Digital implementation of the cellular sensor-computers

Two different kinds of cellular sensor-processor architectures are used nowadays in various applications. The first is the traditional sensor-processor architecture, where the sensor and the processor arrays are mapped into each other. The second is the foveal architecture, in which a small active fovea is navigating in a large sensor array. This second architecture is introduced and compared here. Both of these architectures can be implemented with analog and digital processor arrays. The efficiency of the different implementation types, depending on the used CMOS technology, is analyzed. It turned out, that the finer the technology is, the better to use digital implementation rather than analog