Coding an Dynamic Date Interval that identify duplicates in rows

I have a Table with Three Columns:

SUBSCRIPTION_ID      SERVREQ_HEADER_ID       ACTIVITY_DT
23399505      292601967            27/07/2016
2808783            286971613            28/05/2016
555283            280376335            06/03/2016
23399505      238892180            30/09/2014
555283            280376158            06/03/2016
2808783            263311971            12/08/2015
555283            280375788            06/03/2016
320899            337109024            18/08/2017
320899            337741717            23/08/2017
320899            337461638            21/08/2017
320899            338246919            27/08/2017
2808783            365329629            13/04/2018
2808783            365329641            13/04/2018

This is an event table with SERVREQ_HEADER_ID being the unique indicator

1) I need dynamic code that looks at an SERVREQ_HEADER_ID  event for a particular subscription_id then looks 30 days back  and any records prior to that event adds a duplicate indicator field.

2) Also retrieve the earliest date in that 30 day period and updates the SERVREQ_HEADER_ID  event with a the earliest date in the 30 day interval

Would appreciate any Oracle SQL or PLSQL code help for this.

Many Thanks!!