Geometry-Informed Transmission Strength Scaling in Barrage Relay Networks

Barrage relay networks (BRNs) are a type of ad hoc wireless network with current and proposed uses in military, disaster response, industrial and vehicle-to-vehicle applications. BRNs are designed specifically to operate robustly without any information about the relative positions of other nodes on the network. The nodes in a specific connection within a BRN are determined by the formation of a controlled barrage region (CBR). There is a trade-off between reliability and node utilization or the number of nodes that are reserved for a single unicast link. In a previous paper, we investigated the suitability of BRNs for higher-density networks, which are anticipated with the explosion of network-connected devices. We showed that node utilization increases superlinearly with both node density and the distance between source and destination, calling into question the suitability of BRNs for use in high-density networks. In this paper we use a connected graph model to show that CBR formation algorithm generates a specific geometry in the limit of infinite node density. Applying this geometry to a more physically realistic random channel model (RCM) with finite node density, we propose tuning the model by scaling the transmission power of receivers. The transmit power is based on received signal strength during the sending of request to send (RTS) and clear to send (CTS) packets, used to determine the nodes used in the CBR. We show that in discrete event simulations this signal strength scaling can reduce the utilization while simultaneously improving the probability of CBR formation.