A Proposed Method to Estimate Signal to Noise Plus Interference Levels in Order to Improve the Performance of Mobile Ad Hoc Network Routing Protocols
John P. Mullen · 2005
Because random variations in signal-to-noise plus interference ratio (SNIR) profoundly affect mobile ad hoc network performance, good estimates of mean SNIR are essential for reliable routing. This paper proposes that signal levels be estimated independently from noise plus interference (N+I) levels because 1) N+I levels can be estimated when the signal is absent, 2) signal estimates must allow for missing values when the transmitter cannot be identified, 3) signal and N+I levels vary for different reasons and 4) separate estimates are less variable than direct estimates of SNIR. It also demonstrates the effectiveness of these methods via simulation. MANET routing; wireless propagation; stochastic models; decision rules; power av-Keywords: eraging.