Abstract
As wireless network technology and mobile processor performance improve, small wireless mobile devices like smartphones have become ubiquitous. These devices often leverage GPS for location-based services; however, GPS fails in indoor environments or signal-shaded areas. While IEEE 802.11 (Wi-Fi) based tracking systems can utilize existing infrastructure for indoor positioning, standard Wi-Fi protocols suffer from high power consumption. This paper proposes a low-power tracking system that minimizes energy usage during channel scanning and connection maintenance. We designed and implemented a tracking tag device and measured its performance, demonstrating a 46% reduction in power consumption compared to standard execution.
Introduction
The demand for indoor positioning has grown alongside the proliferation of mobile devices. While GPS is effective outdoors, its signals cannot penetrate buildings. Alternative short-range wireless technologies often require the installation of dedicated infrastructure. Wi-Fi-based tracking is a compelling alternative because Access Points (APs) are already widely deployed. However, the standard IEEE 802.11 operation is not optimized for battery-powered tracking tags. Specifically, the processes of sequential channel scanning and constant connection maintenance (handoffs) are energy-intensive.
The "Energy Tax" of Standard Wi-Fi
Standard Wi-Fi tracking devices waste significant energy on two main tasks:
- Sequential Scanning: Searching all 13 channels in the 2.4GHz band, regardless of whether APs are present.
- Connection Maintenance: Performing frequent handoffs and listening for beacons to maintain a persistent link, which is often unnecessary for periodic location reporting.
As shown in the figure above, the current consumption spikes significantly during channel scanning and handoff events, reaching up to 125mA. These spikes represent the primary targets for optimization in a low-power design.
Proposed Low-Power Methods
To address these inefficiencies, the researchers proposed a two-pronged strategy:
1. Adaptive Channel Scanning
Instead of a blind search, the system maintains an AP List Table that tracks the SSID, MAC address, and "Age" of detected APs. Channels are assigned weights based on how recently and frequently APs were found there. The device prioritizes scanning high-weight channels and stops immediately once enough APs (typically 3 to 5) are found for trilateration, avoiding unnecessary energy expenditure on empty channels.
2. Low-Power Operation Model
The communication flow was redesigned to favor "Sleep" and "Deep Sleep" states. Rather than maintaining a constant connection and performing energy-intensive handoffs as the device moves, the tag only connects to an AP when it needs to transmit location data.
The mathematical models above illustrate how the proposed method removes the Handoff x N term and simplifies the WiFi_Cycle by replacing beacon listening with a more efficient association process.
Implementation and Evaluation
The prototype was built using the GS1011MIE low-power Wi-Fi module, which supports standby modes as low as 4.7µA. The system was evaluated in both a physical testbed (8 APs in an L-shaped indoor area) and a simulation environment (20 APs over a 280m x 210m grid).
Key Results
- Power Efficiency: Simulation results confirmed a 46.4% reduction in total power consumption compared to standard IEEE 802.11 execution.
- Scanning Optimization: In environments with 3 APs, the adaptive method reduced the number of channel scans by 76% compared to a full scan.
- Extended Battery Life: For a device requiring 5 APs for a fix, the predicted operating time on a 370mAh battery increased from roughly 5.5 hours to over 8 hours using the proposed method.
Conclusion
By optimizing channel scanning and eliminating unnecessary connection maintenance, this research demonstrates a Wi-Fi tracking system that significantly extends the battery life of mobile tags. This approach allows for effective indoor positioning using existing infrastructure without the prohibitive energy costs of standard Wi-Fi protocols. Future work will explore the application of these techniques to the 5GHz and 6GHz bands, where the number of channels is even greater.
Summary updated