Evidence of Task-Independent Person-Specific Signatures in EEG Using Subspace Techniques
Electroencephalography (EEG) signals are promising as alternatives to other biometrics owing to their protection against spoofing. Previous studies have focused on capturing individual variability by analyzing task/condition-specific EEG. This work attempts to model biometric signatures independent...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
Institute of Electrical and Electronics Engineers (IEEE)
2021
|
| Online Access: | https://hdl.handle.net/1721.1/138333 |
| _version_ | 1811085897865625600 |
|---|---|
| author | Kumar, Mari Ganesh Narayanan, Shrikanth Sur, Mriganka Murthy, Hema A |
| author2 | Massachusetts Institute of Technology. Department of Brain and Cognitive Sciences |
| author_facet | Massachusetts Institute of Technology. Department of Brain and Cognitive Sciences Kumar, Mari Ganesh Narayanan, Shrikanth Sur, Mriganka Murthy, Hema A |
| author_sort | Kumar, Mari Ganesh |
| collection | MIT |
| description | Electroencephalography (EEG) signals are promising as alternatives to other biometrics owing to their protection against spoofing. Previous studies have focused on capturing individual variability by analyzing task/condition-specific EEG. This work attempts to model biometric signatures independent of task/condition by normalizing the associated variance. Toward this goal, the paper extends ideas from subspace-based text-independent speaker recognition and proposes novel modifications for modeling multi-channel EEG data. The proposed techniques assume that biometric information is present in the entire EEG signal and accumulate statistics across time in a high dimensional space. These high dimensional statistics are then projected to a lower dimensional space where the biometric information is preserved. The lower dimensional embeddings obtained using the proposed approach are shown to be task-independent. The best subspace system identifies individuals with accuracies of 86.4% and 35.9% on datasets with 30 and 920 subjects, respectively, using just nine EEG channels. The paper also provides insights into the subspace model's scalability to unseen tasks and individuals during training and the number of channels needed for subspace modeling. |
| first_indexed | 2024-09-23T13:17:15Z |
| format | Article |
| id | mit-1721.1/138333 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T13:17:15Z |
| publishDate | 2021 |
| publisher | Institute of Electrical and Electronics Engineers (IEEE) |
| record_format | dspace |
| spelling | mit-1721.1/1383332023-06-22T13:38:09Z Evidence of Task-Independent Person-Specific Signatures in EEG Using Subspace Techniques Kumar, Mari Ganesh Narayanan, Shrikanth Sur, Mriganka Murthy, Hema A Massachusetts Institute of Technology. Department of Brain and Cognitive Sciences Electroencephalography (EEG) signals are promising as alternatives to other biometrics owing to their protection against spoofing. Previous studies have focused on capturing individual variability by analyzing task/condition-specific EEG. This work attempts to model biometric signatures independent of task/condition by normalizing the associated variance. Toward this goal, the paper extends ideas from subspace-based text-independent speaker recognition and proposes novel modifications for modeling multi-channel EEG data. The proposed techniques assume that biometric information is present in the entire EEG signal and accumulate statistics across time in a high dimensional space. These high dimensional statistics are then projected to a lower dimensional space where the biometric information is preserved. The lower dimensional embeddings obtained using the proposed approach are shown to be task-independent. The best subspace system identifies individuals with accuracies of 86.4% and 35.9% on datasets with 30 and 920 subjects, respectively, using just nine EEG channels. The paper also provides insights into the subspace model's scalability to unseen tasks and individuals during training and the number of channels needed for subspace modeling. 2021-12-06T18:35:24Z 2021-12-06T18:35:24Z 2021 2021-12-06T18:30:37Z Article http://purl.org/eprint/type/JournalArticle https://hdl.handle.net/1721.1/138333 Kumar, Mari Ganesh, Narayanan, Shrikanth, Sur, Mriganka and Murthy, Hema A. 2021. "Evidence of Task-Independent Person-Specific Signatures in EEG Using Subspace Techniques." IEEE Transactions on Information Forensics and Security, 16. en 10.1109/TIFS.2021.3067998 IEEE Transactions on Information Forensics and Security Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf Institute of Electrical and Electronics Engineers (IEEE) arXiv |
| spellingShingle | Kumar, Mari Ganesh Narayanan, Shrikanth Sur, Mriganka Murthy, Hema A Evidence of Task-Independent Person-Specific Signatures in EEG Using Subspace Techniques |
| title | Evidence of Task-Independent Person-Specific Signatures in EEG Using Subspace Techniques |
| title_full | Evidence of Task-Independent Person-Specific Signatures in EEG Using Subspace Techniques |
| title_fullStr | Evidence of Task-Independent Person-Specific Signatures in EEG Using Subspace Techniques |
| title_full_unstemmed | Evidence of Task-Independent Person-Specific Signatures in EEG Using Subspace Techniques |
| title_short | Evidence of Task-Independent Person-Specific Signatures in EEG Using Subspace Techniques |
| title_sort | evidence of task independent person specific signatures in eeg using subspace techniques |
| url | https://hdl.handle.net/1721.1/138333 |
| work_keys_str_mv | AT kumarmariganesh evidenceoftaskindependentpersonspecificsignaturesineegusingsubspacetechniques AT narayananshrikanth evidenceoftaskindependentpersonspecificsignaturesineegusingsubspacetechniques AT surmriganka evidenceoftaskindependentpersonspecificsignaturesineegusingsubspacetechniques AT murthyhemaa evidenceoftaskindependentpersonspecificsignaturesineegusingsubspacetechniques |