Wideband Calibration Including Baseband Delays for an Array of Spatially Distributed Subarrays

Abstract

Antenna arrays and spatial signal processing techniques have proven to be the most effective countermeasures against radio interference (RFI) and spoofers. Due to the comparably large size of uniform rectangular arrays (URAs), a hidden installation in cars produced for the consumer mass market is not possible, which is a strict requirement by industry and customers, such that the aesthetic design of the car is not infringed. Therefore, the authors of this paper presented a new array concept consisting of two uniform linear arrays (ULAs), which can be spaced several wavelengths apart from each other. With that additional degree of freedom and the reduced footprint of a ULA compared to a URA, this array allows for a concealed installation in the front and/or rear bumpers as well as the side mirrors of a car. As a drawback, the distribution of the ULAs requires cables with a length of several meters from the individual antenna elements to the central processing unit inside the car, which digitizes and processes the signals of all antennas together in the digital domain. Due to that, the signals are digitized with a non-neglectable differential delay in the baseband signals. The authors demonstrated by simulation and experiment, that it is possible to extend blind mitigation algorithms, which are typcially incorporated for RFI mitigation, to reduce the effect of the differential delay in case of radio interference. However, the delay must be estimated to be able to incorporate deterministic spatial processing algorithms as well, such as attitude estimation or spoofer mitigation. To estimate the delays and the overall influence of the receive chain on the digitized signals, calibration methods can be used. State-of-the-art methods are developed for compact arrays such as the aforementioned URA and hence neglect delays in the baseband signals. This paper demonstrates a new approach: A new calibration method is developed, which is able to estimate differences in baseband delay and phase offset between the individual antenna channels of the proposed array. Furthermore, this method is also able to determine the frequency dependent transfer characteristic of the overall receive chain, such that the correlation loss for wideband signals is minimized. This guarantees, that also signals with bandwidths above 10 MHz as for example GPS L5 or Galileo E5a, can be processed by the proposed array. Different approaches for calibration networks and methods will be discussed on their advantages and disadvantages for the new array concept. Subsequently, the best concept will be evaluated.