Classification of RF Signal-Based Indoor Positioning Technologies

Indoor positioning technologies are broadly classified into categories that utilize RF (Radio Frequency) signals, motion sensors, and imaging technologies. Among these, RF signal-based methods are favored for tracking a large number of assets due to their wide applicability and low power consumption requirements. (For more detailed information on the types of indoor positioning technologies, please refer to the [previous article].)

RF signal methods are further categorized based on the use of fixed reference points, known as anchors. Anchor-based methods, such as BLE and UWB, provide relatively high accuracy but necessitate substantial investment in infrastructure setup and involve complex maintenance. Conversely, anchor-free methods offer savings on infrastructure costs but generally deliver lower accuracy and require complex initial configurations. Fingerprinting technology is a typical example of this method, involving the setting of Points of Interest (POI*) and comparing collected signal patterns with current data.

*POI: Point of Interest

IPIN LABS’ Indoor Positioning Technology – Core AI

Each indoor positioning technology presents distinct advantages and disadvantages. Therefore, adopting a hybrid approach that combines the strengths of various technologies is advantageous, as it delivers stable and efficient results across different environments. This innovative approach is referred to as the hybrid method. IPIN LABS has enhanced the accuracy of its indoor positioning solutions through a hybrid approach that combines the cost benefits of anchor-free methods with data analysis from inertial measurement units (IMU), successfully overcoming the limitations of existing technologies.

Maximizing Efficiency and Preventing Accuracy Loss with AI Algorithms

Existing indoor positioning technologies, such as anchor-based (BLE, UWB) and fingerprinting methods, face significant limitations due to their inefficiency in deployment and maintenance. The primary challenge involves reducing the costs and time associated with anchor installation, management, or setting Points of Interest (POI) and data collection. To address this issue, IPIN LABS has adopted an innovative approach using AI, which integrates two core strategies:

1. A deep learning algorithm that broadly utilizes various RF signals without specific anchors, thus reducing infrastructure costs and management tasks.

2. An AI-IMU technology that simplifies the manual pre-data collection process, typically aligned with POI setup, into mere walking, significantly shortening deployment time.

Creating RF SLAM with Self-supervised AI Techniques

SLAM (Simultaneous Localization And Mapping) is a technology where artificial intelligence determines its location (Localization) based on surrounding data while simultaneously constructing a map (Mapping). This process enhances accuracy by mutually reinforcing location measurement and map creation, similar to how a person navigates and understands their position in a new environment by observing their surroundings.

IPIN LABS’ self-supervised AI algorithm learns from various surrounding RF signals (Wi-Fi, BLE) and IMU data, creating an RF SLAM capable of precise location measurement and path tracking. This technology ensures high application efficiency and accuracy across different environmental conditions.

IPIN LABS’ Positioning Technology vs Existing Positioning Technologies

Successfully implementing an indoor positioning system hinges on selecting the right technology and optimizing solution operations. IPIN LABS supports companies in enhancing productivity through accurate and efficient location tracking, employing positioning technology that meets both accuracy and efficiency criteria.

Surpassing Existing Limitations with IPIN LABS’ AI-Based Technology

IPIN LABS’ AI-based indoor positioning technology emerges as an ideal solution for commercializing indoor positioning systems, demonstrating cost efficiency, accuracy, and scalability suitable for various environments. The AI-created RF SLAM successfully addresses and overcomes significant challenges associated with existing technologies, such as high infrastructure costs and complex initial setup requirements. This breakthrough provides companies with a new and efficient pathway to adopt and operate indoor positioning solutions, marking a pivotal advancement in the field.