Extended Abstract (172K)10B.5
RLAN 5 GHz interference to C-band meteorological radars in Europe : solutions, lessons, follow-up
Philippe Tristant, Meteo France, Toulouse, France
RLAN 5 GHz were authorised in the 5 GHz range following a decision made at the World Radiocommunication Conference in 2003 (WRC-07) with an obligation to implement a so-called Dynamic Frequency Selection (DFS) mechanism to ensure protection of radars operating in the 5250-5350 MHz and 5470-5725 MHz bands, meteorological radars in particular.
DFS mechanism parameters were mainly specified on a theoretical basis, included in Equipment standards (IEEE, ETSI,…) and implemented in RLAN equipments designs, allowing for market deployment by about years 2005-2006.
Unlike some Meteorological Services (Canada, Australia, …) that were deeply involved in this process and succeeded in forbidding RLAN in the 5600-5650 MHz band, one should admit that European Meteorological services did not involve themselves in the debate at relevant time, leading to RLAN standardisation in Europe that was not considering meteorological radars specificities and have hence been facing consequences of this lack of involvement, i.e. a number of interference cases.
Following these cases, under the auspices of the European Commission, a deep and detailed analysis work was jointly undertaken by the European Meteorological community (EUMETNET) and the RLAN industry that allowed highlighting the short-comings in ETSI RLAN 5 GHz standard. Recognising the urgency to solve this issue before the point of no-return of a mass-market RLAN deployment, a “package solution” was specified including new RLAN DFS parameters in the 5600-5650 MHz band together with meteorological community commitments as described in a EUMETNET Recommendation, with the aim of allowing for a global and long-term coexistence.
Among other, one can certainly highlight the combination of the DFS Channel Availability Check (CAC) time increase to 10 minutes and detection probability of 99.99% that, together with the “minimum detectable signal” concept applied to meteorological radars (at least 1 signal every 10 minutes) represents the major tool to ensure relevant protection of meteorological radars. Also important is the fact that such solution will allow for future radar developments without a need for new RLAN DFS parameters, that would in any case be totally unrealistic, facing millions RLAN that would more than likely be already deployed.
Obviously, the efficiency of this “package solution” will need to be monitored and verified in the following years, but the European meteorological community is confident that it would provide a satisfactory coexistence between meteorological radars and RLAN solution, although a preferred solution would have been to exclude RLAN from the 5600-5650 MHz band.
This exclusion could have indeed been the result of an initial strong involvement in WRC-03 preparation and one major lesson to consider is that, nowadays, the whole meteorological community, each national meteorological service, radar users and manufacturers cannot afford anymore to leave aside radio-frequency issues and not argue about our essential protection requirements.
The European RLAN issue is certainly symptomatic and without the recent actions within a still positive timeframe, this issue could have seen disastrous conclusions resulting in unsolvable interference cases that would have obliged meteorological services to refarm their meteorological radars from the 5 GHz band, with obvious operational and financial impacts.
Unfortunately, RLAN 5 GHz is one issue, but radio-frequency management is a prospective, long-term and never-ending process, and there are number of other radio applications that are or will be willing using C-Band, with a particular stress on increasing demand for non-licensed and mass-market equipments such as Short-Range devices, Ultra-Wide Band Devices, …
Proponents of such radio applications, e.g. the telecom industry, are powerful, organised and efficiently lobbying radio administration. This is currently the case with the potential use of RLAN 5 GHz in airplanes, as recently proposed by the aviation industry, presenting a new and serious threat to C-Band meteorological radars deployed worldwide.
Session 10B, Testbeds
Thursday, 8 October 2009, 8:30 AM-10:00 AM, Room 18
Previous paper Next paper
Recorded presentation