Application of Noise-Enhanced Detection of Subthreshold Signals for Communication Systems

Abstract

A signal detection system using noise statistical processing is proposed. By approaching the problems of low voltage and high noise from miniaturization of a device from a stochastic point of view, a faint-signal receiving system that can effectively detect subthreshold and noise level signals has been developed. In addition, an alternative to statistical processing is proposed, and would be successfully implemented on a circuit. For the proposed signal detection method, the detection sensitivity was investigated using numerical simulation, and the detection sensitivity was sufficiently high to detect even a signal with a signal-to-inherent-noise ratio of −14 dB. Thus, it is anticipated that the application of this system to an integrated circuit will have a significant impact on signal processing.

Design of a 500-MS/s stochastic signal detection circuit using a non-linearity reduction technique in a 65-nm CMOS process

“…1 (b), and their distribution can be estimated by MonteCarlo analysis under a device mismatch model [1]. Even noises generated in the devices also can be estimated under the assumption of their Gaussian distributions [2,4]. Thus, the mean and standard deviation of input-referred comparator offsets can be estimated and controlled to a certain extent in the circuit design.…”

“…In a previous study [4], we proposed a simple estimation method using two groups of comparators with similar offset distributions and different references, as shown in Fig. 1 (c), for real-time signal detection instead of using Eq.…”

“…This non-linearity reduction technique can be implemented by applying ±σ offset estimated in a Monte-Carlo simulation as reference voltages. The problem caused by a difference between a simulated offset and a real value can be addressed by using a slightly wider margin for a large number of comparators [4].…”

“…A completely new approach to detect a signal using statistics of device noise and mismatch has been proposed by the authors [4]. Recently, a new ADC, referred to as the stochastic ADC, has been reported [5].…”

Optimal condition of memristance enhancement circuit using external voltage source

“…Such manipulation method has been applied to several systems for signal detections, [28][29][30] imaging, 31 and communications. 32 It has been reported in Refs. 33-35 that the state of memristor is controlled by white noise due to the "stochastic resonance" phenomenon.…”