WiFi-Based Passive Bistatic Radar: Data Processing Schemes and Experimental Results

Abstract

The practical feasibility of a WiFi transmissions based passive bistatic radar (PBR) is analyzed here. The required data processing steps are described including the adopted techniques for 1) the control of the signal autocorrelation function (ACF) usually yielding a high sidelobe level, and 2) the removal of the undesired signal contributions which strongly limit the useful dynamic range. The performance of the proposed techniques is firstly evaluated against simulated data generated according to the IEEE 802.11 Standards. Moreover the results are presented against a real data set collected by an experimental setup when using the conventional dual (reference and surveillance) channels PBR receiving scheme. This allows us to demonstrate the potentialities of a WiFi-based PBR for local area surveillance applications, where vehicles and people can be detected and tracked. Based on the digital nature of the exploited signals of opportunity, the attractive possibility is also investigated of avoiding the use of a dedicated receiving channel for the reference signal, by synthesizing it from the surveillance channel. This approach is shown to yield comparable performance with respect to the conventional PBR approach while yielding a remarkable saving in terms of system complexity

Localization of moving targets with a passive radar system based on WiFi transmissions

“…After a fully coherent base-band down-conversion stage, the data are sampled with a sampling frequency of 22 MHz and stored for off-line processing. The WiFi-based passive radar processing scheme presented in [8] -which includes the steps of sidelobes control and disturbance removal -is applied against the surveillance signals using a coherent integration time of 0.5 s. The cancellation stage is performed by adopting the ECA batches algorithm ( [9]) with a batch duration equal to 50 ms over a range of 300 m. The two-dimensional range/Doppler map is then evaluated over consecutive portions of the acquired signals with a fixed displacement of 0.1 s (thus, 10 frames per second are obtained). Target detection is then performed over the three channels by resorting to a simple adaptive detector based on the cell-average CFAR (a cross-shape for the distribution of the reference cells was chosen with a total number of reference cells per frame set to 45,080 and the probability of false alarm was set to 10 -4 ).…”

“…The WiFi access point (AP) providing coverage for an assigned area, might act as an ideal illuminator of opportunity for short range target detection using the passive radar principle, as it has been demonstrated in [4]- [6]. Moreover, effective processing techniques have been already designed to enable the practical operation of the resulting system ( [7]- [9]) and to rise the specific challenges issued by advanced applications, like target classification ([10]- [12]) and localization ( [13]). In particular, the results reported in [13] preliminarily show that, when target localization is required in an assigned area, it is in principle possible to estimate the target x-y coordinates.…”

GNSS based passive bistatic radar for micro-Doppler based classification of helicopters: Experimental validation

“…Real demonstrators [2], [3] have been built to prove the real applicability of this technology. A wide set of possible illuminators of opportunity have been evaluated [4], [5], [6], [7]. Moreover a set of more advanced radar techniques have been proved in this field such as through the wall radar [8], SAR and ISAR imaging [9], [10], reliable detection [11], [12] and Automatic Target Recognition [13].…”

Wifi-based passive bistatic radar by using moving target indicator and least square adaptive filtering

“…We can summarize the method used to counteract these limitations in [12]: weighting networks for DSSS pulses, standard Hamming network and guard interval blanking for OFDM pulses. In [14] we found an experiment of localization and tracking of moving targets and in [15] the description of the data processing scheme adopted, here Colone et al suggest to use the ECA algorithm (an LS algorithm based on delayed versions of the reference signal) together with the above mentioned techniques for the high sidelobes reduction.…”