AF-DCGAN: Amplitude Feature Deep Convolutional GAN for Fingerprint Construction in Indoor Localization Systems

With widely deployed WiFi network and the uniqueness feature (fingerprint) of wireless channel information, fingerprinting based positioning is currently mainstream indoor method, in which fingerprint database construction crucial. However, for accuracy, this approach requires enough data to be sampled at many reference points, consumes excessive efforts time. In paper, we collect Channel State Information (CSI) points by method device-free localization, then convert collected CSI into amplitude maps extend using proposed Amplitude-Feature Deep Convolutional Generative Adversarial Network (AF-DCGAN) model. The use AF-DCGAN accelerates convergence during training phase, substantially increases diversity map. extended both reduces human effort involved accuracy an localization system, as demonstrated experiments.