The Sigma-Delta CNN with Second Order Noise Shaping Property

Abstract: In this paper, a novel spatial domain 2-D sigma-delta modulation using two-layered discrete-time cellular neural networks (DT-CNNs) with multi-stage noise shaping (MASH) property is proposed. Although the sigmadelta modulation is widely used for realizing of analog to digital converter (ADC), sigma-delta concepts are only for 1-D signals. Signal processing in the digital domain is extremely useful in 2-D signals such as image processing, medical imaging, ultrasound imaging and so on. In the proposed architecture, the A-template is used for a digital to analog converter (DAC), the C-template works as an integrator, and the nonlinear output function is used for the bilevel output. Besides, due to the CNN characteristic, each pixel of an image corresponds to the cell of CNN, and each cell is connected spatially by the A-template. Also, the multi stage noise shaping (MUSH) property, which is one of a very effective framework for a high order sigma-delta modulation, is exploited for realizing a second order sigma-delta modulator by cascade connected first-order CNN sigma-delta modulators.