Cell Under Test Research Articles

Convergence Rate of LSMI in Amplitude Heterogeneous Clutter Environment

In this letter, we analyze the convergence rate of loaded sample matrix inversion (LSMI) algorithm in amplitude heterogeneous clutter environment. The probability density function of output signal to interference and noise ratio loss (SINR Loss) is derived. Then we give an approximate expression of average SINR loss. Compared with the case where samples used for estimation of covariance matrix of cell under test (CUT) are independent and identically distributed (i.i.d.) with the snapshot of CUT, if the clutter to noise ratio (CNR) of the training samples is larger than that of CUT, the convergence rate of LSMI is faster and output SINR is higher; conversely, the convergence rate of LSMI is slower and SINR is lower. Simulation validates the theoretical analysis.

Weak Cell Detection in Deep-Submicron SRAMs: A Programmable Detection Technique

Embedded SRAM bit count is constantly growing limiting yield in systems-on-chip (SoCs). As technology scales into deep sub-100-nm feature sizes, the increased defect density and process spreads make stability of embedded SRAMs a major concern. This paper introduces a digitally programmable detection technique, which enables detection of SRAM cells with compromised stability [with data retention faults (DRFs) being a subset]. The technique utilizes a set of cells to modify the bitline voltage, which is applied to a cell under test (CUT). The bitline voltage is digitally programmable and can be varied in wide range, modifying the pass/fail threshold of the technique. Programmability of the detection threshold allows tracking process variations and maintaining the optimal tradeoff between test quality and test yield. The measurement results of a test chip presented in the paper demonstrate the effectiveness of the proposed technique

An approach to knowledge-aided covariance estimation

This paper introduces a parametric covariance estimation scheme for use with space-time adaptive processing (STAP) methods operating in heterogeneous clutter environments. The approach blends both a priori knowledge and data observations within a parameterized model to capture instantaneous characteristics of the cell under test (CUT) and reduce covariance errors leading to detection performance loss. We justify this method using both measured and synthetic data. Performance potential for the specific operating conditions examined herein include: 1) averaged behavior within roughly 2 dB of the optimal filter, 2) 1 dB improvement in exceedance characteristic relative to the optimal filter, highlighting improved instantaneous capability, and 3) impervious ness to corruptive target-like signals in the secondary data (no additional signal-to-interference-plus-noise ratio (SINK) loss, compared with 10 dB or greater loss for the standard STAP implementation), with corresponding detections comparable to the optimal filter case

Robust adaptive matched filtering using the FRACTA algorithm

An effective method is developed for selecting sample snapshots for the training data used to compute the adaptive weights for an adaptive match filter (AMF); specifically a space/time adaptive processing (STAP) airborne radar configuration is considered. In addition, a new systematic robust adaptive algorithm is presented and evaluated against interference scenarios consisting of jamming, nonhomogeneous airborne clutter (generated by the Research Laboratory STAP (RLSTAP) or knowledge-aided sensor signal processing and expert reasoning (KASSPER) high-fidelity clutter models or using the multi-channel airborne radar measurement (MCARM) clutter data base), internal system noise, and outliers (which could take the form of targets themselves). The new algorithm arises from empirical studies of several combinations of performance metrics and processing configurations. For culling the training data, the generalized inner product (GIP) and adaptive power residue (APR) are examined. In addition two types of data processing methods are considered and evaluated: sliding window processing (SWP) and concurrent block processing (CBP). For SWP, a distinct adaptive weight is calculated for each cell-under-test (CUT) in a contiguous set of range cells. For one configuration of CBP, two distinct weights are calculated for a contiguous set of CUTs. For the CBP, the CUTs are in the initial training data and there are no guard cells associated with the CUT as there would be for SWP. Initial studies indicate that the combination of using the fast maximum likelihood (FML) algorithm, reiterative censoring, the APR metric, CBP, the two-weight method, and the adaptive coherence estimation (ACE) metric (we call this the FRACTA algorithm) provides a basis for effective detection of targets in nonhomogeneous interference. For the KASSPER data, FRACTA detects 154 out of 268 targets with one false alarm (P/sub F//spl ap/3/spl times/10/sup -5/) whereas the FML algorithm with SWP detects 11 with one false alarm. The clarvoyant processor (where each range cell's covariance matrix is known) detects 192 targets with one false alarm.