A floating gate CMOS Euclidean distance calculator and its application to hand-written digit recognition

An efficient initialization mechanism of neurons for Winner Takes All Neural Network implemented in the CMOS technology

“…Distance calculation circuit (DCC) is responsible for the calculation of a distance between XðtÞ and W j ðtÞ vector in a given, jth, neuron. In the analog approach this block can be realized either as a voltage-mode [28,29] or a current-mode circuit [30][31][32]. Example circuits of this type have been reported in [32][33][34].…”

Energy Efficient Distance Computing: Application to K-Means Clustering

“…Further, the final addition operation after the multiplication step requires twice the number of bits of the input vector (as it is driven by multiplier outputs). Since this operation is performed iteratively over a large number of input vectors, alternative implementations involving analog current-mode designs of such computations are being actively explored [11,12,14]. The main advantage of this approach results from the fact that the addition/subtraction operation can be mapped to current flowing into/out of a circuit node (following Kirchhoff's current law).…”

“…A digital comparator and subtractor unit computes the absolute difference between the two inputs. It is worth noting here that current-mode analog computing is also a popular mechanism to calculate the absolute difference between two inputs [11,12,14]. However, in that case, the inputs should be represented by analog voltages or currents, which require additional converting units to interface with external digital logic.…”

“…However, despite the theoretical study and application of K-means clustering, efficient hardware implementation does not appear to have yet been achieved despite the simple and repetitive multiplication and addition operations required for distance calculation. Apart from the most intuitive approach to implementing clustering algorithms based on digital logic gates, various distance computing methods using analog [11][12][13][14][15] and mixed-mode design [16,17] have been proposed in the literature. As is well known, the digital-based approach can provide the most accurate and fast operating speed at the cost of area and power, while analog methods have comparable precision characteristics despite their simple structure.…”

Low-Power Analog Associative Processors Employing Resonance-Type Current-Voltage Characteristics

“…Konda et al (1996) and Cauwenberghs & Pedroni (1997) proposed neuron MOSFET (νMOS)-based and charged-based Manhattan-distance evaluation cells, respectively. A νMOS-based Euclidean distance calculator used in a recognition system for handwritten digits was proposed (Vlassis et al, 2001). Kramer et al (1997) also proposed an analog Manhattan-distance-based content-addressable memory (CAM) using the analog non-volatile memory technology.…”