Spectrum Sharing between Small Cells and Satellites - IEEE Xplore

IEEE ICC 2015 - Workshop on Cognitive Radios and Networks for Spectrum Coexistence of Satellite and Terrestrial Systems (CogRaN-Sat)

Spectrum Sharing between Small Cells and Satellites: Opportunities and Challenges Awais Khawar∗ , Ishtiaq Ahmad† , and Ahmed Iyanda Sulyman‡ ∗ Department

of Electrical and Computer Engineering, Virginia Tech, USA for Telecommunications Research, University of South Australia, Australia ‡ Electrical Engineering Department, King Saud University, Saudi Arabia Email: {[email protected], [email protected], [email protected]} † Institute

Abstract—Spectrum sharing between heterogeneous systems such as cellular systems, radars, and satellites - is an emerging area of research as it promises to solve the looming spectrum congestion problem. Spectrum regulators and technical specification bodies are contemplating over the idea of allowing commercial wireless systems to be deployed in satellite bands on a sharing basis. In the United States, the Federal Communications Commission (FCC) has proposed to deploy small cells in the 3.5 GHz satellite band. In this paper, we consider deployment of small cells in satellite bands and address opportunities and challenges that are associated with such a deployment. We focus on the conventional and extended C-band fixed satellite service (FSS) receiving earth stations. We show that small cell operation can result in out-ofband emission and low noise amplifier (LNA) saturation issues for FSS systems. Exclusion zones are required to protect FSS systems. We show that exclusion zones are quite larger when small cells are deployed outdoors as compared to the case when small cells are deployed indoors. Therefore, we propose that small cells should be deployed indoors for the scenarios where FSS systems are geographically colocated. This results in maximum utilization of small cell technology and minimum exclusion zones required to protect FSS receiving earth stations. Index Terms—Spectrum sharing, LNA saturation, out-of-band emissions (OOBE), FSS receiving earth stations, coexistence.

I. I NTRODUCTION Current spectrum allocations are not sufficient to meet the ever growing bandwidth requirements of mobile users. Spectrum sharing is an emerging area of research that has potential to solve the looming spectrum crunch problem that operators are facing around the world. Huge chunks of spectrum are available in unlicensed, for example 500 MHz available in the 5 GHz band, and underutilized licensed bands, for example upto 150 MHz available in the 3.5 GHz band, that can be shared with wireless operators. Moreover, spectrum can be shared in time, frequency, and geography. However, significant research efforts are required to realize spectrum sharing between heterogeneous systems. For example, spectrum sharing between small cells and satellites which has been proposed by the Federal Communications Commission (FCC) in the 3.5 GHz band [1]. The deployment of small cells in satellite bands can result in in-band interference, out-of-band interference, and low noise amplifier (LNA) saturation issues for fixed satellite service (FSS) receiving earth stations in the 3.5 GHz band. FSS earth stations can be protected from

978-1-4673-6305-1/15/$31.00 ©2015 IEEE

harmful interference by allowing small cells to operate outside protection/exclusion zones. In order to increase coverage and capacity, cellular operators are deploying pico/femto cells (or small cells) in their traditional macro cells. A small cell is nothing but a small base station that can be deployed indoors/outdoors to fillin coverage gap and enhance user experience. Small cells can have a 3G or 4G air interface and can easily blend in with existing large macro cells to form a heterogeneous (HetNet) network. It is envisioned that a large macro cell would comprise of a large number of small cells without having any significant interference issues. Small cells offer great potential when deployed in bands that are to be shared with incumbents as they result in less amount of interference as compared to traditional macro cells that operate at high power levels and have wider coverage as compared to small cells. This is one of the motivating reasons behind the FCC’s ruling on small cell operation in the 3.5 GHz band where incumbents are commercial FSS receiving earth station and government radars [2]. In order to protect incumbents from harmful interference produced by small cell operation, the FCC has established strict guidelines for small cells that includes limit on maximum transmit power and antenna gain. The 3.5 GHz band contains C-band satellite systems that are used by content providers and broadcasters around the world to deliver radio and television programs from satellites to receiving earth stations that are then responsible for content dissemination. Therefore, such services need to be protected from any kind of interference that arises due to sharing of satellite band. An efficient and robust interference mitigation algorithm that steers nulls of IMT-Advanced systems towards FSS systems in the 3400-4200 and 4500-4800 MHz bands is proposed in [3]. However, the proposed algorithm is very complex in nature as it requires a central authority/operator for precoding design and is not applicable to small cell architecture which is generally distributed in nature. Moreover, the proposed algorithm relies on the availability of channel state information which is not always easy to obtain especially if the two systems do not cooperate. Protection zones required to protect FSS systems from IMT-Advanced systems in the 3400-3600 MHz band has been evaluated in [4]. In [5] coexistence

1600


II. S YSTEM M ODELS In this section, we discuss typical parameters of FSS receiving earth stations and small cells. In addition, we discuss spectrum sharing scenario between small cells and satellite systems. A. FSS receiving Earth Station Characteristics Globally, and in the United States, the 3600-4200 MHz band is an attractive band for satellite operations and satellite companies have invested heavily in the said band. The 36003700 MHz portion of the band falls in the extended C-band and the rest 3700-4200 MHz band is the standard C-band for satellite operations. The satellites in this band usually have 40 MHz channels that are spaced at 20 MHz intervals. Antennas used by receiving earth stations are generally in the 3.5-4.5 meter diameter range. The emission pattern of satellite antennas are regulated by part 25 of the FCC’s rules, which is not to exceed the following limit ⎧ 29 − 25 log(θ) for 1.5◦ ≤ θ ≤ 7◦ ⎪ ⎪ ⎪ ⎨8 for 7◦ < θ ≤ 9.2◦ GFSS (dBi) = ⎪ 32 − 25 log(θ) for 19.2◦ < θ ≤ 48◦ ⎪ ⎪ ⎩ −10 for 48◦ < θ ≤ 180◦ .

FSS Antenna Performance 25 20 15 Antenna Gain (dbi)

analysis between indoor LTE-hotspots and FSS systems is analyzed in the 3400-3600 MHz band. However, the analysis is limited to in-band interference and the system parameters used are hypothetical. Stochastic geometry has also been used to study the outage probability of FSS systems in a spectrum sharing environment with IMT-Advanced systems [6]. Most of the existing work is limited to analysis of interference from high power IMT-Advanced systems, in traditional macro cell architecture, to FSS systems. In addition, the issues of out-ofband interference and LNA saturation are not addressed. In this paper, we compute exclusion zones that are required to protect FSS earth stations from out-of-band interference and low noise amplifier (LNA) saturation issues that result due to the operation of small cells in satellite band. We show that exclusion zones required for FSS protection can be reduced if small cells are deployed indoors rather than outdoors for scenarios where small cells and FSS earth stations are spatially close. In addition, we highlight some challenges that need to be addressed to realize spectrum sharing between small cells and satellite systems. This paper is organized as follows. In Section II we define system models of small cells, FSS receiving earth station, and our spectrum sharing scenario. In Section III we discuss scenarios in which small cells can cause interference to FSS systems. In Section IV we discuss some of the opportunities and challenges that arise due to spectrum sharing between small cells and FSS systems. In Section V we provide some simulation results that discuss exclusion zones for OOBE and LNA saturation. In Section VI we conclude the paper. In Section VII we discuss some future research directions.

10 5 0 −5 −10 −15 −60

−40

−20 0 20 Elevation Angle (theta)

40

60

Fig. 1. FSS antenna performance according to the FCC rules.

to the corresponding look angle will be used to calculate exclusion zones. The look angle is basically the orientation of FSS receiving earth station relative to the satellite. This is set to maximize receive energy from the satellite. High look angles basically correspond to the case when the satellite is directly above FSS receiving earth station and low look angles correspond to the case when a satellite is located to the far west or east. That is why low-look angles are more susceptible to interference. B. Small Cell Characteristics Small cells are low-power wireless base stations that are used to enhance capacity and coverage of cellular service and can be deployed indoor and/or outdoors. The FCC has proposed to operate small cells with a maximum transmit power of 23 dBm (0.2 W) and with a maximum antenna gain of 7 dBi, making the maximum transmitted EIRP 1W (30 dBm), regardless of bandwidth [1]. Therefore, in this paper we do an analysis with the proposed characteristics of small cells. Of course, size of exclusion zones depends upon the transmitted power levels of small cells. C. Propagation Model In this paper, we consider two deployment choices for small cells: indoors and outdoors. For both the scenarios, the propagation models are described as follows. For indoor deployed small cells, we use indoor-to-outdoor channel model, i.e., channel between an indoor deployed small cell and a FSS receiving earth station. We assume that small cells are deployed on ground floor and thus do not take into consideration the number of floors and floor heights. We consider a distance dependent path loss propagation model which takes into account the attenuation due to internal and external walls and is given as

The gain of satellite antenna according to the above requirements is given in Figure 1, for reference. The antenna gain

1601

α=

e wi Lw e li 1 + dβ

(1)


III. I NTERFERENCE I SSUES In this section we address interference and saturation issues that arise due to the operation of small cells in satellite bands. A. In-band Interference The FCC has proposed to use the 3550-3650 MHz band for commercial use. The 3600-3650 MHz portion of the band falls in the extended C-band and is utilized by many FSS receiving earth station in the U.S. The launch of commercial services in this band will cause in-band interference to FSS systems if proper mitigation methods are not employed. B. Out-of-band Emissions (OOBE)

Fig. 2. Spectrum sharing scenario between small cells and fixed satellite service (FSS) receiving earth stations. Due to the in-band and adjacent band operation of small cells, FSS systems face interference issues ranging from out-of-band emissions to LNA saturation.

where Le and Li account for signal loss due to external and internal walls, respectively, we and wi account for the number of external and internal walls between a small cell and a FSS system, d is the distance between a small cell and a FSS receiving earth station, and β is the path loss exponent. This propagation model along with the values, reported in Section V, is a standard model used often to model indoor to outdoor propagation in network consisting of small cell devices [7]. For future work this model can be modified to incorporate high-rise building scenarios that can take number of floors and floor heights into consideration for coexistence analysis. For outdoor deployed small cells, we use free space path loss model which is given as FSPL = 20 log(d) + 20 log(f ) + 32.45

(2)

where f is the center frequency. This model is an accurate representation of small cells that are deployed on a tower or billboard that have a line-of-sight (LoS) with a FSS system. D. Spectrum Sharing Scenario We assume that small cells are sharing the 3550-3650 MHz band with satellite systems according to the FCC proposal [1]. A typical spectrum sharing scenario in this regard is depicted in Figure 2. Small cells are deployed by operators to increase capacity and coverage and thus work under a central authority as a part of a heterogeneous network (HetNet). As discussed above, the 3550-3650 MHz band contains frequencies that are globally used by satellite systems, thus, small cell operation in satellite bands will result in interference issues for satellite systems. These issues are discussed in detail in the following section.

FSS receiving earth stations in part of the extended C-band (i.e from 3650-4200 MHz) are subject to OOBE from small cells proposed to be deployed in the adjacent 3550-3650 MHz band. These OOBE are due to the signals outside the passband and assigned bandwidth of a small cell which might fall in the pass-band of the adjacent FSS receiving earth station and affect its performance. The level of OOBE depends upon various small cells and satellite parameters including spatial and spectral separation, orientation of small cells relative to satellite main beam and an earth station’s directional antenna, and the elevation angle of a FSS receiving earth station. The effect of OOBE can be reduced by installing filters at small cells and increasing the spatial/spectral separation between small cells and FSS receiving earth stations. Thus, FSS systems immediately next to small cell band would be affected more than those that are far away in frequency. The FCC is evaluating the impact of current emission limit on FSS receiving earth station. The current OOBE limit set by the FCC is 43+10log(P) which is equivalent of -43 dBW/MHz [1]. This is also the limit set for OOBE for LTE devices by ITU and LTE specification in Category A [8], [9]. In addition, ITU and LTE define another stringent limit on OOBE in Category B that is equivalent to -60 dBW/MHz [8], [9]. C. Low Noise Amplifier (LNA) Saturation FSS receiving earth stations are designed to be very sensitive to low power received signals from the serving satellites. For this reason, FSS receiving earth stations are equipped with LNAs to amplify the received signal. Many LNAs are not equipped with filters to reject adjacent band’s unwanted signals. As a result, a strong signal in the adjacent band can over saturate the LNA device. Thus, FSS receiving earth stations can also suffer from LNA overdrive from adjacent band small cells. This problem can be overcome by installing filters between the antenna and the LNA. IV. O PPORTUNITIES AND C HALLENGES Spectrum sharing between small cells and satellite systems offer coverage and capacity enhancement for commercial cellular operators. At the same time, spectrum sharing is challenging to realize from many aspects. Spectrum sharing opportunities and challenges for small cells are addressed in this section.

1602


A. Opportunities Small cells can be deployed in satellite bands in various configurations that includes indoor, outdoor, and hybrid scenarios. In addition small cells can be subscriber-owned or operatorowned. Some of the opportunities by deploying small cell in satellite bands are discussed as follows: • Indoors – Majority of cellular traffic (voice and data) is generated indoors and traditionally carried by macro cells. However, RF signals are too weak to penetrate into buildings and this results in coverage holes. Small cells are envisioned to be first deployed indoors as not only it will result in improved coverage and user experience but will also result in maximum profit for operators. Moreover, economic cost of deployment and easy access to electrical power and backhaul are favorable for deployment of small cells in residential and enterprise areas, high-rise buildings, stadiums, and shopping malls. • Outdoors – Outdoor locations such as cell-edges, small rural towns, bus stops, and other venues that have heavy user presence serve as a motivation for outdoor deployment of small cells. They can be used to fill coverage gaps or provide extra bandwidth. Outdoor small cells can be mounted onto lamp posts, street lights, billboards etc. Outdoor deployment of small cells face challenges such as access rights to mounting locations, electric power, and availability of backhaul links. • Hybrid – Small cells can be deployed both indoors and outdoors to serve users inside and outside buildings in dense urban areas. This deployment choice comes with advantages and disadvantages associated with the two schemes discussed above. However, for spectrum sharing studies, such as small cell interference to FSS earth station all schemes need to be considered. B. Challenges The FCC has proposed that small cells operate in such a way that they do not interfere with FSS operations. This is made sure by having each small cell register with a central database and report its transmission characteristics. The database would then calculate the exclusion zones required for FSS protection and inform small cells accordingly on whether they are allowed to operate. However, this is a very challenging task due to the following reasons: • Deployment Information – Small cell devices that are used in high rise buildings or mounted on billboards or used for point-to-point links have the potential to cause interference to FSS receiving earth station on a larger scale as they’ll be highly likely to operate in-line with the beam of a satellite. Therefore, each small cell should report its exact location (co-ordinates), height, and building floor information. • Propagation Modeling – Propagation modeling for rural, sub-urban, and urban area is quite different due to the presence of crops, forest, buildings, etc. Therefore, the database should take this information into account when calculating exclusion zones.

TABLE I S MALL C ELL PARAMETERS Parameters

Values

Small Cell Maximum Transmit Power

23 dBm

Small Cell Maximum Antenna Gain

7 dBi

Maximum EIRP

30 dBm

TABLE II I NDOOR - TO -O UTDOOR C HANNEL PARAMETERS

•

•

•

•

Parameters

Notations

Values

External Wall Attenuation

Le

15 dB

Internal Wall Attenuation

Li

7 dB

No. of External Walls

we

1

No. of Internal Walls

wi

1

Path Loss Exponent

β

4

Real-time Interference Modeling – Wireless traffic is very dynamic in nature and can have busy and idle periods on a given day. Therefore, exclusion zone calculation should be done in real time. This means exclusion zones won’t be static over time. Security – Wireless links are prone to eavesdroppers and other malicious users that may falsely report their characteristics to a database [10]. Bad data would lead to inaccurate calculation of exclusion zones. Therefore, in order to protect FSS operations, security of the database and information conveyed to database must be ensured. Static vs. Mobile Operations – We have considered static indoor and outdoor small cells but did not specifically address small cells that are mobile. A mobile small cell would be continuously changing its location and thus needs to report its characteristics more often than static small cells. It would be very challenging for the database to accommodate such a user for exclusion zone calculations that is changing its characteristics too rapidly. Power Control and Beamforming – Modern wireless systems frequently use power control and beamforming to maximize its throughput and enhance user experience. Both factors are very important for exclusion zone calculation. For example, a small cell might want to use maximum transmit power for a certain user at a certain angle that is also in line-of-sight to a FSS system. This is an undesirable scenario which is harmful for FSS operations and thus the database has to play its role in avoiding this situation. Therefore, both power control and beamforming information must also be reported to the database. V. S IMULATIONS

In this section, we present simulation results for both outdoor and proposed indoor scenarios. The parameters used in

1603


Outdoor Small Cells

Indoor Small Cells

50

10 Category A (−43 dBW/MHz) Category B (−60 dBW/MHz)

45

8

Exclusion Zone (km)

Exclusion Zone (km)

40 35 30 25 20 15

7 6 5 4 3

10

2

5

1

0

Category A (−43 dBW/MHz) Category B (−60 dBW/MHz)

9

5

10

15

20

25

30

35

40

45

0

50

FSS Earth Station Antenna Off−Axis Angle (deg)

5

10

15

20

25

30

35

40

45

50


Fig. 3. Exclusion zones to protect FSS receiving earth stations from OOBE for Category A and Category B small cells that are deployed outdoors.

Fig. 4. Exclusion zones to protect FSS receiving earth stations from OOBE for Category A and Category B small cells that are deployed indoors.

the simulations are given in Tables I and II. All simulations are performed in MATLAB. We base our calculations on International Telecommunication Union (ITU) agreed interference-tonoise ratio (I/N) limits for FSS earth stations from secondary sources in co-channel and adjacent channel and co-primary sources [11], [12]. Using the ITU-defined interference-to-noise (I/N) limit, small cell parameters, and FSS earth station antenna characteristics a received power limit can be calculated that should not be exceeded.

if a small cell orientation is significantly-off relative to FSS look angle the exclusion zones are very small.

A. Out-of-band Emissions In this section we calculate exclusion zones or protection distances required to protect FSS receiving earth stations from OOBE. We consider two cases involving indoor and outdoor deployment of small cells. Example: Outdoor-Deployed Small Cells – In this example, we provide numerical results for the OOBE produced by small cells that are deployed outdoors. We compute exclusion zones that are required to protect FSS receiving earth stations from outdoor-deployed small cells. In Figure 3, we present required exclusion zones or separation distances (in Kilometers) for various elevation or off-axis angles of the FSS earth station antenna due to OOBE for the outdoor scenario. The exclusion zones are plotted for both Category A and Category B OOBE limits. Note that due to the strict OOBE criteria of Category B, exclusion zones are much smaller than the less-strict criteria of Category A. Moreover, if a small cell orientation is significantly-off relative to FSS look angle the exclusion zones are very small. Example: Indoor-Deployed Small Cells – For indoor-deployed smalls, the exclusion regions are computed in Figure 4 for both Category A and Category B OOBE limits. Note that due to the strict OOBE criteria of Category B, exclusion zones are much smaller than the less-strict criteria of Category A. Moreover,

Comparison of Indoor- vs. Outdoor-Deployment – By careful analysis of Figures 1 and 2 it is observed that the indoor scenario require less separation distances than the outdoor scenario for all elevation angles and for both categories of OOBE limits. For example, at an elevation angle of 5◦ , which is the worst-case scenario, using Category B OOBE limit, the required exclusion zone for outdoor scenario is 48.046 Km and that for the indoor scenario is 9.49 Km. Thus, close to FSS receiving earth station sites it is feasible to deploy small cells in indoor scenarios and at relatively far distances small cells can also be deployed outdoors. This configuration would give maximum benefit of the use of small cells to enhance capacity and coverage and at the same time shrink exclusion zones and OOBE to its minimum. B. Low Noise Amplifier (LNA) Saturation In the last section, we showed that indoor-deployed small cells result in minimum size of exclusion zones. This result is used as a motivation to analyze exclusion zones that are required to protect FSS systems from LNA saturation. Therefore, we consider an indoor-deployed small cell scenario for exclusion zone calculations. The necessary separation distance required to protect FSS systems from LNA saturation are presented in Figure 5 for various elevation angles using indoor scenario. Similar to the OOBE analysis, the separation distances here are also decreasing with the increasing elevation angels. Note that the exclusion zones are presented for the case when there is no filter at the LNA for adjacent channel interference rejection. If such filters are feasible to install they would result in reduced exclusion zone sizes depending upon the gain of the interference rejection filter.

1604


Indoor Small Cells 1100 1000

Exclusion Zone (m)

900 800 700

•

600 500 400 300 200

including channel propagation characteristics (for example scattering, reflection, etc.), indoor and outdoor wall losses for different buildings, and real FSS antenna pattern (measured) are often quite different from the estimated values. Thus, a test-bed can be used to evaluate such a scenario. This will give further insights and better help in understating spectrum sharing scenario. Aggregate Interference Analysis – Interference coming from a large number of indoor and outdoor deployed small cells must be analyzed at FSS earth stations to further refine exclusion zones, required to protect FSS earth stations from in-band, OOBE, and LNA saturation. ACKNOWLEDGMENT

5

10

15

20

25

30

35

40

45

This work was supported by NSTIP strategic technologies programs (no 11-ELE1854-02) in the Kingdom of Saudi Arabia.

50


Fig. 5. Exclusion zones to protect a FSS receiving earth station’s LNA from saturation from indoor-deployed small cells.

VI. C ONCLUSION Future RF spectrum is envisioned to be shared by many entities including small cells, satellites, and radar systems. These systems are heterogeneous in nature from the point of view of their spatial, spectral, and temporal characteristics. Spectrum sharing between heterogeneous systems is challenging from many aspects and interference mitigation and analysis is one of them. In this paper we explore spectrum sharing scenario between small cells and satellite systems. We look at out-of-band interference and LNA saturation issues, for fixed satellite service receiving earth stations, that result due to the operation of small cells in satellite bands. We show that when small cells are mounted outdoors i.e. on to lamp posts/billboards/towers the protection distances required to protect FSS earth stations from OOBE and LNA saturation are much larger than the case when small cells are deployed indoors. Thus, we propose to deploy small cells indoors in areas where FSS earth stations are close in geography and vice versa. Such a measure results in maximum utilization of small cells and minimum exclusion zones required for FSS protection. In addition, we highlight some of the advantages of deploying small cells in satellite bands as compared to high powered macro cells. Moreover, we address some interesting practical issues for efficient spectrum sharing between small cells and satellites. VII. F UTURE R ESEARCH D IRECTIONS In this paper we provided a discussion and some results on spectrum sharing between small cells and FSS earth stations. We envision that future research in the following areas can be very useful to realize spectrum sharing between communication systems and satellites. • Experimental Evaluation – In this paper, we have resorted to simulation-based analysis of spectrum sharing scenario. However, real world spectrum sharing scenario

R EFERENCES [1] Federal Communications Commission (FCC), “FCC proposes innovative small cell use in 3.5 GHz band.” Online: http://www.fcc.gov/document/fcc-proposes-innovative-small-celluse-35-ghz-band, December 12, 2012. [2] S. Sodagari, A. Khawar, T. C. Clancy, and R. McGwier, “A projection based approach for radar and telecommunication systems coexistence,” IEEE Global Communications Conference (GLOBECOM), pp. 5010– 5014, December 2012. [3] J.-W. Lim, H.-S. Jo, H.-G. Yoon, and J.-G. Yook, “Interference mitigation technique for the sharing between IMT-advanced and fixed satellite service,” Journal of Communications and Networks, vol. 9, pp. 159–166, June 2007. [4] S. Aijaz, “Effects of deploying IMT-Advanced systems on fixed satellite services in the 3400-3600 MHz frequency band in Pakistan,” in 2nd International Conference on Advances in Space Technologies (ICAST), pp. 1–5, Nov 2008. [5] Q. Sun and S. Nan, “Coexistence studies between LTE-Hotspot indoor and earth station of fixed satellite service in the band 3400-3600 MHz,” in IEEE 11th International Conference on Signal Processing (ICSP), vol. 3, pp. 2275–2278, Oct 2012. [6] C. Su, X. Han, X. Yan, Q. Zhang, and Z. Feng, “Coexistence analysis between IMT-Advanced system and fixed satellite service system,” in IEEE Military Communications Conference (MILCOM), pp. 1692–1697, Oct 2014. [7] K. Hosseini, J. Hoydis, S. Brink, and M. Debbah, “Massive MIMO and small cells: How to densify heterogeneous networks,” in IEEE International Communications Conference, 2013. [8] 3GPP TS 36.104 version 11.3.1 Release 11, “Lte: Evolved universal terrestrial radio access (E-UTRA): Base station radio transmission and reception,” Feb 2013. [9] Recommendation ITU-R SM.329-12, “Unwanted emissions in the spurious domain,” Sep 2012. [10] A. Khawar, Spectrum Sensing Security in Cognitive Radio networks. PhD thesis, University of Maryland at College Park, December 2010. [11] Recommendation ITU-R SF.1006, “Determination of the interference potential between earth stations of the fixed-satellite service and stations in the fixed service,” 1993. [12] Recommendation ITU-R S.1432, “Apportionment of the allowable error performance degradations to fixed-satellite service (FSS) hypothetical reference digital paths arising from time invariant interference for systems operating below 30 GHz,” 2006.

1605