An experimental evaluation of the performance of two-branch space and polarization diversity schemes at 1800 MHz

Abstract

Determines the mean signal level and envelope cross-correlation of 1800 MHz base station signals received in two-branch spatial and polarization diversity schemes. Measurements have been conducted with the experimental base site located in (i) two urban sites, (ii) a residential area, (iii) a rural area, and (iv) near a motorway. In each location, the effect of the random orientation of a typical mobile radio telephone handset has been studied by examining the characteristics of signals received from a mobile collinear antenna inclined at angles of 0/spl deg/, 30/spl deg/, 45/spl deg/, 60/spl deg/, and 90/spl deg/ to the vertical. Furthermore, the diversity gain at 90% signal reliability has been evaluated for each diversity scheme by simulating selection, equal-gain and maximal-ratio combining techniques using the recorded signals as inputs. Results have shown that 20/spl lambda/ separation in the horizontal plane or 15/spl lambda/ in the vertical plane is sufficient to obtain a cross-correlation of less than 0.7 for most of the time at 1800 MHz. Similar cross-correlation results were obtained for polarization diversity. When the antenna is inclined at 45/spl deg/, a 6 dB degradation in signal level was recorded for space diversity schemes. However, the diversify gain is unaffected by tilt and remains unchanged at 5-6 dB for horizontal and 3.5-4.5 dB for vertical separation. For polarization diversity, only a little degradation is experienced because most of the energy lost on the vertical branch is recovered on the horizontal branch. The diversity gain is between 1-2 dB at 0/spl deg/ tilt and increases to 3-5.2 dB at 45/spl deg/. >