Wireless Sensor Inertial Streaming Without Acknowledgment Resets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Wireless sensor systems face data loss and inaccuracies due to packet loss during transmission between sensor units and master units, especially when processing inertial data, which affects the calculation of integral quantities like orientation and velocity, leading to errors and instability in sensor fusion algorithms.
Innovation Solution
Implementing an inertial integration transmitter at the sensor unit that continuously integrates and transmits orientation and velocity data without resetting integration intervals, allowing the master unit to reconstruct original increments using a dedicated receiver architecture, and eliminating the need for acknowledgments to reduce processing and record-keeping requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If the sensor unit continuously integrates and transmits inertial data without resetting integration intervals, then processing demands on the sensor unit are reduced, but reliability of data transmission is worsened due to packet loss
Solution Approach 1:
The patent inverts the traditional approach by having the master unit (receiver) perform the integration of inertial data rather than the sensor unit (transmitter). The sensor unit simply transmits raw acceleration and angular velocity data without performing integration, while the master unit integrates the received data to compute orientation, velocity, and position. This inversion reduces processing demands on the battery-powered sensor unit while maintaining data reliability through the continuous transmission approach.
2Productivity
If acknowledgments are eliminated to reduce processing requirements, then processing and record-keeping loads are reduced, but data loss is increased due to inability to detect lost packets
Solution Approach 1:
The patent applies preliminary action by having the sensor unit continuously integrate inertial data in advance and store the integrated values locally before transmission. When packets are lost during transmission, the master unit can request retransmission of specific time intervals, and the sensor unit can provide the pre-computed integrated values from its local storage, ensuring data completeness without requiring complex acknowledgment protocols for every packet.
3Measurement precision
If integration intervals are reset at each acknowledgment to ensure data accuracy, then measurement precision is improved, but device complexity increases due to record-keeping requirements
Solution Approach 1:
The patent extracts the complex integration and record-keeping functions from the sensor unit and relocates them to the master unit. The sensor unit's role is simplified to merely transmitting raw sensor data and pre-computed integrated values when requested. The master unit performs the integration of acceleration and angular velocity data, maintains the integration intervals, and manages all record-keeping, thereby achieving measurement precision without imposing record-keeping complexity on the battery-powered sensor unit.
4Adaptability or versatility
If sensor fusion algorithms are run at the application processor side to combine inertial data with additional information, then adaptability is improved, but loss of time increases due to wireless transmission requirements
Solution Approach 1:
The patent merges the inertial data stream with additional information sources (map-matching, WiFi, GSM) at the master unit/application processor side. By continuously transmitting inertial data at high rates and performing sensor fusion algorithms at the master unit where multiple data sources are available, the system achieves adaptability through comprehensive data integration while minimizing transmission time through efficient continuous streaming rather than periodic batch transmissions.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A system and method of communicating efficiently and reliably between a sensor unit and a master unit, especially in a wireless system, is provided. The disclosed techniques require only minimal processing by the sensor unit for communications purposes an d as such, the described system and method can be implemented in systems with tight constraints on the sensor unit side in terms of required memory, power consumption, and cost, such as in wearable sensor systems.