Laser satellite communication: precise estimation of the satellite position and synchronization and the stabilization of the beam on the receiving satellite

This paper analyses a number of systems and control issues associated with the design of a low orbit communications satellite network. This paper discusses the adaptation of ISDN ATM protocols to the constraints inherent with laser communications. The main proposed change is to replace the link-by-link off band routing by in band arrow routing. The first field is coded using several laser colors, such as to trigger, without delays, the switching of the incoming cell, from the incoming receiver to the outgoing laser. The laser beam is generated and steered by an array of semiconductor lasers and concentrated by a telescope. Monopulse measurement of the direction of incoming light is performed using several arrays of photodiodes. The whole concept is based on precise pointing of the transmitting beam towards the receiving satellite. It also requires a very precise estimation of the position and synchronization of the satellite nodes. This requires solving of three control problems. The first control problem involves the estimation of the position and synchronization of the satellite using the measurement of the propagation time and range from the other transmitting satellites. This problem is solved by a Kalman filter. The second control problem involves the estimation of the satellite attitude and its stabilization. Since precise attitude tracking of the receiving satellite demands much electrical power, an alternative low power consumption method is proposed. This proposal is based on the concept of estimation rather than precise tracking of the receiving satellite. Since the relative positions of both satellites are known, the attitude of the satellite can be estimated very precisely. In order to avoid the problem of alignment of the transmitter and receiver channels, the estimation of the pointing is performed, when possible, using the reflection of the transmitted fight on a total reduction prism placed on the receiving satellite. The pointing of the beam is estimated using a Kalman observer. Third the beam must be steered towards the receiving satellite. A problem encountered in trying to steer the laser beam is the propagation delay introduced by transmitting the beam over large distances. Two techniques are used to solve this problem. The first consists in using the prediction of the beam pointing as the feedback in the control loop. The second method consists of modeling the delay using Padre's approximation and using neural controller technique to handle the non-minimum phase associated with the control problem.