Performance of IEE 802.15.4 Mac layer for lab & field test for high speed train
Tutor / director / evaluatorBarba Martí, Antonio
Document typeMaster thesis
Rights accessOpen Access
Several proposals are now in research and development for wireless sensor networks(WSN) to apply on railway systems for high speed lanes and others slow and medium speed systems, some are working between communications from train to Local Operation Point(LOP) and then to the Network Operation Center (NOC) and others from a specific sensorized network to the LOP and then to NOC, we are focus in between LOP and sensors. In order to monitor, control, secure, adapt, focus, and to have the best performance in wireless Sensor Networks (WSN), we need to ask Why Quality of Service (QoS)? Because in Telecommunications, we need to assure our Protocols, links, in order to comply with user requeriments. In this report we present an approach for high speed train, in order to assure links between sensors to the sink or coordinator (in order to process this signal as input for alarms and monitoring track enviroments). So then we present a QoS state of art, to achieve this we present a brief description of the IEEE 802.15.4 trying to focus on the MAC layer, CSMA/CA mechanism to access the channel and send packets. For QoS, we compare some other proposals, also with mechanisms for queuing packets and for both beacon and beaconless modes, also compare super frame structure, and other parameters. We use different metrics for performance as delay, throughput, PER(Packet error rate), BER (Bit error rate), PLR (Packet loss ratio), focus on finding the best scenario for high speed trains. We present some laboratory and field results developed by SENER and UPC, using Telegesis and Daintree equipments of ZigBee Alliance. The chips used in this is EM250 and EM250RCM these are developed in EMBER we are using this sensor nodes to sense temperature, light and we adapt a potentiometer in order to sense a special vibration. On the field tests there are trains crossing the sensor links, so we focus on this results. Also we compare laboratory and field results to take out conclusions of QoS and performance.