Authors: Kwangkeun Song, Dongbin Oh



The popularity of bike computer devices has grown in recent years. These devices generate a wealth of data in the form of Flexible and Interoperable Data Transfer (FIT) files, which can be used to store fitness related data efficiently. However, the recovery of corrupted FIT files remains a significant challenge due to their inherent structure. The format relies on a chain of messages stored sequentially, with each message referencing previous data to parse the subsequent record. As a result, the recovery of data situated between corrupted portions be- comes notably challenging. This study introduces an efficient, and robust method for dense recovery of corrupted files. Our approach combines multiple phases of data carving techniques to maximize data recovery. By employing this method, investigators can effectively access crucial information including accident reconstruc- tion, and criminal activities. The proposed methods demonstrate higher recovery rate through the proof-of- concept and real-world experiments, proving its utility and reliability in the field of digital forensics.