Impact of on-body channel models on positioning success rate with UWB Wireless Body Area Networks

Abstract

In this paper, we aim to evaluate the positioning success rate of nodes placed on the body using different scheduling strategies at the Media Access Control (MAC) layer with Ultra Wide Band (UWB) Wireless Body Area Networks (WBAN) and under three different channel models. For this purpose, each node calculates its relative position with the estimation of its distances with the on-body anchors. Accordingly, the distance between two nodes can be estimated with the transmission of three packets, as defined by the the '3-Way ranging' protocol (3-WR). However, these transactions can be affected by the WBAN channel leading into a packet loss and therefore positioning errors. In this work, we consider a PHY layer based on Impulse-Radio UWB (IR-UWB) with three different channels: (a) a theoretical path loss channel model based on the on-body CM3 channel (Anechoic chamber), (b) a simulated channel calculated with the PyLayers ray-tracing simulator and (c) experimental traces obtained by measurement. Moreover, we analyze the positioning success rate using three scheduling strategies (Single node localization (P2P), Broadcast Single node localization (P2P-B) and Aggregated & Broadcast (A&B)) with a MAC layer based on time division multiple access (TDMA) and under a realistic pedestrian walking scenario. Our results show that the scheduling strategy with A&B let the nodes to estimate more positions even through channels with slow and fast fading.