Challenges of spectrum sensing in cognitive radios. Public CWC & VTT GIGA Seminar 08 4th December 2008

Transcription

1 Challenges of spectrum sensing in cognitive radios Marja Matinmikko Public CWC & VTT GIGA Seminar 08 4th December 2008

2 Outline Introduction Current spectrum use Challenges Performance metrics Interference considerations Time aspects Example Conclusions References 2

3 Introduction Cognitive radio has the capability to sense and be aware of its operational environment and adjust its radio operating parameters accordingly Cognitive radio adds intelligence on top of software defined radio (SDR) in which RF operating parameters, e.g. frequency range, modulation type, or output power, can be set or altered by software Cognitive radio has the potential to improve the efficiency in resource use (spectrum) by opportunistically exploiting available resources Cognitive radio is to be understood as a technique applicable to a variety of systems 3

4 Current spectrum use Radio regulations of the ITU have internationally governed the use of radio spectrum since 1906 Static spectrum allocations for different services (e.g. mobile, fixed, broadcast) Major goal: Protection from interference Two aspects of efficiency in spectrum use Spectrum occupancy: What is the utilization rate of the spectrum over time (%)? Spectral efficiency: How efficiently a system uses its spectrum (bit/s/hz)? Measurements on current spectrum use indicate that some parts of spectrum have low occupancy in some areas Average spectrum occupancy measured only 5.5% with maximum 13.1% 1% in urban and minimum 1% in rural area in US [SSC05] Cellular mobile communication systems are very efficient in spectrum use (both spectrum occupancy and spectral efficiency) ITU = International Telecommunication Union 4

5 Current spectrum use A paradigm shift from static spectrum allocations to more dynamic and liberal use is envisioned Spectrum requirement planning is usually made for worst case scenario but spectrum availability is a problem E.g. IMT systems* at WRC-07 Cognitive radio is a potential technique to improve spectrum occupancy WRC-11 Agenda item 1.19 to consider regulatory measures and their relevance, in order to enable the introduction of software-defined radio and cognitive radio systems, based on the results of ITU-R studies, in accordance with Resolution 956 [COM6/18] (WRC-07) WRC = World Radiocommunication Conference *M. Matinmikko, J. Huschke, T. Irnich, N. Matoba, J. Ojala, P. Ojanen, H. Takagi, B. Walke & H. Yoshino. Spectrum Requirement Calculation for IMT-Advanced. In: H. Takagi & B. Walke (ed.) Spectrum Requirement Planning in Wireless Communications: Model and Methodology for IMT-Advanced. d John Wiley & Sons, 2008, pp

6 Challenges Cognitive radio is not allowed to cause harmful interference to the primary users of the spectrum Identification of primary users is critical Challenges for cognitive radio High probability of detection of primary users Uncertainty due to radio wave propagation Own performance (no protection from interference guaranteed) Time (e.g. sensing time, data transmission time) Range (e.g. sensing range, interference range, communication range) Transmission Control data 6

7 Performance metrics Receiver operating characteristics (ROC) Probability of detection how well cognitive radio notices the presence of primary users? Probability of false alarm how well cognitive radio can exploit spectrum opportunities? Time Time between failures in detection how often primary user tolerates interruption? Sensing time how much time is used for monitoring the spectrum use vs. actual data transmission? Sensing rate how often spectrum must be sensed? Transmission Traditional measures: throughput, delay, reliability and precedence [ETSI01] performance of cognitive radio is not guaranteed! Suitability of spectrum opportunities how good are the spectrum opportunities? 7

8 No talk zone, no SUs allowed when PU transmitting due to interference from SU Interference considerations Interference range (PU) PU tx PU rx Sensing range SU rx Communication range of PU SU tx Interference range (SU) Communication range of SU Requirement for sensing: Sensing range Interference range + Communication range of PU 8

9 Interference considerations Interference is caused at the primary user receivers while spectrum sensing can identify only primary user transmitters Some systems use the spectrum passively without transmitting at all (e.g. radio astronomy service) Current systems s (e.g. TV receivers) e are not designed ed to tolerate opportunistic spectrum use on their band or adjacent bands Some services (e.g. satellite) are critical and must be protected from interference at all times in all locations of the Earth Cognitive radio has to obey the rules of the specific band 9

10 Time aspects 10

11 Example of performance metrics Joint pro obability of detection Q d N FFT = 512, n CR = 1, M = 1 N = 512, n = 2, M = 1 FFT CR N = 512, n = 3, M = 1 FFT CR N FFT = 512, n CR = 1, M = 8 N FFT = 512, n CR = 2, M = 8 N = 512, n = 3, M = 8 FFT CR Joint probability of false alarm Q f Theoretical ROC for Welch s periodogram, see [Sarvanko08]. 11

12 Conclusions Cognitive radio offers potential to increase spectrum occupancy by identifying and exploiting spectrum opportunities Future systems will be intelligent in terms of spectrum use but existing systems cannot be changed Introduction of cognitive radio capabilities in future wireless networks should have no impact on the existing networks if they are to be deployed on the same bands Characteristics and restrictions of existing systems must be taken into account in the cognitive radio development Technical challenges vs. regulatory constraints Now is the time! Cooperation in all forms will become crucial 12

13 References [ETSI01] Digital cellular telecommunications system (Phase 2+) (GSM); General Packet Radio Service (GPRS); Service description; Stage 2, 3GPP TS version 7.5.0, Release 1998, ETSI, January [Matinmikko08a] M. Matinmikko, M. Höyhtyä, M. Mustonen, H. Sarvanko, A. Hekkala, M. Katz, A. Mämmelä, M. Kiviranta, & A. Kautio. Cognitive radio: an intelligent wireless communication system, VTT, Oulu, 154 p, VTT-R Available: / /ju uut/ SS_ Research _ Report.pdf pd [Matinmikko08b] M. Matinmikko, J. Huschke, T. Irnich, N. Matoba, J. Ojala, P. Ojanen, H. Takagi, B. Walke & H. Yoshino. Spectrum Requirement Calculation for IMT- Advanced. In: H. Takagi & B. Walke (ed.) Spectrum Requirement Planning in Wireless Communications: Model and Methodology for IMT-Advanced. John Wiley & Sons, 2008, pp [Sarvanko08] H. Sarvanko, M. Mustonen, A. Hekkala, A. Mämmelä, M. Matinmikko, and M. Katz, Performance evaluation of spectrum sensing using Welch s periodogram in cognitive radios, In Proc: IEEE CogART 08, 14th February 2008, Aalborg, Denmark. [SSC05] Spectrum occupancy measurements, Shared Spectrum Company (SSC), Available: 13