Proximity Monitoring Using RSSI Temporal Features
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing proximity monitoring systems using RSSI-based solutions for contact tracing are less accurate due to reliance on spatial features alone, lacking effective handling of temporal variations and environmental factors, which hampers real-time social distancing enforcement and contact tracing efficiency.
Innovation Solution
A system that utilizes a machine learning approach combining Received Signal Strength Indicator (RSSI) with temporal features to classify proximity between devices, implementing decision aggregation and unique alert mechanisms for real-time social distancing reminders, and integrating with a server framework for enhanced accuracy and false positive reduction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If RSSI-based solutions are used for proximity monitoring, then the system can operate with simple hardware, but the accuracy of proximity detection deteriorates due to inability to handle temporal variations and environmental factors
Solution Approach 1:
The patent transitions from using only spatial features (RSSI values at single points) to incorporating temporal dimensions by collecting RSSI values over multiple time instances. This temporal dimension allows the system to distinguish between real proximity changes and transient signal variations, thereby improving accuracy without adding complex hardware.
Solution Approach 2:
The system performs preliminary actions by collecting and storing RSSI values over a predetermined time window before making proximity decisions. This preliminary data collection enables the system to later analyze temporal patterns and make more accurate proximity determinations, resolving the accuracy issue while maintaining simple hardware architecture.
2Ease of manufacture
If RSSI-based proximity monitoring is implemented, then the system can be deployed easily, but false positive rate increases due to inability to distinguish real proximity from transient signal variations
Solution Approach 1:
The patent implements periodic action by continuously collecting RSSI values at multiple time instances within a predetermined time window. This periodic sampling allows the system to observe temporal patterns and distinguish between genuine proximity events and transient signal variations, thereby reducing false positives while maintaining easy deployment through simple periodic measurements.
Solution Approach 2:
The system uses feedback mechanisms by comparing current RSSI measurements with historical data from the time window. This feedback loop enables the system to learn from past signal patterns and make more reliable proximity determinations, reducing false positives without complicating the deployment process.
3Measurement precision
If temporal features are added to RSSI analysis, then proximity detection accuracy improves, but computational complexity increases
Solution Approach 1:
The patent applies partial action by collecting RSSI values at a limited number of predetermined time instances within a time window, rather than continuously or excessively sampling. This partial temporal sampling provides sufficient temporal information for accurate proximity detection while keeping computational complexity manageable, resolving the contradiction between accuracy improvement and computational cost.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The system provides real-time alerts and improved accuracy in proximity monitoring, reducing false positives and enhancing user adherence to social distancing norms, thereby aiding in infection control and contact tracing.
Implementation Method 1
one or more data packets are transmitted and received, in real-time, from one or more secondary computing devices
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
Conventionally, Received Signal Strength Indicator (RSSI)-based solutions have been extensively devised in the domains of indoor localization and context-aware applications. These solutions are primarily based on a path-loss attenuation model, with customizations on RSSI processing and are usually regression-based. Further, existing solutions for distance and proximity estimation incorporate data features derived only from the RSSI values themselves with additional features like frequency of occurrence of certain RSSI values thus are less accurate. Present disclosure provides systems and methods that implement a classification model that uses RSSI as well as temporal features derived from the received data packets. The model uses data from multiple devices in different environments for training and can execute proximity decisions on the device itself. The method of the present disclosure monitoring proximity between a plurality of devices implements/uses an effective protocol for decision aggregation to suppress false positive proximity events generated and further stabilizes device's response.