Reconstruction of chaotic signals in wireless sensor networks

The problem of chaotic signal reconstruction in wireless sensor networks is studied. The sensors observe a common chaotic signal in a sensing field. The observations are gathered in a fusion center which is responsible for reconstructing the chaotic signal. Due to the bandwidth constraint of sensors, the observations are quantized and the quantized data are sent to the fusion center. The fusion center combines the received data and employs an unscented Kalman filtering (UKF) algorithm to reconstruct the chaotic signal. The results show that this method can recover the chaotic signal effectively and achieve close performance to the benchmark case where the observations are not quantized. The UKF is also compared with the optimal best linear unbiased estimator (BLUE).