Design and Implementation of a High-Speed Low-Power K-Nearest-Neighbor-Based Algorithm for Detecting Micro-Single-Event Latchups

In an irradiation environment, such as space, single-event latchup (SEL) is a radiation effect that may cause an abnormal high-current state in semiconductor devices, particularly in commercial-off-the-shelf (COTS) integrated circuits (ICs). On the other hand, the micro-SEL (
$\mu $
-SEL) is lower current and more localized and, therefore, more difficult to detect. Recent reported work embodying machine learning (ML) to detect
$\mu $
-SELs is only partially successful (70% accuracy), yet it has nonreal-time detection (50 s) and high power dissipation (82–132 mW). In this article, we propose a novel software-hardware co-optimized high-speed low-power k-nearest-neighbor-based (KNN-based) algorithm to protect a COTS IC against
$\mu $
-SELs. Furthermore, to obtain the highest
$\mu $
-SEL detection accuracy, yet with low hardware and power overhead, we propose two features of
$\mu $
-SEL characteristics in addition to the reported seven. The two proposed features are very worthy—the optimum requires five features (of which two are proposed). When implemented in a field-programmable gate array (FPGA) clocked at 10 MHz, our prototype achieves high (95%)
$\mu $
-SEL detection accuracy, yet real-time (110
$\mu $
s) detection and low (47 mW) power dissipation—equivalent to 36% improved accuracy, yet ~450
$000\times $
faster detection and 43%–64% lower power dissipation over the reported work. This renders our proposed approach appropriate for power-constrained space applications.