Outdoor RSSI Experiments and the Two-Ray Propagation Model

The experiment setup showing the contribution of the line of sight (LOS) and reflected rays 

In this experiments we measured the Received Signal Strength Index (RSSI) as function of the nodes distance. The sensors, two TelosB boards equipped with directional antennas, were placed in an open space free of obstacles at 1.3 m (4.26 ft) above the ground and set to transmit at +0 dBm (i.e. 1 milliwatt, the maximum transmission power allowed by the transceiver). The following figure reports the RSSI values averaged over 200 radio packets. The RSSI values are plotted together with the signal power computed using the Friis' free space equation and the Two-Ray fading model.

RSSI measured with directional antenna in an open field 

The Two-Ray model takes into account the contribution of the line of sight (LOS) signal and the ray reflected from the ground. The two signals are out of phase with respect to each other because of two effects:

1) The second ray changes phase when reflected from the ground (180º phase shift). This occurs because the directional antenna used in the experiment operates in vertical polarization (i.e. the E-field is perpendicular to the ground) and produces radio waves that bounce off the ground with opposite phase.

2) The reflected ray travels a longer distance than the LOS component and arrives at the receiver with a phase delay. Since the length difference between the two paths changes with the distance (and so does the phase delay), the contribution of the reflected wave can be constructive or destructive at different locations. In particular, the contribution is oscillatory within the first 25m (82 ft), constructive between 50 (164 ft) and 150 meters (492 ft) and destructive after 150 meters. After such distance the power signal decays approximately as 1/r4 (instead than 1/r2 as in the free space propagation). The maximum transmission range for the directional antennas tested was about 350 m (1150 ft). 

In the next plot, the experiment was repeated using the omnidirectional Planar-Inverted F Antenna (PIFA) integrated on the TelosB boards. Again, the Two-Ray model predicts the received power very accurately. Looking at the plot, we note the deep fade null around 25 m (82 ft): although the maximum transmission range was approximately 140 m (460 ft), at that location more than 90% of the radio packets were lost.

RSSI measured with omni-directional antenna (TelosB) in an open field

Effects of antenna polarization

The directional antenna in the experiments above works in linear polarization. In our experiments we found that using this antenna to communicate with the omni-directional antenna integrated on the TelosB (PIFA) works no better that using just two omnidirectional antennas (green line in the plot below). Better results are obtained when using TelosB to communicate with a directional antenna that works in circular polarization. The following plot reports the RSSI values measured for different antenna combinations.

Effect of different antenna polarization on the RSSI