Radar for Indoor Monitoring by Moeness Amin

Author:Moeness Amin
Language: eng
Format: epub
Publisher: CRC Press LLC

FIGURE 8.9

Example of spectrograms for different movements performed by the same subject to compare monostatic and bistatic data: picking up an object—monostatic (a) and bistatic (b); walking—monostatic (c) and bistatic (d).

FIGURE 8.10

Feature space plot for samples related to different actions: (a) monostatic data; (b) bistatic data.

FIGURE 8.11

Feature space plot for samples related to two actions (picking up an object and standing and sitting) performed by two subjects: (a) monostatic data; (b) bistatic data.

An NB classifier was applied to the features extracted to evaluate the success rate of the monostatic configuration in comparison with the bistatic result. For the monostatic classification case, the features from the four different actions using empirically extracted features provided a classification success rate of 71% using 60% training. The training set was greater in this example, but the challenge is a four-class problem resulting in a greater potential for misclassifying each result. In contrast, the bistatic result from the same experiments was found to have a success rate of 88% with the same level of training. This increase is significant and clearly shows some of the advantages of using a bistatic solution.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.