Although broadcasting using tree structure established in a network is a well known and widely used technique, it is typically claimed to be inappropriate for ad hoc networks, being the maintained tree very sensitive to network changes. On the contrary this paper presents an efficient tree based broadcasting scheme, which is reliable and stable even in case of the ever changing network structure of the ad hoc networks. To achieve this, first, a novel method is presented to maintain a spanning tree in an ad hoc network in a fully distributed, on-line and asynchronous way. Once the tree is established the broadcast itself is performed based on this tree. Some further improvements on the basic algorithm are also presented that reduce the resource requirements even more, increase the stability of the tree, enable the mobility of the nodes to be taken into account and make the method more configurable. As it is shown by simulation, the obtained broadcast scheme is stable, reliable and it uses small amount of resources: the acyclic structure of the broadcast tree ensures that the nodes get the broadcast messages only once, so the broadcast needs little bandwidth and the nodes need not store the recent broadcast messages, reducing the computational and memory requirements. As a byproduct a technique is proposed to measure the mobility of the nodes. This technique needs no additional GPS device or any geographical information but it is based on the stability of the links of the node.
The paper presents the detector system developed by Datalist Systems, Ltd. (previously ANTE Innovative Technologies) for the NEAT-II spectrometer at HZB. We present initial concept, design and implementation highlights as well as the first results of measurements such as position resolution. The initial concept called for modular architecture with 416 3He detector tubes organized into thirteen 32-tube modules that can be independently installed and removed to and from the detector vacuum chamber for ease of maintenance. The unalloyed aluminum mechanical support modules for four 8-tube units each also house the air-boxes that contain the front-end electronics (preamplifiers) that need to be on atmospheric pressure. The modules have been manufactured and partly assembled in Hungary and then fully assembled and installed on site by Datalist Systems crew. The signal processing and data acquisition solution is based on low time constant (~60 ns) preamplifier electronics and sampling ADC's running at 50 MS/s (i.e. a sample every 20 ns) for all 832 data channels. The preamplifiers are proprietary, developed specifically for the NEAT spectrometer, while the ADC's and the FPGA's that further process the data are based on National Instruments products. The data acquisition system comprises 26 FPGA modules each serving 16 tubes (providing for up to 50 kHz count rate per individual tube) and it is organized into two PXI chassis and two data acquisition computers that perform post-processing, event classification and provide appropriate preview of the collected data. The data acquisition software based on Event Recording principles provides a single point of contact for the scientific software with an Event Record List with absolute timestamps of 100ns resolution, timing data of 100 ns resolution for the seven discs chopper system as well as classification data that can be used for flexible data filtering in off-line analysis of the gathered data. A unique 3-tier system of filtering criteria of events is in operation: a hard threshold in the FPGA's to reduce the effect of noise, a pulse-shape based classification to eliminate gamma sensitivity and an additional flexible feature based classification to filter out pileup and other unwanted phenomena. This ensures high count rates (50kHz per tube, 1MHz overall) while maintaining good quality of measurements (e.g. position resolution).The first measurement results show that the delivered detector system meets the initial requirements of 20 mm position resolution along the 2000mm long detector tubes. This is partly due to the innovative event classification system that provides vital pulse shape data that can be used for sophisticated position resolution algorithms implemented on the DAQ computers. .
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.