2 November 2017 in IEEE Transactions on Industrial Informatics
A new method for attack detection of smart grids with wind power generators using reservoir computing (RC) is introduced in this paper. RC is an energy-efficient computing paradigm within the field of neuromorphic computing and the delayed feedback networks (DFNs) implementation of RC has shown superior performance in many classification tasks. The combination of temporal encoding, DFN, and a multi-layer perceptron (MLP) as the output readout layer is shown to yield performance improvement over existing attack detection methods such as MLPs, support vector machines (SVM), and conventional state vector estimation (SVE) in terms of attack detection in smart grids. The proposed algorithms are shown to be more robust than MLP and SVE in dealing with different variables such as the amplitude of the attack, attack types, and the number of compromised measurements in smart grids. The attack detection rate for the proposed RC based system is higher than 99%, based on the accuracy metric for the average of 10000 simulations.
30 October 2019 in IEEE Transactions on Emerging Topics in Computational Intelligence
Spiking neural networks have been widely used for supervised pattern recognition exploring the underlying spatio-temporal correlation. Meanwhile, spatio-temporal correlation manifests significantly between different components in a smart grid making the spiking neural network a desirable candidate for false data injection attack detection. In this paper, we develop a spiking-neural-network-based technique for dynamic cyber-attack detection in a smart grid. This is achieved through judiciously integrating spiking neurons with a special recurrent neural network called the delayed feedback reservoir computing. The inter-spike interval encoding is also explored in the precise-spike-driven synaptic plasticity based training process. The simulation results suggest that the introduced method outperforms multi-layer perceptrons and can achieve a significantly better performance compared to the state-of-the-art techniques. Furthermore, our analysis indicates that the delay value in the delayed feedback reservoir will have a substantial impact on the overall system performance.
23 October 2017 in IEEE Journal on Emerging and Selected Topics in Circuits and Systems
With the rapid advancement in the complementary metal-oxide-semiconductor (CMOS) technology, the performance of neuromorphic computing systems that employed the emerging devices has entered a new era of high accuracy and energy efficient operations. As one of the most fundamental components in computing systems, encoding scheme characterizes the relationship between the stimulus and the individual or ensemble neuronal responses. Although rate encoding is easier to implement and with high error tolerance, temporal encoding offers high data density and energy efficiency. In our work, an inter-spike interval (ISI) based resistive crossbar neuromorphic design, built with the standard CMOS technology, is proposed. Our proposed decoder exhibits not only computation accuracy but also robustness. Another worth-mentioning achievement is the data compressor in our work which is based on the spike temporal encoding scheme. In this paper, design and performance analysis for proposed ISI based resistive crossbar are elaborated; as well as the application of our proposed design. 24 patterns are adopted to encode the sensory information which are successfully represented by 7 inter-spike intervals leading to significantly high compression rate. To evaluate the performance, a test bench of video frames consisting one person rotating her head from 0 to 75 degrees with an increment of 15 degrees has been employed. The results showed that the ISI code has better performance in both recognition rate and converging speed.
23 December 2020 in IEEE Transactions on Green Communications and Networking
This paper introduces a novel energy efficient spectrum sensing method for multiple-input multiple-output and orthogonal frequency division multiplexing (MIMO-OFDM) systems in dynamic spectrum sharing (DSS) environments. To be specific, a spiking reservoir computing (RC) based approach is introduced for spectrum sensing of MIMO-OFDM systems to take advantage of the spatial and temporal correlations of the environment. In particular, we develop a spiking delayed feedback reservoir (DFR). DFRs are a new class of recurrent neural networks (RNNs) which tackle the vanishing gradient problem of traditional RNNs while spiking neurons are energy efficient and biologically plausible neuron models. We extend our introduced model in time and space domains to capture the spatial and temporal correlations in DSS environments. Conditional generative adversarial networks (cGANs) are introduced to tackle the data scarcity issue arose from applying machine learning techniques to MIMO-OFDM systems where training data is limited. Simulation results suggest that the probability of detection of the introduced RC based spectrum sensing at the low signal-to-noise (SNR) regime is significantly higher than the state-of-the-art techniques and the computational complexity is also reduced compared with traditional RNNs.