The Effects of Ranging Noise on Multihop Localization: An Empirical Study from UC Berkeley Abon

Outline Motivation Approach Terms Deployment setup Simulation and experimental results Contribution

Motivation Eliminate the discrepancy in simulation and real-world performance. Capture the ranging characteristic in real world Discover the effect of each component in simulation model

Approach Use the combination of existing simulation model to compare with the real-world performance Determine the critical part in different localization deployment Provide a new technique for simulation model

Term ~ Noisy Disk model Two components : Noise component indicates the distribution of the error between the measured distance and the actual distance (e.g., Gaussian, uniform) Connectivity component indicates the maximum distance dmax between two nodes at which a distance estimate can be obtained.

Parametric model Gaussian noise is generated with the function N(dij, σ) Unit Disk connectivity is generated using the inequality dij ≤ dmax The typical data collection technique for ranging is to place a transmitter and receiver at several known distances and measure the response, although this technique doesn’t account for several sources of noise such as node variability.

Statistical sampling We generate data for simulation by randomly drawing measurements from an empirical data set M(δ, ε) to be the empirical distribution of all observed ranging estimates for distances in the interval [δ − ε, δ + ε ]. A ranging estimate for simulation by using the error of a random sample from M(dij, ε)

Statistical sampling We generate a ranging estimate dij for simulation by using the error of a random sample from M(dij, ε). d ij =d ij +( d − d a )

Deployment setup The first is a 49 node network over a 13x13m asphalt area localized using ultrasound (Medusa node). The others are 25 and 49 node networks over a 50x50m grassy area localized using RSS (Chipcon CC1000 FSK radio).

Deployment setup

Topology generated by APS

Simulation Two techniques for simulation: Parametric models Statistical sampling

Results

The 49 node RSS deployment is well predicted by both the */G and */S simulations but not the */N simulations. This trend indicates that noise is the dominant cause of the localization error in this deployment.

Results 49 node ultrasound deployment in Figure 3(c) is well predicted by the S/* simulations but not the D/* simulations. This indicates that the ultrasound connectivity is different than the Unit Disk model, and these deviations dominate noise as the source of error in this deployment.

Results The 25 node RSS deployment shows a similar trend; the S/* simulations predict observed error better than the D/* simulations, but no connectivity/noise combination correctly predicts all the error in this deployment. This indicates that ranging characteristics besides noise and connectivity are causing localization error.

What kind of model you need Parametric model Identify a small set of ranging characteristic Useful in theoretical analysis Need to be evaluated for every new noise characteristic Need to estimate parameters from data

What kind of model you need Statistical sampling New models do not need to be created for new empirical distribution. But it does not reveal insights about the data Need to identify which data subset is important You can extend the parameters in the model

Conclusion At low node density, for ultra sonic ranging, connectivity model dominates the error; for RSS ranging, connectivity and noise model deviation in radio all affect the error. At high node density, noise disk model predicts deployment error fairly well.

Contribution This study suggests a top-down approach to evaluating models by comparing each model’s predictions with empirical observations of localization deployments. This is in contrast with the commonly used bottom-up approach for deriving models by analyzing raw empirical data. A bottom- up approach is useful for identifying and characterizing the few most important features of empirical data and building them into a model. A top-down approach can evaluate whether the model is a sufficient representation of those features, and whether that set of features is sufficient to represent the empirical data.

Thank you