A good understanding of how solar bursts affect cellular communications will help in the design of future generations of wireless systems, the researchers say. Their report is published March 7 and appears in the March-April issue of the journal, Radio Science, published by the American Geophysical Union.
The study was possible only because of an archive of data on solar radio bursts that has been assembled by the National Oceanic and Atmospheric Administration (NOAA) from observations made around the world by the U.S. Air Force and other entities and that is now maintained by the National Geophysical Data Center (NGDC) of NOAA in Boulder, Colorado. The first detections of solar radio bursts (at much lower frequencies) were made inadvertently in 1942 by some of the earliest radars deployed during World War II. After the war, solar radio studies became a recognized field of astronomical research, and the Air Force was active in collecting data, since the bursts continued to affect radar.
Because cellular communication has greatly expanded in recent years, the researchers looked back at the last four decades (1960- 1999) of NGDC data in the context of noise levels found in wireless communication systems. This data interval covered slightly more than four solar cycles, including the solar maximum in 1989-1991 which occurred before cellular communications became ubiquitous around the world. The researchers note that the number and location of collection points varied over time, and the instruments used to measure solar radio bursts have improved significantly since the early years. They do not believe that these variations affect the main results of their study.
Radio wave energy received from the Sun is measured in solar flux units (SFU), with one SFU equaling 10^-22 [1/10 followed by 22 zeros] watts per square meter of receptor area per hertz. During a burst, the energy received may be as high as 100,000 SFU, with the energy also depending upon the frequency measured. In the study, the scientists at Bell Laboratories, together with Gary of NJIT, sought to determine how often bursts of at least 1,000 SFU have occurred over the years, this being the level that can potentially disrupt cell communications by covering conversation with noise or causing calls to be dropped.
Counting the number of solar bursts was difficult, since the same event may have been recorded by several monitoring stations, often on different frequencies, and separate events may also have occurred close in time to one another. The researchers' analysis suggested that on 12 minutes was the minimum interval between what they would regard as separate solar bursts, and they limited their study to the frequency range of 1-20 gigahertz (Ghz). Most present-day cell phone transmitters currently operate in the band from 900 megahertz (MHz) to around 3 GHz.
The analysis of the data by the research team, which also included Dr. Bala Balachandran and Dr. David Thomson, then of Bell Labs, revealed that solar radio bursts of 1,000 SFU can occur on 10-20 days per year, on average, with higher rates and stronger bursts during solar maximum periods and lower, weaker ones during solar minimum periods. The effect of bursts on wireless communications is dependent upon the orientation of cell antennas, with those pointing east-west more susceptible mornings and evenings than at noon. Therefore, any given cell site might be affected by solar radio bursts only every 40-80 days, or several times per year on average. But any single burst could affect a large service area, since number of cell sites are likely to be pointed in the direction of the
Sun when an event occurs. Furthermore, the impacts on service, in terms of increased noise levels and call disruptions, would be expected to be more frequent during the years of maximum solar activity.
The study was supported in part by Lucent Technologies and in part by the Space Weather Program of the National Science Foundation at the New Jersey Institute of Technology.
Journal
Radio Science