Validation and Receiver Design for a Random Point Process Model of Atmospheric Radio Noise.
Abstract
An investigation of low frequency atmospheric radio noise (ARN) indicates that the return stroke from lightning discharges is the major source of the noise. A model of ARN is then presented which is based on the return stroke. The model consists of the sum of two marked Poisson processes. The sum of marked Poisson process can be thought of as a single marked Poisson process with a transformed rate and a mixture density on the marks. The complete statistical description of this process is then derived and a minimum probability of error processor is designed using the ARN model as the noise. The theoretical amplitude probability distribution (APD) is then derived for the output of a quadrature envelope detector using the ARN model as an input. The theoretical APD curves were then compared to a measured APD curve from CCIR Report 322 and two measured APD curves from an article by K. Furutsu and T. Ishida. The linear mean square error (MSE)between the theoretical and measured curves when plotted on Rayleigh paper is .018, and .015 for the Furutsu and Ishida data and .018 for the CCIR data. Based on the first order statistics, the model is found to be a good representation of low frequency ARN. It is therefore recommended that the model be used to evaluate receiver performance in the low frequency channel and that a more thorough validation study be done including a validation of higher order statistics. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 01, 1979
- Accession Number
- ADA080214
Entities
People
- John F. Stach
Organizations
- Air Force Institute of Technology