High-Density Random Access for High-Speed Satellite Data Services

In recent years there has been a strong resurgence in interest in satellite communications. The European Space Agency (ESA), for example is investigating satellite-based wireless data services using multi-spot, high-capacity geo-stationary satellites. Such a system will require a satellite throughput capacity on the order of a terabyte/s in order to be economically competitive with ground-based wireless services. A modern satellite, operating with multiple spot beams, is capable of covering a large geographic area and servicing a very large number of users. Difficulties arise when the traffic through the satellite increases to the point where basic access and communications methods no longer suffice to service the data demand. This difficulty is compounded by the long round-trip delays between transmissions on the up-link and reception of acknowledgements on the down-link, which prevents rapid coordination between transmitters. The receiver also needs to be able to resolve signal collisions and be able to estimate and track channel conditions for a multitude of user terminals. These challenges represent a are the major focus of the proposed research project.The most recent variations of random access contention resolution transmit multiple copies of a data packet in randomly distributed time slots throughout a transmission frame. This approach statistically increases the rate of packet collisions, but the probability that at least one copy of a packet is received collision-free is increased. This packet can then be cancelled from the other time slots where it was also transmitted, allowing for iterative peel-off decoding of the entire frame. Theoretical studies have shown that this method can achieve 80% of the single user capacity for practical scenarios, which represents a vast improvement over the 37% achievable with a basic slotted ALOHA protocol. However, the channel itself, is a multiple access channel, and as such it has a theoretical capacity which is superior to that of any multiplexed channel. Exploitation of this advantage requires the incorporation of joint detection principles. Multi-packet reception(MPR) will be addressed with iterative interference cancellation, which has recently been shown to be able to approach the multiple access channel capacity in theory. However, substantial problems remain before such MPR receivers can be deployed. This project focuses on the challenges that arise when realistic channel propagation models are used, in particular channels with highly dynamic behavior. It is anticipated that the fundamental results from this work will lead to increased utilization of spectral resources and robustness of networks. We have been collaborating with the German Aerospace Institute (DLR), and this grant will accelerate this collaboration, and lead to a strong research program in resource-efficient data communications at Dalhousie University.