{"@type": "dcat:Dataset", "accessLevel": "public", "accrualPeriodicity": "irregular", "bureauCode": ["026:00"], "contactPoint": {"@type": "vcard:Contact", "fn": "SCOTT POLL", "hasEmail": "mailto:scott.d.poll@nasa.gov"}, "description": "Many diagnostic datasets suffer from the adverse\r\neffects of spikes that are embedded in data and noise. For\r\nexample, this is true for electrical power system data where\r\nthe switches, relays, and inverters are major contributors to\r\nthese effects. Spikes are mostly harmful to the analysis of\r\ndata in that they throw off real-time detection of abnormal\r\nconditions, and classification of faults. Since noise and\r\nspikes are mixed together and embedded within the data,\r\nremoval of the unwanted signals from the data is not always\r\neasy and may result in losing the integrity of the\r\ninformation carried by the data. Additionally, in some\r\napplications noise and spikes need to be filtered\r\nindependently. The proposed algorithm is a multi-resolution\r\nfiltering approach based on Haar wavelets that is capable of\r\nremoving spikes while incurring insignificant damage to\r\nother data. In particular, noise in the data, which is a useful\r\nindicator that a sensor is healthy and not stuck, can be\r\npreserved using our approach. Presented here is the\r\ntheoretical background with some examples from a realistic\r\ntestbed.", "distribution": [{"@type": "dcat:Distribution", "description": "2010IEEE_Sheybani_RemovingSpikesWaveletFilter.pdf", "downloadURL": "https://c3.nasa.gov/dashlink/static/media/publication/2010IEEE_Sheybani_RemovingSpikesWaveletFilter.pdf", "format": "PDF", "mediaType": "application/pdf", "title": "2010IEEE_Sheybani_RemovingSpikesWaveletFilter.pdf"}], "identifier": "DASHLINK_871", "issued": "2013-12-19", "keyword": ["ames", "dashlink", "nasa"], "landingPage": "https://c3.nasa.gov/dashlink/resources/871/", "modified": "2025-04-01", "programCode": ["026:029"], "publisher": {"@type": "org:Organization", "name": "Dashlink"}, "title": "Removing Spikes While Preserving Data and Noise using Wavelet Filter Banks"}