The recent emergence of Machine-to-Machine networks and the Internet of Things has generated interest in technologies that can underpin networks of low power devices that sporadically exchange data. Both the research and the standardization communities have been actively participating in seeking and defining technologies of the underlying structure of future networks. These Low Power Wide Area Networks (LPWAN) will support large number of nodes with long-range wireless communications on the one hand, while guaranteeing a long life for very cheap sensing devices on the other. Among the many applications that can be supported on these network, there is a wide range of monitoring applications which include low power tracking and radiolocation. The radiolocation application enables a device to determine its position using signals from base stations. It is similar in spirit to GPS except that the signals of radiolocation are provided by terrestrial base stations instead of satellites. This contribution focuses on one such LPWAN technology, called Long Range Wide Area Networks (LoRaWAN) which incorporates the necessary features for radiolocation purposes. A LoRaWAN consists of a central controller which controls several gateways, that in turn listen to incoming data sent by low powered end-devices over sub-gigahertz unlicensed spectra. The geographical placement is such that a signal sent by one end-device is received by at least three gateways. Multilateration for the location of an end-device can be performed by using a combination of both the strength of the signals and the time difference of arrivals measured at the multiple gateways which are in the transmission range of this end-device. In order to feature LoRaWAN in city scaled scenarios, it is very important to design a proper positioning scheme for gateways and end-devices to assess the related performances in several network conditions. Given the lack of simulation instruments targeting scalable LoRaWAN deployments, this contribution introduces an event based simulation environment that has been developed to characterize the performances of very large LoRaWAN scenarios. Through huge simulation campaigns, the capacity of such network has been thoroughly evaluated in order to predict the impact of low power radiolocation for future tracking applications. Simulation results have also validated the expected probability models for the LoRaWAN throughput and have provided a realistic description of more complex application scenarios.