Memristive cellular neural network
A system includes a sensor, a plurality of cells, and a processor. The sensor includes a plurality of transducers arranged on a planar surface and includes a first transducer and a second transducer. The first transducer is configured to produce an analog output signal corresponding to a detected input signal. The cells are arranged in a network and include a first and a second cell and are disposed proximate the sensor in a three-dimensional stacking fashion. The first transducer and the second transducer are electrically coupled to the first cell and the second cell in one-to-one relation. The first cell includes a plurality of inputs and a first cell output. Each input is coupled to an output of a corresponding plurality of neighboring cells. The first cell includes a first memristor and a bridge circuit configured to receive the analog output signal and provide a current corresponding to the detected input signal in a pixel-parallel fashion. The processor is coupled to the network and generates a parameter associated with a weight for each cell of the plurality of cells.
1 . A smart sensor system comprising:
a sensor having a plurality of transducers arranged on a planar surface, the plurality of transducers including a first transducer and a second transducer wherein the first transducer is configured to produce an analog output signal corresponding to a detected input signal;
a plurality of cells arranged in a network, the network being a memristive cellular neural network, the plurality of cells including a first cell and a second cell, the plurality of cells disposed proximate to the sensor wherein the first transducer is electrically coupled to the first cell and the second transducer is electrically coupled to the second cell in a one-to-one relation via 3D stacking, wherein the first cell includes a plurality of inputs and a first cell output, wherein each input of the plurality of inputs is coupled to an output of a corresponding plurality of neighboring cells of the plurality of cells, the first cell includes a first memristor and a bridge circuit configured to receive the analog output signal and provide a current corresponding to the detected input signal; and
a processor coupled to the network and configured to generate a parameter associated with a weight for each cell of the plurality of cells.
2 . The smart sensor system of claim 1 wherein the processor is configured to generate the parameter based on the output of the plurality of neighboring cells.
3 . The smart sensor system of claim 1 wherein the processor is configured to generate the parameter based on inputs of the plurality of neighboring cells.
4 . The smart sensor system of claim 1 wherein the network is configured to execute analog in-memory computing.
5 . The smart sensor system of claim 1 wherein the first cell and the second cell are configured to execute in parallel.
6 . The smart sensor system of claim 1 wherein the first memristor includes a drift memristor.
7 . The smart sensor system of claim 1 wherein the first memristor includes tantalum and hafnium oxide.
8 . The smart sensor system of claim 1 wherein the first cell includes a second memristor coupled to the bridge circuit.
9 . The smart sensor system of claim 1 wherein the bridge circuit includes a differential amplifier having a differential input corresponding to a voltage determined by a resistance of the first memristor.
10 . The smart sensor system of claim 1 including an interface processor coupled to the processor, the interface processor configured to execute an interface program using an operating parameter of the network.
11 . The smart sensor system of claim 1 wherein the sensor is disposed on a first layer and the plurality of cells is disposed on a second layer, and wherein the first layer and the second layer are in 3D stacked alignment.
12 . The smart sensor system of claim 1 wherein the sensor and the plurality of cells are fabricated in a very large-scale integration (VLSI) circuit.