Deploying a Wireless Sensor Network on an Active Volcano Summary: In this paper the author describes the importance of the

sensor network applications. Author described the most important application i.e., Deploying a Wireless sensor network on an active volcano, which requires addressing high data rates and high data fidelity. The authors application mainly depends on triggered event detection and reliable data retrieval to meet bandwidth and data quality demands. The authors group mainly focussed on tiny, low-power wireless sensor nodes for geophysical studies. They deployed larger more capable network on Volcan Reventador, Which comprised 16 stations equipped with seismic and acoustic sensors. Each station consisted of a Moteiv TMoteSky wireless sensor network node, an 8-dBi 2.4-GHz external omnidirectional antenna, a seismometer, a microphone, and a custom hardware interface board. The TMote Sky features a Texas Instruments MSP430 microcontroller, 48 Kbytes of program memory, 10 Kbytes of static RAM, 1 Mbyte of external flash memory, and a 2.4-GHz Chipcon CC2420 IEEE 802.15.4 radio. The authors team built a custom hardware board to integrate the TMote Sky with the seismoacoustic sensors.The board features up to four Texas Instruments AD7710 analog-to-digital converters (ADCs), providing resolution of up to 24 bits per channel. A pair of alkaline D cell batteries powered each sensor node. Each node samples two or four channels of seismoacoustic data at 100 Hz, storing the data in local flash memory. They developed a reliable data-collection protocol, called Fetch, to retrieve buffered data from each node over a multihop network. They developed a Java-based GUI for monitoring the networks behavior and manually setting parameters, such as sampling rates and event-detection thresholds. In addition, the GUI was responsible for controlling data collection following a triggered event, moving significant complexity out of the sensor network. During transmission, a sensor node fragments each requested block into several chunks, each of which is sent in a single radio message. The base-station laptop retrieves a block by flooding a request to the network using Drip, a variant of the TinyOS Trickle6 data-dissemination protocol. The target node replies by sending the requested chunks over a multihop path to the base station. Our system constructs the routing tree using MultiHopLQI, a variant of the TinyOS MintRoute4 routing protocol modified to select routes based on the CC2420 link quality indicator (LQI) metric. The network captured 230 volcanic events letting us evaluate the performance of large scale sensor networks for collecting high-resolution volcanic data. This paper poses some drawbacks they are listed below: Since the volcanic signal data analysis focus on discrete events the authors group designed a network to capture the time limited events which doesnt answer all the questions regarding its activity. To get complete information we need complete waveforms spanning long time intervals. As the authors group used low radio bandwidth wireless sensor nodes with which it is impossible for them to do complete analysis. The authors group are using four AD7710 analog-to-digital converters (ADCs), providing resolution of up to 24 bits per channel. Although the MSP430 microcontroller provides on-board ADCs, (unsuitable since they provide only 16 bits of resolution, and Seismoacoustic signals require an aggressive filter centered around 50 Hz.).The on board ADCs which are not used at all consume some amount of power which is being wasted unnecessarily. Since power consumption here plays a major role they should have used another microcontroller that provides several configurable ports that easily support external devices, and the large amount of flash memory. Since they are using pair of alkaline D cell batteries to power each sensor node, one should go to the remote location and change the battery regularly it is difficult for one going to such remote locations just for changing the batteries as it is a volcanic eruption zone where there is threat to life. During their three-week deployment, they changed batteries on the entire sensor array twice. If it is a continuous process i.e. if one want to continuously monitor the volcano for few years one person should continuously go and change the battery frequently which is undesirable. They might have used the solar cells or any other alternative for that.

In FreeWave radio modems they used 9-dBi directional Yagi antennae to establish a long distance radio link between the sensor network and the observatory. The biggest disadvantage of the Yagi design is that the antenna is obtrusive .The antenna is limited to having a gain in the range of the 6 to 18 dBi range. Yagi design only achieves high gain over a rather narrow bandwidth.For larger bandwidths the gain is very low. The TMote Skys flash memory fills in roughly 20 minutes when recording two channels of data at 100 Hz. If we want to record four channels of data the TMote Skys flash memory would not have been sufficient. This buffer is not sufficient to fit every volcanic event only some events can fit in this buffer .Thus the authors group failed to record every volcanic event. The main drawback was when the laptop initiates data collection from the entire network it also collects data from the nodes that didnt report the event. Such data is un useful for us . There is no concept of optimization in data collection path. They are gathering data from unwanted nodes they should have adopted methods that only collects data from nodes which reports the event, and also fetching 60 seconds of data from 16 nodes in the network takes one hour since nodes can only buffer 20 minutes of eruption data, here each node has to pause sampling and reporting events until it has uploaded its data. This latency prevents the network from capturing all events. This kind of buffering prevents the continuous capturing of events of a volcanic activity. In order to provide data of extremely high fidelity samples must be accurately time stamped to allow comparisons between nodes and between networks because a single corrupt sample can invalidate whole record. Small differences in sampling rates between two nodes give wrong information. Since we are using low radio bandwidth radios, such as the Chipcon CC2420, have data rates of 30Kbps .Overheads caused by packet framing, medium access control (MAC), and multihop routing reduce the achievable data rate to less than 10 Kbps, even in a single-hop network. This is undesirable. The authors group used a Flooding Time Synchronization Protocol (FTSP) to establish a global clock Across the network which is supposed to have high accuracy but FTSP failed because the, nodes reported inaccurate global times, preventing some data from being correctly time stamped. The authors group disabled the sampling while collecting the data which disabled them to collect data from two back to back events. They were unable to give the priority to the events for instance if a small seismic event triggered the data collection and they missed a large explosion shortly after that. Thus they completely failed on prioritization. The Authors Group retrieved data from the network only 61 percent of the time. There were short outages occurred because there was no proper for laptop round the clock as a result of which they were unable to record each and every event. The author mentioned that the longest continuous network outage was due to a software component failure, which took the system offline for three days. His team went to the deployment site to reprogram nodes manually. It would be not possible every time to go the deployment site because sometimes the volcano may be exploding so there should be a mechanism or the nodes should programmed automatically not manually

Linear configurations can also affect achievable network bandwidth, which degrades when data must be transmitted over multiple hops. Node failure poses a serious problem in sparse networks because a single failure can obscure a large portion of the network.