PROPAGATION MEASUREMENTS FOR SATELLITE RADIO RECEPTION INSIDE BUILDINGS

Swept CW signals (from 700 to 1800 MHz) were received inside six build ings of brick, corrugated sheet-metal, wood-frame, mobile-home, and re inforced concrete-wall construction. A transmitter antenna was mounted outdoors on top of an 18 m tower to simulate a satellite, and a linea rly scanned directional receiver antenna was used to probe the spatial , spectral, and temporal variability of the signal indoors. Levels wer e found to have much structure in the spatial and frequency domain, bu t were relatively stable in time. Typically, people moving nearby prod uced variations of less than 0.5 dB, whereas a person blocking the tra nsmission path produced fades of 6 to 10 dB. At an average position in a room, losses increased with frequency from 6 to 12 dB, but could be mitigated to just 2 to 6 dB by moving the antenna typically less than 30 cm. Severe losses (17.5 dB, mitigated to 12.5 dB) were observed in the concrete-wall building, which also exhibited the longest multipat h delays (>100 ns). Losses inside a mobile home were even larger (>20 dB) and were independent of antenna orientation. The median trough-to- crest distance was near 40 cm; the median trough length decreased from 30 cm at -3 dB to 10 cm at -18 dB and was direction independent. The power-frequency distortion increased with the logarithm of the bandwid th, but could be reduced by moving to a position of higher power. Only the losses showed a clear frequency dependence.