This work is part of a series of papers describing in detail the design and characterization of the LSPE-STRIP, a microwave telescope operating in the Q- and W-bands which is foreseen to be installed at the Observatorio del Teide in Tenerife. The paper aims to describe the Pointing Reconstruction Model (PRM) and the prototype Star Tracker, which will be mounted on LSPE-STRIP. Pointing reconstruction is a crucial step in deriving sky maps of foreground emissions. The PRM will be in charge of integrating the information on the instantaneous attitude provided by the telescope control system, encoded in two control angles, to obtain the actual pointing direction and focal plane orientation of the telescope. The PRM encodes various non-idealities in the telescope setup from eight configuration angles. The Star Tracker, plus the observation of an artificial source installed on a drone and possibly observations of point sources of known positions, will be used to calibrate the configuration angles of the PRM. We study the pointing error produced by incorrectly calibrating configuration angles by comparing surveys with different realizations of systematic pointing errors against the ideal case. In this way, we validated the required ≈ 1 arcmin maximum systematic pointing error in the LSPE-STRIP survey as the worst effect of the pointing error, in this case, is two orders of magnitude below the instrumental sensitivity. After a description of the main structure and operations of the Start Tracker, we present the results of a campaign of actual sky observations carried out on a prototype of the Star Tracker aimed at assessing the final Star Tracker accuracy. From the point of view of performance, the Star Tracker prototype fully represents the final Star Tracker, the main differences being related to several implementation details. The results show a Star Tracker RMS accuracy is ≈ 3 arcsec while the systematic error is below 10 arcsec. From those results, we analyzed the problem of reconstructing the PRM configuration angles by simulating a calibration survey. Given the need to intercalibrate the offset of the Start Tracker pointing direction with respect to the focal plane pointing direction, we simulated two possible intercalibration strategies: one by simulating intercalibration with the use of observations of planets, the second by observing a drone carrying an optical beacon and a radio beacon. In the first case, the accuracy is determined by the level of 1/f instrumental plus atmospheric noise, determining the S/N by which the planet can be observed. A very conservative S/N=10 case and a more likely S/N=50 case have been considered, allowing for an intercalibration accuracy respectively of 1 arcmin and 1/3 arcmin. In the second case, the most important source of error is the correct evaluation of the parallaxes between the telescope and the Star Tracker. Our analysis shows that the intercalibration accuracy will be between 0.25 arcmin and 1 arcmin in the worst cases.