In the setting of lightweight IoT devices and Wireless Sensor Networks (WSNs), a (t, n) Secret Sharing (SS) scheme has been proposed to distribute secrets over n shares where n is the number of disjoint paths between the sender and the receiver and a threshold value t is the number of shares required to reconstruct the original secret. However, studies in NDSS’21 and SPACE’22 have shown that all SS schemes are vulnerable to Network Data Remanence (NDR) attacks under practical scenarios. In this work, we propose a countermeasure to address the vulnerability of SS schemes to NDR attacks. The proposed solution is a Physically Related Function (PReF) based Threshold Changeable SS-Scheme that allows for the threshold value to be modified within a range of integers [t, u] such that \(t \le u \) and \( u \le n \) without requiring the shares to be updated. The PReF can be thought of as a representation or model of Strong Physically Unclonable Functions (PUFs) in a way that abstracts away some of the specific physical details. The modified threshold value is securely transmitted via Physically Related Functions (PReF). Further in the work, we have presented the experimental results of the proposed scheme, where the scheme has been implemented by replicating two Raspberry-pi devices one as a sender and the other as a receiver, and both are connected in a wireless network. The results demonstrate that the probability of having an NDR-planned attack on this new scheme is significantly reduced by a factor of \(\frac{1}{n}\) as compared with the basic SS-scheme, where n is the number of disjoint paths available between the sender and the receiver.