A Dual-Encoder-Condensed Convolution Method for High-Precision Indoor Positioning

We study the problem of indoor positioning, which is a fundamental service in managing and analyzing objects environments. Unpredictable signal interference sources increase degeneration accuracy robustness existing solutions. Deep learning approaches have recently been widely studied to overcome these challenges attain better performance. In this paper, we aim develop efficient algorithms, such as dual-encoder-condensed convolution (DECC) method, can achieve high-precision positioning for services. particular, firstly, convolutional module add original channel state information location information. Secondly, explore differences between different antennas, adopt dual-encoder stacking mechanism parallel calculation. Thirdly, two kernels speed up convergence. Performance studies on scenario urban canyon datasets demonstrate efficiency effectiveness our new approach.