Multi-target tracking for merged measurements of automotive narrow-band radar sensors

Novel Deterministic Detection and Estimation Algorithms for Colocated Multiple-Input Multiple-Output Radars

“…4). In fact, if we assume that the intensity of the echo received by each VA from the i-th target detected in the l-th frequency bin does not change from antenna to antenna, the complex amplitude C i [p, q, l] observed on the (p, q) VA can be expressed as (see (9) and (10)) here, λ = c/f 0 is the wavelength associated with the start frequency (see (14)), (p R , q R ) is the couple of integers identifying the selected reference VA,…”

“…In this paragraph, the accuracy of the STDREC algorithm employed by the RPE is analysed for two specific static scenarios. The first scenario is characterized by a single detectable target (a small metal disk 14 4 for all the employed radar devices (the data referring to the i-th position are collected in the column identified by T i , with i = 1, 2, ..., 10). The second scenario, instead, is characterized by the presence of an overall number of targets ranging from 1 to 9 (so that 1 ≤ L ≤ 9).…”

“…In addition, they exploit the iterative estimation techniques developed in [10] and based on a serial cancellation approach for evaluating the parameters of multiple overlapped sinusoids or multiple overlapped complex exponentials in the presence of additive noise. The devised algorithms are able to accurately detect and estimate multiple close targets, and to solve the problem of merged measurements or unresolved measurements [11]- [14], in the sense that targets generating merged measurements in the range domain are resolved in the estimation of their angular coordinates.…”

360°-Environment sensing and signal processing for an automotive pre-crash application

“…The sensors need special signal processing, due to the effect of measurement merging, which is not in the scope of this paper. For further information on this topic refer to [13].…”