Xiao Xian-Ci

Cyclic spectral features based modulation recognition

Modulation recognition of an intercepted communication signal is a fundamental problem of electromagnetic signal monitoring task arising in many fields, such as electronic surveillance and broadcasting control. A cyclic spectral features based neural network modulation recognition method is proposed. Because of the use of cyclic spectral features and the application of neural network classifier, the proposed method can efficiently recognize almost all currently used modulation types. Some computer simulation results are also reported.

Neural Volterra filter for chaotic time series prediction

A new second-order neural Volterra filter (SONVF) with conjugate gradient (CG) algorithm is proposed to predict chaotic time series based on phase space delay-coordinate reconstruction of chaotic dynamics system in this paper, where the neuron activation functions are introduced to constraint Volterra series terms for improving the nonlinear approximation of second-order Volterra filter (SOVF). The SONVF with CG algorithm improves the accuracy of prediction without increasing the computation complexity. Meanwhile, the difficulty of neuron number determination does not exist here. Experimental results show that the proposed filter can predict chaotic time series effectively and one-step and multi-step prediction performances are obviously superior to those of SOVF, which demonstrate that the proposed SONVF is feasible and effective.

Predicting Hyper-Chaotic Time Series Using Adaptive Higher-Order Nonlinear Filter

A newly proposed method, i.e. the adaptive higher-order nonlinear finite impulse response (HONFIR) filter based on higher-order sparse Volterra series expansions, is introduced to predict hyper-chaotic time series. The effectiveness of using the adaptive HONFIR filter for making one-step and multi-step predictions is tested based on very few data points by computer-generated hyper-chaotic time series including the Mackey-Glass equation and four-dimensional nonlinear dynamical system. A comparison is made with some neural networks for predicting the Mackey-Glass hyper-chaotic time series. Numerical simulation results show that the adaptive HONFIR filter proposed here is a very powerful tool for making prediction of hyper-chaotic time series.

AR modeling-based features extraction of multiple signals for modulation recognition

Modulation recognition of multiple signals is a new topic in the communication signal processing area. A novel feature-based modulation recognition method for multiple signals is proposed. The main idea of this method is that AR modeling is employed to extract instantaneous frequency and bandwidth features of each signal to be recognized from the mixed waveform intercepted by a receiver, and then a MLP neural network classifier is followed to identify the modulation types. To assess the efficiency of these features, computer simulations are also performed.

The fourth-order cumulants based spectral estimation method and its application to direction-finding

Traditional approaches of spatial spectral estimation are usually based on the second-order statistics. The higher-order cumulants and the poly-spectrum contain more information and are capable of reducing the Gaussian noise. In this paper, we present a new spectrum estimation method for direction-finding, the FOMUSIC algorithm, which is based on the eigen-structure analysis of the fourth-order cumulants. The derivation of the algorithm is given in detail and its performance is illustrated by both the computer simulations and the experiments of a direction-finding system. The obtained results demonstrate that the fourth-order cumulants based method outperforms the traditional methods, especially when the noise is an unknown colored one.

Local discrete cosine transformation domain Volterra prediction of chaotic time series

In this paper a local discrete cosine transformation (DCT) domain Volterra prediction method is proposed to predict chaotic time series, where the DCT is used to lessen the complexity of solving the coefficient matrix. Numerical simulation results show that the proposed prediction method can effectively predict chaotic time series and improve the prediction accuracy compared with the traditional local linear prediction methods.

Study on fractal features of modulation signals

Based on fractal theory, the note presents a novel method of modulation signals classification that adopts box dimension and information dimension extracted from received signals as features of classification. These features contain the characteristics of magnitude, frequency and phase of signals, and collect discriminatory information among various modulation modes. They are effective features in classification sense, and are insensitive to noises interfering. The theoretical analysis also proves the above conclusion. The classifier design is very simple based on such features. The simulation results show that the performances of signal classification are superior.

Second-order sampling-based fast recovery of bandpass signals

Fast recovery and frequency-shifting of real bandpass signal based on second-order sampling is discussed. Using FFT and complex filtering, a real band-pass signal can be recovered as an analytic signal whose central frequency is within two times its bandwidth, and phase property is unchanged. Lastly, using simple computation, the original frequency can be acquired. Computer simulation shows the correctness of the method.

Individual communication transmitter identification based on multifractal analysis

In this letter, the communication transmitter transient signals are analyzed based on the time-variant hierarchy exponents of multifractal analysis. The species of optimized sample set is selected as the template of transmitter identification, so that the individual communication transmitter identification can be realized. The turn-on signals of four transmitters are used in the simulation. The experimental results show that the multifractal character of transmitter transient signals is an effective character of individual transmitter identification.

Noise reduction method based on weighted manifold decomposition

A noise reduction method based on weighted manifold decomposition is proposed in this paper, which does not need knowledge of the chaotic dynamics and choosing number of eigenvalues. The simulation indicates that the performance of this method can increase the signal-to-noise ratio of noisy chaotic time series.