{ "id": 255940, "title": "Nyquist Theorem", "url": "/competitions/g2net-gravitational-wave-detection/discussion/255940", "author_name": "", "post_date": "2021-07-29T21:56:40.402972600Z", "votes": 8, "comment_count": 4, "views": 0, "content": "

Please note that the sampling rate of data given is 2048 (2048 data points every second).

\n

Therefore, as per the Nyquist theorem, the max observable frequency in the data will be half of the sampling rate. ie. 1024 Hz or 1.024 KHz

\n

Anything above 1024 Hz would be alias of frequencies within 1 KHz

\n

I just started out on the competition and I saw kernels using 2048 as Fmax for spectrogram plots. I'm not sure how it impacts the plot or any calculations but it should be 1024 instead.

\n

Reference: https://www.youtube.com/watch?v=0slziGiwZOg

", "messages": [ { "id": "1404457", "postDate": "07/29/2021 21:56:40", "content": "

Please note that the sampling rate of data given is 2048 (2048 data points every second).

\n

Therefore, as per the Nyquist theorem, the max observable frequency in the data will be half of the sampling rate. ie. 1024 Hz or 1.024 KHz

\n

Anything above 1024 Hz would be alias of frequencies within 1 KHz

\n

I just started out on the competition and I saw kernels using 2048 as Fmax for spectrogram plots. I'm not sure how it impacts the plot or any calculations but it should be 1024 instead.

\n

Reference: https://www.youtube.com/watch?v=0slziGiwZOg

", "rawMarkdown": "Please note that the sampling rate of data given is 2048 (2048 data points every second).\n\nTherefore, as per the Nyquist theorem, the max observable frequency in the data will be half of the sampling rate. ie. 1024 Hz or 1.024 KHz\n\nAnything above 1024 Hz would be alias of frequencies within 1 KHz\n\nI just started out on the competition and I saw kernels using 2048 as Fmax for spectrogram plots. I'm not sure how it impacts the plot or any calculations but it should be 1024 instead.\n\nReference: https://www.youtube.com/watch?v=0slziGiwZOg", "votes": null }, { "id": "1404680", "postDate": "07/30/2021 06:52:41", "content": "

I agree, according to the Kotelnikov theorem, the maximum frequency with such sampling is also 1024

", "rawMarkdown": "I agree, according to the [Kotelnikov theorem](https://ru.wikipedia.org/wiki/%D0%A2%D0%B5%D0%BE%D1%80%D0%B5%D0%BC%D0%B0_%D0%9A%D0%BE%D1%82%D0%B5%D0%BB%D1%8C%D0%BD%D0%B8%D0%BA%D0%BE%D0%B2%D0%B0), the maximum frequency with such sampling is also 1024", "votes": null }, { "id": "1404772", "postDate": "07/30/2021 08:25:32", "content": "

Some public kernels assume a 4096 sampling rate and 1second length. In that case using fmax = 2048 is fine.

", "rawMarkdown": "Some public kernels assume a 4096 sampling rate and 1second length. In that case using fmax = 2048 is fine.", "votes": null }, { "id": "1404800", "postDate": "07/30/2021 08:45:25", "content": "

It seems to me that information about frequencies above 1024 is already lost there. For example, you have an 8k (sampling) audio recording. Even if you open it with 64k sampling, frequencies above 4k will not appear there.

", "rawMarkdown": "It seems to me that **information** about frequencies above 1024 is already **lost** there. For example, you have an 8k (sampling) audio recording. Even if you open it with 64k sampling, frequencies above 4k will not appear there.", "votes": null }, { "id": "1561287", "postDate": "10/27/2021 13:12:27", "content": "

Hey All,

\n

Thank you all for taking part in our competition. The participation has been overwhelmingly positive. We are currently conducting a survey to gauge the demographic and outreach achieved. Kindly spare 2min and fill in this survey https://forms.gle/QP9L16niPexozyhu5.

\n

Thank you all,

\n

Regards,
\nChris

", "rawMarkdown": "Hey All,\n\nThank you all for taking part in our competition. The participation has been overwhelmingly positive. We are currently conducting a survey to gauge the demographic and outreach achieved. Kindly spare 2min and fill in this survey https://forms.gle/QP9L16niPexozyhu5.\n\nThank you all,\n\nRegards,\nChris", "votes": null } ], "comments": [ { "id": 1404680, "author_name": "sapr3s", "author_url": "", "post_date": "07/30/2021 06:52:41", "content": "

I agree, according to the Kotelnikov theorem, the maximum frequency with such sampling is also 1024

", "votes": null, "replies": [] }, { "id": 1404772, "author_name": "cpmpml", "author_url": "", "post_date": "07/30/2021 08:25:32", "content": "

Some public kernels assume a 4096 sampling rate and 1second length. In that case using fmax = 2048 is fine.

", "votes": null, "replies": [ { "id": 1404800, "author_name": "sapr3s", "author_url": "", "post_date": "07/30/2021 08:45:25", "content": "

It seems to me that information about frequencies above 1024 is already lost there. For example, you have an 8k (sampling) audio recording. Even if you open it with 64k sampling, frequencies above 4k will not appear there.

", "votes": null, "replies": [] } ] }, { "id": 1561287, "author_name": "zerafachris", "author_url": "", "post_date": "10/27/2021 13:12:27", "content": "

Hey All,

\n

Thank you all for taking part in our competition. The participation has been overwhelmingly positive. We are currently conducting a survey to gauge the demographic and outreach achieved. Kindly spare 2min and fill in this survey https://forms.gle/QP9L16niPexozyhu5.

\n

Thank you all,

\n

Regards,
\nChris

", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "1404457": "Please note that the sampling rate of data given is 2048 (2048 data points every second).\n\nTherefore, as per the Nyquist theorem, the max observable frequency in the data will be half of the sampling rate. ie. 1024 Hz or 1.024 KHz\n\nAnything above 1024 Hz would be alias of frequencies within 1 KHz\n\nI just started out on the competition and I saw kernels using 2048 as Fmax for spectrogram plots. I'm not sure how it impacts the plot or any calculations but it should be 1024 instead.\n\nReference: https://www.youtube.com/watch?v=0slziGiwZOg", "1404680": "I agree, according to the [Kotelnikov theorem](https://ru.wikipedia.org/wiki/%D0%A2%D0%B5%D0%BE%D1%80%D0%B5%D0%BC%D0%B0_%D0%9A%D0%BE%D1%82%D0%B5%D0%BB%D1%8C%D0%BD%D0%B8%D0%BA%D0%BE%D0%B2%D0%B0), the maximum frequency with such sampling is also 1024", "1404772": "Some public kernels assume a 4096 sampling rate and 1second length. In that case using fmax = 2048 is fine.", "1404800": "It seems to me that **information** about frequencies above 1024 is already **lost** there. For example, you have an 8k (sampling) audio recording. Even if you open it with 64k sampling, frequencies above 4k will not appear there.", "1561287": "Hey All,\n\nThank you all for taking part in our competition. The participation has been overwhelmingly positive. We are currently conducting a survey to gauge the demographic and outreach achieved. Kindly spare 2min and fill in this survey https://forms.gle/QP9L16niPexozyhu5.\n\nThank you all,\n\nRegards,\nChris" }, "source": "meta" }