Neuro-fuzzy sensor's linearization based FPGA

Abstract

Nonlinear sensors and digital solutions are used in many embedded system designs. As the input/output characteristic of most sensors is nonlinear in nature, obtaining data from a nonlinear sensor by using an optimized device has always been a design challenge. This paper aims to propose a new Adapive Neuro-Fuzzy Inference System (ANFIS) digital architecture based Field Programmable Gate Array (FPGA) to linearize the sensor's characteristic. The ANFIS linearizer in synthesized and optimized in view to digital linearization. The performance of the developed architecture is examined with comparison to ANFIS software model and with two other designed architectures based FPGA using classical techniques. The results show the successful of the proposed architecture by demonstrating the capability of the operation to linearize a temperature sensor characteristic.

Linearization of the sensors characteristics: a review

“…Several research articles have reported to compensate nonlinearity, temperature, hysteresis, drift due to aging of various sensors by different ANN structures such as adaptive linear neural network (Islam et al, 2006;Islam and Saha, 2007), multilayer perceptron neural network (Khan et al, 2003;Khan and Islam, 2011;Kumar et al, 2015;Tarikul Islam et al, 2015), computationally efficient Chebyshev neural network , Laguerre neural network (LaNN) [137], fully connected cascade (FCC) neural network (Cotton and Wilamowski, 2011), fuzzy logic (Teodorescu), neuro-fuzzy architecture (Bouhedda, 2013), support vector machine (Xiaodong, 2008;Patra et al, 2011), covariance 3. Some of the works reported the hardware implementation of the optimized ANN models using a microcontroller, or FPGA (O'Droma and Mgebrishvili, 2005;Islam and Saha, 2007;Patra et al, 2011), or basic analog signal conditioning block.…”

Novel Information-Driven Smoothing Spline Linearization Method for High-Precision Displacement Sensors Based on Information Criterions

“…A comprehensive review for the implementation of calibration systems for highprecision displacement sensors could be found in the literature [18]. Currently, most of the attention has been paid to the design of high-precision nonlinearity measurement systems to improve the nonlinearity measurement accuracy [19][20][21][22][23] and the development of high-performance software algorithms in processing systems like microcontrollers to correct the nonlinearity more efficiently [24][25][26].…”

“…For the implementation of nonlinearity correction algorithms, Erdem compared the computational performance of six linearization algorithms on microcontrollers with a nonlinear optical displacement sensor and offered insights into the selection and implementation of an optimal linearization algorithm [24]; Bouhedda proposed a neuro-fuzzy architecture for linearization on FPGA (Field Programmable Gate Array) and reduced the required computational resource by over 40% compared to traditional computation methods [25]; Sonowal implemented four linearization algorithms, including piecewise linearization, lookup table, interpolation, and artificial neural networks, on FPGA and conducted a thorough comparison of their accuracy, speed of operation, and cost of implementation [26].…”