Online Activity Matching using Wireless Sensor Network

The main objective in this project is to explore the application of Wireless Sensor Networks to carry out online activity matching for sports coaching applications. We desire to identify one or more ideal algorithm to match the movement of a trainer and to find spatial and temporal differences. This type of system can certainly help the trainer in group lessons where it really is challenging for the trainer to maintain a record of all the trainees. In this project, we take into account fitness like movements such as experienced in aerobic training. We restrict ourselves just to one sensor node on the trainer and one sensor node on the trainee, however this well extends to more sensors per trainer and trainee. Additionally, it scales nicely to more trainees per trainer. We have utilized Sun SPOT as wireless sensor nodes, and extend the group of sensors to get additional inertial measurement capabilities. The accelerometer and gyroscope sensor are put to use to get the movements.

Watch a video on What are Wireless Sensor Networks?

Contents

1 Introduction
1.1 Problem Desciption
1.2 Solution Overview
2 Related Work
2.1 Distance based similarity measures
2.2 Correlation based similarity measures
2.3 Coherence based similarity measures
3 System Architecture
4 Preprocessing
4.1 Noise removal
4.2 Gravity detection
4.2.1 Principles of Kalman filters
4.2.2 Gravity from the accelerometer
4.2.3 Angular rate of change approximation
4.2.4 Improved estimation of gravity
5 Segmentation
5.1 Movement detection
5.2 Direction change detection
5.3 Fusion
6 Comparing Movements
6.1 Classic DTW
6.2 Existing Optimizations
6.3 Fast Incremental Dynamic Time Warping Algorithm
6.3.1 Flexible Endpoint
6.3.2 Distance Calculation
6.4 Path Analysis
6.5 Evaluation
7 Implementation
7.1 Hardware
7.2 Software
7.2.1 Overview
7.2.2 Sampling
7.2.4 Segmentation
7.2.5 Wireless Communication…….

Download URL 2: Visit Now

Source: University of Twente

Leave a Comment