Time interval based aggregation on timeseries

Working with Data
, , ,
1

Hi,
I need assistance with performing aggregation on a timeseries collection in MongoDB 6. The collection structure includes fields like

     {
        "Timestamp": {
          "$date": "2024-03-12T11:11:32.000Z"
        },
        "PointId": "12",
        "Value": {
          "$numberLong": "57"
        }
      }

query:

db.timeseriesCollection.aggregate([{
    $match: {
      "Timestamp": { $gte:  ISODate("2023-02-01T00:00:00.000Z"), $lt: ISODate("2025-10-31T00:00:00.000Z") },
      "PointId": "12"
    }
  },
  {
    $group: {
      _id: {
	    PointId: "$PointId",
        Timestamp: "$Timestamp"
      },
      startTimestamp: { $min: "$Timestamp" },
      avg: { $avg: "$Value" },
    }
  },
  {
    $project: {
      _id: 0,
      id: "$_id.PointId",
      intervalStart: "$_id.Timestamp",
      avg: 1,
    }
  },
  {
    $sort: { intervalStart: 1 }
  }
])

on this collection I want to aggregate data within a specified time range, such as from ‘2023-02-02T08:45:00.000Z’ to ‘2025-10-31T00:00:00.000Z’, and group it into intervals of different durations, such as minutely, hourly, daily, weekly, and yearly. For example, if the interval is set to monthly, I expect the output to include timestamps for each month within the range, starting from ‘2023-02-02T08:45:00.000Z’ and followed by
for eg:
2023-02-02T00:00:00.000Z
2023-03-02T00:00:00.000Z
2023-04-02T00:00:00.000Z
…
daily:
2023-02-02T08:45:00.000Z’
2023-02-03T00:00:00.000Z
2023-02-04T00:00:00.000Z
2023-02-05T00:00:00.000Z …

weekly:
2023-02-02T08:45:00.000Z
2023-02-05T00:00:00.000Z (Sunday as start of the week)
2023-02-12T00:00:00.000Z
2023-02-19T00:00:00.000Z …

I’ve attempted various solutions but haven’t achieved the desired results. Any assistance in crafting the MongoDB aggregation query would be greatly appreciated. Thank you!"