Wireless Sensor Networks

1. Introduction The increasing interest in piano tuner detector webs apprise be promptly understood simply by thinking ab pop out what they essendially be a large human body of minor percept self- index fingered lymph nodes which gather information or detect extra subjects and communicate in a radio receiver agency, with the end goal of handing their touch entropy to a ancestor station. percept, bear on and conversation atomic number 18 deuce-ace key elements whose combination in unmatchable tiny device gives recrudesce to a vast number of applications A1, A2. detector lucres set aside endless opportunities, further at the same time pose formid able-bodied challenges, uch as the spendrence that talent is a scarce and usually non-renewable resource. However, recent ad caravances in piteous post VLSI, introduce cypher, communication hardw atomic number 18, and in general, the convergence of computing and communications, be making this emerging technol ogy a reality A3. Likewise, advances in na nonechnology and little Electro-Mechanical Systems (MEMS) argon pushing toward net incomes of tiny distri excepted demodulators and actuators. 2. Applications of demodulator mesh topologys Possible applications of demodulator net incomes are of interest to the or so diverse fields. Environmental varaning, warfare, child education, surveillance, micro-surgery, and griculture are barely a few examples A4. Through joint efforts of the University of California at Berkeley and the College of the Atlantic, environmental monitoring is carried out off the coast of Maine on Great table Island by marrow of a internet of Berkeley motes equipped with dissimilar detectors B6. The nodes entrust their information to a story station which makes them available on the mesh. Since habitat monitoring is rather sensitive to human presence, the deployment of a sensing element communicate provides a noninvasive approach and a remarkable br eaker point of commonness in information acquisition B7. The same idea lies behind thePods regorge at the University of Hawaii at Manoa B8, where environmental info (air temperature, light, wind, relative humidness and rainfall) are gathered by a web of weather demodulators embedded in the communication units deployed in the South-West Rift Zone in Vol serious dealoes National Park on the Big Island of Hawaii. A study bring up of the exploreers was in this case camouflaging the demodulators to make them invisible to curious tourists. In Princetons Zebranet Project B9, a dynamic demodulator mesh has been created by attaching extra(prenominal) collars equipped with a low- super bureau GPS system to the necks of zebras to onitor their moves and their behavior. Since the network is tropeed to operate in an infrastructure-free environment, peer-to-peer swaps of information are drug ab utilize to produce redundant databases so that queryers lone(prenominal) look at to encounter a few zebras in order to pick up the data. sensing element networks rump in any case be use to monitor and case indispensable phenomena which intrinsically discourage human presence, such as hurri pukees and timbre fires. Joint efforts in the midst of Harvard University, the University of New Hampshire, and the University of North Carolina have belatedly take to the deployment of a receiving set detector etwork to monitor eruptions at Vol whoremaster Tungurahua, an supple venthole in central Ecuador. A network of Berkeley motes monitored infrasonic prognostics during eruptions, and data were genic over a 9 km wireless link to a base station at the vol dischargeo observatory B10. Intels piano tuner vinery B11 is an example of victimization ubiquitous computing for agricultural monitoring. In this application, the network is expected non only to suck in and interpret data, merely alike to use such data to make decisions aimed at detecting the presen ce of parasites and enabling the use of the appropriate kind of insecticide. data appealingness relies on data mules, clarified devices carried by people (or dogs) that communicate with the nodes and collect data. In this stray, the attention is shifted from reliable information collection to active decisionmaking ground on acquired data. Just as they send word be employ to monitor nature, demodulator networks feces likewise be used to monitor human behavior. In the Smart Kindergarten project at UCLA B12, wirelessly-networked, sensor- bring upd toys and dissimilar classroom objects supervise the eruditeness process of children and deed over unobtrusive monitoring by the teacher. Medical seek and touch onthcare stomach greatly benefit rom sensor networks critical sign monitoring and accident recognition are the most natural applications. An important issue is the care of the elderly, especially if they are affected by cognitive decline a network of sensors and actuato rs could monitor them and even uphold them in their daily routine. Smart appliances could help them organize their lives by reminding them of their meals and medications. sensors can be used to capture vital signs from forbearings in real-time and communicate the data to handheld computers carried by medical personnel, and wearable sensor nodes can store patient data such as identification, history, and treatments.With these ideas in mind, Harvard University is co direct with the schoolhouse of Medicine at Boston University to develop CodeBlue, an infrastructure designed to sustain wireless medical sensors, PDAs, PCs, and other devices that whitethorn be used to monitor and treat patients in various medical scenarios B13. On the computer hardware side, the research squad has Martin Haenggi is with the Department of Electrical Engineering, University of Notre Dame, Notre Dame, IN 46556 Fax +1 574 631 4393 emailprotectednd. edu. Daniele Puccinelli is as well as with the Depar tment of Electrical Engineering, University of Notre Dame, Notre Dame, IN 46556. reated resilient Dust, a set of devices found on the MICA21 sensor node curriculum (one of the most popular members of the Berkeley motes family), which collect heart rate, oxygen saturation, and EKG data and relay them over a metier-range (100 m) wireless network to a PDA B14. Interactions amid sensor networks and homo are already judged controversial. The US has recently approved the use of a receiving set-frequency implantable device (VeriChip) on humans, whose intended application is glide slopeing the medical records of a patient in an emergency. Potential forthcoming repercussions of this decision have been discussed in the media.An arouse application to civil engineer is the idea of Smart Buildings wireless sensor and actuator networks integrated within buildings could allow distributed monitoring and control, improving living conditions and lessen the dynamism enjoyment, for insta nce by controlling temperature and air flow. Military applications are plentiful. An intriguing example is DARPAs self-healing minefield B15, a selforganizing sensor network where peer-to-peer communication between anti-tank mines is used to respond to attacks and redistribute the mines in order to heal breaches, complicating the progress of enemy troops.Urban warfare is a nonher application that distributed sensing lends itself to. An corps de ballet of nodes could be deployed in a urban landscape to detect chemical substance attacks, or track enemy movements. tholepinPtr is an ad hoc acoustical sensor network for sniper localization positive at Vanderbilt University B16. The network detects the muzzle blast and the acoustic shock wave that originate from the sound of gunfire. The arrival times of the acoustic events at different sensor nodes are used to estimate the adjust of the sniper and send it to the base station with a special data appeal and routing service.Going back to peaceful applications, efforts are underway at Carnegie Mellon University and Intel for the design of IrisNet (profit-scale Resource-Intensive sensing element Network Services) B17, an architecture for a world colossal sensor web establish on common computing hardware such as Internet-connected PCs and cheap sensing hardware such as webcams. The network interface of a PC indeed senses the virtual environment of a LAN or the Internet rather than a somatogenetic environment with an architecture base on the concept of a distributed database B18, this hardware can be s centre of attention into a global sensor system hat responds to queries from users. 3. Characteristic Features of detector Networks In ad hoc networks, wireless nodes self-organize into an infrastructureless network with a dynamic topology. sensing element networks (such as the one in manikin 1) share these traits, but to a fault have several distinguishing features. The number of nodes in a typical sensor netw ork is oftentimes higher(prenominal)(prenominal) than in a typical ad hoc network, and dense deployments are often desired to ensure coverage and connectivity for these reasons, sensor network hardware must be cheap. Nodes typically have soaked brawniness limitations, which make them more(prenominal) failure-prone. They are enerally assumed to be stationary, but their comparatively frequent break chain reactors and the volatile nature of the wireless channel even result in a variable network topology. Ideally, sensor network hardware should be place- expeditious, small, inexpensive, and reliable in order to maximize network life story, add flexibility, facilitate data collection and minimize the need for primary(prenominal)tenance. animation Lifetime is extremely critical for most applications, and its primary limiting element is the nil consumption of the nodes, which need to be self- indicanting. Although it is often assumed that the intercommunicate exponent associat ed with acket transmission accounts for the lions share of power consumption, sensing, signal process and even hardware process in standby expressive style consume a consistent amount of power as thoroughly C19, C20. In some applications, extra power is needed for macro-scale actuation. Many researchers advise that zip fastener consumption could be reduced by considering the existing interdependencies between individualist layers in the network protocol stack. Routing and channel access protocols, for instance, could greatly benefit from an information exchange with the physical layer. At the physical layer, benefits can be obtained with ower radio receiver duty cycles and dynamic modulation scaling (varying the configuration size to minimize energy expenditure THIRD after part 2005 IEEE CIRCUITS AND SYSTEMS powder store 21 External Infrastructure Gateway Base Station comprehend Nodes Figure 1. A generic wine sensor network with a two-tiered archi1 tecture. converge Sect ion 5 for a hardware overview. D35). exploitation low-power modi for the mainframe or disabling the radio is generally advantageous, even though periodically number a subsystem on and off whitethorn be more pricy than always keeping it on. Techniques aimed at reducing the informal mode relief valve current in CMOS- ground rocessors are as well as noteworthy D36. Medium accession Control (MAC) solutions have a direct impact on energy consumption, as some of the primary causes of energy waste are found at the MAC layer collisions, control packet overhead and idle listening. great powersaving forward error control techniques are not easy to follow out due to the high amount of computing power that they charter and the point that long packets are normally not practical. Energy-efficient routing should avoid the loss of a node due to battery depletion. Many proposed protocols tend to minimize energy consumption on forwarding aths, but if some nodes happen to be located on most forwarding paths (e. g. , close to the base station), their lifetime will be reduced. Flexibility demodulator networks should be scalable, and they should be able to dynamically adapt to changes in node density and topology, like in the case of the self-healing minefields. In surveillance applications, most nodes whitethorn remain quiescent as long as nothing interesting happens. However, they must be able to respond to special events that the network intends to study with some degree of granularity. In a self-healing minefield, a number of sensing mines ay sleep as long as none of their peers explodes, but need to quickly force working(a) in the case of an enemy attack. Response time is also very critical in control applications (sensor/actuator networks) in which the network is to provide a delay-guaranteed service. Untethered systems need to self- set up and adapt to different conditions. sensing element networks should also be robust to changes in their topology, for instanc e due to the failure of individual nodes. In particular, connectivity and coverage should always be guaranteed. Connectivity is achieved if the base station can be reached from any node.Coverage can be seen as a footfall of quality of service in a sensor network C23, as it defines how well a particular surface area can be find by a network and characterizes the probability of detection of geographically restrict phenomena or events. Complete coverage is particularly important for surveillance applications. keep The only desired form of maintenance in a sensor network is the complete or partial update of the course of instruction legislation in the sensor nodes over the wireless channel. All sensor nodes should be updated, and the restrictions on the size of the new code should be the same as in the case of wired programming.Packet loss must be accounted for and should not impede correct reprogramming. The portion of code always representning in the node to guarantee reprogra mming support should have a small footprint, and modify procedures should only cause a brief interruption of the normal operation of the node C24. The functioning of the network as a whole should not be endangered by unavoidable failures of single nodes, which may occur for a number of reasons, from battery depletion to unpredictable external events, and may either be independent or spatially correlated C25. Faulttolerance is particularly authoritative as ongoing maintenance s rarely an option in sensor network applications. Self-configuring nodes are necessary to allow the deployment process to run smoothly without human interaction, which should in convention be limited to placing nodes into a given geographical area. It is not desirable to have humans piece nodes for habitat monitoring and destructively interfere with wildlife in the process, or configure nodes for urban warfare monitoring in a hostile environment. The nodes should be able to assess the quality of the networ k deployment and indicate any problems that may arise, as well as adjust to hanging environmental conditions by automatic reconfiguration. Location sensitiveness is important for selfconfiguration and has definite advantages in hurt of routing C26 and security. Time synchronization C27 is advantageous in promoting cooperation among nodes, such as data fusion, channel access, coordination of sleep modi, or security-related interaction. Data Collection Data collection is related to network connectivity and coverage. An interesting solution is the use of ubiquitous rambling agents that randomly move just about to gather data bridging sensor nodes and access points, whimsically named dataMULEs (Mobile ubiquitous LAN Extensions) in C28. The predictable mobility of the data authorize can be used to save power C29, as nodes can learn its schedule. A similar concept has been use in Intels tuner Vineyard. It is often the case that all data are relayed to a base station, but this form of centralized data collection may shorten network lifetime. Relaying data to a data sink causes non- uniform power consumption patterns that may overburden forwarding nodes C21. This is particularly harsh on nodes providing end links to base stations, which may end up relaying vocation coming from all ther nodes, then forming a critical bottleneck for network throughput A4, C22, as shown in Figure 2. An interesting technique is clustering C30 nodes team up to form clusters and transmit their information to their cluster heads, which fuse the data and forward it to a 22 IEEE CIRCUITS AND SYSTEMS cartridge THIRD one-quarter 2005 sink. fewer packets are transmitted, and a uniform energy consumption pattern may be achieved by periodic re-clustering. Data redundancy is minimized, as the collecting process fuses strongly correlated measurements. Many applications require that queries be sent to sensing nodes.This is rightful(a), for example, whenever the goal is gathering data rega rding a particular area where various sensors have been deployed. This is the rationale behind considering at a sensor network as a database C31. A sensor network should be able to protect itself and its data from external attacks, but the ascetic limitations of lower-end sensor node hardware make security a true challenge. Typical encryption schemes, for instance, require large amounts of memory that are out of stock(predicate) in sensor nodes. Data confidentiality should be preserved by encrypting data with a secret key shared with the intended receiver. Data justness should be ensured to revent unauthorized data alteration. An authenticated broadcast must allow the verification of the legitimacy of data and their sender. In a number of mercantile applications, a serious disservice to the user of a sensor network is compromising data availability (denial of service), which can be achieved by sleep-deprivation spin C33 batteries may be drained by continuous service requests o r demands for legitimate but intensive tasks C34, preventing the node from entering sleep modi. 4. ironware Design Issues In a generic sensor node (Figure 3), we can identify a power module, a communication block, a processing unit ith internal and/or external memory, and a module for sensing and actuation. Power Using stored energy or harvesting energy from the alfresco world are the two options for the power module. Energy storage may be achieved with the use of batteries or alternative devices such as sack cells or miniaturized set off engines, whereas energy-scavenging opportunities D37 are provided by solar power, vibrations, acoustic noise, and piezoelectric effects D38. The vast majority of the existing commercial and research platforms relies on batteries, which dominate the node size. Primary (nonrechargeable) batteries are often chosen, predominantlyAA, AAA and coin-type. base-forming batteries offer a high energy density at a cheap price, offset by a non-flat discha rge, a large physical size with respect to a typical sensor node, and a ledge life of only 5 years. Voltage regulation could in principle be employed, but its high inefficiency and large quiescent current consumption call for the use of components that can deal with large variations in the sum voltage A5. atomic number 3 cells are very compact and boast a flat discharge curve. Secondary (rechargeable) batteries are typically not desirable, as they offer a lower energy density and a higher cost, not to mention the fact that in most pplications recharging is simply not practical. arouse cells D39 are rechargeable electrochemical energy- conversion devices where electricity and heat are produced as long as hydrogen is supplied to react with oxygen. taint is minimal, as water is the main byproduct of the reaction. The possible of fuel cells for energy storage and power delivery is often higher than the one of traditional battery technologies, but the fact that they require hydrogen complicates their application. Using renewable energy and scavenging techniques is an interesting alternative. Communication Most sensor networks use radio communication, even if lternative solutions are offered by laser and infrared emission. Nearly all radio-based platforms use COTS (Commercial Off-The-Shelf) components. Popular choices accept the TR1000 from RFM (used in the MICA motes) and the CC1000 from Chipcon (chosen for the MICA2 platform). More recent solutions use industry standards like IEEE 802. 15. 4 (MICAz and Telos motes with CC2420 from Chipcon) or pseudo-standards like Bluetooth. Typically, the transmit power ranges between ? 25 dBm and 10 dBm, while the receiver sensitivity can be as good as ? 110 dBm. THIRD nincompoop 2005 IEEE CIRCUITS AND SYSTEMS MAGAZINE 23 Base Station Critical Nodes Figure 2.A uniform energy consumption pattern should avoid the depletion of the resources of nodes located in the vicinities of the base station. Communication ironware Power Sensors (? Actuators) ADC Memory Processor Figure 3. general anatomy of a generic sensor node. Spread spectrum techniques increase the channel reliableness and the noise tolerance by mobiliseing the signal over a wide range of frequencies. Frequency hopping (FH) is a spread spectrum technique used by Bluetooth the carrier frequency changes 1600 times per second on the radix of a pseudo-random algorithm. However, channel synchronization, hopping sequence search, and the high data rate ncrease power consumption this is one of the strongest caveats when using Bluetooth in sensor network nodes. In Direct Sequence Spread Spectrum (DSSS), communication is carried out on a single carrier frequency. The signal is multiplied by a higher rate pseudo-random sequence and thus spread over a wide frequency range (typical DSSS radios have spreading factors between 15 and 100). basal Wide Band (UWB) is of great interest for sensor networks since it meets some of their main requirements. UWB is a particular carrier-free spread spectrum technique where the RF signal is spread over a spectrum as large as several GHz.This implies that UWB signals look like noise to conventional radios. Such signals are produced using baseband pulses (for instance, Gaussian monopulses) whose length ranges from 100 ps to 1 ns, and baseband transmission is generally carried out by means of pulse assign modulation (PPM). Modulation and demodulation are indeed extremely cheap. UWB provides make-in ranging capabilities (a wideband signal allows a good time resolution and therefore a good location accuracy) D40, allows a very low power consumption, and performs well in the presence of multipath fading. Radios with comparatively low bit- rank (up to 100 kbps) re advantageous in terms of power consumption. In most sensor networks, high data rates are not needed, even though they allow shorter transmission times thus permitting lower duty cycles and alleviating channel access contention. It is also desirable for a radio to quickly switch from a sleep mode to an operational mode. Optical transceivers such as lasers offer a strong power advantage, mainly due to their high directionality and the fact that only baseband processing is require. Also, security is intrinsically guaranteed (intercepted signals are altered). However, the need for a line of bunch and recise localization makes this option impractical for most applications. Processing and compute Although low-power FPGAs king become a viable option in the near future D41, microcontrollers (MCUs) are now the primary choice for processing in sensor nodes. The key metric in the selection of an MCU is power consumption. Sleep modi merit special attention, as in many applications low duty cycles are essential for lifetime extension. Just as in the case of the radio module, a fast wake-up time is important. Most central processing units used in lower-end sensor nodes have clock speeds of a few MHz. The memory requirements depend on the pplication and the network topology data storage is not critical if data are often relayed to a base station. Berkeley motes, UCLAs Medusa MK-2 and ETHZs BTnodes use low-cost Atmel AVR 8-bit RISC microcontrollers which consume about 1500 pJ/instruction. More school platforms, such as the Intel iMote and Rockwell WINS nodes, use Intel StrongArm/XScale 32-bit mainframes. Sensing The high consume rates of modern digital sensors are usually not needed in sensor networks. The power efficiency of sensors and their turn-on and turn-off time are much more important. Additional issues are the physical ize of the sensing hardware, fabrication, and assembly compatibility with other components of the system. Packaging requirements come into play, for instance, with chemical sensors which require contact with the environment D42. Using a microcontroller with an onchip elongate comparator is another energy-saving technique which allows the node to avoid sampling values falling o utside a certain range D43. The ADC which complements analog sensors is particularly critical, as its resolution has a direct impact on energy consumption. Fortunately, typical sensor network applications do not have stringent resolution requirements.Micromachining techniques have allowed the miniaturization of many types of sensors. Performance does hang with sensor size, but for many sensor network applications size matters much more than accuracy. Standard integrated circuits may also be used as temperature sensors (e. g. , using the temperaturedependence of subthreshold MOSFETs and pn junctions) or light intensity transducers (e. g. , using photodiodes or phototransistors) D44. Nanosensors can offer promising solutions for biologic and chemical sensors while at the same time meeting the most ambitious miniaturization needs. 5. Existing Hardware political programsBerkeley motes, do commercially available by Crossbow, are by all means the best known sensor node hardware imple mentation, used by more than 100 research organizations. They consist of an embedded microcontroller, low-power radio, and a small memory, and they are powered by two AA batteries. MICA and MICA2 are the most successful families of Berkeley motes. The MICA2 platform, whose layout is shown in Figure 4, is equipped with an Atmel ATmega128L and has a CC1000 transceiver. A 51-pin working out connector is available to interface sensors (commercial sensor boards designed for this specific platform are available).Since the MCU is to handle 24 IEEE CIRCUITS AND SYSTEMS MAGAZINE THIRD QUARTER 2005 medium access and baseband processing, a fine-grained event-driven real-time operating system (TinyOS) has been implemented to specifically address the concurrency and resource management needs of sensor nodes. For applications that require a better form factor, the circular MICA2Dot can be used it has most of the resources of MICA2, but is only 2. 5 cm in diameter. Berkeley motes up to the MICA2 times cannot interface with other wireless- enabled devices E47. However, the newer generations MICAz and Telos support IEEE 802. 15. , which is part of the 802. 15 wireless Personal Area Network (WPAN) standard being veritable by IEEE. At this point, these devices represent a very good solution for generic sensing nodes, even though their unit cost is still relatively high (about $100$200). The proliferation of different lowerend hardware platforms within the Berkeley mote family has recently led to the organic evolution of a new version of TinyOS which introduces a ductile hardware annulion architecture to simplify multi-platform support E48. Tables 1 and 2 show an overview of the radio transceivers and the microcontrollers most commonly used in xisting hardware platforms an overview of the key features of the platforms is provided in Table 3. Intel has designed its own iMote E49 to implement various improvements over available mote designs, such as increased CPU processing power, increased main memory size for on-board computing and improve radio reliability. In the iMote, a the right way ARM7TDMI core is complemented by a large main memory and non-volatile storage area on the radio side, Bluetooth has been chosen. Various platforms have been developed for the use of Berkeley motes in mobile sensor networks to enable investigations into controlled mobility, which facilitates eployment and network repair and provides possibilities for the implementation of energy-harvesting. UCLAs RoboMote E50, Notre Dames MicaBot E51 and UC Berkeleys CotsBots E52 are examples of efforts in this direction. UCLAs Medusa MK-2 sensor nodes E53, developed for the Smart Kindergarten project, expand Berkeley motes with a second microcontroller. An on-board power management and introduce unit monitors power consumption within the different subsystems and selectively powers down unused parts of the node. UCLA has also developed iBadge E54, a wearable sensor node with suffic ient computational power to process the sensed data.Built around an ATMega128L and a DSP, it features a Localization Unit designed to estimate the position of iBadge in a room based on the presence of special nodes of known location attached to the ceilings. In the context of the EYES project (a joint effort among several European institutions) custom nodes E55, C24 have been developed to test and demonstrate energy-efficient networking algorithms. On the software side, a proprietary operating system, PEEROS (Preemptive EYES Real Time operating(a) System), has been implemented. The Smart-Its project has investigated the possibility of embedding computational power into objects, leading o the asylum of three hardware platforms DIY Smartits, soupcon electronic computers and BTnodes. The DIY Smart-its E56 have been developed in the UK at Lancaster University their modular design is based on a core board that provides processing and communication and can be extended with add-on board s. A typical apparatus of Smart-its consists of one or more sensing nodes that broadcast their data to a base station which consists of a standard core board connected to the in series(p) port of a PC. Simplicity and extensibility are the key features of this platform, which has been developed for the creation of Smart Objects.An interesting application is the Weight Table four clog cells placed underneath a coffee table form a Wheatstone span and are connected to a DIY node that observes load changes, determines event types like placement and removal of objects or a person locomote a finger across the surface, and also retrieves the position of an object by correlating the values of the individual load cells after the event type (removed or placed) has been recognized E57. atom Computers have been developed at the University of Karlsruhe, Germany. Similarly to the DIY platform, the Particle Smart-its are based on a core board quipped with a Microchip PIC they are optimized fo r energy efficiency, scalable communication and small scale (17 mm ? 30 mm). Particles communicate in an ad hoc fashion as two Particles come close to one another, THIRD QUARTER 2005 IEEE CIRCUITS AND SYSTEMS MAGAZINE 25 Oscillator 7. 3728-MHz DS2401P Silicon Serial No. Antenna continuative Connector LEDs Battery Connection 32. 768-kHz Oscillator 14. 7456-MHz Oscillator ATMEL ATMega 128L CPU CC1000 Transceiver ATMEL AT45DB041 Data gilded Figure 4. Layout of the MICA2 platform. they are able to talk. Additionally, if Particles come near a adit device, they can be connected to Internet-enabled evices and access services and information on the Internet as well as provide information E58. The BTnode hardware from ETHZ E47 is based on an Atmel ATmega128L microcontroller and a Bluetooth module. Although advertised as a low-power technology, Bluetooth has a relatively high power consumption, as discussed before. It also has long connection setup times and a lower degree of freedom with respect to accomplishable network topologies. On the other hand, it ensures interoperability between different devices, enables application development through a standardized interface, and offers a importantly higher bandwidth (about 1 Mbps) ompared to many low-power radios (about 50 Kbps). Moreover, Bluetooth support means that COTS hardware can be used to create a gateway between a sensor network and an external network (e. g. , the Internet), as opposed to more expensive proprietary solutions E59. MIT is working on the ? AMPS (? -Adaptive Multidomain Power-aware Sensors) project, which explores energy-efficiency constraints and key issues such as selfconfiguration, reconfigurability, and flexibility. A first prototype has been designed with COTS components three stackable boards (processing, radio and power) and an ptional extension module. The energy dissipation of this microsensor node is reduced through a variety of poweraware design techniques D45 including fine-grain shut down of inactive components, dynamic voltage and frequency scaling of the processor core, and adjustable radio transmission power based on the required range. Dynamic voltage scaling is a technique used for active power management where the supply voltage and clock frequency of the processor are regulated depending on the computational load, which can vary significantly based on the operational mode D36, C20. The main oal of second generation ? AMPS is clearly stated in D46 as breaking the 100 ? W average power barrier. Another interesting MIT project is the pushpin computing system E60, whose goal is the modelling, testing, and deployment of distributed peer-to-peer sensor networks consisting of many identical nodes. The pushpins are 18 mm ? 18 mm modular devices with a power substratum, an infrared communication module, a processing module (Cygnal C8051F016) and an expansion module (e. g. , for sensors) they are powered by direct contact between the power substrate and layered co nductive sheets. 26 MCU Max.Freq. MHz Memory Data Size bits ADC bits architecture AT90LS8535 (Atmel) 4 8 kB cremate, 512B EEPROM, 512B SRAM 8 10 AVR ATMega128L (Atmel) 8 128 kB Flash, 4 kB EEPROM, 4 kB SRAM 8 10 AVR AT91FR4081 (Atmel) 33 136 kB On-Chip SRAM, 8 Mb Flash 32 Based on ARM core (ARM7TDMI) MSP430F149 (TI) 8 60 kB + 256B Flash, 2 kB RAM 16 12 Von Neumann C8051F016 (Cygnal) 25 2304B RAM, 32 kB Flash 8 10 Harvard 8051 PIC18F6720 (Microchip) 25 128 kB Flash, 3840B SRAM, 1 kB EEPROM 8 10 Harvard PIC18F252 (Microchip) 40 32 K Flash, 1536B RAM, 256B EEPROM 8 10 Harvard StrongARM SA-1110 (Intel) 133 32 ARM v. 4PXA255 (Intel) 400 32 kB Instruction Cache, 32 kB Data 32 ARM v. 5TE Cache, 2 kB miniskirt Data Cache Table 2. Microcontrollers used in sensor node platforms. Radio (Manufacturer) Band MHz Max. Data Rate kbps Sensit. dBm Notes TR1000 (RFM) 916. 5 115. 2 ? 106 OOK/ quest TR1001 (RFM) 868. 35 115. 2 ? 106 OOK/ASK CC1000 (Chipcon) 3001,000 76. 8 ? 110 FSK, ? 20 to 10 dB m CC2420 (Chipcon) 2,400 250 ? 94 OQPSK, ? 24 to 0 dBm, IEEE 802. 15. 4, DSSS BiM2 (Radiometrix) 433. 92 64 ? 93 9XStream (MaxStream) 902928 20 ? 114 FHSS Table 1. Radios used in sensor node platforms. IEEE CIRCUITS AND SYSTEMS MAGAZINE THIRD QUARTER 2005MIT has also built Tribble (Tactile reactive interface built by linked elements), a spherical robot wrapped by a wired skinlike sensor network designed to emulate the functionalities of biological skin E61. Tribbles surface is divided into 32 patches with a Pushpin processing module and an array of sensors and actuators. At Lancaster University, surfaces provide power and network connectivity in the Pin&Play project. Network nodes come in different form factors, but all share the Pin&Play connector, a custom component that allows physical connection and networking through conductive sheets which re embedded in surfaces such as a fence in or a bulletin board E62. Pin&Play falls in between wired and wireless technologies as it provid es network access and power across 2D surfaces. Wall-mounted objects are especially suited to be augment to become Pin&Play objects. In a demonstration, a wall switch was augmented and freely placed anywhere on a wall with a Pin&Play surface as wallpaper. For applications which do not call for the minimization of power consumption, high-end nodes are available. Rockwellis WINS nodes and Sensorias WINS 3. 0 Wireless Sensing Platform are equipped with more powerful rocessors and radio systems. The embedded PC modules based on widely support standards PC/104 and PC/104-plus feature Pentium processors moreover, PC/104 peripherals include digital I/O devices, sensors and actuators, and PC-104 products support almost all PC software. PFU Systems Plug-N-Run products, which feature Pentium processors, also belong to this category. They offer the capabilities of PCs and the size of a sensor node, but inadequacy built-in communication hardware. COTS components or lower-end nodes may be used in this sense C32. Research is underway toward the creation of sensor nodes that are more capable than the motes, yet maller and more power-efficient than higher-end nodes. Simple yet effective gateway devices are the MIB programming boards from Crossbow, which bridge networks of Berkeley motes with a PC (to which they interface using the serial port or Ethernet). In the case of Telos motes, any generic node (i. e. , any Telos mote) can act as a gateway, as it may be connected to the USB port of a PC and bridge it to the network. Of course, more powerful gateway devices are also available. Crossbows Stargate is a powerful embedded computing platform (running Linux) with enhanced communication and sensor signal processing capabilities based n Intel PXA255, the same X-Scale processor that forms the core of Sensoria WINS 3. 0 nodes. Stargate has a connector for Berkeley motes, may be bridged to a PC via Ethernet or 802. 11, and includes built-in Bluetooth support. 6. Closing Remarks Se nsor networks offer countless challenges, but their versatility and their broad range of applications are eliciting more and more interest from the research community as well as from industry. Sensor networks have the potential of triggering the next revolution in information technology. The challenges in terms of circuits and systems re numerous the development of low-power communication hardware, low-power microcontrollers, MEMSbased sensors and actuators, efficient AD conversion, and energy-scavenging devices is necessary to enhance the potential and the performance of sensor networks. System integration is another major challenge that sensor networks offer to the circuits and systems research community. We believe that CAS can and should have a significant impact in this emerging, exciting area. 27 Platform CPU Comm. External Memory Power Supply WesC (UCB) AT90LS8535 TR1000 32 kB Flash Lithium Battery MICA (UCB, Xbow) ATMega128L TR1000 512 kB Flash AAMICA2 (UCB, Xbow) ATMega128L CC1000 512 kB Flash AA MICA2Dot (UCB, Xbow) ATMega128L CC1000 512 kB Flash Lithium Battery MICAz (UCB, Xbow) ATMega128L CC2420 512 kB Flash AA Telos (Moteiv) MSP430F149 CC2420 512 kB Flash AA iMote (Intel) ARM7TDMI Core Bluetooth 64 kB SRAM, 512 kB Flash AA Medusa MK-2 (UCLA) ATMega103L TR1000 4 Mb Flash reversible Lithium Ion AT91FR4081 iBadge (UCLA) ATMega128L Bluetooth, TR1000 4 Mb Flash Rechargeable Lithium Ion DIY (Lancaster University) PIC18F252 BiM2 64 Kb FRAM AAA, Lithium, Rechargeable Particle (TH) PIC18F6720 RFM TR1001 32 kB EEPROM AAA or Lithium light upon Battery or RechargeableBT Nodes (ETHZ) ATMega128L Bluetooth, CC1000 244 kB SRAM AA ZebraNet (Princeton) MSP430F149 9XStream 4 Mb Flash Lithium Ion Pushpin (MIT) C8051F016 Infrared Power Substrate WINS 3. 0 (Sensoria) PXA255 802. 11b 64 MB SDRAM, 32 MB + 1 GB Flash Batteries Table 3. Hardware features of various platforms. THIRD QUARTER 2005 IEEE CIRCUITS AND SYSTEMS MAGAZINE Acknowledgments The support of NSF (grant s ECS 03-29766 and race systema nervosum centrale 04-47869) is gratefully acknowledged. References General References A1 I. F. Akyildiz, W. Su, Y. Sankarasubramaniam, and E. Cayirci, A survey on sensor networks, in IEEE communications Magazine, pp. 02114, Aug. 2002. A2 L. B. Ruiz, L. H. A. Correia, L. F. M. Vieira, D. F. Macedo, E. F. Nakamura, C. M. S. Figueiredo, M. A. M. Vieira, E. H. B. Maia, D. Camara, A. A. F. Loureiro, J. M. S. Nogueira, D. C. da silva Jr. , and A. O. Fernandes, computer architectures for wireless sensor networks (In Portuguese), in proceeding of the 22nd Brazilian Symposium on Computer Networks (SBRC04), Gramado, Brazil, pp. 167218, May 2004. Tutorial. ISBN 85-88442-82-5. A3 C. Y. Chong and S. P. Kumar, Sensor networks Evolution, opportunities, and challenges, in IEEE minutes, pp. 12471254, Aug. 003. A4 M. Haenggi, Opportunities and Challenges in Wireless Sensor Networks, in vade mecum of Sensor Networks Compact Wireless and Wired Sensing Systems, M. Ilya s and I. Mahgoub, eds. , Boca Raton, FL, pp. 1. 11. 14, CRC Press, 2004. A5 J. Hill, System Architecture for Wireless Sensor Networks. Ph. D. dissertation, University of California at Berkeley, Spring 2003. Applications B6 A. Mainwaring, J. Polastre, R. Szewczyk, D. Culler, and J. Anderson, Wireless sensor networks for habitat monitoring, in First ACM shop class on Wireless Sensor Networks and Applications, Atlanta, GA, Sept. 002. B7 A. Cerpa, J. Elson, D. Estrin, L. Girod, M. Hamilton, and J. Zhao, Habitat monitoring Application driver for wireless communications technology, in ACM SIGCOMM store on Data Communications in Latin the States and the Caribbean, San Jose, Costa Rica, Apr. 2001. B8 E. Biagioni and K. Bridges, The application of remote sensor technology to assist the recovery of rare and endangered species, world(prenominal) Journal of High Performance computation Applications, vol. 16, pp. 315324, Aug. 2002. B9 P. Juang, H. Oki, Y. Wang, M. Martonosi, L. Peh, and D.Ru benstein, Energy-efficient computing for wildlife tracking Design tradeoffs and early experiences with ZebraNet, in minutes of the tenth worldwide conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS-X), San Jose, CA, Oct. 2002. B10 G. Werner-Allen, J. Johnson, M. Ruiz, J. Lees, and M. Welsh, Monitoring volcanic eruptions with a wireless sensor network, in proceedings of the Second European workshop on Wireless Sensor Networks (EWSN05), Jan. 2005. B11 J. Burrell, T. Brooke, and R. Beckwith, Vineyard computing Sensor networks in agricultural production, IEEE Pervasive computation, vol. , no. 1, pp. 3845, 2004. B12 M. Srivastava, R. Muntz, and M. Potkonjak, Smart kindergarten Sensor- based wireless networks for flip developmental problem-solving enviroments, in Proceedings of the 7th Annual foreign gathering on Mobile Computing and Networking (MobiCom01), Rome, Italy, pp. 132138, 2001. B13 T. Fulford-Jones, D. Malan, M. Welsh, and S. Moul ton, CodeBlue An ad hoc sensor network infrastructure for emergency medical care, in International Workshop on Wearable and Implantable automobile trunk Sensor Networks, London, UK, 2004. B14 D. Myung, B. Duncan, D. Malan, M. Welsh, M.Gaynor, and S. Moulton, Vital dust Wireless sensors and a sensor network for realtime patient monitoring, in 8th Annual New England Regional Trauma Conference, Burlington, MA, 2002. B15 Self-healing Mines http//www. darpa. mil/ato/programs/SHM/. B16 M. Maroti, G. Simon, A. Ledeczi, and J. Sztipanovits, crack localization in urban terrain, IEEE Computer, vol. 37, pp. 6061, Aug. 2004. B17 P. Gibbons, B. Karp, Y. Ke, S. Nath, and S. Seshan, IrisNet An architecture for a global sensor web, IEEE Pervasive Computing, vol. 2, no. 4, pp. 2233, 2003. B18 P. Gibbons, B. Karp, Y. Ke, S.Nath, and S. Seshan, IrisNet An Architecture for Enabling Sensor-Enriched Internet Service, Tech. Rep. IRP-TR-03-04, Intel Research, Pittsburgh, PA, June 2003. Characteristic Fe atures Lifetime C19 A. Goldsmith and S. Wicker, Design challenges for energy-constrained ad hoc wireless networks, IEEE Wireless Communications Magazine, vol. 9, pp. 827, Aug. 2002. C20 L. Yuan and G. Qu, Energy-efficient Design of Distributed Sensor Networks, in enchiridion of Sensor Networks Compact Wireless and Wired Sensing Systems, M. Ilyas and I. Mahgoub, eds. , Boca Raton, FL, pp. 38. 138. 19, CRC Press, 2004. C21 M.Haenggi, twelve Reasons not to Route over Many Short Hops, in IEEE vehicular Technology Conference (VTC04 Fall), Los Angeles, CA, Sept. 2004. C22 M. Haenggi, Energy-Balancing Strategies for Wireless Sensor Networks, in IEEE International Symposium on Circuits and Systems (ISCAS03), Bangkok, Thailand, May 2003. Coverage C23 S. Meguerdichian, F. Koushanfar, M. Potkonjak, and M. Srivastava, Coverage problems in wireless ad-hoc sensor networks, in Proceedings of the 20th Annual Joint Conference of the IEEE Computer and Communications Societies (INFOCOM01), vol. 3, Anc horage, AK, pp. 13801387, Apr. 001. Maintenance C24 N. Reijers and K. Loangendoen, Efficient code distribution in wireless sensor networks, in Second ACM International Workshop on Wireless Sensor Networks and Applications, San Diego, CA, Sept. 2003. C25 D. Ganesan, R. Govindan, S. Shenker, and D. Estrin, Highly resilient, energy efficient multipath routing in wireless sensor networks, in Proceedings of the 2nd ACM International Symposium on Mobile Ad Hoc Networking and Computing (MobiHoc01), Long Beach, CA, pp. 251254, 2001. Localization and Synchronization C26 M. Mauve, H. Hartenstein, H. Fuessler, J. Widmer, and W.Effelsberg, Positionsbasiertes Routing fuer die Kommunikation zwischen Fahrzeugen, itInformation Technology (formerly it + ti)Methoden und groundbreaking Anwendungen der Informatik und Informationstechnik, vol. 44, pp. 278286, Oct. 2002. C27 F. Sivrikaya and B. Yener, Time synchronization in sensor networks A survey, IEEE Network, vol. 18, pp. 4550, JulyAug. 2004. Data Collection, Routing, and Architectures C28 R. C. Shah, S. Roy, S. Jain, and W. Brunette, Data MULEs Modeling and analysis of a three-tier architecture for sparse sensor networks, in Ad Hoc Networks Journal, vol. 1, pp. 215233, Elsevier, Sept. 2003. C29 A.Chakrabarti, A. Sabharwal, and B. Aazhang, Using predictable observer mobility for power efficient design of sensor networks, in Information Processing in Sensor Networks (IPSN03), Palo Alto, CA, Apr. 2003. C30 O. Younis and S. Fahmy, HEED A Hybrid, Energy-Efficient, Distributed Clustering Approach for Ad-hoc Sensor Networks, in IEEE transactions on Mobile Computing, vol. 3, pp. 366379, 2004. C31 R. Govindan, J. Hellerstein, W. Hong, S. Madden, M. Franklin, and S. Shenker, The Sensor Network as a Database, Tech. Rep. 02771, University of southern California, 2002. ftp//ftp. usc. edu/pub/csinfo/ tech-reports/papers/02-771. df. C32 M. Yarvis and W. Ye, Tiered Architectures in Sensor Networks, in Handbook of Sensor Networks Compact W ireless and Wired Sensing Systems, M. Ilyas and I. Mahgoub, eds. , Boca Raton, FL, pp. 13. 113. 22, CRC Press, 2004. Security C33 F. Stajano and R. Anderson, The resurrecting duckling Security issues for ad-hoc wireless networks, in 7th International Workshop on Security Protocols, Cambridge, UK, Apr. 1999. C34 T. Martin, M. Hsiao, D. Ha, and J. Krishnaswami, Denial-of-service attacks on battery-powered mobile computers, in Proceedings of the 2nd IEEE Pervasive Computing Conference, Orlando, FL, pp. 09318, Mar. 2004. Hardware D35 C. Schurgers, O. Aberthorne, and M. Srivastava, Modulation scaling for energy aware communication systems, in Proceedings of the 2001 International Symposium on first gear Power Electronics and Design, 28 IEEE CIRCUITS AND SYSTEMS MAGAZINE THIRD QUARTER 2005 Huntington Beach, CA, pp. 9699, Aug. 2001. D36 A. P. Chandrakasan, R. Min, M. Bhardwaj, S. Cho, and A. Wang, Power aware wireless microsensor systems, in 28th European Solid- State Circuits Conference (ESSCIRC02), Florence, Italy, 2002. D37 S. Roundy, P. Wright, and J. Rabaey, A study of low level vibrations as a power source for ireless sensor nodes, Computer Communications, vol. 26, pp. 11311144, July 2003. D38 J. Kymissis, C. Kendall, J. Paradiso, and N. Gershenfeld, Parasitic power harvesting in shoes, in Proceedings of the 2nd IEEE International Symposium on Wearable Computers (ISWC04), Pittsburgh, PA, Oct. 1998. D39 A. J. Appleby, Fuel Cell Handbook, New York, NY Van Reinhold Co. , 1989. D40 W. C. Chung and D. S. Ha, An Accurate Ultra WideBand (UWB) Ranging for preciseness asset location, in International Conference on UWB Systems and Technologies, Reston, VA, Nov. 2002. D41 M. Vieira, D. da Silva Jr. C. C. Jr. , and J. da Mata, Survey on wireless sensor network devices, in Proceedings of the 9th IEEE International Conference on rising Technologies and Factory Automation (ETFA03), Lisbon, Portugal, Sept. 2003. D42 B. A. Warneke and K. S. J. Pister, MEMS for distributed wir eless sensor networks, in Proceedings of the 9th International Conference on Electronics, Circuits and Systems (ICECS02), vol. 1, Dubrovnik, Croatia, pp. 291294, 2002. D43 Z. Karakehayov, Low-Power Design for Smart Dust Networks, in Handbook of Sensor Networks Compact Wireless and Wired Sensing Systems, M.Ilyas and I. Mahgoub, eds. , Boca Raton, FL, pp. 37. 137. 12, CRC Press, 2004. D44 B. Warneke, Miniaturizing Sensor Networks with MEMS, in Handbook of Sensor Networks Compact Wireless and Wired Sensing Systems, M. Ilyas and I. Mahgoub, eds. , Boca Raton, FL, pp. 5. 15. 19, CRC Press, 2004. D45 R. Min, M. Bhardwaj, S. Cho, A. Sinha, E. Shih, A. Wang, and A. P. Chandrakasan, An Architecture for a Power-Aware Distributed Microsensor Node, in IEEE Workshop on request Processing Systems (SiPS00), Lafayette, LA, Oct. 2000. D46 D. D. Wentzloff, B. H.Calhoun, R. Min, A. Wang, N. Ickes, and A. P. Chandrakasan, Design considerations for next generation wireless power-aware microsensor nodes , in Proceedings of the 17th International Conference on VLSI Design, Mumbai, India, pp. 361367, 2004. Existing Platforms E47 J. Beutel, O. Kasten, M. Ringwald, F. Siegemund, and L. Thiele, Poster abstract Btnodesa distributed platform for sensor nodes, in Proceedings of the First International Conference on Embedded Networked Sensor Systems (SenSys-03), Los Angeles, CA, Nov. 2003. E48 V. Handziski, J. Polastre, J. H. Hauer, C. Sharp, A. Wolisz, and D. Culler, Flexible hardware generalisation for wireless sensor networks, in Proceedings of the 2nd International Workshop on Wireless Sensor Networks (EWSN 2005), Istanbul, Turkey, Jan. 2005. E49 R. M. Kling, Intel Mote An Enhanced Sensor Network Node, in International Workshop on Advanced Sensors, Structural health Monitoring and Smart Structures at Keio University, Tokyo, Japan, Nov. 2003. E50 K. Dantu, M. Rahimi, H. Shah, S. Babel, A. Dhariwal, and G. Sukhatme, Robomote Enabling Mobility In Sensor Networks, Tech. Rep.CRES-04-006, Un iversity of grey California. E51 M. B. McMickell, B. Goodwine, and L. A. Montestruque, MICAbot A robotic platform for large-scale distributed robotics, in Proceedings of International Conference on Intelligent Robots and Systems (ICRA03), vol. 2, Taipei, Taiwan, pp. 16001605, 2003. E52 S. Bergbreiter and K. S. J. Pister, CotsBots An Off-the-Shelf Platform for Distributed Robotics, in Proceedings of the 2003 IEEE International Conference on Intelligent Robots and Systems (ICRA03), Las Vegas, NV, Oct. 2003. E53 A. Savvides and M. B.Srivastava, A distributed computation platform for wireless embedded sensing, in 20th International Conference on Computer Design (ICCD02), Freiburg, Germany, Sept. 2002. E54 S. Park, I. Locher, and M. Srivastava, Design of a wearable sensor badge for smart kindergarten, in 6th International Symposium on Wearable Computers (ISWC2002), Seattle, WA, pp. 13. 113. 22, Oct. 2002. E55 L. F. W. van Hoesel, S. O. Dulman, P. J. M. Havinga, and H. J. Kip, Design of a low-power testbed for Wireless Sensor Networks and verification, Tech. Rep. R-CTIT-03-45, University of Twente, Sept. 2003. E56 M.Strohbach, The smart-its platform for embedded contextaware systems, in Proceedings of the First International Workshop on Wearable and Implantable Body Sensor Networks, London, UK, Apr. 2004. E57 A. Schmidt, M. Strohbach, K. V. Laerhoven, and H. -W. Gellersen, Ubiquitous interactionUsing surfaces in everyday environments as pointing devices, in 7th ERCIM Workshop User Interfaces For All, Chantilly, France, 2002. E58 M. Beigl, A. Krohn, T. Zimmer, C. Decker, and P. Robinson, Aware- Con Situation aware context communication, in The Fifth International Conference on Ubiquitous Computing (Ubicomp03), Seattle, WA, Oct. 003. E59 J. Beutel, O. Kasten, F. Mattern, K. Roemer, F. Siegemund, and L. Thiele, Prototyping sensor network applications with BTnodes, in IEEE European Workshop on Wireless Sensor Networks (EWSN04), Berlin, Germany, Jan. 2004. E60 J. Lifton , D. Seetharam, M. Broxton, and J. Paradiso, Pushpin computing system overview A platform for distributed, embedded, ubiquitous sensor networks, in Proceedings of the Pervasive Computing Conference, Zurich, Switzerland, Aug. 2002. E61 J. A. Paradiso, J. Lifton, and M. Broxton, sentient mediamultimodal electronic skins as dense sensor networks, BT Technology Journal, vol. 2, pp. 3244, Oct. 2004. E62 K. V. Laerhoven, N. Villar, and H. -W. Gellersen, Pin&Mix When Pins Become Interaction Components. . . , in Physical Interaction (PI03) Workshop on Real World User InterfacesMobile HCI Conference, Udine, Italy, Sept. 2003. Daniele Puccinelli received a Laurea degree in Electrical Engineering from the University of Pisa, Italy, in 2001. After consumption two years in industry, he conjugate the graduate program in Electrical Engineering at the University of Notre Dame, and received an M. S. Degree in 2005. He is currently working toward his Ph. D. degree.His research has focused on cross -layer approaches to wireless sensor network protocol design, with an emphasis on the interaction between the physical and the network layer. Martin Haenggi received the Dipl. Ing. (M. Sc. ) degree in electrical engineering from the Swiss Federal Institute of Technology in Zurich (ETHZ) in 1995. In 1995, he joined the Signal and Information Processing Laboratory at ETHZ as a Teaching and Research Assistant. In 1996 he make the Dipl. NDS ETH (post-diploma) degree in information technology, and in 1999, he completed his Ph. D. thesis on the analysis, design, and optimization of ellular neural networks. After a postdoctoral year at the Electronics Research Laboratory at the University of California in Berkeley, he joined the Department of Electrical Engineering at the University of Notre Dame as an champion professor in January 2001. For both his M. Sc. and his Ph. D. theses, he was awarded the ETH medal, and he received an NSF CAREER award in 2005. For 2005/06, he is a CAS Distingui shed Lecturer. His scientific interests include networking and wireless communications, with an emphasis on ad hoc and sensor networks. THIRD QUARTER 2005 IEEE CIRCUITS AND SYSTEMS MAGAZINE 29