Strategy for WiFi interference detection in weather radar applications

Abstract

In the current context of intensive spectrum use by communications systems, WiFi systems have been allowed to use bands previously reserved for weather radars, as opportunity users. Some drawbacks in spectrum management make WiFi systems a source of interference that degrades the quality of observables obtained by C-band weather radars. In this work we present a strategy to detect these interfering WiFi packets at the output signal of the radar matched filter. The strategy is based on a delay and correlate algorithm that exploits the periodic structure of the WiFi packets preamble, periodicity that remains unchanged even though the signal is distorted when passing through the radar reception stages. We formulate the detection strategy as a hypothesis test that uses the squared modulus of the auto-correlation as the statistic, extended to a constant false alarm (CFAR) formulation to cope with the unknown noise power. We evaluate analytically and through numerical simulations the performance of the test in terms of detection probability. We also perform a series of controlled experiments using real-world weather radar data collected by Argentinian C-band RMA radars. The results show a high detection rate both when WiFi interference is in regions where there is only noise and when it is in regions where there is also a meteorological target.