Multiple bits distributed moving horizon state estimation for wireless sensor networks

Wireless sensor networks distributed signal processing has to operate under stringent energy and bandwidth limitations. Usually, the observations need to be quantized before they are sent to fusion center (FC). This paper we introduce a multiple bits distributed moving Horizon state estimation approach based on a multiple state single output Gaussian-Markov model. Each sensor node has a thresholds book, then the sensors quantize their measurements to several bits by using the thresholds book which are sent to fusion center. After receiving these information, fusion center make final estimation for system states. Compared with single bit distributed moving horizon state estimation (DMHE), this method ensures that FC need not to send the estimate information back to sensor nodes and provides higher precision of state estimation.