1、 Recommendation ITU-R SM.1880-2 (09/2017) Spectrum occupancy measurements and evaluation SM Series Spectrum management ii Rec. ITU-R SM.1880-2 Foreword The role of the Radiocommunication Sector is to ensure the rational, equitable, efficient and economical use of the radio-frequency spectrum by all
2、radiocommunication services, including satellite services, and carry out studies without limit of frequency range on the basis of which Recommendations are adopted. The regulatory and policy functions of the Radiocommunication Sector are performed by World and Regional Radiocommunication Conferences
3、 and Radiocommunication Assemblies supported by Study Groups. Policy on Intellectual Property Right (IPR) ITU-R policy on IPR is described in the Common Patent Policy for ITU-T/ITU-R/ISO/IEC referenced in Annex 1 of Resolution ITU-R 1. Forms to be used for the submission of patent statements and lic
4、ensing declarations by patent holders are available from http:/www.itu.int/ITU-R/go/patents/en where the Guidelines for Implementation of the Common Patent Policy for ITU-T/ITU-R/ISO/IEC and the ITU-R patent information database can also be found. Series of ITU-R Recommendations (Also available onli
5、ne at http:/www.itu.int/publ/R-REC/en) Series Title BO Satellite delivery BR Recording for production, archival and play-out; film for television BS Broadcasting service (sound) BT Broadcasting service (television) F Fixed service M Mobile, radiodetermination, amateur and related satellite services
6、P Radiowave propagation RA Radio astronomy RS Remote sensing systems S Fixed-satellite service SA Space applications and meteorology SF Frequency sharing and coordination between fixed-satellite and fixed service systems SM Spectrum management SNG Satellite news gathering TF Time signals and frequen
7、cy standards emissions V Vocabulary and related subjects Note: This ITU-R Recommendation was approved in English under the procedure detailed in Resolution ITU-R 1. Electronic Publication Geneva, 2017 ITU 2017 All rights reserved. No part of this publication may be reproduced, by any means whatsoeve
8、r, without written permission of ITU. Rec. ITU-R SM.1880-2 1 RECOMMENDATION ITU-R SM.1880-2 Spectrum occupancy measurement and evaluation (2011-2015-2017) Scope Although automatic occupancy measurement will not completely replace manual observations, it is still well suited for most cases. Frequency
9、 channel occupancy, as well as frequency band occupancy, should have a certain level of accuracy in order to be compared or merged if necessary. By using the technique and proper method a more efficient use of existing equipment is possible. Keywords Spectrum occupancy measurements, frequency channe
10、l occupancy, revisit time, busy hour The ITU Radiocommunication Assembly, considering a) that the increasing demand of radiocommunication services requires the most efficient use of the radio-frequency spectrum; b) that good spectrum management can only satisfactorily proceed if the spectrum manager
11、s are adequately informed on the current usage of the spectrum and the trends in its demand; c) that results of spectrum occupancy measurements would provide important inputs into: frequency allotments and assignments; verification of complaints concerning channel blocking; establishment of the degr
12、ee of efficiency of spectrum usage; d) that information obtained from frequency assignment databases does not reveal the degree of loading on each frequency channel; e) that some administrations assign the same frequency to more than one user for shared use; f) that it is desirable to compare measur
13、ement results from different countries in border areas or for instance in the aeronautical or maritime mobile services bands; g) that automatic monitoring equipment is now in use by administrations, including methods for the analysis of records, and a number of parameters can be evaluated, which are
14、 of considerable value in enabling more efficient utilization of the spectrum; h) that in designing an automated system to gather occupancy data for use in spectrum management, one must determine what parameters are to be measured, the relationship among these parameters and how often measurements h
15、ave to be taken to ensure the data are statistically significant; i) that measurement procedures and techniques should be harmonized to facilitate the exchange of measurement results between various countries; j) that successful merging or combining monitoring data not only depends on the data forma
16、t in which the data is stored but also on the environmental and technical conditions under which the data is gathered, 2 Rec. ITU-R SM.1880-2 recognizing a) that various principles and methods of spectrum occupancy measurements are in use in the different countries; b) that one particular method exi
17、sts to get the high-accuracy frequency channel occupancy data and that such data usually is the basic to form the frequency band occupancy, recommends 1 that the measurement procedures and techniques specified in Annexes 1 and 2 should be used for spectrum occupancy measurements; 2 that both Report
18、ITU-R SM.2256 and the ITU Handbook on Spectrum Monitoring in force should be used as guidance for spectrum occupancy measurements and the equipment should satisfy the requirement mentioned in that Handbook; 3 that a common data format, that is a line-based ASCII file derived from the radio monitorin
19、g data format (RMDF), should be used following Recommendation ITU-R SM.1809. Annex 1 1 Introduction Annex 1 describes frequency channel occupancy measurements performed with a receiver or spectrum analyser. The signal strength of each frequency step is stored. By means of post-processing the percent
20、age of time that the signal is above a certain threshold level is determined. An example of the procedure for such post-processing is presented in Report ITU-R SM.2256 (Annex 1). Different users of a channel often produce different field-strength values at the receiver. This makes it possible to cal
21、culate and present the occupancy caused by different users. 2 Definitions Frequency channel occupancy measurements: Measurements of channels, not necessarily separated by the same channel distance, and possibly spread over several different frequency bands to determine whether the channel is occupie
22、d or not. The goal is to measure as many channels as possible in a time as short as possible. Revisit time: The time taken to visit all the channels to be measured (whether or not occupied) and return to the first channel. Observation time: The time needed by the system to perform the necessary meas
23、urements on one channel. This includes any processing overheads such as storing the results to memory/disk. Maximum number of channels: The maximum number of channels which can be visited in the revisit time. Transmission length: The average length of individual radio transmission duration. Integrat
24、ion time: Time interval for which an individual occupancy estimate is made. Normally 5 or 15 minutes. Duration of monitoring: The total time during which the occupancy measurements are carried out. Rec. ITU-R SM.1880-2 3 Preset threshold level for measurement: If a signal is received above the thres
25、hold level, the channel is considered to be occupied. Busy hour: The highest level of occupancy of a channel in a 60-min period. 3 Requirements 3.1 Equipment A suitable system capable of making frequency channel occupancy measurements by using frequency band registrations will consist of a radio rec
26、eiver or spectrum analyser, appropriate antenna, cable, a PC/controller, with interface adaptor, suitable acquisition and post-processing software. Other features may include GPS, for mobile/nomadic operation of the station, communications modem, for remote control and data exchange, system calibrat
27、ion for traceable field strength measurements, antenna switches, filters and attenuators, for multiple band and/or strong EMF exposure environments. 3.2 Site considerations Site should be chosen such as that the expected signal strength for the emissions of interest is above the expected threshold l
28、evel. The relation between these two parameters will define an area within which the measurement performed is of relevance to any station operating above a certain effective radiated power (e.r.p.) level or effective isotropic radiated power (e.i.r.p.) level. The expected signal strength can be eval
29、uated considering the licensed stations at the region, their emission profile and using simulation software. The threshold can be estimated considering the system sensitivity (noise floor) or previous measurements performed under similar conditions with the same equipment and configuration. If no pr
30、eliminary information is available, a site survey using portable equipment could be performed. This is especially important if the equipment placement is of definitive nature and future relocations may not be easily performed. Measurement results should ideally be accompanied by a report of analysis
31、 performed to select the site, indicating the area and the emitters that are expected to be considered. 3.3 Time related parameters There is a relationship between observation time, number of channels, average transmission length and the duration of monitoring. The revisit time is directly dependent
32、 on the observation time and the number of channels. Also the processing time (data transfer between receiver and controller) influences the revisit time and should be kept as short as possible. Revisit time = (Observation time Number of channels of identical bandwidth) + Processing time The observa
33、tion time per channel is limited by the scanning speed of the monitoring equipment. In order to maintain a reasonably short revisit time with relatively slow equipment, the number of channels to be measured must be reduced. 4 Rec. ITU-R SM.1880-2 Whenever applying the above equation to spectrum anal
34、ysers, when the RBW is set equal to the channel bandwidth, the Number of channels can be considered as the number of bins1 per sweep and the observation time as the dwell time per bin. On FFT analysers the principle still applies, especially if the number of channels to be scanned is greater than th
35、e FFT size and some sweep still being performed. On this case however, the number of scanned channels should be divided by the number of channels evaluated on each single FFT. The monitoring system needs to scan at an acceptable speed in order to detect individual short transmissions. There are prin
36、cipally two different approaches to obtain channel occupancy figures: a) Capturing every transmission in the band under observation. This approach requires a maximum revisit time that is half the minimum on or off time of any transmission in the band, whichever is shorter. This method delivers an ac
37、curacy that is independent of the occupancy result and may allow shorter monitoring duration. b) Statistical approach: Especially when considering bursts of digital systems, the minimum transmission time might be too small for the practical application of the above principle. However, if the monitor
38、ing time is long enough to provide enough samples, the occupancy result will be correct even with far longer revisit times, because the statistical probability of catching a transmission vs. the probability of missing it is the same as the duty cycle of the transmission. The accuracy of the statisti
39、cal approach, however, depends on the value of occupancy as described below. The duration of monitoring is a combination of the revisit time, typical transmission lengths expected, number of channels to be scanned and the wanted accuracy of the results. The duration of monitoring should be long enou
40、gh to allow the monitoring of all relevant emissions. If no time distribution pattern is known, initial evaluations should consider at least 24 h or multiples of 24 h. One week of monitoring gives the difference in occupancy over the various days of the week and occupancy during the week end. Seven
41、periods of 24 h spread of a longer period of time (e.g. one year) gives more reliable occupancy information. 3.4 Accuracy, statistical confidence level and required number of samples From a statistics point of view, the result of spectrum occupancy measurement is an estimate or statistical value, th
42、us it has accuracy and reliability attributes. Accuracy reflects the control of error, usually measuring with relative accuracy or relative error as well as absolute error, and reliability indicates the confidence of the result marking with confidence level. Measurement itself can be deemed as a sam
43、pling process from a given population, and result analysis is essentially as a process of estimating the population using the limited samples. In practice, the result of the collection and processing of data is abbreviated as SO, but it is not a true value as just mentioned. Even in cases where, dur
44、ing the integration time, the monitoring equipment provides only a small number of data samples, calculation of the occupancy estimate will give a number of values characterizing the radio-channel occupancy to a greater or lesser degree. However, such values will correspond to the true value of occu
45、pancy (SO) only when averaged over a large number of integration times, while the individual evaluations may deviate considerably from SO. On the other hand, under computer control it is possible to collect a large number of samples, more than is necessary for determining occupancy with the required
46、 accuracy. There is an optimum number 1 Bins in statistics refer to groups (or categories or classes) of data that fall within a certain range of values. Rec. ITU-R SM.1880-2 5 of samples, beyond which additional data may not significantly improve results. The optimum number of samples needed is dis
47、cussed in the current Report ITU-R SM.2256. There is no linear relationship between accuracy and revisit time. In the case of measuring 100 channels with a revisit time of 1 s, which is a practical value, the number of channels can be increased to 1 000 with a revisit time of 10 s without affecting
48、the confidence level/accuracy too much. There is a linear relationship between the occupancy and the number of samples required to achieve a desired confidence level. The lower the occupancy, the more samples will be needed. Table 1 compares independent sampling that is the simplest case using centr
49、al limit theorem and dependent sampling using a first order Markov-chain differ little from many more complicated mathematical models. TABLE 1 Number of dependent and independent samples required to achieve 10% relative accuracy and a 95% confidence level at various occupancy percentages (assumes a 1 s sampling period) Occupancy (%) Number of required independent samples Number of required dependent samples Required hours of dependent sampling 6.67 5 368 16 641 4.6 10 3 461 10 730 3.0 15 2 117 6 563 1.8 20 1 535 4 759 1.3 30 849 2 632 0.72 40 573 1 777 0.5 50 381 1 182 0.32 60 253
