Doc.: IEEE 802.11-03/xxxr0 Submission November 2004 Kiran Challapali, Philips Spectrum Agile Radios Kiran Challapali Philips [email protected].

November 2004

Kiran Challapali, Philips

doc.: IEEE 802.11-03/xxxr0

Submission

Spectrum Agile Radios

Kiran Challapali

Philips

[email protected]

November 2004



Submission

Motivation

• Rapid growth of wireless communications over the past several years – Unlicensed (WLAN/WPAN) and licensed (Cellular) bands

– Recently, wireless internet access using cellular networks• available in 14 US cities, $80/month, 60-80 kpbs*

– Vast and growing demand for spectrum-based communication links

• In some measurements, only about 10% of the allocated spectrum is in use at any given time

• Spectrum access and efficiency becoming a critical public policy issue

November 2004



Submission

Outline of talk1. Introduction

– Goal of the talk

– What are Spectrum Agile/Cognitive Radios?2. Applications of agile radios

– Wireless internet and others

3. FCC policy modernization

– Spectrum usage models

– Actual spectrum usage (measurements)

4. Spectrum agile radios

– Dynamic spectrum management concepts and key considerations

5. Standardization

6. Results

7. Discussions and Summary

November 2004



Submission

What are spectrum agile radios?• Cognitive radio

– Cognition (Merriam-Webster)• to become acquainted with, to know, to come to know

• the act or process of knowing including awareness and judgment

– Cognitive radios term originally coined by Joseph Mitola III• Includes learning and reasoning

• Spectrum agile radios– Radios that are:

• Aware of their environment

• Adapt transmission characteristics (based on environment)

– Our definition close to FCC’s definition

1. Introduction

November 2004



Submission

2. Applications

November 2004



Submission

2. Applications

Source: BusinessWeek online

November 2004



Submission

Remote patient monitoring

• Motivation

– Healthcare costs are staggering

– Aging population, shortage of trained staff in hospitals

• Advantages of remote monitoring

– Patients spend less time in hospitals, reducing costs

– Quality of care improved due to continuous monitoring

– Staying at home improves quality of life

2. Applications

November 2004



Submission

Public safety

• Communications failure during 9-11

• All cellular communications were down for several hours

– Cellular networks did not have enough capacity when needed

• Public safety communications also failed

– Fire and police could not communicate

• Public safety under State control (no interoperability)

2. Applications

November 2004



Submission


November 2004



Submission

Source: FCC Website• Spectrum reform becoming a global issue


November 2004



Submission

FCC’s Spectrum policy task force

• Findings– Spectrum access versus

scarcity

– New methods as a solution to access

– Interference tolerance

– Need to define rights and responsibilities

• Recommendations– Modernizing the regulatory

model

– Increase access to spectrum

– New interference management tech.

– Legislative recommendations

• Setup by Chairman Powell, headed by Paul Kolozdy

• Submitted findings and recommendations Dec. 2002


November 2004



Submission

• Spectrum usage in NYC during RNC ~ Sept 1st 2004.– Measurements by Share Spectrum Company

• Result: 16 % duty cycle, 30 MHz - 3 GHz, over 24 hours – Actually, even lower (< 10%) occupancies

Observation

Frequency in MHz

500

1

0 1000 1500 2000 2500 30000

Duty

Cyc

le

TV CellTVFM UnlicensedPCS

Source: Shared Spectrum Company


November 2004



Submission

Spectrum usage models

Source: Paul Kolodzy @ ISART 2003


November 2004



Submission

FCC NPRM on Cognitive Radio• Issued on December 30 2003• Scope

– Facilitating Opportunities for Flexible, Efficient, and Reliable Spectrum Use Employing Cognitive Radio Technologies

– Authorization and Use of Software Defined Radios

• A simpler definition of CR

• CR use in four scenarios discussed– Licensee can make more efficient use of their own spectrum– Secondary markets: Based on agreements between licensees and third

parties– Co-primaries: Automated frequency coordination– Unlicensed operation: Opportunistic usage (no explicit agreements)


November 2004



Submission

FCC NPRM on Unlicensed operation in TV Bands

• Issued on May 13, 2004• Scope

– Unlicensed operations in the TV Broadcast Bands

• New unlicensed devices will not cause interference to– Analog and Digital TV Broadcast stations– LPTV, TV Translator, TV Booster Stations, Auxiliary Operations,

and Wireless Microphones– Land Mobile Radio Services (PLMRS and CMRS)

• Operation in – Channels 5 to 13 and 21 to 51 (except ch. 37) (76 to 698 MHz)– In addition, Channels 14 to 20 permitted in rural areas


November 2004



Submission

FCC NPRM on Unlicensed operation in TV Bands (Contd.)

• Two types of devices permitted

• Personal/portable devices– Peak power 100mW

– Devices must receive a control signal indicating which channels are vacant

• Fixed access devices– Peak power 1W

– Devices must either

• Include GPS and means to know which channels are vacant

• Professionally installed to operate in vacant channels


November 2004



Submission

4. Spectrum agile radios

November 2004



Submission

Key considerations• How to harness spectrum “white spaces”?

– Are reliable wireless services without licensed frequency allocation possible?

– Can devices automatically find “white spaces”?

• How can they avoid harmful interference to licensees?

• What if receivers are silent?

– How can devices share spectrum efficiently?

• How to avoid many competing (interfering) devices in the same band?

– What are the hardware/software design challenges?

– How to ensure compliance?

• FCC currently does not monitor emissions

• Solution: Cognitive/Agile Radio technologies

– Several approaches to dynamic spectrum sharing

4. Agile radios

November 2004



Submission

Research directions• Several technical communities answering key questions

• Joseph Mitola III

• DARPA XG program– For defense needs, architecture based on Policy language

• NSF ProWIN NeTS program– Significant investment on test-beds

• Wireless Industry, Standards– IEEE 802.22 WG

– IEEE 802.16h SG

4. Agile radios

November 2004



Submission

Definition of spectrum agile radios

• Definition– A new paradigm for wireless communications.

– The physical and medium access layers of spectrum agile radios adapt their transmission characteristics to the external radio environment, while retaining the flexibility to react to evolving FCC policies.

• Our version of Cognitive Radio, and,

Aware/Adaptive

Learning/Reasoning

4. Agile radios

November 2004



Submission

spectrum agile radio

=

flexible re-configurable radio

+

smart protocols and algorithms

Definition of spectrum agile radios

(“quickly adapts transmission characteristics” )

(“aware of spectrum usage in vicinity,

makes intelligent decisions on that basis, and

reacts to evolving FCC policies”)

4. Agile radios

November 2004



Submission

Flexible PHY layer

Opportunity manager

Opp

ort

un

ity

iden

tifier

Policy interaction

Spectrum agile radio: Key components

4. Agile radios

November 2004



Submission

Spectrum agile radio: Key components

TXpower

frequency[MHz]

time

Multicarrier - Spectrum AgileRadio for Broadband Applications

6 MHz TV channel

spectrum opportunities(unused TV channels)

adaptive sub-carrier bandwidth

(usage afteridentification)

adaptiveTX power

multicarrier transmissionsin unused TV-channels

4. Agile radios

November 2004



Submission

Dynamic spectrum sharing concepts• Primary (vertical) sharing

– Finding and using spectrum white space

• Secondary (horizontal) sharing – Dissimilar networks then sharing the spectrum efficiently

4. Agile radios

November 2004



Submission

Primary sharing approaches• Self-sensing (or real-time measurements)

– Are measurements reliable

– Does diversity solve the problem

• GPS based location + Retrieval from database– Rely on up-to-date databases (difficult) and Machine readability

– Lacks adaptability

• Over the air control channel– Cost (infrastructure and receiver)

• Professional installation– Lacks adaptability

4. Agile radios

November 2004



Submission

Secondary sharing approaches• Capacity suffers if no co-ordination at all

• Some approaches– Rules based sharing, e.g. Etiquette (Stefan Mangold)

• WiFi Co-existence Task Group

• Voluntary set of rules, e.g. LBT, max air time, etc.

– Explicit coordination

• Over the air (separate co-ordination channel)

– Rutgers proposal

• Via internet (could be broker assisted)

• Regional aggregation (statistical multiplexing)

– Stevens proposal

4. Agile radios

November 2004



Submission

5. IEEE Standards

November 2004



Submission

IEEE Standardization• IEEE 802.22

– A new Working Group formed in IEEE – Regional Area Network TV Band Specification

• includes mechanisms to protect incumbent licensees from harmful interference.

• Specifically,• SCOPE: This standard specifies the air interface, including the medium access

control layer (MAC) and physical layer (PHY), of fixed point-to-multipoint regional area networks operating in the VHF/UHF TV broadcast bands between 54 MHz and 862 MHz.

• PURPOSE: This standard is intended to enable deployment of interoperable 802 multivendor regional area network products, to facilitate competition in broadband access by providing alternatives to wireline broadband access and extending the deployability of such systems into diverse geographic areas, including sparsely populated rural areas, while preventing harmful interference to incumbent licensed services in the TV broadcast bands.

November 2004



Submission

IEEE Standardization

• IEEE 802.16 – A new Task Group h proposed– PAR and 5 Criteria drafted– Will likely be voted on in November (this) meeting

• SCOPE (DRAFT): This amendment specifies improved mechanisms, as policies and medium access control enhancements, to enable coexistence among license-exempt systems based on IEEE Standard 802.16 and to facilitate the coexistence of such systems with primary users.

November 2004



Submission

6. Results

November 2004



Submission

802.11k medium sensing time histogram• With 802.11k, stations can report new measurement results

• After sensing, results are reported with standardized frames

• Histogram helps identifying opportunities

21 89 62 31 03 02

Reported densities are indicators[0..255] for probability p of occurrence:

Density:

1 65432Bin:

Density(bin)p(bin) , bin=1...6255 SIFS DIFS

PIFS aSlotTime

density

time

CCA idle time histogramBin Interval = aSlotTimeBin Offset = SIFSNumber of Bins = 6

1 654

3

2

bin

p(bin) 1

Density Vector(m=var.)

21 89 62 31 03 02

Reported densities are indicators[0..255] for probability p of occurrence:

Density:

1 65432Bin:

Density(bin)p(bin) , bin=1...6255 SIFS DIFS

PIFS aSlotTime

density

time

CCA idle time histogramBin Interval = aSlotTimeBin Offset = SIFSNumber of Bins = 6

1 654

3

2

bin

p(bin) 1

Density Vector(m=var.)

6. Results

Author: Stefan Mangold/Zhun Zhong

November 2004



Submission

Examples …heavily used no opportunity

unlicensed (4

channels)

licensed spectrum, used with

deterministic pattern

opportunity

6. Results


November 2004



Submission

802.11k MSTH measurement results• Simulation of high channel load (many stations)

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.50

0.1

0.2

0.3

prob

(id

le d

ur.)

idle duration [ms]

CCAidle

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 10

0.01

0.02

0.03

0.04

0.05

prob

(bu

sy d

ur.)

busy duration [ms]

CCAbusy

typical geom. distribution high channel load,CSMA medium access`

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.50

0.1

0.2

0.3

prob

(id

le d

ur.)

idle duration [ms]

CCAidle

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 10

0.01

0.02

0.03

0.04

0.05

prob

(bu

sy d

ur.)

busy duration [ms]

CCAbusy

typical geom. distribution high channel load,CSMA medium access`

6. Results


November 2004



Submission

Spectrum agile radio: cyclic spectrum for analog TV

2/

2/

2)2

()2

(1

limˆ T

T

ti

Tx dtetxtx

TR

deRfS fixx

2)(ˆ)(ˆ

6. Results

November 2004



Submission

Spectrum agile radio: Cyclic correlation for digital TV

6. Results

November 2004



Submission

November 2004



Submission

Further reading …[1] MANGOLD, S. AND CHALLAPALI, K. (2003) Coexistence of Wireless Networks in Unlicensed Frequency

Bands. Wireless World Research Forum #9 Zurich Switzerland July 2003.

[2] CHALLAPALI, K. AND MANGOLD, S. AND ZHONG, Z. (2004) Spectrum Agile Radio: Detecting Spectrum Opportunities. International Symposium on Advanced Radio Technologies 2004 Boulder Colorado USA, Mar 2004.

[3] CHALLAPALI, K. AND BIRRU, D. AND MANGOLD, S. (2004) Spectrum Agile Radio for Broadband Applications. EETimes In Focus Article Aug 23.

[4] MANGOLD, S. AND ZHONG, Z. AND CHALLAPALI, K. (2004) Spectrum Agile Radio: Radio Resource Measurements for Opportunistic Spectrum Usage. IEEE Globecom 2004 Dallas TX, USA, Nov 2004.

[5] MANGOLD, S. AND ZHONG, Z. AND HIERTZ, G. AND WALKE, B. (2004) IEEE 802.11e/802.11k Wireless LAN - Spectrum Awareness for Distributed Resource Sharing. Special Issue on Emerging WLAN Technologies and Applications. Wireless Communications and Mobile Computing. New York USA: John Wiley & Sons.

[6] CHOU C.T., SAI SHANKAR N, KIM, H., and SHIN, K.G., “What and How Much to Gain from Spectrum Agility?”, Submitted to IEEE/ACM Trans. On Networking. June 2004

[7] SAI SHANKAR N, CHOU, C.T., CHALLAPALI, K., and MANGOLD, S., “Spectrum Agile Radio: Capacity and QoS Implications of Dynamic Spectrum Management”, Submitted to IEEE ICC 2005, Seoul, South Korea

[8] XING, Y., CHANDRAMOULI, R., SAI SHANKAR N., and MANGOLD, S., “Analysis and Performance of a Fair Channel Access Protocol for Open Spectrum Wireless Networks”, Submitted to IEEE ICC 2005, Seoul, South Korea

[9] MANGOLD, S. AND SAI SHANKAR N., S. AND BERLEMANN, L. (2005) Spectrum Agile Radio: A Society of Machines with Value-Orientation. Submission to IEEE European Wireless 2005.

6. Results

November 2004



Submission

7. Summary

• A new paradigm for wireless communications

– Harness unused spectrum, thereby,

– Enable many new applications

• Substantial interest in many technical

communities

Doc.: IEEE 802.11-03/xxxr0 Submission November 2004 Kiran Challapali, Philips Spectrum Agile Radios Kiran Challapali Philips [email protected].

Documents