Implementation of channel equalization of indoor positioning system based on MB-OFDM

Abstract

Precise indoor positioning is becoming increasingly important in commercial, military and public service applications for tracking people and asset. Ultra Wide Band (UWB) can provide high accuracy position with strong anti-jamming and low power consumption. By combining orthogonal frequency-division multiplexing (OFDM) with multi-band, MB-OFDM systems can capture multi-path energy more efficiently than single-band direct sequence UWB (DS-UWB). MB-OFDM modulation is selected in indoor positioning system based on UWB. Channel estimation and equalization are very important and can not be ignored for indoor positioning system based on UWB. There will be a 3dB loss in signal-to-noise ratio (SNR) approximately under no channel estimation. Through the analysis of architecture, signal and channel model of MB-OFDM indoor positioning systems, we studied the channel estimation techniques based on preamble training sequences and pilot sub-carriers respectively. Further more, the linear estimations of least square (LS) and minimum mean square error (MMSE) are analyzed and compared under different UWB channel conditions. A suitable algorithm is selected and hardware implementation scheme is proposed. Some most significant modules for channel estimation and equalization have been validated with Xilinx Virtex IV FPGA. It has been shown by some modelsim simulations that this channel estimation and equalization approach has some merits, such as simple implementation and less consumption of resource. It can meet the requirement of system design and some good performances can be achieved.