Inside view

15 August 2013

The ASI's antenna and satellite beacon receiver. (Insert): radio wave propagation experiment

The ASI's antenna and satellite beacon receiver. (Insert): radio wave propagation experiment

Lorenzo Luini (left) and Carlo Capsoni (right)

Lorenzo Luini (left) and Carlo Capsoni (right)

Why is rainfall significant for radio?

Rainfall is the dominant impairment for propagation of electromagnetic waves in the atmosphere for systems operating at frequencies above 10 GHz. Besides causing secondary effects such as depolarisation, scintillation and interference due to scattering, the presence of rain along the radio path induces extra attenuation which dramatically increases with frequency. So it is important to know the local rainfall statistics to predict its effects when designing a radio link. Owing to its fast time variability, rain intensity must be measured within a short time interval (the integration time) to properly catch the most intense rainy periods – those of which could cause the outage of the radio link.

How do you deal with rain data?

The complementary cumulative distribution function (CCDF) of a quantity expresses the probability of the quantity exceeding a given value in a reference period (typically a year). The rain rate CCDF varies from site to site because of the different climatic features and the influence from local orography (hills and mountains) and topography. This is the main input of any model aimed at predicting rain attenuation statistics, i.e. the probability to exceed given rain attenuation values in the average year. In the design phase this allows the determination of the power margin required to guarantee a desired availability or quality of service (QoS) of the communication system.

What was the aim of your work?

Rain rate CCDFs with 1-minute integration time are needed for path attenuation prediction models for systems above 10 GHz. Unfortunately, this data is not commonly available worldwide, while rainfall accumulations with longer integration times (typically 30 minutes or more) are more easily retrievable from local meteorological agencies. As a consequence, efforts have been made to develop models that can convert rain rate CCDFs to shorter integration times from the available long integration time data. In our Letter, we have shown that, thanks to their physical soundness, some models originally developed to operate on yearly rain rate CCDFs can be successfully applied, with slight modifications, to derive 1-minute integrated rainfall statistics on a monthly basis.

What were the main challenges?

The major challenge in predicting rain rate CCDFs with 1-minute integration time from temporally coarser data is retrieving what has been filtered out by the time averaging process – the intense precipitation rates. The most accurate, reliable and globally applicable way to do this is to employ models relying on key physical features of rainfall (formation process, type of event, evolution in time etc.). Specifically, we developed and used the EXCELL RSC (Rainfall Statistics Conversion) model, which simulates the integration process of a rain gauge in a rainfall environment evolving in time. This environment is reproduced using synthetic rain cells, whose type and occurrence varies from site to site. As a result, EXCELL RSC is applicable worldwide and, thanks to its solid physical basis, offers a very good prediction performance.

How is the work developing?

Our on-going work addresses the upgrade of a prediction model of path rain attenuation, SC (Startiform/Convective) EXCELL, to best use the detailed information provided by monthly rain intensity statistics. This involves identification of the other model input parameters, such as the vertical profile of precipitation and the type of rain cells to be used on a monthly basis. The idea is to produce a very flexible prediction tool adaptable to any request and any climate and, also, to improve the overall performance prediction of the model.

What are the wider goals of your group?

Our research group has a lot of experience in wave propagation from microwave to optical wavelengths. The theoretical activity is mainly devoted to: description of the radio channel and development of models of the various propagation impairments caused by adverse atmospheric conditions (e.g. rain and fog attenuation, depolarization, interference due to hydrometeor scatter, turbulence); simulation of rain and cloud fields; design of advanced telecommunication systems and fade mitigation techniques. This is all part of an effort to improve the performance prediction and application flexibility of worldwide propagation impairments models.

The group’s theoretical activity is supported by our experimental work. We have been involved in most European satellite propagation experiments since 1979 with the experimental station located at Spino d'Adda. Looking ahead we will be involved as Principal Investigator of the Italian Space Agency (ASI), in the Alphasat - Aldo Paraboni propagation experiment. The satellite was launched on July 25th this year.

What do you think the future holds in this area?

We expect increasing demand from end users for larger bandwidth to support multimedia services to drive an increase in the carrier microwave frequencies; up to W band in the near future. This is for fixed and mobile wireless systems, space communications and indoor applications. A lot of work will be needed to accurately describe the radio channel characteristics because attenuation caused by rain is not the only effect to increase dramatically with frequency. There are phenomena caused by other atmospheric constituents (e.g. clouds and gases such as oxygen and water vapour) that need to be taken into account. Moreover, a definite interest is appearing in free space optic communications, in fields from indoor to space science applications, because of the extremely large bandwidth, high security, low power consumption and immunity to electromagnetic interference.

Further reading

This article is based on the Letter: Prediction of monthly rainfall statistics from data with long integration time (new window).

Applescores group:http://applescores.ws.dei.polimi.it/ (new window)

A PDF version (new window) of this feature article is also available.

Journal content

Cover of Electronics Letters, Volume 49, Issue 25

View all content

Browse or search all papers in the latest or past issues of Electronics Letters on the IET Digital Library.