Search

In this paper, we focus on the prediction issue of the nonlinear chaotic time series. In particular, we introduce the modified echo state network (M-ESN) to predict the time series of nonlinear chaotic system. Thereinto, to solve the ill-conditioned output weight matrix caused by vast neurons in hidden layer of the ESN, we introduce the hybrid regularized network (HRN) based on the l2/3 regularization and the l2 regularization. Thereinto, the l2/3 regularization plays the role to produce the sparse output weight and import the oracle property to the prediction model, while the l2 regularization plays the role to shrink the amplitude of the output weight matrix produced by the former to further improve the generalization ability of the HRN. Therefore, the M-ESN produces the sparse output weight matrix, enjoys the oracle property and has well generalization ability for the nonlinear chaotic time series. Besides, considering that the random input weight matrix may affect the whole system performance, we have investigated the plasticity property of the neuron and developed a fine-tuning strategy of the input weight matrix based on the Bienenstock, Cooper, Munro (BCM) theory. The input weight matrix in M-ESN is fine-tuned in the unsupervised training process. We introduce and explain its calculation process in detail. Finally, we evaluate the M-ESN by the time series of the classical nonlinear Lorenz chaotic system, the time series of the nonlinear Lv chaotic system, the time series of the Hindmarsh–Rose neuron model and the time series of the rainfall series in Hefei city in past 50 years. The simulation results indicate the M-ESN has well one-step and multistep prediction performances for nonlinear chaotic time series. [ABSTRACT FROM AUTHOR]

The IEEE 802.11ah standard is designed to overcome the limited transmission range in conventional wireless local area network (WLAN) technologies and connect hundreds and thousands of sensors in the Internet of Things (IoT). Therefore, as a promising communication technology, it will play an important role in the IoT in the future. In this paper, to support the adaptive transmissions in the IoT, we focus on the channel prediction issue based on the IEEE 802.11ah standard. In particular, we introduce an adaptive echo state network (AESN) to predict the channel state information (CSI) in the orthogonal frequency division multiplexing (OFDM) systems of the IEEE 802.11ah standard. The AESN includes a basic echo state network (ESN) model and an adaptive elastic network. Thereinto, the adaptive elastic network is utilized to estimate the output weight matrix in the ESN. Therefore, the AESN produces the well sparse output weight matrix, enjoys the oracle property, and offers the excellent prediction performance. In the simulation section, an outdoor scenario with the Rayleigh channel and an indoor scenario with the Nakagami-m channel are considered based on the OFDM systems of the IEEE 802.11ah technology. In the simulations, we firstly use the recurrence quantification analysis (RQA) to explore the local predictability of the CSI samples. Then, the AESN is further evaluated in the one-step prediction, the multistep prediction and the robustness test. The results indicate that by using the AESN, we can effectively predict the CSI to support the adaptive transmissions in the IoT. [ABSTRACT FROM AUTHOR]

Developing fault detection is very important for improving the equipment reliability and saving energy consumption. As compared with neural networks, extreme learning machine (ELM) is based on statistical learning theory, which has advantages of better classification ability and generalization performance. This paper presents a novel approach for fault detection and diagnosis based on Firefly-Chaos Algorithm and Extreme Learning Machine. The experiment result indicates this proposed method is effective for Analog Circuit fault detection and diagnosis and the model generalization ability is favorable. [ABSTRACT FROM AUTHOR]