Efficient Restricted-Access Building Drawing Regeneration from Passive Radio Localization

Wireless networks have become a vital factor in our modern life. However, besides providing connectivity for many civil applications, wireless networks are the subjects of radio localization and user tracking. This work introduces a novel technique that exploits passive radio localization on multiple users for regenerating a restricted-access building drawing. Initially, from the data points generated by our built-in signal-to-image-based localization, we generate data points in an empty building drawing. Then by estimating the absolute distance among nearby data points, we group data points by a rectangle which represents the accessible area by human walk. Finally, we connect the center points of the generated rectangles to illustrate the accessible areas. The evaluation results show that the overall accessible area regeneration accuracy can reach up to 80%. Our technique can apply to many potential applications. For example, the police can safely determine the entrance/exit routes or walkable places to intrude in the drug raid mission. Another case is to expose the common visit places of the criminals or reveal the functions of an area in the restricted-access building.

I. Introduction

Massive MIMO and directional transmission techniques will become the main-stream antenna design and data transmission technologies in 5G and beyond [1], [2]. It enables high-accuracy localization by training on beam signal space and ex-ploiting spatial-temporal signal processing. These techniques can come with several applications, including rescuing victims from disaster trapped inside the wreckage and locating escaped criminals or wanted persons. Indoor localization via wireless signal has been an important and popular research topic [1]–[4]. One of the common methods is to use the relative location between two devices via peer-to-peer feedback to increase localization accuracy [5]. However, that approach may cause significant errors if there are errors in communications or signals are too weak. To deal with large errors under dynamic scenarios, machine learning is the prominent choice. However, this technique often requires large training data [6], [7], which is challenging to acquire under unknown-structure buildings. Meanwhile, ground truth data collection might be insufficient if the user density is too sparse. Recently, several studies explored a signal-based shape drawing approach. For example, the authors in [8] propose to compute the discrete average of multiple shapes by implementing progressive meta-morphosis pixel adding and suppressing. However, this approach can make huge errors if the image is distorted with noise pixels. In order to eliminate the noise interference, attention-based real image restoration is proposed in [9]. However, the approach requires lots of training data with distorted images, which is difficult to collect when the drawing of the building is unknown. Fig. 1.

By exploiting passive radio localization, an adversary can acquire the raw drawing of a restricted-access building without physical intrusion, e.g., entrance/exit routes, and walkable areas.