Neural network approach to short-term fade prediction on satellite links

Chambers, A.P.; Otung, Ifiok

doi:10.1049/el:20052711

Cited by 8 publications

(8 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Portsmouth University model (Grémont et al, 1999) is an adaptive ARMA(3,3) whose parameters are updated in real time by a Recursive Least Squares (RLS) algorithm (Ljung, 1999). The ADALINE model (Chambers and Otung, 2005) is a single neuron with a linear activation function. The learning parameter was empirically set at 10 −4 .…”

Section: Comparison and Resultsmentioning

confidence: 99%

“…Since the uplink is more affected by attenuation than the downlink (because it generally operates at a higher frequency), the downlink attenuation is predicted and then scaled to the uplink frequency in order to estimate the future propagation conditions of the uplink. Various prediction models have been proposed, such as a linear regression (Dossi, 1990), a first order stochastic equation (Manning, 1990(Manning, , 1991, Markov chains (Castanet et al, 2003;Fiebig, 2002), an adaptive linear filter (Grémont et al, 1999), a neural networks (Chambers and Otung, 2005) or a model based on the fade slope ( Van de Kamp, 2002). A comparison was performed with Olympus 20 GHz data (Sect.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Short-term prediction of rain attenuation level and volatility in Earth-to-Satellite links at EHF band

Montera¹,

Mallet²,

Barthès³

et al. 2008

Nonlin. Processes Geophys.

View full text Add to dashboard Cite

Abstract. This paper shows how nonlinear models originally developed in the finance field can be used to predict rain attenuation level and volatility in Earth-to-Satellite links operating at the Extremely High Frequencies band (EHF,. A common approach to solving this problem is to consider that the prediction error corresponds only to scintillations, whose variance is assumed to be constant. Nevertheless, this assumption does not seem to be realistic because of the heteroscedasticity of error time series: the variance of the prediction error is found to be time-varying and has to be modeled. Since rain attenuation time series behave similarly to certain stocks or foreign exchange rates, a switching ARIMA/GARCH model was implemented. The originality of this model is that not only the attenuation level, but also the error conditional distribution are predicted. It allows an accurate upper-bound of the future attenuation to be estimated in real time that minimizes the cost of Fade Mitigation Techniques (FMT) and therefore enables the communication system to reach a high percentage of availability. The performance of the switching ARIMA/GARCH model was estimated using a measurement database of the Olympus satellite 20/30 GHz beacons and this model is shown to outperform significantly other existing models.The model also includes frequency scaling from the downlink frequency to the uplink frequency. The attenuation effects (gases, clouds and rain) are first separated with a neural network and then scaled using specific scaling factors. As to the resulting uplink prediction error, the error contribution of the frequency scaling step is shown to be larger than that of the downlink prediction, indicating that further study should focus on improving the accuracy of the scaling factor.

show abstract

Section: Comparison and Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Short-term prediction of rain attenuation level and volatility in Earth-to-Satellite links at EHF band

Montera¹,

Mallet²,

Barthès³

et al. 2008

Nonlin. Processes Geophys.

View full text Add to dashboard Cite

show abstract

“…There are different methods for the fade predicting depending on the model assumed for the channel, see [24]- [26]. For the channel model assumed in this paper, we can exploit (3) to derive the MMSE estimator which is used to estimate the A(t + t d ) based on the observed A(t).…”

Section: Switching Based On Snr Predictionmentioning

confidence: 99%

Multiple Gateway Transmit Diversity in Q/V Band Feeder Links

Gharanjik

Shankar

Arapoglou³

et al. 2015

IEEE Trans. Commun.

View full text Add to dashboard Cite

Abstract-Design of high bandwidth and reliable feeder links is central towards provisioning new services on the user link of a multibeam satellite communication (SatCom) system. Towards this, utilization of the Q/V band and an exploitation of multiple gateways (GW) as a transmit diversity measure for overcoming severe propagation effects are being considered. In this context, this contribution deals with the design of a feeder link comprising N + P GWs (N active and P redundant GWs). Towards provisioning the desired availability, a novel switching scheme is analysed and practical aspects such as prediction based switching and switching rate are discussed. Unlike most relevant works, a dynamic rain attenuation model is used to derive analytically average outage probability in the fundamental 1 + 1 gateway case. Building on this result, an analysis for the N + P scenario leading to a quantification of the end−to−end performance is provided. This analysis aids system sizing by illustrating the interplay between the number of active and redundant gateways on the chosen metrics : average outage and average switching rate.

show abstract

“…Given that the ADALINE and ARMA predictors perform comparably well in [9], especially at low values of k; and since the Gauss-Newton update method of the ARMA predictor converges more quickly than the gradient method of the ADALINE predictor, the advantages of both methods can be exploited.…”

Section: Hybrid Predictormentioning

confidence: 99%

“…A previous work introduced a predictor based on an adaptive linear element (ADALINE) [9], where it was demonstrated that the slope averaging predictors perform poorly in comparison with the signal processing and statistical types.…”

Section: Introductionmentioning

confidence: 99%

Near‐optimum short‐term fade prediction on satellite links at Ka and V‐bands

Chambers¹,

Otung²

2007

Satell Commun Network

View full text Add to dashboard Cite

SUMMARYSeveral short-term predictors of rain attenuation are implemented and tested using data recorded from a satellite link in Southern England, and a comparison is made in terms of the root-mean-square error and the cumulative distribution of under-predictions. A hybrid of an autoregressive moving average and adaptive linear element predictor is created that makes use of Gauss-Newton and gradient direction coefficient updates and exhibits the best prediction error performance of all prediction methods in the majority of cases.

show abstract

Neural network approach to short-term fade prediction on satellite links

Cited by 8 publications

References 3 publications

Short-term prediction of rain attenuation level and volatility in Earth-to-Satellite links at EHF band

Short-term prediction of rain attenuation level and volatility in Earth-to-Satellite links at EHF band

Multiple Gateway Transmit Diversity in Q/V Band Feeder Links

Near‐optimum short‐term fade prediction on satellite links at Ka and V‐bands

Contact Info

Product

Resources

About