

For a suburban environment, the path loss according

to the Okumura–Hata model is given by

L ¼ 69:55þ 26:16 log10 fc ½MHz� � 13:82 log10 hb ½m�,� aþ (44:9� 6:55 log10 hb ½m�) log10 d ½km�

� 2 log10

fc ½MHz�28

� �2

� 5:4

!,

a¼(1:1log10(fc ½MHz�)�0:7)hm ½m��(1:56log10 fc ½MHz��0:8),

where L is the path loss in decibels; fc is the carrier

frequency in megahertz; hb is the base station height in

meters, which corresponds to the height of the CR trans-

mitter; TX; hm is the mobile height, which corresponds to

the TV receiver height in meters; and d is the distance

between the CR TX and TV receiver in kilometers.

The minimum distance between a CR device and a

TV receiver is given in Table 2. For 4-W fixed devices,

we assume a height of 30 m, as this is supposed to illus-

trate the IEEE 802.22 base station. For the 100-mW

device, we assume a height of 2 m. The carrier fre-

quency is equal to 650 MHz, and the TV receiver

antenna height is set to 10 m. The difference between

the minimum required distance for the cochannel and

adjacent channel interference is due to the difference

between EcoRCR and Ead

RCR.

Simulations Based on

Real Statistics from Norway

To illustrate the advantage of TV receiver registration, we

compute the gray space amount at two municipalities in

TV WHITE SPACES ARE THE MOST PROMISINGSPECTRUM IN WHICH COGNITIVE RADIOSCAN BE DEPLOYED.

MARCH 2012 | IEEE VEHICULAR TECHNOLOGY MAGAZINE ||| 27

Norway based on the household statistics in these munici-

palities. The areas chosen are Vinje and Lillehammer. The

land area and population density of these municipalities

range from 478.2 to 3,106 km2 and 1/km2 to 54/km2. The

household locations of these areas are shown in Figure 2.

Choosing these areas is regarded as appropriate to ana-

lyze the suburban and rural areas in Norway, where we

consider that exploitation of white and gray spaces will

have the most impact.

TV Broadcast Station Information

An area may contain households that receive TV signals

from different TV stations. By using the coverage plan-

ning map by Norway Television [9], the strongest TV TX

in each area is found. For instance, in Vinje, the

strongest broadcast station is Rauland, which uses

Channels 25, 27, 32, 35, and 42. Using the information

from Norwegian Post and Telecommunications Author-

ity (NPT), additionally, there are 12 TV broadcast sta-

tions that are connected to Rauland in a single-

frequency network (SFN), where all broadcast stations

use the same channels [10].

In our simulations, we assume that all households in dif-

ferent areas receive TV service from either the strongest TV

broadcast station or from another station in the same SFN.

In reality, there may be exceptions (e.g., if a TV receiver is at

the edge of the area, it may receive signals from other broad-

cast stations that are not in the same SFN). However, these

exceptions are neglected in this article.

Topology and TV Receiver Locations

The information on household location is retrieved from

Statistics Norway (SSB) [11], who provides the statistics

for different grid sizes: 100 m, 250 m, and 1 km. In this arti-

cle, we have used the statistics based on a grid size of 1

km. For each grid element, the information provides the

number of households within 1 km2.

The method used to map the information from SSB to

the grid leads to an overestimation of the area. The area is

converted to a rectangular M 3 N grid, but the original

shape of the municipality may be different. We compen-

sate for this expansion by marking the number of grid ele-

ments to be equal to the expansion factor at certain

borders of the grid as not being part of the municipality.

Protection of TV Receivers

In Table 2, the minimum distance is given for cochannel

and adjacent channel usage for the two CR devices. The

10

10

60

50

40

30

20

20 30 40 50 60

10

5 10 15 20 25

5

0

5

(a) (b)

FIGURE 2 Household locations for (a) Vinje and (b) Lillehammer are illustrated by black squares. Each square represents a 1-km grid size.

TABLE 2 Minimum required distance between a CR device and a TV receiver.

Channel Type PTCRðdBmÞCR FieldStrength

Minimum RequiredPath Loss L (dB)

MinimumDistance

4-W equivalentisotropically radiatedpower (EIRP)

Cochannel 36 140.8 123.8 7.35 km

Adjacent channel 73.8 281 m100-mW EIRP Cochannel 120 124.3 107.8 910 m

Adjacent channel 57.8 62 m

28 ||| IEEE VEHICULAR TECHNOLOGY MAGAZINE | MARCH 2012

resolution obtained on the household statistics is 1 km2.

Thus, we have to protect each TV receiver by a minimum

distance of 1 km. The distance used in the simulations is

obtained by ceiling these distances to the closest kilometer.

When this is done, each TV receiver is protected by a

square of grid elements, where each grid element is 1 km2,

with the TV receiver location in the middle and the number

of grid elements surrounding the TV receiver equal to

(2 min �distanced e)2 � 1.

Amount of White and Gray Spaces

In this article, we have assumed that a broadcast station

uses five MUXs to provide service to each area. Each of

the MUXs contain several TV channels and occupy a band-

width of 8 MHz. The total amount of spectrum used for TV

broadcasting in Norway is 320 MHz. White space is

defined as the spectrum that is neither occupied by a MUX

nor by the channel’s adjacent channels.

The maximum number of adjacent channels is ten.

Therefore, the maximum white space amount is

320� 5 3 8� 10 3 8 ¼ 200 MHz. Gray space is defined as

the channels used for broadcasting in the area and the

adjacent channels. The maximum gray space amount is

5 3 8þ 10 3 8 ¼ 120 MHz.

Results

We now provide the simulation results from the munici-

palities. It is assumed that 98% of all households have a TV

receiver. For KL1, all households are pro-

tected. For KL2, it is assumed that 98% of

the households receive MUX1, whereas

15% of the households receive other MUXs.

This is based on statistics from the DVB-T

provider in Norway. We give the results in

terms of percentage of the total area that con-

tain a certain amount of gray space for the dif-

ferent knowledge levels and municipalities.

Table 3 gives the results obtained for the

municipalities of Vinje for a 4-W CR device

and a 100-mW CR device. Table 4 gives the

same for the municipality Lillehammer.

In the results, we have listed the amount

of total white space at the different loca-

tions as well as the gray space available for

the different knowledge levels. For example,

in Vinje, 100% of the locations contain

200 MHz of white spaces. When considering

KL1 and a 4-W device, there is an addition of

80 MHz of gray space at 40% of the loca-

tions. As expected, the gray space amount

increases with KL2.

Discussion

The hypothesis of this article is that the

knowledge of the TV receivers will

increase the spectrum available for a CR device. The sim-

ulations in the previous section verify the hypothesis to

a large degree. We now discuss the results obtained in

this article.

Knowledge Levels

Depending on the CR device considered, knowledge of the

TV receiver locations increases the available spectrum by

80–120 MHz in certain locations. In this article, we have

not taken into account the amount of households that can

actually utilize the frequencies made available by the TV

receiver information.

KL1 is the easiest to implement, needing only TV

receiver location and the information on the channels

available at those locations. The drawback is, of course,

the limited amount of locations containing gray space.

KL2 requires information about the location and

subscriber information of the TV receiver. Although the

subscriber information is business sensitive, our results

TABLE 3 White space amount (WSA) and gray space amount in Vinjefor a 4-W and 100-mW CR device.

DevicePercentageof Area WSA (MHz) KL1 (MHz) KL2 (MHz)

20 200 80 1124 W 40 200 80 88

70 200 — 8090 200 — 6420 200 120 120

100 mW 40 200 120 12070 200 — 12090 200 — 96

TABLE 4 White space amount (WSA) and gray space amountin Lillehammer for a 4-W and 100-mW CR device.

DevicePercentageof Area WSA (MHz) KL1 (MHz) KL2 (MHz)

20 200 80 804 W 40 200 — 80

70 200 — 6490 200 — —20 200 120 120

100 mW 40 200 — 9670 200 — —90 200 — —

THE COMMON FACTOR FOR THESE ISSUES ISTHAT THE REGISTRATION OF TV RECEIVERLOCATION TOGETHER WITH TERRAINKNOWLEDGE WOULD SOLVE THESE ISSUESTO A LARGE EXTENT.

MARCH 2012 | IEEE VEHICULAR TECHNOLOGY MAGAZINE ||| 29

show that this information can be justified by the addi-

tional amount of gray space found.

Business Aspect

The main argument for opening the white space for public

use is that the need for spectrum is increasing, while at

the same time, the spectrum used for TV broadcasting is

severely underutilized. The results found in this article

have value for both the license holder and a service

provider. The license holder may be interested to know

how ineffective the utilization of the frequencies are at cer-

tain locations so that the license holder can provide addi-

tional services to its customers or lease out the spectrum

to other service providers.

For a service provider, the gray space can provide

additional capacity to its service. But the amount that

he can obtain and the location where the spectrum can

be utilized define the value of the gray space. If no one

can utilize the spectrum, the service provider will not

make a profit. To clearly estimate the potential of gray

space for a service provider, a more thorough investi-

gation of household locations and business locations

are necessary.

Conclusions

The goal of this article is to investigate the potential

increase in available spectrum for cognitive devices in the

TV bands by registering TV receiver information in addi-

tion to the conventional approach to only register TV

broadcast station information. The spectrum made avail-

able with this method was defined as gray space, and by

operating within the carrier-to-interference levels given by

Ofcom, we provided estimations of the amount of gray

space based on household statistics from Norway.

Author Information

Hemdan Bezabih (hemdanb@gmail.com) received her

M.S. degree in electronics and computer technology

from the University of Oslo in 2011. She is currently

working as a radio network planning consultant.

Brage Ellingsæter (brage@unik.no) received his M.S.

degree in electrical and computer engineering from Nor-

wegian University of Science and Technology in 2010. He

is currently a Ph.D. candidate at the University of Oslo

working on resource allocation in CR. His other research

interests include game theory, communication theory,

and wireless networks.

Josef Noll (josef@unik.no) is a professor at the

University of Oslo in the area of mobile services. His

work concentrates on the working areas such as mobile-

based trust and authentication, personalized and con-

text-aware service provisioning, and evolution toward

beyond 3G systems. He is also a CTO in Movation. He is

a founding member of the Center for Wireless Innova-

tion, a group consisting of seven universities and

university colleagues in Norway. He is a project leader

of the European Union (EU) Artemis nSHIELD project

and was a leader of several other EU and Eurescom

projects. He is a reviewer of several EU FP6/FP7 proj-

ects, an evaluator of national and EU research pro-

grams, a steering board member of Special Interest

Groups, and coeditor of Semantic Services and User

Profiles/Profiling publications.

Torleiv Maseng (torleiv.maseng@ffi.no) was involved

in the start-up of the new private mobile operator in

1992–1994 at NetCom GSM in Norway, where he had

technical responsibility. Since 1994, he has been a chair

in radio communications at the University of Lund, Swe-

den. In 1996, he took up his employment at Norwegian

Defense Research Establishment located at Kjeller. Since

2005, he has been professor II at the University of Oslo.

He is a director of research at the Norwegian Defense

Research Establishment, where he is responsible for

communications and information systems. He worked as

a scientist at SINTEF in Trondheim for ten years and was

involved in the design and standardization of GSM. For

seven years, he was a scientist at the NC3A NATO

research center in The Hague. He is the author of more

than 200 papers, holds patents, and is a member of the

Norwegian Technical Academy of Science. He has re-

ceived awards for outstanding research and has ar-

ranged large international conferences.

References[1] Cisco (2011, Feb.). Cisco visual networking index: Global mobile data

traffic forecast update, 2010–2015. [Online]. Available: http://www.cisco.com/en(US/solutions/collateral/ns341/ns525/ns537/ns705/ns827/white_paper_c11-520862.pdf

[2] J. Mitola, III and G. Q. Maguire, Jr., ‘‘Cognitive radio: Making soft-ware defined radios more personal,’’ personal communication, IEEE,Aug. 1999.

[3] FCC. (2002, Nov.). Report of the spectrum efficiency workinggroup. [Online]. Available: http://www.ictregulationtoolkit.org/en/Publication.2831.html

[4] L. Berlemann and S. Mangold, Cognitive Radio and Dynamic SpectrumAccess. Hoboken, NJ: Wiley, 2002.

[5] FCC. (2010, Sept.). Second memorandum opinion and rules, fcc-10-

174. [Online]. Available: http://transition.fcc.gov/Daily_releases/daily_busi-ness/2010/db0923/fcc-10-174A1.pdf

[6] Ofcom. (2009, Nov.). Digital dividend: Geolocation for cognitiveaccess. [Online]. Available: http://stakeholders.ofcom.org.uk/binaries/consultations/cogaccess/summary/cogacess.pdf

[7] Committee on Radio Astronomy Frequencies. Conversion formulae.[Online]. Available: http://www.fermentas.com/techinfo/nucleicacids/maplambda.htm

[8] S. R. Saunders and A. Aragn-Zavala, Antennas and Propagation forWireless Communication Systems, 2nd ed. Hoboken, NJ: Wiley, 2007.

[9] Norwegian Television (NTV). [Online]. Available: www.ntv.no[10] Norwegian Post and Telecommunications (NPT). [Online]. Available:

www.npt.no[11] Statistics Norway (SSB). [Online]. Available: www.ssb.no

THE GOAL OF THIS ARTICLE IS TO INVESTIGATETHE POTENTIAL INCREASE IN AVAILABLESPECTRUM THAT A CR DEVICE CAN UTILIZEDUE TO TV RECEIVER KNOWLEDGE.

30 ||| IEEE VEHICULAR TECHNOLOGY MAGAZINE | MARCH 2012