Predicting Network Reliability for Satellite-Based Communication Systems

Abstract

Satellite communication networks with on-board processing (OBP) satellites can provide high-speed da-ta transmission rates and global service coverage with reduced propagation delays. In addition, a use of soft-ware defined radio (SDR) in a small satellite can support flexible small satellite communications and offer flexible and adaptive communication protocols. OBP and SDR systems are implemented in static random-access memory (SRAM)-based field-programmable gate arrays (FPGAs) that are the most representative devices for reprogrammable platforms. However, the SRAM, a volatile memory, is very vulnerable to the space radiation environments and the most common damages are single event upsets (SEUs) that generate the OBP and SDR system malfunctions or system failures. In communication channels, high frequency carri-ers in the channel between satellites and terrestrial gateways are extremely susceptible to weather attenuation and other atmospheric turbulences, which induce unavoidably high bit error rates (BERs) in each communi-cation channel. Besides, with growing demands of wireless network service, bursty traffic in packet transmis-sions will increase, which lead high packet loss ratio (PLR). Such factors have degraded the reliability of satel-lite communications networks over time. This thesis suggests a prediction model for OBP and SDR system failure rates, and a means of analyzing the quantitative reliability of the satellite communication network systems. The first subject presents an OBP system adopting Triple Modular Redundancy with the concept of mitigation windows and external scrubber, and then suggests a mathematical model that predicts the OBP system failure rate by only using the information of system configuration resources. Our mathematical deri-vation can estimate on-board processor system reliability as a function of the SEU rate, the number of miti-gation windows, and on-board processor shield thickness. The second subject proposes a means of analyzing the quantitative reliability of the satellite communi-cation network systems. We identify the four major factors that affect the quality of network services: the OBP states, uplink channels, downlink channels, and uplink packet collision losses. Based on these four fac-tors, a Markov model is derived to analyze the probability distributions of various network states. Based on the developed model, a method is suggested for iteratively updating the reliability distribution of network sys-tems affected by changes in the four factors as well as network access time changes. Finally, the third subject contains a derivation of a Markov model presenting reliability of the small satellite network with respect to SDR structures, transmitted signal powers on uplink/downlink channels, code rates, and packet collisions through an enhanced random access (RA) protocol. Our model provides the quantitative network reliability in terms of SDR structures with bad space radiation environments, signal-to-noise ratios (SNRs) on uplink and downlink channels, and PLRs through an enhanced RA protocol.