1. Introduction
The field of computer science is constantly evolving to process larger data sets and maintain higher levels of connectivity. At same time, advances in miniaturization allow for increased mobility and accessibility. Body Area Networks represent the natural union between connectivity and miniaturization. A Body Area Network (BAN) is defined formally as a system of devices in close proximity to a persons body that cooperate for the benefit of the user. The BBC's Jo Twist gave a more informal definition of Body Area Networks in her article title When technology gets personal:
Inanimate objects will start to interact with us: we will be surrounded – on streets, in homes, in appliances, on our bodies and possibly in our heads -by things that "think".
Forget local area networks - these will be body area networks.
Twist makes the possibility of BAN sound more like science fiction than a real possibility, but several experts in the field expect to see BAN in production for general use by 2010 . While this might seem like an aggressive estimate, when put into context with the history and development of BAN up to this point it becomes a much more achievable goal. In the paper we will start off introducing the reader to the history and development of BAN. We will cover the medical heritage of BAN and how the technology grew from a simple generalization of the concept of Body Sensor Networks (BSN). We will investigate current applications of BAN with an emphasis on applications in the medical sector. As we cover applications of BAN, we will spend a portion of the paper identifying some of technical problems facing BAN. Finally, we will conclude the paper with several solutions currently in development and how they hope to address and overcome the challenges inherent to BAN.
Wearable health monitoring systems integrated into a telemedicine system are novel information technology that will be able to support early detection of abnormal conditions and prevention of its serious consequences . Many patients can benefit from continuous monitoring as a part of a diagnostic procedure, optimal maintenance of a chronic condition or during supervised recovery from an acute event or surgical procedure. Important limitations for wider acceptance of the existing systems for continuous monitoring are: a) unwieldy wires between sensors and a processing unit, b) lack of system integration of individual sensors, c) interference on a wireless communication channel shared by multiple devices, and d) nonexistent support for massive data collection and knowledge discovery. Traditionally, personal medical monitoring systems, such as Holter monitors, have been used only to collect data for off-line processing. Systems with multiple sensors for physical rehabilitation feature unwieldy wires between electrodes and the monitoring system. These
wires may limit the patient's activity and level of comfort and thus negatively influence the measured results. A wearable health-monitoring device using a Personal Area Network (PAN) or Body Area Network (BAN) can be integrated into a user's clothing .This system organization, however, is unsuitable for lengthy, continuous monitoring, particularly during normal activity , intensive training or computer-assisted rehabilitation . Recent technology advances in wireless networking, micro-fabrication, and integration of physical sensors, embedded microcontrollers and radio interfaces on a single chip, promise a new generation of wireless sensors suitable for many applications.
However, the existing telemetric devices either use wireless communication channels exclusively to transfer raw data from sensors to the monitoring station, or use standard high-level wireless protocols such as Bluetooth that are too complex, power demanding, and prone to interference by other devices operating in the same frequency range. These characteristics limit their use for prolonged wearable monitoring. Simple, accurate means of monitoring daily activities outside of the laboratory are not available; at the present, only estimates can be obtained from questionnaires, measures of heart rate, video assessment, and use of pedometers or accelerometers .Finally, records from individual monitoring sessions are rarely integrated into research databases that would provide support for data mining and knowledge discovery relevant to specific conditions and patient categories. Increased system processing power allows sophisticated real-time data processing within the confines of the wearable system. As a result, such wearable system can support biofeedback and generation of warnings. The use of biofeedback techniques has gained increased attention among researchers in the field of physical medicine and tele-rehabilitation. Intensive practice schedules have been shown to be important for recovery of motor function. Unfortunately, an aggressive approach to rehabilitation involving extensive therapist-supervised motor training is not a realistic expectation in today's health care system where individuals are typically seen as outpatients about twice a week for no longer than 30–45 min. Wearable technology and biofeedback systems appear to be a valid alternative, as they reduce the extensive time to setup a patient before each session and require limited time involvement of physicians and therapists. Furthermore, Wear able technology could potentially address a second factor that hinders enthusiasm for rehabilitation, namely the fact that setting up a patient for the procedure is rather time-consuming. This is because tethered sensors need to be positioned on the subject, attached to the equipment, and a software application needs to be started before each session. Wearable technology allows sensors that will be positioned on the subject for prolonged periods, therefore eliminating the need to position them for every training session. Instead, a personal server such as a PDA can almost instantly initiate a new training session whenever the subject is ready and willing to exercise. In addition to home rehabilitation, this setting also may be beneficial in the clinical setting, where precious time of physicians and therapists could be saved. Moreover, the system can issue timely warnings or alarms to the patient, or to a specialized medical response service in the event of significant deviations of the norm or medical emergencies. However, as for all systems, regular, routine maintenance (verifying configuration and thresholds) by a specialist is required.
Typical examples of possible applications include stroke rehabilitation, physical rehabilitation after hip or knee surgeries, myocardial infarction rehabilitation, and traumatic brain injury rehabilitation. The assessment of the effectiveness of rehabilitation procedures has been limited to the laboratory setting; relatively little is known about rehabilitation in real-life situations. Miniature, wireless, wearable technology offers a tremendous opportunity to address this issue. We propose a wireless BAN composed of off-the-shelf sensor platforms with application-specific signal conditioning modules. In this paper, we present general system architecture and describe a recently developed activity sensor "ActiS". ActiS is based on a standard wireless sensor platform and a custom sensor board with a one-channel bio amplifier and two accelerometers. As a heart sensor, ActiS can be used to monitor heart activity and position of the upper trunk. The same sensor can be used to monitor position and activity of upper and lower extremities. A wearable system with ActiS sensors would also allow one to assess metabolic rate and cumulative energy expenditure as a valuable parameter in the management of many medical conditions. An early version of the ActiS has been based on a custom developed wireless intelligent sensor and custom wireless protocols in the license-free 900 MHz Scientific and Medical Instruments (ISM) band. Our initial experience indicated the importance of standard sensor platforms with ample processing power, minute power consumption, and standard software support. Such platforms were not available on the market during the design of our first prototype system. The recent introduction of an IEEE standard for low-power personal area networks (802.15.4) and ZigBee protocol stack, as well as new ZigBee compliant Telos
Figure: Wireless Body Area Network of Intelligent Sensors for Patient Monitoring
2. History and Development of BAN
BAN technology is still an emerging technology, and as such it has a very short history. BAN technology emerges as the natural byproduct of existing sensor network technology and biomedical engineering. Professor Guang-Zhong Yang was the first person to formally define the phrase "Body Sensor Network" (BSN) with publication of his book Body Sensor Networks in 2006. BSN technology represents the lower bound of power and bandwidth from the BAN use case scenarios. However, BAN technology is quite flexible and there are many potential uses for BAN technology in addition to BSNs.
Some of the more common use cases for BAN technology are:
• Body Sensor Networks (BSN)
• Sports and Fitness Monitoring
• Wireless Audio
• Mobile Device Integration
• Personal Video Devices
Each of these use cases have unique requirements in terms of bandwidth, latency, power usage, and signal distance. IEEE 802.15 is the working group for Wireless Personal Area Networks (WPAN) . The WPAN working group realized the need for a standard for use with devices inside and around close proximity to the human body. IEEE 802.15 established Task Group #6 to develop the standards for BAN. The BAN task group has drafted a (private) standard that encompasses a large range of possible devices.
In this way, the task group has given application and device developers the decision of how to balance data rate and power. Figure 1, below, describes the ideal position for BAN in the power vs data rate spectrum.
Figure : - Data Rate vs Power
As you can see the range of BAN devices can vary greatly in terms of bandwidth and power consumption. The BAN draft requirements, displayed below, add a common set of requirements as to ensure that all devices conform to a similar set of behaviors yet still encompass a wide variety of devices as previously mentioned.
Table 1 - BAN Draft Specifications
2.1 System Architecture
Continuous technological advances in integrated circuits, wireless communication, and sensors enable development of miniature, non-invasive physiological sensors that communicate wirelessly with a personal server and subsequently through the Internet with a remote emergency, weather forecast or medical database server; using baseline (medical database), sensor (WBAN) and environmental (emergency or weather forecast) information, algorithms may result in patient-specific recommendations. The personal server, running on a PDA or a 3 G cell phone, provides the human-computer interface and communicates with the remote server(s). Figure 1 shows a generalized overview of a multi-tier system architecture; the lowest level encompasses a set of intelligent physiological sensors; the second level is the personal server (Internet enabled PDA, cell-phone, or home computer); and the third level encompasses a network of remote health care servers and related services (Caregiver, Physician, Clinic, Emergency, Weather). Each level represents a fairly complex subsystem with a local hierarchy employed to ensure efficiency, portability, security, and reduced cost. Figure 2 illustrates an example of information flow in an integrated WBAN system.
Wireless
Sensor level
A WBAN can include a number of physiological sensors depending on the end-user application. Information of several sensors can be combined to generate new information such as total energy expenditure. An extensive set of physiological sensors may include the following:
• an ECG (electrocardiogram) sensor for monitoring heart activity
• an EMG (electromyography) sensor for monitoring muscle activity
• an EEG (electroencephalography) sensor for monitoring brain electrical activity
• a blood pressure sensor
• a tilt sensor for monitoring trunk position
• a breathing sensor for monitoring respiration
• movement sensors used to estimate user's activity
• a "smart sock" sensor or a sensor equipped shoe insole used to delineate phases of individual steps .
Figure:-Data flow in an integrated WWBAN
These physiological sensors typically generate analog signals that are interfaced to standard wireless network platforms that provide computational, storage, and communication capabilities. Multiple physiological sensors can share a single wireless network node. In addition, physiological sensors can be interfaced with an intelligent sensor board that provides on-sensor processing capability and communicates with a standard wireless network platform through serial interfaces. The wireless sensor nodes should satisfy the following requirements: minimal weight, miniature form-factor, low-power operation to permit prolonged continuous monitoring, seamless integration into a WBAN, standard based interface protocols, and patient-specific calibration, tuning, and customization. These requirements represent a challenging task, but we believe a crucial one if we want to move beyond 'stovepipe' systems in healthcare where one vendor creates all components. Only hybrid systems implemented by combining off-the-shelf, commodity hardware and software components, manufactured by different vendors promise proliferation and dramatic cost reduction. The wireless network nodes can be implemented as tiny patches or incorporated into clothes or shoes. The network nodes continuously collect and process raw information, store them locally, and send them to the personal server. Type and nature of a healthcare application will determine the frequency of relevant events (sampling, processing, storing, and communicating). Ideally, sensors periodically transmit their status and events, therefore significantly reducing power consumption and extending battery life. When local analysis of data is inconclusive or indicates an emergency situation, the upper level in the hierarchy can issue a request to transfer raw signals to the upper levels where advanced processing and storage is available.
Personal server level
The personal server performs the following tasks:
• Initialization, configuration, and synchronization of WBAN nodes
• Control and monitor operation of WBAN nodes
• Collection of sensor readings from physiological sensors
• Processing and integration of data from various physiological sensors providing better insight into the users state
• Providing an audio and graphical user-interface that can be used to relay early warnings or guidance (e.g., during rehabilitation)
• Secure communication with remote healthcare provider servers in the upper level using Internet services.
The personal server can be implemented on an off-theshelf Internet-enabled PDA (Personal Digital Assistant) or 3 G cell phone, or on a home personal computer. Multiple configurations are possible depending on the type of wireless
network employed. For example, the personal server can communicate with individual WBAN nodes using the Zigbee wireless protocol that provides low-power network operation and supports virtually an unlimited number of network nodes. A network coordinator, attached to the personal server, can perform some of the pre-processing and synchronization tasks. Other communication scenarios are also possible. For example, the personal server running on a Bluetooth or WLAN enabled PDA can communicate with remote upper-level services through a home computer; the computer then serves as a gateway (Figure 1).
Relying on off-the-shelf mobile computing platforms is crucial, as these platforms will continue to grow in their capabilities and quality of services. The challenging tasks are to develop robust applications that provide simple and intuitive services (WBAN setup, data fusion, questionnaires describing detailed symptoms, activities, secure and reliable communication with remote medical servers, etc). Total information integration will allow patients to receive directions from their healthcare providers based on their current conditions.
Medical services
We envision various medical services in the top level of the tiered hierarchy. A healthcare provider runs a service that automatically collects data from individual patients, integrates the data into a patient's medical record, processes them, and issues recommendations, if necessary. These recommendations are also documented in the electronic medical record. If the received data are out of range or indicate an imminent medical condition, an emergency service can be notified (this can also be done locally at the personal server level). The exact location of the patient can be determined based on the Internet access entry point or directly if the personal server is equipped with a GPS sensor. Medical professionals can monitor the activity of the patient and issue altered guidance based on the new information, other prior known and relevant patient data, and the patient's environment (e.g., location and weather conditions).
The large amount of data collected through such services will allow quantitative analysis of various conditions and patterns. For example, suggested targets for stride and forces of hip replacement patients could be suggested according to the previous history, external temperature, time of the day, gender, and current physiological parameters (e.g., heart rate). Moreover, the results could be stored in research databases that will allow researchers to quantify the contribution of each parameter to a given condition if adequate numbers of patients are studied in this manner. Again, it is important to emphasize that the proposed approach requires seamless integration of large amounts of data into a research database in order to be able to perform meaningful statistical analyses.
ActiS – Activity Sensor
The ActiS sensor was developed specifically for WBAN based wearable computer-assisted, rehabilitation applications. With this concept in mind, we integrated a one-channel bio-amplifier and three accelerometer channels with a low power microcontroller into an intelligent signal processing board that can be used as an extension of a standard wireless sensor platform. ActiS consists of a standard sensor platform, Telos, from Moteiv and a custom Intelligent Signal Processing Module – ISPM (Figure 3). A block diagram of the sensor node is shown in Figure4. The Telos platform is an ideal fit for this application due to small footprint and open source system software support. A second generation of the Telos platform features an 8 MHz MSP430F1611 microcontroller with integrated 10 KB of RAM and 48 KB of flash memory, a USB (Universal Serial Bus) interface for programming and communication, and an integrated wireless ZigBee compliant radio with on-board antenna [11]. In addition, the Telos platform includes humidity, temperature, and light sensors that could be used as ambient sensors. The Telos platform features a 10-pin expansion connector that allows one UART (Universal Asynchronous Receiver Transmitter) and I2 C interface, two general-purpose I/O lines, and three analog input lines.
The ISPM extends the capabilities of Telos by adding two perpendicular dual axis accelerometers (Analog Devices ADXL202) and a bio-amplifier with a signal conditioning circuit. The ISPM has its own MSP430F1232 processor for sampling and low-level data processing. This microcontroller was selected primarily for its compact size and ultra low power operation. Other features that were desirable for this design were the 10-bit ADC and the timer capture/ compare registers that are used for acquisition of data from the accelerometers. The F1232 has hardware UART that is used for communications with Telos.
The ISPM's two ADXL202 accelerometers cover all three axes of motion. One ADXL202 is mounted directly on the ISPM board and collects data for the X and Y axes in the same plane. The second ADXL202 is mounted on a daughter card that extends vertically from the ISPM. The user's physiological state is monitored using an onboard bio-amplifier implemented using an instrumentation amplifier with a signal conditioning circuit. The bioamplifier could be used for electromyogram (EMG) or electrocardiogram (ECG) monitoring. The output of the signal conditioning circuit is connected to the local microcontroller as well as to the microcontroller on the Telos board via the expansion connector. The AD converter on the Telos board has a higher resolution (12 bit) than the F1232 on the ISPM (10 bit). This configuration gives flexibility of utilizing either microcontroller to process physiological signals.
Figure:- Block diagram of the activity sensor (Telos platform and ISPM module)
An example application of the ActiS sensor as motion sensor on an ankle is given in Figure 5. This figure also visualizes the main components of acceleration during slow movements as projections of the gravity force (g) on the accelerometer's reference axes – Ax and Ay. Rotations of the sensor in the vertical plane (Θ) can be estimated as Θ = arctan(Ax / Ay). A compensation for non-ideal vertical placement can be achieved using the second accelerometer (not mounted in this photo) at 90-degree angle. Instead of calculating the angular position, many systems use off-the-shelf gyroscopes to measure angular velocity for the detection of gait phases. A typical example of step detection is illustrated in Figure 6
Issues and Applications
WBAN systems can capitalize on recent technological advances that have enabled new methods for studying human activity and motion, making extended activity analysis more feasible. However, before WBAN becomes a widely accepted concept, a number of challenging system design and social issues should be resolved. If resolved successfully, WBAN systems will open a whole range of possible new applications that can significantly influence our lives.
System Design Issues
The development of pedometers and Micro-Electro Mechanical Systems (MEMS) accelerometers and gyroscopes show great promise in the design of wearable sensors. The main system design issues include:
• types of sensors
• power source
• size and weight of sensors
• wireless communication range and transmission characteristics of wearable sensors
• sensor location and mounting
• seamless system configuration
• automatic uploads to the patient's electronic medical record
• intuitive and simple user interface
Types of sensors
As for sensors, accelerometers and gyroscopes offer greater sensitivity and are more applicable for monitoring of motion since they generate continuous output. Bout net a found that frequency of human induced activity ranges from 1 to 18 Hz. Sampling rates in the existing projects vary from 10 – 100 Hz. Almost all projects in the last five years use MEMS accelerometers or a combination of accelerometers and gyroscopes . As examples of full sets of sensors for research purposes, "MIThril" and Shoe Integrated Gait Sensor (SIGS) [26] systems feature 3 axes of gyroscopes, 3 axes of accelerometers, two piezoelectric sensors, two electric field sensors, two resistive band sensors, and four force sensitive resistors. These sensors can be mounted on the back of a shoe and in a shoe insole, respectively. Researchers at University of Washington School of Nursing have used off-the-shelf triaxis accelerometer modules to study physical movement in COPD (Chronic Obstructive Pulmonary Disease) patients. Both Lancaster University, UK, and ETH Zurich, Switzerland, has developed custom hardware realizing arrays of inertial sensor networks. Lancaster used an array of 30 two-axis accelerometers. Similarly, ETH Zurich used a modular harness design. The majority of foot-contact pedometers are designed to count steps only. Although they have been studied for use in complex energy estimation and have even shown a high degree of accuracy for walking / running activities they are not well suited for rehabilitation.
Power source, size/weight, and transmission characteristics
To be unobtrusive, the sensors must be lightweight with small form factor. The size and weight of sensors is predominantly determined by the size and weight of batteries. Requirements for extended battery life directly oppose the requirement for small form factor and low weight. This implies that sensors have to be extremely power efficient, as frequent battery changes for multiple WBAN sensors would likely hamper users' acceptance and increase the cost. In addition, low power consumption is very important as we move toward future generations of implantable sensors that would ideally be self-powered, using energy extracted from the environment. The radio communication poses the most significant energy consumption problem. Intelligent on-sensor signal processing has the potential to save power by transmitting the processed data rather than raw signals, and consequently to extend battery life. A careful trade-off between communication and computation is crucial for an optimal design. It appears that the most promising wireless standard for WBAN applications is Zig Bee, as it represents an emerging wireless technology for the low power, short-range, wireless sensors.
Location of Sensors
Although the purpose of the measurement does influence sensor location, researchers seem to disagree on the ideal body location for sensors. A motion sensor attached to an ankle is the most discriminative single position for state recognition, while a combination of hip and ankle sensors discriminates the states even more. In a study of the relationship between metabolic energy expenditure and various activities, researchers at Eindhoven University of Technology, the Netherlands, placed tri-axial accelerometers on a subject's back waist line . Krause et al use two accelerometers on the Sense Wear arm band . Lee et al placed accelerometer sensors in the subject's thigh pocket in order to measure angular position and velocity of the thigh. Doing so, they were able to accurately monitor a subject's activity and with the assistance of gyro-scopes and compass headings were able to successfully estimate a subject's change in location. Some systems employ large arrays of wearable sensors. Laerhoven et al developed a loose fitting lab coat and trousers consisting of 30 sensors; Kern et al developed tighter fitting modular harnesses including a total of 48 sensors. Sensor attachment is also a critical factor, since the movement of loosely attached sensors creates spurious oscillations after an abrupt movement that can generate false events or mask real events.
Seamless system configuration
The intelligent WBAN sensors should allow users to easily assemble a robust ad-hoc WBAN, depending on the user's state of health. We can imagine standard off-the-shelf sensors, manufactured by different vendors, and sold "over-the-counter" . Each sensor should be able to identify itself and declare its operational range and functionality. In addition, they should support easy customization for a given application.
Algorithms
Application-specific algorithms mostly use digital signal pre-processing combined with a variety of artificial intelligence techniques to model user's states and activity in each state. Digital signal processing include filters to resolve high and low frequency components of a signal, wavelet transform algorithms to correlate heel-strike and toe-off (steps) to angular velocity measured via gyroscopes, power spectrum analysis and a Gaussian model to classify activity types . Artificial intelligence techniques may include fuzzy logic and Kohonen self-organizing maps . Some systems use physiological signals to improve context identification . It has been shown that the activity-induced energy expenditure (AEE) is well correlated with the sum of integrals of the high frequency component of each individual axis . Most of the algorithms in the open literature are not executed in real-time, or require powerful computing platforms such as laptops for real-time analysis.
Social Issues
Social issues of WBAN systems include privacy/security and legal issues. Due to communication of health-related information between sensors and servers, all communication over WBAN and Internet should be encrypted to protect user's privacy. Legal regulation will be necessary to regulate access to patient-identifiable information.
2.2Body area sensors
It can enable novel applications in and beyond healthcare, but research must address obstacles such as size, cost, compatibility, and perceived value before networks that use such sensors can become widespread.
Inroads into coordinated, intelligent computing are enabling sensor networks that monitor environments, systems, and complex interactions in a range of applications. Body area sensor networks (BASNs), for example, promise novel uses in healthcare, fitness, and entertainment. Each BASN consists of multiple interconnected nodes on, near, or within a human body, which together provide sensing, processing, and communication capabilities. BASNs have tremendous potential to transform how people interact with and benefit from information technology, but their practical adoption must overcome formidable technical and social challenges, as the “Requirements for Widespread Adoption” sidebar describes. These challenges have far-reaching implications but offer many immediate opportunities for system design and implementation.
Although BASNs share many of these challenges and opportunities with general wireless sensor networks (WSNs) and can therefore build off the body of knowledge associated with them many BASN-specific research and design questions have emerged that require new lines of inquiry. For example, to achieve social acceptance, BASN nodes must be extremely noninvasive, and a BASN must have fewer and smaller nodes relative to a conventional WSN. Smaller nodes imply smaller batteries, creating strict tradeoffs between the energy consumed by processing, storage, and communication resources and the fidelity, throughput, and latency required by applications. Packaging and placement are also essential design considerations, since BASN nodes can be neither prominent nor uncomfortable. As with any technology, economic concerns can affect BASN adoption. To amortize nonrecurring engineering costs, each BASN platform will require either significant volume in single application or aggregate volume across-several applications, creating design tradeoffs between application-specific optimizations and general-purpose programmability. Finally, value to the user will ultimately determine the technology’s success. BASNs must effectively transmit and transform sensed phenomena into valuable information and do so while meeting other system requirements, such as energy efficiency. A BASN’s value therefore rests in large part on its ability to selectively process and deliver information at fidelity levels and rates appropriate to the data’s destination, whether that is to a runner curious about her heart rate or a physician needing a patient’s electrocardiogram. These disparate application requirements call for the ability to aggregate hierarchical information and integrate BASN systems into the existing information technology infrastructure.
Application areas
Because of demonstrated need and market demand, BASN research thus far has concentrated on healthcare applications, addressing the weaknesses of traditional patient data collection, such as imprecision (qualitative observation) and under sampling (infrequent assessment). In contrast, BASNs can continuously capture quantitative data from a variety of sensors for longer periods. By addressing challenges such as the energy-fidelity tradeoff, BASNs will enable telehealth applications medicine beyond the confines of hospitals and clinics1 and, because of their human-centricity, will facilitate highly personalized and individual care.
As Figure 1 illustrates, BASNs integrated with higher-level infrastructure will likely excel in healthcare scenarios, serving the interests of multiple stakeholders. In addition to delay-insensitive applications such as longitudinal assessment, BASNs that can offer real-time sensing, processing, and control will augment and preserve body functions and human life. BASN researchers are already working to improve deep brain stimulation, heart regulation, drug delivery, and prosthetic actuation. BASN technology will also help protect those exposed to potentially life-threatening environments, such as soldiers, first responders, and deep-sea and space explorers. Finally, BASNs are well positioned to benefit from the intersection of two formerly disparate application areas. Physiological and biokinetic sensing applications are increasing as athletes and fitness enthusiasts seek to improve human performance, while gaming systems are pushing their envelope by integrating more sophisticated interfaces based on human movement. With the crossing of these markets, BASNs are well positioned to deliver the biofeedback and interactivity necessary for next-generation fitness and entertainment applications.
Body Area Sensing
As Figure 2 shows, BASN nodes create an interface to humans, typically encapsulating an energy source, one or more sensors, a mixed-signal processor, and a communication transceiver. Some nodes also support data storage or feedback control to body-based actuators, such as an insulin pump or robotic prosthetic. Although BASN and WSN nodes have similar functional architecture, differences in their operational characteristics—sensing, signal processing, communication, caching, feedback control, and energy harvesting—present unique challenges and opportunities for BASN nodes.
Sensors
Sensing is fundamental to all sensor networks, and its quality depends heavily on industry advances in signal conditioning, micro electromechanical systems (MEMS), and nanotechnology. Sensors fall into three categories. Physiological sensors measure ambulatory blood pressure, continuous glucose monitoring, core body temperature, blood oxygen, and signals related to respiratory inductive plethysmography, electrocardiography (ECG), electroencephalography (EEG), and electromyography (EMG). Biokinetic sensors measure acceleration and angular rate of rotation derived from human movement.
Ambient sensors measure environmental phenomena, such as humidity, light, sound pressure level, and temperature. Although the number of sensors in the BASN in Figure 1 might seem unrealistic, BASN users are likely to tolerate and accept some degree of burden if they perceive enough value in doing so.
Sensors in typical WSNs are numerous, homogeneous, and generally insensitive to placement error. BASN sensors, in contrast, are few, heterogeneous, and require specific placement. Indeed, ineffective placement or unintended displacement from movement can significantly degrade the captured data’s quality. Such requirements call for strategies that will minimize and detect placement error, such as better packaging combined with on-node signal classification. Commercial sensors exhibit a wide range of power supply requirements, calibration parameters, output interfaces, and data rates.
Figure 3 shows the power consumption and data rate across a sampling of commercial systems for continuous, ambulatory monitoring. Engineering BASN nodes to accommodate this breadth of sensing requirements could necessitate an application-specific approach that minimizes the design space, improves efficiency, and amortizes cost over a single application. Likewise, BASN nodes designed with a high degree of configurability could amortize cost over a much larger range of applications, including those unforeseen.
Signal processing
Signal processing is needed to extract valuable information from captured data that stems from transient events, such as falls, as well as from trends, such as the onset of fever. BASNs may need to concurrently capture, process, and forward information to different stakeholders. Time critical information from both events and trends would go immediately to emergency services, for example, but information that is not sensitive to delays would go to the physician for review later on.
Figure 4 shows the power consumption of wireless transceivers and micro-processors in popular BASN and WSN platforms. It underlines two characteristics of existing embedded technology: Processing data at a given rate consumes less power on average than transmitting the data wirelessly, and reducing the data rate will reduce power consumption for both wireless transceivers and microprocessors. These characteristics create a tradeoff between processing and communication: On-node signal processing will consume power to extract information, but it will also reduce in-network data rate and power consumption. Arbitrary data-rate reduction will lower the transmitted information’s fidelity, and for loss compression schemes, a rate-distortion analysis would need to define the limits of such a reduction. Low-power computational techniques such as dynamic voltage-frequency scaling or dynamic power management will create opportunities for dynamic adjustment of algorithmic complexity, and therefore trade off energy and fidelity based on an application’s predefined or situational needs. Context awareness and predictive models might better inform and guide processes that control data reduction. Resource constraints challenge BASNs, including integer- only math, limited memory (<>
Communication
Communication is essential to node coordination. BASNs are unique in that they attempt to restrict the communication radius to the body’s periphery. Limiting transmission range reduces a node’s power consumption, decreases interference among adjacent BASNs, and helps maintain privacy. WSNs typically communicate over radiative radiofrequency (RF) channels between 850 MHz and 2.4 GHz. Unlike WSNs, wireless BASNs are challenged by the dramatic attenuation of transmitted signals resulting from body shadowing the body’s line-of-sight absorption of RF energy, which, coupled with movement, causes significant and highly variable path loss. Preserving quality of service (QoS) over traditional wireless links could require one of several approaches, including adaptive channel coding; transmission power scaling; multiple input, multiple output; novel transceiver architecture; and QoS-aware media access protocols. Ultra wide band communication could help mitigate aspects of this problem in the near future.2 Technologies such as smart textiles, magnetic induction, 3 and body-coupled communication 4 also shows long-term promise. In smart textiles, wires are embedded in clothing, thereby reducing communication power overhead and simplifying networking schemes.5 Cost, ease of cleaning, and manufacturer standardization could limit market uptake. Magnetic induction uses magnetic near-field effects to communicate between two coils of wire. Near-field communication typically suffers less path loss than radiative communication, but coil dimensions complicate packaging. Despite this complication, implantable and swallowed sensors have exploited this communication technology. Body coupled communication uses the human body as a channel. BASN transceivers of this nature are either in contact with, or capacitive coupled to, the skin. Body-coupled communication is appealing because little radiated energy is detectable beyond the human body, channels are highly stable, and energy requirements are low. However, additional research will need to determine the safety of this approach. Future BASNs might implement several types of transceivers to serve situational needs. For example, a sensor node could employ both lower data rate, lower power communication transceivers in parallel with higher data rate, and higher power transceivers for both longitudinal and critical communication needs. Transceiver diversity could also help mitigate body shadowing.
Storage
The microelectronics industry is exploring lower power nonvolatile memory such as MRAM and RRAM. Consequently, the availability of on-node storage might enhance BASN functionality. Because long-term data collection often needs no real-time aggregation, on-node storage is a reasonable solution for archiving data, thereby increasing battery life. Longitudinal assessment is insensitive to delay metrics that challenge time-critical monitoring. Some applications might choose to cache data until body channel conditions are more favorable for transmission. Consequently, conditional caching could prolong battery life, decrease form factor, or decrease bit errors. On-node storage could also be used to archive data for signal classification. By storing biokinetic gait patterns over time, for example, a BASN could learn to classify healthy gait from pathological gait. Such an archive could inform the signal-processing routines needed to detect longitudinal trends (recovery from surgery) and instantaneous events (falls).
Feedback control
BASNs open exciting opportunities for augmenting and assisting bodily functions. Medical devices such as deep brain stimulators now run in an open-loop mode because no local feedback is available from the brain’s central cortex to adjust the stimulator’s excitation cycles. The accurate and reliable assessment of tremor through body area sensors could change that by empowering feedback tremor control. The deployment and control of prosthetics or remote robotic assistive devices is another possible application. EMG signals from the eyelid or jaw might be used to control a device that assists or replaces a limb or to activate a robotic device that opens doors or controls simple household appliances. Other forms of feedback control include drug delivery and blood glucose regulation facilitated by implantable biochemical sensors. Clearly, if BASNs are to control or help assess life-critical physiological events, they must be reliable. Unlike traditional WSNs, the failure of one BASN sensor could threaten life. Such applications will require fail-safe, fault-tolerant design principles.
Energy harvesting
Although the microelectronics industry has faithfully adhered to Moore’s frenetic pace, advancements in commercial battery technology have been gradual. To remain must continue to increase energy density, and investments in increased energy density must have commensurate levels of investment in battery safety particularly in light of recent battery recalls.
The high energy density of lithium-based batteries is helping power many portable consumer technologies. Such batteries work well for handheld electronics, but their capacity is limited in diminutive BASN enclosures. The need to replace or recharge batteries frequently makes BASN useless desirable. Super capacitors and carbon-nano tube-based energy stores have great potential to improve battery capacity, but have not yet matured to commercial availability. Energy harvesting taking energy from ambient sources, such as sunlight or vibration—is an attractive solution to energy woes. Recharging batteries with harvested energy could not only extend battery life, but also simplify BASN use. Research challenges are formidable because of node placement variability and uncertainty about the user’s exposure to ambient energy. These realities severely constrain opportunities.
Figure 5 shows the results of our investigation to estimate the average power that a BASN user can harvest per hour per day. For each of seven energy-harvesting sources,6 we correlated the amount of power available per square centimeter with that source’s availability during common human activities. We then compiled statistics from the US Department of Labor’s 2007 “American Time Use Survey” (www.bls.gov/tus) on the percentage of Americans engaged in each activity at a given hour. The figure shows an optimistic view of haves table power, thus defining an upper bound for a system’s power consumption from harvesting alone. The total power shown is available only if someone deploys all sources (at 1 cm2 each) simultaneously and combines their power output. The data also illustrates a pronounced blackout period that renders these BASN nodes nearly powerless.
Finally, the available harvestable power will differ substantially among individuals, which means it will be necessary to carefully match application profiles to activity levels in the target demographic. Energy-harvesting sources vary widely in the energy available per area. For example, a solar panel in full outdoor sunlight provides up to 15 mW per square centimeter, but the same device generates only 10 μW in indoor lighting for the same area. Both placement and packaging would be affected by such variation. Thus, although increasing battery life through harvesting would revolutionize BASNs, more research is needed to create highly efficient hybrid solutions that incorporate energy generation and storage.
Body Area Networking
Networking among devices in, on, and around the body poses unique challenges for resource allocation, sensor fusion, hierarchical cooperation, QoS, coexistence, and privacy. On the one hand, minimalistic networking techniques increase system runtime and reduce obtrusiveness; on the other, sacrificing QoS or privacy is unacceptable for life-critical or sensitive medical applications. BASNs introduce a wide range of application scenarios, yet it is not certain if a unified network solution is preferable over application-specific protocols and topologies.
Unlike conventional WSNs, BASNs are generally smaller (fewer nodes and less area covered) and have fewer opportunities for redundancy. Scalability can lead to inefficiencies when working with the two to 10 nodes typical of a BASN. Adding sensor and path redundancy to address node failure and network congestion might not be a viable strategy for a BASN seeking to minimize form factor and resource usage. Consequently, the focus must be on generating intelligent and cooperative QoS for the nodes. On-body and in-body (implantable) networks exhibit heterogeneity because of placement constraints and sensor requirements. Wear ability requirements can vary drastically across applications. Some call for multiple wired networks in a single garment; others call for multiple wirelessly networked devices securely attached at various body locations; and still others call for ultra miniature, biocompatible implanted devices with less frequent communication to the outside world. BASNs also have a distinctly hierarchical nature. They capture large quantities of data continuously and naturalistically, which microprocessors must process to extract actionable information. Data processing must be hierarchical to exploit the asymmetry of resources, preserve system efficiency, and ensure that data is available when needed a practical energy source for BASNs, battery technology
Figure 6 shows the levels and their respective requirements for data processing, archiving, and management. During data fusion, systems can detect or react to notable occurrences from dynamic data, explicit queries, and so on. Specific reactions might include heightening the state of awareness, collecting data at a higher fidelity for closer inspection, forwarding events to higher levels, or even immediate response. Aside from a BASN’s inherent characteristics, designers must consider the desired destination of sensed information. Stand-alone BASNs route data for storage or for processing to another location in or on the body; other BASNs move data from the body through a gateway into other ambient networks. An example of integration with existing wireless technologies is an assisted-living facility, in which each resident’s BASN wirelessly communicates to a back-end medical network. All the BASNs must maintain sufficient QoS by cooperating to mitigate network interference and transmit relevant information for further processing and presentation. BASNs should also encrypt information to ensure that only trusted stakeholders, such as physicians or caregivers, have access to it.
Hierarchical aggregation
Data processing at the sensor node reveals information specific to the sensor’s locality. Information, however, might also come from relationships between data collected at multiple sensors over time. The body area aggregator has the important role of combining data from multiple sensors on the body. The aggregator typically possesses a richer collection of resources and a greater energy capacity than the BASN nodes. In addition to its role as a data fusion center, the aggregator creates a bridge between the nodes and higher level infrastructure. It can also offer user interfacing and can possess its own sensing capabilities. The convergence of wireless technologies, such as Bluetooth, cellular, and IEEE 802.11; interactive user interfaces such as touch screens; and highly capable embedded microprocessors, such as the ARM 11 and OMAP, make newer mobile phones and personal digital assistants attractive hosts for body area aggregation. At the body aggregator, data processing must reveal relationships among a body’s sensors. With progressively richer resources, more sophisticated and dedicated data mining systems could uncover information related to small and large populations. Each successive hierarchical level must aggregate more data by supporting higher data rates, making more general inferences, and archiving more information. Consequently, hardware and software will need to interoperate through multiple levels of infrastructure to share information. Moreover, information gained at each level will provide feedback to and inform the refinement of classification schemes, feature-detection algorithms, and sensor coordination, placement, and design.
Topology
Star and star-mesh hybrid topologies show promise for meeting wear ability, size, and data-fusion needs.7 Both the star and star-mesh hybrid topologies exploit the resource asymmetry (aggregator versus node) and hierarchical nature of BASNs. In a star network, all peripheral nodes connect to the body aggregator, which allows for high data throughput and simplified routing. Having a central coordinator also means having a single point of failure, however. To address that weakness, a star-mesh hybrid topology extends the traditional star approach and creates mesh networking among central coordinators in multiple star networks. The failure of a single coordinator can trigger the reorganization of nodes and coordinators with minimal service interruption. Star-mesh hybrid topologies could also link aggregators and bridge networks from the body area to a wider area.
Coordination
Standards will help guide industry efforts by making it easier to fulfill the promise of compatible and interoperable networked technology. Fortunately, the IEEE 802.15.4 and the IEEE 802.15.6 working groups are leading the effort to address scalable, body-area network coordination.8 ZigBee technology, such as the CC2420,employs the 802.15.4 Medium Access Control protocol to allocate guaranteed time slots to specific nodes. In this protocol, the coordinator prevents nodes from transmitting during other nodes’ reserved slots. Collision avoidance and network coordination will be essential to maintaining QoS in both WSNs and BASNs. Standardizing BASNs is not easy. Because of their hierarchical nature, they exhibit significant communication asymmetry, and all BASN nodes exist within range of each other, so they are likely to hear the entire network’s transmissions. Another challenge is that BASN nodes will likely exhibit differences in transmitted data rates. Finally BASNs that span application types, such as life-critical and non-life-critical, must coexist, and will require some scheme for prioritizing and encrypting messages. Addressing all these challenges is likely to require new approaches to media access and protocol design.
BASNs are enabling human-centric sensing for a variety of intriguing applications in healthcare, fitness, and entertainment, but such networks must demonstrate enough value for users to overcome inhibitions related to inconvenience, invasiveness, and general discomfort. In his bestselling book, Visions: How Science Will Revolutionize the 21st Century and Beyond (Oxford University Press, 1999), futurist Michio Kaku described wearable technologies that will “silently monitor” heart rhythm, detect irregularities, and alert emergency personnel in the event of a heart attack. This vision, not far removed from current research efforts, illustrates the promise of BASNs in this important area. But the promise of this technology should not be restricted to one area. Fitness and entertainment are taking new directions that are also well-suited to BASN architecture. The same architecture that captures body motion for medical assessment is equally adept at capturing body motion for a videogame. New sensors will only increase the breadth of potential applications and market opportunities and propel this technology into applications formerly depicted only in science fiction.
3. Applications in Healthcare
As previously mentioned, BANs have grown as a refinement of BSN. As such, BSN remain the most thought out applications of BAN. In his summary of the BAN task group's findings thus far, Stefan Drude, a researcher at Phillips, outlined the possible needs the group had found for the very low BSN devices. BSN devices refine the general requirements by restricting themselves to a much smaller range (<>
First, the human body itself can become a channel for short range communication, thus removing the need for a traditional antenna. By removing the requirement of an additional antenna, the power consumption of BSN devices shrinks to 0.1 - 1.0 mW. At this low of power, the human body is actually capable of generating enough excess energy that the devices could "scavenge" the required energy directly from the host's body, removing the restriction on traditional power sources (like batteries) .
BAN technology is not one that is unique to Mr. Drude and the members of the BAN task group, this exact use case scenario has been thoroughly described by Microsoft in their patent titled, Method and apparatus for transmitting power and data using the human body. In the following subsections, we will investigate systems that utilize the BSN technology to accomplish higher level tasks.
3.1 Managed Body Sensor Networks
A managed body sensor network (MBSN) is defined as a system where the third party makes decisions based the data collected from one or many BSN. We will discuss MobiHealh and CodeBlue, two managed BSN that are approaching development of managed BSN from two different perspectives.
In 2003, two researchers from the University of Twente published a paper entitled "Continuous monitoring of vital constants for mobile users: the MobiHealth approach."The paper described the increasing demand of resources placed on the medical community, the rising costs of in-patient care, and the relative lack of out-patient monitoring. The paper defined "extra-BAN communication" (EBAN) as communication between a BAN and another network. The solution paper provided was MobiHealth, a BSN with EBAN connectivity to a 2.5/3G networks to provide out-patient monitoring of patients vital signs. Through this infrastructure the MobiHealth designers were able to provide sensor information to qualified medical professionals, where multiple patients’ data could be monitored in an aggregate form.MobiHealth is simply one example of a managed BSN. Harvard University's Code Blue represents another example of BSN currently in the trial stages. Like MobiHealth, CodeBlue provides an infrastructure for multiple patients monitoring through EBAN communication. However, CodeBlue takes a more middleware approach to BSN instead of the packaged solution that MobiHealth provides. By providing a middleware layer, the CodeBlue project allows developers to specify the modules to use. In this way, CodeBlue is rather flexible at runtime. Two examples given by the MobiHealth team are emergency response and monitoring limb movement in stroke patient rehabilitation. Both scenarios have very different requirements both from a sensor perspective, and a timeliness perspective however the platform is able scale to accommodate both accordingly.
3.2 Autonomous Body Sensor Networks
Autonomous body sensor networks (ABSN) and MBSN share the same goals, but they accomplish them in different ways. While a MBSN will relies on reading sensor information and delivering it to a third party for decision making and intervention, ABSN take a more proactive approach. ABSN introduce actuators in addition to the sensors to allow the BSN to effect change on the users body. In addition to the actuators, ABSN contain more intelligent sensors that contain enough intelligence to complete their own tasks independently. Human is a project developed in Belgium that aims to bring ABSN to the mainstream.
The design of Human is relatively simple, any node in the mesh-network are able to talk to any other node in the network. There is a predefined "central" node that is designated for all EBAN communication. The central node also publishes information on any services that the ABSN provides external access to. An example ABSN diagram can be seen below in figure 2.
Figure 2 - Example ABSN Diagram
3.3 Case Study: Cardiac Monitoring
The most effective way of describing the current state of BSN is to actually describe a case study as a representative sample of the progress of BSN. In 2007 Zheng et al. published "A wearable mobihealth care system supporting real-time diagnosis and alarm," a paper describing a MBSN using the MobiHealth infrastructure mentioned in the above section. We will briefly cover their design and implementation. This will help lead into our discussion of the challenges associated with BAN.
The MobiHealth cardiac monitoring system implemented by Zheng et al. had one goal a few simple design principles and improvements over older MobiHealth products. The goal of the system was to "provide long-term continuous monitoring of vital signs for high-risk cardiovascular patients." The project aimed for tight integration with GPS, which allowed system dispatchers to know the exact location of patients in distress. The project aimed to have a user friendly design that minimized the impact the monitoring system had on the patients. They accomplished this task using a "Wearable Shirt" comprised of smart fabric. The smart fabric was designed not only to provide sensor information wirelessly with the MBSN, but also to be resistant to casual wear and cleaning. The final system design was to provide online diagnosis and three separate levels of alarm on the local device. In this way, the design blended a little bit of ABSN technology into the system, by allowing the communication node to selectively raise events to dispatch only on anomalies, increasing the autonomy of the system.
4. Challenges Associated with BAN
BAN technology is still emerging and there are a lot of problems left to solve. Setting aside ethical issues like privacy, there are still plenty of technical challenges that we must overcome before BAN will become an effective solution. The BAN draft submissions have defined solutions for a lot of the basic wireless network protocols, but there is still a large amount of research that must be done to effectively propagate a signal in and around the human body. The last challenge BAN technology faces is actually a problem of Human-Computer Interaction (HCI) and how to make the technology usable.
4.1 Signal & Path Performance
As one might expect, the signal and path loss inside the human body is drastically different than the rules in plain space. That said the rules governing signal and path loss remain the same. Researchers have been able to model signal loss throughout the human body, however the more interesting research involves using the human body as a transmission medium for electrical signals. Marc Wegmueller et al. have attempted to model the conductivity and permittivity of signals sent from one area of the body to another. A full summary of their research is beyond the scope of this paper, but it is worth noting that in the frequency range of 10 kHz to 1MHz, for every 5 cm between the transmitter and receiver there is an increase in attenuation by 6 to 9 dB. Other factors lowered or raised these constants, such as the geometry of the path, the amount of fat, and the presence of joints.
4.2 Usability
Given the close proximity of users to the BAN technology, the demands on usability are exceptionally high. In section 3.3 we discussed Zheng et al. and the MobiHealth framework, we will again refer to the study as they represent some of the most advanced HCI design in the BAN field. Zheng's group decided to use advances in textile manufacturing to sensing wearable shirts that would actively monitor the wearer.
Interestingly enough, Zheng's group also found a usability fault in the EPI-MEDICS design, as the system would record ECG data and raise alarms as required, but it would only do so when requested by the patient. Zheng's group classified this as a usability flaw, as the usefulness of emergency detection sensors is in their detection of emergencies that are not planned.
5.Summary
Definition
WBAN or BAN, short for (Wireless) Body Area Network, consists of a set of mobile and compact intercommunicating sensors, either wearable or implanted into the human body, which monitor vital body parameters and movements. These devices, communicating through wireless technologies, transmit data from the body to a home base station, from where the data can be forwarded to a hospital, clinic or elsewhere, real-time. The WBAN technology is still in its primitive stage and is being widely researched. The technology, once accepted and adopted, is expected to be a breakthrough invention in healthcare, leading to concepts like telemedicine and m Health becoming real.
Applications
Initial applications of WBANs are expected to appear primarily in the health care domain, especially for continuous monitoring and logging vital parameters of patients suffering from chronic diseases such as diabetes, asthma and heart attacks.
§ A WBAN network in place on a patient can alert the hospital, even before he has a heart attack, through measuring changes in his vital signs.
§ A WBAN network on a diabetic patient could auto inject insulin though a pump, as soon as his insulin level declines, thus making the patient ‘doctor-free’ and virtually healthy.
Other applications of this technology include sports, military, or security. Extending the technology to new areas could also assist communication by seamless exchanges of information between individuals, or between individual and machines. Imagine businesspeople exchanging business cards, just with a handshake, with the help of BAN sensors. These applications might become reality with the WBAN implementation very soon.
Challenges
Before we start leveraging the positives of BAN the following issues need to be addressed:
§ Interoperability: WBAN systems would have to ensure seamless data transfer across standards such as Bluetooth, ZigBee etc. to promote information exchange, plug and play device interaction. Further, the systems would have to be scalable, ensure efficient migration across networks and offer uninterrupted connectivity.
§ System Devices: The sensors used in WBAN would have to be low on complexity, small in form factor, light in weight, power efficient, easy to use and reconfigurable. Further, the storage devices need to facilitate remote storage and viewing of patient data as well as access to external processing and analysis tools via the Internet.
§ System and device-level security: Considerable effort would be required to make BAN transmission secure and accurate. It would have to be made sure that the patient’s data is only derived from each patient’s dedicated BAN system and is not mixed up with other patient’s data. Further, the data generated from WBAN should have secure and limited access.
§ Invasion of privacy: People might consider the WBAN technology as a potential threat to freedom, if the applications go beyond ‘secure’ medical usage. Social acceptance would be key to this technology finding a wider application.
§ Sensor validation: Pervasive sensing devices are subject to inherent communication and hardware constraints including unreliable wired/wireless network links, interference and limited power reserves. This may result in erroneous datasets being transmitted back to the end user. It is of the utmost importance especially within a healthcare domain that all sensor readings are validated. This helps to reduce false alarm generation and to identify possible weaknesses within the hardware and software design
§ Data consistency: Data residing on multiple mobile devices and wireless patient motes need to be collected and analysed in a seamless fashion. Within Body Area Networks, vital patient datasets may be fragmented over a number of nodes and across a number of networked PCs or Laptops. If a medical practitioner’s mobile device does not contain all known information then the quality of patient care may degrade
6.Conclusion
A wearable Wireless Body Area Network (WBAN) of physiological sensors integrated into a telemedical system holds the promise to become a key infrastructure element in remotely supervised, home-based patient rehabilitation. It has the potential to provide a better and less expensive alternative for rehabilitation healthcare and may provide benefit to patients, physicians, and society through continuous monitoring in the ambulatory setting, early detection of abnormal conditions, supervise drehabilitation, and potential knowledge discovery through data mining of all gathered information. Continuous monitoring with early detection likely has the potential to provide patients with an increased level of confidence, which in turn may improve quality of life. In addition, ambulatory monitoring will allow patients to engage in normal activities of daily life, rather than staying at home or close to specialized medical services. Last but not least, inclusion of continuous monitoring data into medical databases will allow integrated analysis of all data to optimize individualized care and provide knowledge discovery through integrated data mining. Indeed, with the current technological trend toward integration of processors and wireless interfaces, we will soon have coincided intelligent sensors. They will be applied as skin patches, seamlessly integrated into a personal monitoring system, and worn for extended periods of time.
No comments:
Post a Comment