A source location privacy protocol‐based energy‐efficient and link‐reliable multi‐scale bifurcated deep Capsnet routing in social Internet of Things

Summary Source location privacy is a developing research topic in the social Internet of Things. Source location privacy holds paramount importance in security critical wireless sensor network applications like tracking and monitoring. Several methods have been proposed for source location privacy in the social Internet of Things, but the existing methods have some issues such as improper path selection, the transmission of duplicate messages, and low network lifetime. To overcome these issues, a source location privacy protocol based on energy‐efficient and link‐reliable multi‐scale bifurcated deep Capsnet routing in the social Internet of Things is proposed in this manuscript. At first, the optimal route for the source is selected with the help of energy‐efficient and link‐reliable routing, this method helps to avoid improper path selection. To estimate the quality of the selected optimal path, the multi‐scale bifurcated deep Capsnet is applied. The introduced method is executed in MATLAB. The introduced method's performance is estimated with the aid of several performances evaluating metrics like sensitivity, energy consumption, network lifetime, safety period, and delay. The “energy‐efficient and link‐reliable multi‐scale bifurcated deep capsule network” (E2MSDcaps) introduces a novel approach for SLP in SIoT. It identifies optimal paths and also incorporates a mechanism for accurately evaluating path quality. Utilizing the energy‐efficient and link‐reliable routing (E2LR) technique enhances SLP by selecting optimal paths, reducing duplicate messages (DM), and maintaining network balance. A novel technique, multi‐scale bifurcated deep capsule networks (MSDcaps), is presented to precisely assess the quality of the selected pathway, categorizing it as either excellent or poor.