Spectrum Management and

Cognitive Radios Alessandro Guidotti, XXIV ciclo

Bologna, 24-25/01/2012

DEIS – Fondazione Ugo Bordoni

Is spectrum lacking?

• Command & Control spectrum allocation model

– Static spectrum allocation

• Ever increasing demand for new wireless services

– Broadband access to anyone, anywhere, at anytime

Commonly shared feeling of spectrum scarcity

Great amount of

underused resources

Is spectrum lacking?

• Two main approaches to this issue

– Flexible spectrum access

• Cognitive Radios

– Release of spectrum bands thanks to more

efficient wireless technologies

• Digital swtich-on process (DVB-T)

My Ph. D. activity Spectrum Management

&

Cognitive Radios

Cognitive

Radios

Spectrum

Release

Mutual interference between DVB-T

and LTEAmount of spectrum potentially available

as White Space

Geolocation databases

Characterization of the Hidden Node Margin

Cooperative energy detection algorithms

Coexistence between CRs and TETRA

Cognitive Satellite Terrestrial Radios

Stochastic Geometry for network interference

characterization

CEPT SE43

CEPT

CPG PT-D

SUPELEC

FP7 Project

Introduction of broadband services in

the 800 MHz band

Network interference

• Modeling of network interference is a key issue in

next-generation systems

– Interference mitigation, control of e.m. emissions, etc.

– CRs: network performance, harmful interference control

• Three main factors

1. Spatial distribution of the nodes

2. Transmission characteristics

3. Propagation characteristics

Spatial distribution

• Deterministic model

– Known locations

– Locations constrained to fixed structures (e.g.,

hexagonal grid)

• Cognitive Radio networks

– Inherent unplanned, irregular and random distribution

– No deterministic approach

– Extensive and complex system-level simulations

• Advanced mathematical tools for statistical

modeling

– Spatial average performance metrics

Stochastic Geometry

• Network: snapshot of a stationary random model in

the Euclidean space

• Nodes locations: realization of a Point Process

• d-dimensional Euclidean space

• set S of all the possible sequences of points in it,

s.t. any Φ in S is finite and simple

PP: r.v. whose values are the point

patterns Φ taken from S

-10 -8 -6 -4 -2 0 2 4 6 8 10

-10

-8

-6

-4

-2

0

2

4

6

8

10

X [m]

Y [

m]

Poisson Point Process

• The homogeneous PPP is the most used model

• Spatial invariance properties

– Superposition, thinning, mapping

– Slivnyak’s Theorem: removing a point does not change the

distribution of the other points

Number of nodes: Poisson r.v.

Locations: uniformly distributed

Interference as PPP

• Exponential power-decaying path-loss model, α>2

• Objective: closed-form of the MGF in b(o,R1,R2)

– Coverage probability, average downlink rate, ASEP

• Previous works: closed-form in b(o,R) with

Rayleigh fading only, no modulation

Moment Generating Function

• Procedure

– Condition on having k nodes

• The single interference terms are i.i.d.

– Solve the expectations on modulation

and node activity

– Decondition on having k nodes

– Restrict integration bounds to b(o,R1,R2)

Moment Generating Function

• Generalized modulation coefficients, nodes

activity and network layout

Moment Generating Function

• Nakagami-m fading channels Rayleigh fading

• Series expansion of Γ(a,x) and Hypergeometric functions:

Moment Generating Function

• Lower values for decreasing α

• Lower values for increasing p, λ

α=4 α=5 Nakagami-m, m0=2

Rayleigh

• Slight dependence on the

channel model

Average downlink rate

• BSs distributed according to homogeneous PPP

in Rd with density λ

• The intended node (y≡o) communicates with the

closest BS, all the other BSs act as interferers

• Previous works: three-fold integral expression

with Rayleigh fading and α=4

Average downlink rate

• Proposed model: MGF-based approach

Rate: function of the MGF of the aggregate

interference and the intended link fading

Average downlink rate

Meijer-G function properties

• Single-integral with Meijer-G function

• Generalized fading channels

• Different channel models on intended and interfering

links

Average downlink rate

• Nakagami-m fading

• Lognormal fading

Hermite polynomials

Average downlink rate

• The rate significantly

improves for higher α

α=4 α=5 Nakagami-m vs Rayleigh, m0=2

Rayleigh

• Slight improvements for

increasing m0 (Rayleigh)

Results

• Coverage

– Single-integral, Rayleigh, α=4 closed-form,

generalized fading, α>2

• Average downlink rate

– Three-fold integral, Rayleigh, α=4 single-integral,

generalized fading, α>2

• Average Symbol Error Probability

– Semi-analytical model single-integral

– Synchronous case upper bounds the asynchronous

• Many parameters of interest in closed-form

Future works

• MIMO systems

– Statistical characterization of uncorrelated paths

• Cooperative sensing

– Theoretical expression with generalized fading

• Relay networks

– Effect of multi-hop communications

• Cognitive Satellite networks

– Aggregate interference at Ground Stations

International activities

• CEPT CPG PT-D – WRC-12 agenda items 1.5, 1.7, 1.8, 1.17, 1.18, 1.20, and 1.25

• CEPT SE43 – Technical and operational requirements for the operation of

cognitive radio systems in the „white spaces‟ of the frequency

band 470-790 MHz

• CoRaSat (FP7)

– Cognitive Radio for Satellite Communications

• Cost Action IC0902 – Cognitive Radio and Networking for Cooperative Coexistence of

Heterogeneous Wireless Networks

Personal contributions • A. Guidotti, M. Di Renzo, G. E. Corazza, and F. Santucci, “Simplified Expression of Coverage Probability and

Average Rate of Cellular Networks Using Stochastic Geometry,” IEEE Transactions on Communications (to be

submitted), 2012

• A. Guidotti, M. Di Renzo, G. E. Corazza, and F. Santucci, “Simplified Expression of the Average Rate of Cellular

Networks Using Stochastic Geometry,” IEEE International Conference on Communication 2012 (ICC 2012), June

2012

• C. Merola, A. Guidotti, M. Di Renzo, F. Santucci, and G. E. Corazza, “Average Symbol Error Probability in the

Presence of Network Interference and Noise,” IEEE International Conference on Communication 2012 (ICC 2012),

June 2012

• A. Guidotti, D. Guiducci, M. Barbiroli, C. Carciofi, P. Grazioso, and G. Riva, “Coexistence and Mutual Interference

between Mobile and Broadcasting Systems,” IEEE Vehicular Technology Conference - Spring 2011 (VTC Spring

2011), May 2011

• A. Guidotti, P. Grazioso, M. Barbiroli, C. Carciofi, D. Guiducci, and G. Riva, “Analysis of coexistence and mutual

interference between mobile and digital television systems,” Systemas y Telematica (S&T), October 2010

• M. Barbiroli, A. Guidotti, P. Grazioso, and G. Riva, “Coexistence and mutual interference between mobile radio and

broadcast systems – Preliminary simulation results,” COST 2100 TD(09)709, February 2009

• “WRC-12 AI 1.17 - Broadcasting versus mobile: theoretical analysis of mutual interference,” Tech. Rep.

CPGPTD(10)109, Italian Contribution to CPG PT-D - Agenda Item 1.17, Aug. 2010

• “WRC-12 AI 1.17 - Options for method A1,” Tech. Rep. CPGPTD(10)156, Italian Contribution to CPG PT-D -

Agenda Item 1.17, Aug. 2010

• “WRC-12 AI 1.17 - Options for method A1,” Tech. Rep. CPGPTD(11)33, Italian Contribution to CPG PT-D -

Agenda Item 1.17, Aug. 2011

• M. Barbiroli, C. Carciofi, D. Guiducci, A. Guidotti, E. Tarantino, and G. Riva, “Advantages and disadvantages of the

introduction of IMT systems in the 800 MHz band,” International Conference on Electromagnetics in Advanced

Applications 2011 (ICEAA 2011), September 2010

Personal contributions • “Availability of spectrum for white space devices in the band 470-790 MHz,” Tech. Rep. SE43(10)22, Italian

Contribution to CEPT SE43, Jan. 2010

• “Calculation of the Hidden Node Margin in a real Italian scenario,” Tech. Rep. SE43(10)123, Italian Contribution to

CEPT SE43, Aug. 2010

• M. Barbiroli, C. Carciofi, A. Guidotti, and D. Guiducci, “Evaluation and Analysis of the Hidden Node Margin for

Cognitive Radio System Operation in a Real Scenario,” European Conference on Antennas and Propagation 2011

(EuCAP 2011), April 2011

• S. Kandeepan, L. De Nardis, M.-G. Di Benedetto, A. Guidotti, and G. E. Corazza, “Cognitive Satellite Terrestrial

Radios,” IEEE Global Communications Conference (GLOBECOM 2010), December 2010

• “Enforcement issues related to the geolocation database approach,” Tech. Rep. SE43(10)96, Italian Contribution to

CEPT SE43, Jun. 2010

• “Translation of the information provided to the geolocation database into elements of authorisation to the WSD,”

Tech. Rep. SE43(10)124, Italian Contribution to CEPT SE43, Aug. 2010

• “Proposal of an application methodology to combine geo-location database and sensing,” Tech. Rep. SE43(11)38,

Italian Contribution to CEPT SE43, Jul. 2011

• “Performance analysis of cooperative energy detection sensing,” Tech. Rep. SE43(11)37, Italian Contribution to

CEPT SE43, Jul. 2011

• “Feasibility of autonomous operation of WSDs using sensing,” Tech. Rep. SE43(11)62, Italian Contribution to

CEPT SE43, Sep. 2011

• “Consideration of coexistence between WSD and TETRA,” Tech. Rep. SE43(11)39, Italian Contribution to CEPT

SE43, Jul. 2011

• R. Suffritti, G. E. Corazza, A. Guidotti, V. Petrini, D. Tarchi, A. Vanelli-Coralli, and M. Di Renzo, “Cognitive hybrid

satellite-terrestrial systems,” International Conference on Cognitive Radio and Advanced Spectrum Management

2011 (CogART 2011), October 2011

Short courses

• Long Term Evolution e 4G

• Compatibilità Elettromagnetica LS

• Nuove frontiere nella gestione dello spettro radio

• SatNEx Summer School

• Newcom++ Spring School

– Cognitive Wireless Communication Networks

• COST2100/CONET/Newcom++ Training School

– Cooperating Objects and Wireless Sensor Networks

Thank you!