In this paper a simplified hardware implementation of a CNN softmax layer is proposed. Initially the softmax activation function is analyzed in terms of required accuracy and certain optimizations are proposed. Subsequently the proposed hardware architecture is evaluated in terms of the introduced approximation error. Finally the proposed circuits are synthesized in a 90-nm 1.0 V CMOS standard-cell library using Synopsys Design Compiler. Comparisons reveal significant reduction up to 47% and 43% for certain cases, in terms of area × delay product over prior art. Area savings are achieved with no performance penalty.
This paper presents techniques for low-power addition/subtraction in the logarithmic number system (LNS) and quantifies their impact on digital filter VLSI implementation. The impact of partitioning the look-up tables required for LNS addition/subtraction on complexity, performance, and power dissipation of the corresponding circuits is quantified. Two design parameters are exploited to minimize complexity, namely the LNS base and the organization of the LNS word. A roundoff noise model is used to demonstrate the impact of base and word length on the signal-to-noise ratio of the output of finite impulse response (FIR) filters. In addition, techniques for the low-power implementation of an LNS multiply accumulate (MAC) units are investigated. Furthermore, it is shown that the proposed techniques can be extended to cotransformation-based circuits that employ interpolators. The results are demonstrated by evaluating the power dissipation, complexity and performance of several FIR filter configurations comprising one, two or four MAC units. Simulations of placed and routed VLSI LNS-based digital filters using a 90-nm 1.0 V CMOS standard-cell library reveal that significant power dissipation savings are possible by using optimized LNS circuits at no performance penalty, when compared to linear fixed-point two's-complement equivalents.
In this paper a simplified hardware implementation of a CNN softmax-like layer is proposed. Initially the softmax activation function is analyzed in terms of required numerical accuracy and certain optimizations are proposed. A proposed adaptable hardware architecture is evaluated in terms of the introduced error due to the proposed softmax-like function. The proposed architecture can be adopted to the accuracy required by the application by retaining or eliminating certain terms of the approximation thus allowing to explore accuracy for complexity trade-offs. Furthermore, the proposed circuits are synthesized in a 90 nm 1.0 V CMOS standard-cell library using Synopsys Design Compiler. Comparisons reveal that significant reduction is achieved in area × delay and power × delay products for certain cases, respectively, over prior art. Area and power savings are achieved with respect to performance and accuracy.
In this paper the logarithmic Number System (LNS) is adopted to implement Long-Short Term Memory (LSTM), the basic component of a deep learning network type. Initially, piece wise approximations to activation functions σ and tanh are proposed and evaluated in LNS. Secondly, LNS multipliers and adders are implemented for wordlengths of 9,10 and 11 bits. The circuits are implemented in an 90-nm 1.0 V CMOS standard-cell library and quantitative results are reported. Results demonstrate that LNS is a good candidate for data representation and processing in deep learning networks, as area reduction of up to 36% is possible.
26 July 2021 in IEEE Consumer Electronics Magazine
Currently, hospitals and health care sectors employ low-cost IoT (Internet of Thing) based remote health monitoring systems and labs in order to collect a subjects or patients real-time data. Such a process can be helpful for the early detection of a healthy newborn life and even vital in the case of saving these lives. In this paper, a preliminary implementation of a system monitoring the fetus heart rate has been designed and implemented as a mobile wearable measuring system with remote sensing. The proposed implementation turns out to be an efficient combination of simplicity and cost effectiveness and is accompanied with preliminary accurate measurements of fetus heart rate. The developed board uses an ESP-8266 transceiver module and is programmed to transmit the data efficiently to a remote server station using a IEEE 802.11 b/g/n - based wireless network. The patients data can then be monitored using an Android-based or satellite phone, or even any well-known internet browser connected to the specific network certifying in many occasions the health safety distance required due to various situations including that of COVID-19.
18 February 2009 in IEEE Transactions on Circuits and Systems I: Regular Papers
A graph-based technique is introduced for the design of a class of residue arithmetic multipliers, as well as a family of new high-radix digit adders. A proposed design technique derives simple high-radix modulo-rn multipliers by optimally selecting among the variety of introduced digit adders the ones that compose a minimal-area multiplier. The proposed technique minimizes multiplier complexity by selecting digit adders that observe the constraints imposed on the maximum values of the various intermediate digits. The proposed technique leads to significant area and time improvements over previously published architectures for practical modulus cases.