Citation Context

...ason is that most sensor network applications have very different data, communication, and lifetime requirements. Unlike traditional data collection applications such as environmental monitoring [7, 48, 51], medical deployments are characterized by mobile nodes with varying data rates and few opportunities for in-network aggregation. In addition, medical sensor networks are less concerned with maximizing individual node lifetimes, since it is acceptable to recharge devices or change batteries on a relatively frequent basis. As a result, many of the significant advances in communication models [27, 53], time synchronization [39, 15], and energy management [44] should be reevaluated given these new requirements. This is not to say that we must start from scratch; rather, we believe it is best to borrow from prior systems as much as possible and invent new technology only as needed. A number of other research projects are exploring medical sensor networks. Most of these projects are concerned with developing wearable medical sensors [33, 54, 46], while others have developed infrastructures for monitoring individual patients during daily activity [30], at home [12], or at a hospital [31]. In c...

Citation Context

...to the enclosure. The Pluto mote, battery, enclosure and wristband weigh just 30.5g, whereas the Telos (with AA batteries and no enclosure) weighs 61g. 4 The CodeBlue Architecture The previous section described our work on medical sensor hardware. However, supporting the diverse requirements for medical sensor networks also requires that we take a fresh look at the software environment, routing protocols, and query interfaces. In this section we describe the design and architecture for CodeBlue, a protocol and middleware framework for medical sensor networks. CodeBlue is implemented in TinyOS [24] and provides protocols for integrating wireless medical sensors and end-user devices such as PDAs and laptops. CodeBlue is intended to act as an “information plane” tying together a wide range of wireless devices used in medical settings. CodeBlue is based on a publish/subscribe routing framework, allowing multiple sensor devices to relay data to all receivers that have registered an interest in that data. This communication model fits naturally with the needs of medical applications where a number of caregivers may be interested in sensor data from overlapping groups of patients. A discovery...

Citation Context

... subscribe to the broadcast channel to receive this information. Note that the metadata information about a node is static and is not updated frequently (the current update interval is 30 seconds). It would be straightforward to reduce the number of broadcast messages by performing in-network aggregation of this metadata. 4.3 CodeBlue query interface The CodeBlue Query (CBQ) layer allows receiving devices to establish communication pathways by specifying the sensors, data rates, and optional filter conditions that should be used for data transfer. Similar to Directed Diffusion [27] and TinyDB [36], CBQ is intended to provide a very simple means of expressing data requirements in a CodeBlue network. A CBQ query is generated by an end-user device (such as a PDA or laptop) and instructs CodeBlue nodes to publish data that meets the query conditions on a specific ADMR channel. 4.3.1 Query structure CBQ does not provide a textual interface for issuing queries; rather, queries can be issued using the GUI described in Section 4.6. A CBQ query is specified by the tuple 〈S, τ, chan, ρ, C, p〉. S represents the set of node IDs that should report data for this query and τ is the sensor type repres...

Citation Context

...ates as follows. Every CodeBlue node maintains a node table indexed by the publisher node ID. Each node table entry contains the path cost from the publisher to the current node, as well as the previous hop in the best path from the publisher. Whenever an ADMR message is received, the node table entry corresponding to the publisher is consulted. If the estimated path cost from the publisher to the current node is lower than the node table entry (or no node table entry exists), the new previous hop and path cost fields are updated accordingly. Routing costs can be estimated in a number of ways [10, 53]. We use an estimator of the total path delivery ratio (PDR) from the originating node. This estimate is based on an empirical model that maps the CC2420 radio’s Link Quality Indicator (LQI) to an estimated link delivery ratio (LDR), using extensive measurements from our 30-node sensor network testbed. The total path loss can be calculated as ∏ l∈L LDR(l) where LDR(l) represents the link delivery ratio for link l (estimated from LQI of the received message), for all links L along the path from the originator to the current node. The PDR is carried in the header of each ADMR message and is upda...

Citation Context

...ason is that most sensor network applications have very different data, communication, and lifetime requirements. Unlike traditional data collection applications such as environmental monitoring [7, 48, 51], medical deployments are characterized by mobile nodes with varying data rates and few opportunities for in-network aggregation. In addition, medical sensor networks are less concerned with maximizing individual node lifetimes, since it is acceptable to recharge devices or change batteries on a relatively frequent basis. As a result, many of the significant advances in communication models [27, 53], time synchronization [39, 15], and energy management [44] should be reevaluated given these new requirements. This is not to say that we must start from scratch; rather, we believe it is best to borrow from prior systems as much as possible and invent new technology only as needed. A number of other research projects are exploring medical sensor networks. Most of these projects are concerned with developing wearable medical sensors [33, 54, 46], while others have developed infrastructures for monitoring individual patients during daily activity [30], at home [12], or at a hospital [31]. In c...

Citation Context

...k applications have very different data, communication, and lifetime requirements. Unlike traditional data collection applications such as environmental monitoring [7, 48, 51], medical deployments are characterized by mobile nodes with varying data rates and few opportunities for in-network aggregation. In addition, medical sensor networks are less concerned with maximizing individual node lifetimes, since it is acceptable to recharge devices or change batteries on a relatively frequent basis. As a result, many of the significant advances in communication models [27, 53], time synchronization [39, 15], and energy management [44] should be reevaluated given these new requirements. This is not to say that we must start from scratch; rather, we believe it is best to borrow from prior systems as much as possible and invent new technology only as needed. A number of other research projects are exploring medical sensor networks. Most of these projects are concerned with developing wearable medical sensors [33, 54, 46], while others have developed infrastructures for monitoring individual patients during daily activity [30], at home [12], or at a hospital [31]. In contrast, our focus is to develo...

Citation Context

...nt data, communication, and lifetime requirements. Unlike traditional data collection applications such as environmental monitoring [7, 48, 51], medical deployments are characterized by mobile nodes with varying data rates and few opportunities for in-network aggregation. In addition, medical sensor networks are less concerned with maximizing individual node lifetimes, since it is acceptable to recharge devices or change batteries on a relatively frequent basis. As a result, many of the significant advances in communication models [27, 53], time synchronization [39, 15], and energy management [44] should be reevaluated given these new requirements. This is not to say that we must start from scratch; rather, we believe it is best to borrow from prior systems as much as possible and invent new technology only as needed. A number of other research projects are exploring medical sensor networks. Most of these projects are concerned with developing wearable medical sensors [33, 54, 46], while others have developed infrastructures for monitoring individual patients during daily activity [30], at home [12], or at a hospital [31]. In contrast, our focus is to develop a robust, scalable infrast...

Citation Context

...clude effective congestion management, reliable networking, and security. In the following section we present background on medical sensor networks and discuss related work. In Section 3 we detail our medical sensor hardware designs. Section 4 describes the CodeBlue protocol architecture and prototype implementation. In Section 5 we present initial results evaluating the performance of the CodeBlue system on our 30-node indoor sensor network testbed. Finally, Section 6 discusses future work and concludes. 2 Motivation and Background Medical care is an oft-cited application for sensor networks [29, 11]. The ability to augment medical telemetry with tiny, wearable, wireless sensors would have a profound impact on many aspects of clinical practice. Emergency medical care, triage, and intensive care can all benefit from continuous vital sign monitoring, especially immediate notification of patient deterioration. Sensor data can be integrated into electronic patient care records and retrieved for later analysis. In a wide range of clinical studies, especially those involving ambulatory or at-home monitoring, wireless sensors would permit data acquisition at higher resolution and for longer dura...

Citation Context

...sts of two main components. The first is the coordinator that receives messages from the radio, handles various internal commands (e.g. for debugging), and forwards queries to the CBQ component. CBQ maintains a table of running queries, as well as a sorted queue of query execution events. Each event contains a pointer to a query as well as the time until the next event. This design allows us to use a single timer to drive the execution of all queries. 4.3.2 Discussion CBQ’s implementation is greatly simplified by the use of the underlying publish/subscribe layer. Unlike TinyDB [36] and Cougar [55], the query engine is not responsible for maintaining routing paths, nor is it concerned with how the routing topology may affect results. However, the clean separation between the CBQ and ADMR layers results in some inefficiency. For example, both CBQ and ADMR perform separate broadcast floods, the former for advertising node metadata and the latter for establishing routing paths. It is clear that a simple cross-layer optimization could be performed to combine these floods: for example, ADMR could solicit a message payload from CBQ to include in its periodic path establishment transmissions. ...

Citation Context

...k applications have very different data, communication, and lifetime requirements. Unlike traditional data collection applications such as environmental monitoring [7, 48, 51], medical deployments are characterized by mobile nodes with varying data rates and few opportunities for in-network aggregation. In addition, medical sensor networks are less concerned with maximizing individual node lifetimes, since it is acceptable to recharge devices or change batteries on a relatively frequent basis. As a result, many of the significant advances in communication models [27, 53], time synchronization [39, 15], and energy management [44] should be reevaluated given these new requirements. This is not to say that we must start from scratch; rather, we believe it is best to borrow from prior systems as much as possible and invent new technology only as needed. A number of other research projects are exploring medical sensor networks. Most of these projects are concerned with developing wearable medical sensors [33, 54, 46], while others have developed infrastructures for monitoring individual patients during daily activity [30], at home [12], or at a hospital [31]. In contrast, our focus is to develo...

Citation Context

...Health Insurance Portability and Accountability Act (HIPAA). Recent work on private-key and public-key cryptography schemes for sensor networks [29, 22, 38] is applicable here, but must be integrated into an appropriate authentication and authorization framework. 2.2 Related work Many of the aforementioned requirements have not yet been adequately addressed by the sensor network community. The chief reason is that most sensor network applications have very different data, communication, and lifetime requirements. Unlike traditional data collection applications such as environmental monitoring [7, 48, 51], medical deployments are characterized by mobile nodes with varying data rates and few opportunities for in-network aggregation. In addition, medical sensor networks are less concerned with maximizing individual node lifetimes, since it is acceptable to recharge devices or change batteries on a relatively frequent basis. As a result, many of the significant advances in communication models [27, 53], time synchronization [39, 15], and energy management [44] should be reevaluated given these new requirements. This is not to say that we must start from scratch; rather, we believe it is best to b...

Citation Context

...Health Insurance Portability and Accountability Act (HIPAA). Recent work on private-key and public-key cryptography schemes for sensor networks [29, 22, 38] is applicable here, but must be integrated into an appropriate authentication and authorization framework. 2.2 Related work Many of the aforementioned requirements have not yet been adequately addressed by the sensor network community. The chief reason is that most sensor network applications have very different data, communication, and lifetime requirements. Unlike traditional data collection applications such as environmental monitoring [7, 48, 51], medical deployments are characterized by mobile nodes with varying data rates and few opportunities for in-network aggregation. In addition, medical sensor networks are less concerned with maximizing individual node lifetimes, since it is acceptable to recharge devices or change batteries on a relatively frequent basis. As a result, many of the significant advances in communication models [27, 53], time synchronization [39, 15], and energy management [44] should be reevaluated given these new requirements. This is not to say that we must start from scratch; rather, we believe it is best to b...

Citation Context

...evice mobility: Both patients and caregivers are mobile, requiring that the communication layer adapt rapidly to changes in link quality. For example, if a multihop routing protocol is in use, it should quickly find new routes when a doctor moves from room to room during rounds. Security: Aside from the obvious security considerations with sensitive patient data, United States law mandates that medical devices meet the privacy requirements of the 1996 Health Insurance Portability and Accountability Act (HIPAA). Recent work on private-key and public-key cryptography schemes for sensor networks [29, 22, 38] is applicable here, but must be integrated into an appropriate authentication and authorization framework. 2.2 Related work Many of the aforementioned requirements have not yet been adequately addressed by the sensor network community. The chief reason is that most sensor network applications have very different data, communication, and lifetime requirements. Unlike traditional data collection applications such as environmental monitoring [7, 48, 51], medical deployments are characterized by mobile nodes with varying data rates and few opportunities for in-network aggregation. In addition, me...

Citation Context

... requires relatively high data rates, reliable communication, and multiple receivers (e.g. PDAs carried by doctors and nurses). Moreover, unlike many sensor network applications, medical monitoring cannot make use of traditional in-network aggregation since it is not generally meaningful to combine data from multiple patients. This paper presents our initial experiences with a prototype medical sensor network platform, called CodeBlue. We have developed a range of medical sensors integrated with the commonly-used Mica2 [8], MicaZ [9] and Telos [41] mote designs. These include a pulse oximeter [37], two-lead electrocardiogram (EKG) [17], and a specialized motion-analysis sensor board. In addition, we have developed a small form factor variant of the Telos mote specifically for wearable use. The CodeBlue software framework provides protocols for device discovery, publish/subscribe multihop routing, and a simple query interface allowing caregivers to request data from groups of patients. In addition to monitoring patient vital signs, CodeBlue also integrates an RF-based localization system, called MoteTrack [34], to track the location of patients and caregivers. This capability is especia...

Citation Context

...evice mobility: Both patients and caregivers are mobile, requiring that the communication layer adapt rapidly to changes in link quality. For example, if a multihop routing protocol is in use, it should quickly find new routes when a doctor moves from room to room during rounds. Security: Aside from the obvious security considerations with sensitive patient data, United States law mandates that medical devices meet the privacy requirements of the 1996 Health Insurance Portability and Accountability Act (HIPAA). Recent work on private-key and public-key cryptography schemes for sensor networks [29, 22, 38] is applicable here, but must be integrated into an appropriate authentication and authorization framework. 2.2 Related work Many of the aforementioned requirements have not yet been adequately addressed by the sensor network community. The chief reason is that most sensor network applications have very different data, communication, and lifetime requirements. Unlike traditional data collection applications such as environmental monitoring [7, 48, 51], medical deployments are characterized by mobile nodes with varying data rates and few opportunities for in-network aggregation. In addition, me...

Citation Context

...a2 [8], MicaZ [9] and Telos [41] mote designs. These include a pulse oximeter [37], two-lead electrocardiogram (EKG) [17], and a specialized motion-analysis sensor board. In addition, we have developed a small form factor variant of the Telos mote specifically for wearable use. The CodeBlue software framework provides protocols for device discovery, publish/subscribe multihop routing, and a simple query interface allowing caregivers to request data from groups of patients. In addition to monitoring patient vital signs, CodeBlue also integrates an RF-based localization system, called MoteTrack [34], to track the location of patients and caregivers. This capability is especially valuable in large hospital settings. We present an initial evaluation of the CodeBlue prototype, demonstrating its scalability and robustness as the data rates, number of simultaneous queries, and transmitting sensors are varied. We also study the effect of node mobility, fairness across multiple simultaneous paths, and patterns of packet loss, confirming the system’s ability to maintain stable routes despite variations in node location and data rate. We are collaborating with several hospitals and medical resear...

Citation Context

...Health Insurance Portability and Accountability Act (HIPAA). Recent work on private-key and public-key cryptography schemes for sensor networks [29, 22, 38] is applicable here, but must be integrated into an appropriate authentication and authorization framework. 2.2 Related work Many of the aforementioned requirements have not yet been adequately addressed by the sensor network community. The chief reason is that most sensor network applications have very different data, communication, and lifetime requirements. Unlike traditional data collection applications such as environmental monitoring [7, 48, 51], medical deployments are characterized by mobile nodes with varying data rates and few opportunities for in-network aggregation. In addition, medical sensor networks are less concerned with maximizing individual node lifetimes, since it is acceptable to recharge devices or change batteries on a relatively frequent basis. As a result, many of the significant advances in communication models [27, 53], time synchronization [39, 15], and energy management [44] should be reevaluated given these new requirements. This is not to say that we must start from scratch; rather, we believe it is best to b...

Citation Context

...ength, TOS_Msg* msg); event result_t sendDone(TOS_MsgPtr msg, result_t success); event TOS_MsgPtr receive(TOS_MsgPtr m, uint16_t channel, uint16_t srcAddr); } Figure 6: The TinyADMR software interface. when establishing routing paths. In the medical scenario we expect both patients and caregivers to be mobile. Many patients may be ambulatory and free to roam about the hospital ward. Even those confined to hospital beds may be transferred between wards or temporarily moved for surgery or imaging. The CodeBlue routing layer is based on the Adaptive DemandDriven Multicast Routing (ADMR) protocol [28]. We selected ADMR because it is simple and has been extensively studied in simulation. As far as we are aware, ours is the first implementation of ADMR to be developed and tested on real hardware, and certainly the first using motes and TinyOS. We describe the ADMR protocol only briefly here; more details can be found in [28]. The TinyADMR component provides a PubSub interface that exposes the commands and events shown in Figure 6. The publish and subscribe commands allow a node to state that it wishes to associate with a particular channel, while leave terminates a publish or subscribe reque...

Citation Context

...that reliable routing is required for all medical data; rather, the system should allow each query to specify its reliability needs in terms of acceptable loss, data rate, or jitter. Another area worth exploring is the impact of bandwidth limitations and effective techniques for sharing bandwidth across patient sensors. For example, each CodeBlue query could specify a data priority that would allow certain messages (say, an alert from a critical patient) to have higher priority than others in the presence of radio congestion. This approach can be combined with rate-limiting congestion control [14, 25] to bound the bandwidth usage of patient sensors. An important shortcoming of the current CodeBlue prototype is its lack of security. We have already begun to explore the integration of private-key encryption [29] along with a public-key protocol for key distribution [38, 22] in CodeBlue. The privacy and security requirements for medical care are complex and differ depending on the scenario. For example, HIPAA privacy regulations need not be enforced during life-saving procedures. Nevertheless, we intend to integrate some form of end-to-end security into the next version of the CodeBlue system...

Citation Context

...-capture systems use a wired data logger carried in a waist harness; a multitude of wires runs from the harness to various sensors positioned on body segments of interest (typically the arms, legs, back and torso). Clearly, the use of wearable wireless sensors would greatly simplify data collection and would allow patients to wear the sensors for longer periods of time since the bulky data logger and leads would be eliminated. Three sensor types are commonly used for motion analysis studies in the field: accelerometers, gyroscopes, and surface electrodes for electromyographic (EMG) recordings [4, 40, 5]. Triaxial accelerometers measure the orientation and movement of each body segment. Gyroscopes measure angular velocity and combined with accelerometer data can be used to accurately determine limb position [35, 20]. Surface EMG electrodes capture the electrical field generated by depolarized zones traveling along the muscle fibers during a muscle contraction. The root mean square value of the EMG data is roughly proportional to the force exerted by the monitored muscle. Thus analysis of the patterns of EMG activity can lead to the identification of motor tasks and their characteristics [3]. ...

Citation Context

...clude effective congestion management, reliable networking, and security. In the following section we present background on medical sensor networks and discuss related work. In Section 3 we detail our medical sensor hardware designs. Section 4 describes the CodeBlue protocol architecture and prototype implementation. In Section 5 we present initial results evaluating the performance of the CodeBlue system on our 30-node indoor sensor network testbed. Finally, Section 6 discusses future work and concludes. 2 Motivation and Background Medical care is an oft-cited application for sensor networks [29, 11]. The ability to augment medical telemetry with tiny, wearable, wireless sensors would have a profound impact on many aspects of clinical practice. Emergency medical care, triage, and intensive care can all benefit from continuous vital sign monitoring, especially immediate notification of patient deterioration. Sensor data can be integrated into electronic patient care records and retrieved for later analysis. In a wide range of clinical studies, especially those involving ambulatory or at-home monitoring, wireless sensors would permit data acquisition at higher resolution and for longer dura...

Citation Context

... in communication models [27, 53], time synchronization [39, 15], and energy management [44] should be reevaluated given these new requirements. This is not to say that we must start from scratch; rather, we believe it is best to borrow from prior systems as much as possible and invent new technology only as needed. A number of other research projects are exploring medical sensor networks. Most of these projects are concerned with developing wearable medical sensors [33, 54, 46], while others have developed infrastructures for monitoring individual patients during daily activity [30], at home [12], or at a hospital [31]. In contrast, our focus is to develop a robust, scalable infrastructure for deploying sensor networks in a range of medical settings. More closely related to our efforts are systems for enabling large numbers of medical sensors to be used for disaster response. The SMART [43], AID-N [52], and WiiSARD [32] teams are among several funded through a US National Library of Medicine effort to develop new technologies for disaster management. The AID-N group is making use of our sensor designs, and the SMART team has developed a mote-based EKG [46] that is largely equivalent t...

Citation Context

...eliable communication, and multiple receivers (e.g. PDAs carried by doctors and nurses). Moreover, unlike many sensor network applications, medical monitoring cannot make use of traditional in-network aggregation since it is not generally meaningful to combine data from multiple patients. This paper presents our initial experiences with a prototype medical sensor network platform, called CodeBlue. We have developed a range of medical sensors integrated with the commonly-used Mica2 [8], MicaZ [9] and Telos [41] mote designs. These include a pulse oximeter [37], two-lead electrocardiogram (EKG) [17], and a specialized motion-analysis sensor board. In addition, we have developed a small form factor variant of the Telos mote specifically for wearable use. The CodeBlue software framework provides protocols for device discovery, publish/subscribe multihop routing, and a simple query interface allowing caregivers to request data from groups of patients. In addition to monitoring patient vital signs, CodeBlue also integrates an RF-based localization system, called MoteTrack [34], to track the location of patients and caregivers. This capability is especially valuable in large hospital settings...

Citation Context

... acceptable to recharge devices or change batteries on a relatively frequent basis. As a result, many of the significant advances in communication models [27, 53], time synchronization [39, 15], and energy management [44] should be reevaluated given these new requirements. This is not to say that we must start from scratch; rather, we believe it is best to borrow from prior systems as much as possible and invent new technology only as needed. A number of other research projects are exploring medical sensor networks. Most of these projects are concerned with developing wearable medical sensors [33, 54, 46], while others have developed infrastructures for monitoring individual patients during daily activity [30], at home [12], or at a hospital [31]. In contrast, our focus is to develop a robust, scalable infrastructure for deploying sensor networks in a range of medical settings. More closely related to our efforts are systems for enabling large numbers of medical sensors to be used for disaster response. The SMART [43], AID-N [52], and WiiSARD [32] teams are among several funded through a US National Library of Medicine effort to develop new technologies for disaster management. The AID-N group...

Citation Context

...arly, the use of wearable wireless sensors would greatly simplify data collection and would allow patients to wear the sensors for longer periods of time since the bulky data logger and leads would be eliminated. Three sensor types are commonly used for motion analysis studies in the field: accelerometers, gyroscopes, and surface electrodes for electromyographic (EMG) recordings [4, 40, 5]. Triaxial accelerometers measure the orientation and movement of each body segment. Gyroscopes measure angular velocity and combined with accelerometer data can be used to accurately determine limb position [35, 20]. Surface EMG electrodes capture the electrical field generated by depolarized zones traveling along the muscle fibers during a muscle contraction. The root mean square value of the EMG data is roughly proportional to the force exerted by the monitored muscle. Thus analysis of the patterns of EMG activity can lead to the identification of motor tasks and their characteristics [3]. 3.3.2 Mercury motion analysis sensor board Our Mercury motion analysis sensor board (Figure 1(c)) interfaces to the Telos mote and incorporates a 2g/6g 3-axis accelerometer (STMicroelectronics model LIS3L02AQ), a sin...

Citation Context

...y the low-power radios already incorporated into CodeBlue sensor nodes and end-user devices. In our building, MoteTrack achieves an 80th percentile location error of about 1 m, which is generally accurate enough to locate a patient or caregiver when necessary. In CodeBlue, MoteTrack is simply treated as another sensor type that reports the (x, y, z) location of the device when queried. MoteTrack is an empirical localization scheme that matches the radio “signature” acquired by a roaming device with a database mapping signatures to known locations. MoteTrack improves upon systems such as RADAR [2] in that it does not require a central server to maintain the signature database; rather, this information is stored on the set of beacon nodes that are distributed throughout the area (e.g. a hospital). Each beacon node (which is simply a mote that may be connected to mains power) periodically transmits radio messages at a range of frequencies and transmission power levels (see Figure 8). A mobile node listens for these beacons and acquires a signature that consists of the average received signal strength (RSSI) for each beacon node, frequency, and power level. The signature is compared to a ...

Citation Context

...arly, the use of wearable wireless sensors would greatly simplify data collection and would allow patients to wear the sensors for longer periods of time since the bulky data logger and leads would be eliminated. Three sensor types are commonly used for motion analysis studies in the field: accelerometers, gyroscopes, and surface electrodes for electromyographic (EMG) recordings [4, 40, 5]. Triaxial accelerometers measure the orientation and movement of each body segment. Gyroscopes measure angular velocity and combined with accelerometer data can be used to accurately determine limb position [35, 20]. Surface EMG electrodes capture the electrical field generated by depolarized zones traveling along the muscle fibers during a muscle contraction. The root mean square value of the EMG data is roughly proportional to the force exerted by the monitored muscle. Thus analysis of the patterns of EMG activity can lead to the identification of motor tasks and their characteristics [3]. 3.3.2 Mercury motion analysis sensor board Our Mercury motion analysis sensor board (Figure 1(c)) interfaces to the Telos mote and incorporates a 2g/6g 3-axis accelerometer (STMicroelectronics model LIS3L02AQ), a sin...

Citation Context

...icant advances in communication models [27, 53], time synchronization [39, 15], and energy management [44] should be reevaluated given these new requirements. This is not to say that we must start from scratch; rather, we believe it is best to borrow from prior systems as much as possible and invent new technology only as needed. A number of other research projects are exploring medical sensor networks. Most of these projects are concerned with developing wearable medical sensors [33, 54, 46], while others have developed infrastructures for monitoring individual patients during daily activity [30], at home [12], or at a hospital [31]. In contrast, our focus is to develop a robust, scalable infrastructure for deploying sensor networks in a range of medical settings. More closely related to our efforts are systems for enabling large numbers of medical sensors to be used for disaster response. The SMART [43], AID-N [52], and WiiSARD [32] teams are among several funded through a US National Library of Medicine effort to develop new technologies for disaster management. The AID-N group is making use of our sensor designs, and the SMART team has developed a mote-based EKG [46] that is largel...

Citation Context

...that reliable routing is required for all medical data; rather, the system should allow each query to specify its reliability needs in terms of acceptable loss, data rate, or jitter. Another area worth exploring is the impact of bandwidth limitations and effective techniques for sharing bandwidth across patient sensors. For example, each CodeBlue query could specify a data priority that would allow certain messages (say, an alert from a critical patient) to have higher priority than others in the presence of radio congestion. This approach can be combined with rate-limiting congestion control [14, 25] to bound the bandwidth usage of patient sensors. An important shortcoming of the current CodeBlue prototype is its lack of security. We have already begun to explore the integration of private-key encryption [29] along with a public-key protocol for key distribution [38, 22] in CodeBlue. The privacy and security requirements for medical care are complex and differ depending on the scenario. For example, HIPAA privacy regulations need not be enforced during life-saving procedures. Nevertheless, we intend to integrate some form of end-to-end security into the next version of the CodeBlue system...

Citation Context

... and an optoelectronic sensor opposite. By detecting the amount of light absorbed by hemoglobin in the blood at two different wavelengths (typically 650nm and 805nm), the level of oxygen saturation can be measured. In addition, heart rate can be determined from the pattern of light absorption over time, since blood vessels contract and expand with the patient’s pulse. Computation of HR and SpO2 from the light transmission waveforms can be performed using standard digital signal processing (DSP) techniques. Sophisticated algorithms have been developed to mitigate errors due to motion artifacts [45]. 3.1.2 Mote-based pulse oximeter In developing a mote-based pulse oximeter, we were fortunate that there exist several available products that provide self-contained logic for driving the LEDs and performing the HR and SpO2 calculations. We initially considered the Dolphin Medical [13] OEM 601 and 701 units, credit-card sized boards that contain all of the required signal processing logic. However, they did not meet our requirements due to a current consumption of over 100 mA and an operating voltage of 5 V. The smallest and lowest-power OEM module that we are aware of is the BCI Medical Micr...

Citation Context

...-capture systems use a wired data logger carried in a waist harness; a multitude of wires runs from the harness to various sensors positioned on body segments of interest (typically the arms, legs, back and torso). Clearly, the use of wearable wireless sensors would greatly simplify data collection and would allow patients to wear the sensors for longer periods of time since the bulky data logger and leads would be eliminated. Three sensor types are commonly used for motion analysis studies in the field: accelerometers, gyroscopes, and surface electrodes for electromyographic (EMG) recordings [4, 40, 5]. Triaxial accelerometers measure the orientation and movement of each body segment. Gyroscopes measure angular velocity and combined with accelerometer data can be used to accurately determine limb position [35, 20]. Surface EMG electrodes capture the electrical field generated by depolarized zones traveling along the muscle fibers during a muscle contraction. The root mean square value of the EMG data is roughly proportional to the force exerted by the monitored muscle. Thus analysis of the patterns of EMG activity can lead to the identification of motor tasks and their characteristics [3]. ...

Citation Context

...-capture systems use a wired data logger carried in a waist harness; a multitude of wires runs from the harness to various sensors positioned on body segments of interest (typically the arms, legs, back and torso). Clearly, the use of wearable wireless sensors would greatly simplify data collection and would allow patients to wear the sensors for longer periods of time since the bulky data logger and leads would be eliminated. Three sensor types are commonly used for motion analysis studies in the field: accelerometers, gyroscopes, and surface electrodes for electromyographic (EMG) recordings [4, 40, 5]. Triaxial accelerometers measure the orientation and movement of each body segment. Gyroscopes measure angular velocity and combined with accelerometer data can be used to accurately determine limb position [35, 20]. Surface EMG electrodes capture the electrical field generated by depolarized zones traveling along the muscle fibers during a muscle contraction. The root mean square value of the EMG data is roughly proportional to the force exerted by the monitored muscle. Thus analysis of the patterns of EMG activity can lead to the identification of motor tasks and their characteristics [3]. ...

Citation Context

...s [27, 53], time synchronization [39, 15], and energy management [44] should be reevaluated given these new requirements. This is not to say that we must start from scratch; rather, we believe it is best to borrow from prior systems as much as possible and invent new technology only as needed. A number of other research projects are exploring medical sensor networks. Most of these projects are concerned with developing wearable medical sensors [33, 54, 46], while others have developed infrastructures for monitoring individual patients during daily activity [30], at home [12], or at a hospital [31]. In contrast, our focus is to develop a robust, scalable infrastructure for deploying sensor networks in a range of medical settings. More closely related to our efforts are systems for enabling large numbers of medical sensors to be used for disaster response. The SMART [43], AID-N [52], and WiiSARD [32] teams are among several funded through a US National Library of Medicine effort to develop new technologies for disaster management. The AID-N group is making use of our sensor designs, and the SMART team has developed a mote-based EKG [46] that is largely equivalent to our design described ...

Citation Context

... material describing the communication, routing, discovery, or data query mechanisms used by these systems. 3 Wireless Medical Sensors Medical applications of sensor networks require new hardware designs. In this section we detail three mote-based medical sensors that we have developed: a mote-based pulse oximeter, two-lead electrocardiograph (EKG), and a special-purpose motion-analysis sensorboard. We also describe Pluto, our custom mote design for wearable applications. 3.1 Pulse oximeter Pulse oximetry has been in use as a medical diagnostic technique since its invention in the early 1970s [49]. This non-invasive technology is used to reliably assess two key patient health metrics: heart rate (HR) and blood oxygen saturation (SpO2). These parameters yield critical information, particularly in emergencies when a sudden change in the heart rate or reduction in blood oxygenation can indicate a need for urgent medical intervention. Pulse oximetry can provide advance warning of the onset of hypoxemia even before the patient manifests physical symptoms. 3.1.1 Technology Pulse oximetry involves the projection of infrared and nearinfrared light through blood vessels near the skin. Pulse oxi...

Citation Context

... acceptable to recharge devices or change batteries on a relatively frequent basis. As a result, many of the significant advances in communication models [27, 53], time synchronization [39, 15], and energy management [44] should be reevaluated given these new requirements. This is not to say that we must start from scratch; rather, we believe it is best to borrow from prior systems as much as possible and invent new technology only as needed. A number of other research projects are exploring medical sensor networks. Most of these projects are concerned with developing wearable medical sensors [33, 54, 46], while others have developed infrastructures for monitoring individual patients during daily activity [30], at home [12], or at a hospital [31]. In contrast, our focus is to develop a robust, scalable infrastructure for deploying sensor networks in a range of medical settings. More closely related to our efforts are systems for enabling large numbers of medical sensors to be used for disaster response. The SMART [43], AID-N [52], and WiiSARD [32] teams are among several funded through a US National Library of Medicine effort to develop new technologies for disaster management. The AID-N group...

Citation Context

...0, 5]. Triaxial accelerometers measure the orientation and movement of each body segment. Gyroscopes measure angular velocity and combined with accelerometer data can be used to accurately determine limb position [35, 20]. Surface EMG electrodes capture the electrical field generated by depolarized zones traveling along the muscle fibers during a muscle contraction. The root mean square value of the EMG data is roughly proportional to the force exerted by the monitored muscle. Thus analysis of the patterns of EMG activity can lead to the identification of motor tasks and their characteristics [3]. 3.3.2 Mercury motion analysis sensor board Our Mercury motion analysis sensor board (Figure 1(c)) interfaces to the Telos mote and incorporates a 2g/6g 3-axis accelerometer (STMicroelectronics model LIS3L02AQ), a single-axis gyroscope (Analog Devices ADXRS300) and one EMG unit (MP1A.20.A0DM.60 from Motion Lab Systems, Inc.). The board includes a number of operational amplifiers to enhance signal quality together with voltage conditioning ICs to power the gyroscope and passive filters to eliminate noise. Signals are routed through the board to five ADC ports on the Telos mote. This is a proto...