Total Variation Based Joint Detection and State Estimation for Wireless Communication in Smart Grids

Abstract

A novel total variation (TV) framework is conceived for joint detection and dynamic state estimation (JDSE) for wireless transmission from the measurement devices to the control center in a smart grid. The proposed scheme employs a TV regularization based decoder in conjunction with a Kalman filter-based dynamic power system state estimator to minimize the detection error for transmission of the measurements over a fading wireless channel. A novel application of the Viterbi algorithm is proposed for TV detection of the received measurement vectors. Furthermore, the proposed JDSE scheme is also extended to a system in which each measurement is quantized to a single bit. This reduced-bandwidth-based TV-JDSE leads to a significant bandwidth efficiency improvement in the smart grid and to improved state estimation. Our simulation results provided for the standard IEEE-14 bus test system under different operating conditions demonstrate improved performance in comparison to conventional techniques and they are capable of approaching the ideal Clairvoyant Kalman filter benchmark.

Keywords

• Smart grids
• dynamic state estimation
• Kalman filter
• total variation
• Viterbi algorithm
• wireless communication