Top Banner
Cognitive Radio in a Frequency- Planned Environment: Can it Work? Erik Axell Erik G. Larsson and Mikael Skoglund, IEEE GLOBECOM, 2007
12

Cognitive Radio in a Frequency-Planned Environment: Can it Work? Erik Axell Erik G. Larsson and Mikael Skoglund, IEEE GLOBECOM, 2007.

Mar 31, 2015

Download

Documents

Hailey Suttle
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Cognitive Radio in a Frequency-Planned Environment: Can it Work? Erik Axell Erik G. Larsson and Mikael Skoglund, IEEE GLOBECOM, 2007.

Cognitive Radio in a Frequency-Planned Environment: Can it Work?

Erik Axell

Erik G. Larsson and Mikael Skoglund, IEEE GLOBECOM, 2007

Page 2: Cognitive Radio in a Frequency-Planned Environment: Can it Work? Erik Axell Erik G. Larsson and Mikael Skoglund, IEEE GLOBECOM, 2007.

SENDORA – SEnsor Network for Dynamic and cOgnitive Radio Access

Linköping University THALES Communications Institut Eurocom Kungliga Tekniska Högskolan Helsinki University of Technology Norwegian University of Science and Technology Telenor ASA Universitat de València Università degli Studi di Roma

Page 3: Cognitive Radio in a Frequency-Planned Environment: Can it Work? Erik Axell Erik G. Larsson and Mikael Skoglund, IEEE GLOBECOM, 2007.

The Scenario

Page 4: Cognitive Radio in a Frequency-Planned Environment: Can it Work? Erik Axell Erik G. Larsson and Mikael Skoglund, IEEE GLOBECOM, 2007.

Some Fundamental Questions

Cognitive radios must transmit at very low power. How low must it be?

Cognitive radios must expect very low C/(I+N) from primary system. What orders of magnitude?

At an acceptable loss in primary system coverage, how large is the “area of cognitive operation”?

Page 5: Cognitive Radio in a Frequency-Planned Environment: Can it Work? Erik Axell Erik G. Larsson and Mikael Skoglund, IEEE GLOBECOM, 2007.

Basic Assumptions and Some Terminology Path loss and lognormal fading

n-reuse frequency planning for primary system

SINR for primary system, with primary base station power P0

-SINR is the SINR achieved with probability 1-

Page 6: Cognitive Radio in a Frequency-Planned Environment: Can it Work? Erik Axell Erik G. Larsson and Mikael Skoglund, IEEE GLOBECOM, 2007.

Primary System Operating Point

Primary system is either noise limited, interference limited, or in-between.

Quantify this in terms of ratio between noise and co-channel interference at the primary cell border:

- [dB]: purely noise limited

[dB]: purely interference limited

Page 7: Cognitive Radio in a Frequency-Planned Environment: Can it Work? Erik Axell Erik G. Larsson and Mikael Skoglund, IEEE GLOBECOM, 2007.

With Cognitive Users SINR is now

Suppose we can accept reduction in primary cell radius from r to r’. Then assume M cognitive users, each with power P, uniformly distributed in circular ring.

How large can the circular ring, the ”area of cognitive operation,” be?

Page 8: Cognitive Radio in a Frequency-Planned Environment: Can it Work? Erik Axell Erik G. Larsson and Mikael Skoglund, IEEE GLOBECOM, 2007.

Feasible Operating Regions, with 7-reuse

,, CP NI

C

NI

C

Page 9: Cognitive Radio in a Frequency-Planned Environment: Can it Work? Erik Axell Erik G. Larsson and Mikael Skoglund, IEEE GLOBECOM, 2007.

Feasible Operating Regions, with 21-reuse

Page 10: Cognitive Radio in a Frequency-Planned Environment: Can it Work? Erik Axell Erik G. Larsson and Mikael Skoglund, IEEE GLOBECOM, 2007.

“Forbidden Region,” 12-reuse

Page 11: Cognitive Radio in a Frequency-Planned Environment: Can it Work? Erik Axell Erik G. Larsson and Mikael Skoglund, IEEE GLOBECOM, 2007.

Asymptotic Sensitivity Requirements

The additional (compared to primary receivers) sensitivity requirement is

and satisfies

Page 12: Cognitive Radio in a Frequency-Planned Environment: Can it Work? Erik Axell Erik G. Larsson and Mikael Skoglund, IEEE GLOBECOM, 2007.

Main Conclusions

If cognitive users are to be introduced, they must Be few in numbers, since the aggregate power scales with the

number of devices Transmit with extremely low power, such as -30 dB below the

primary system Have very sensitive radios, in the order of 20-30 dB more sensitive

than the primary system radios.