Body Sensor Networks

Abstract: Recent technological advancements and ultra large scale integration of devices lead to the development of Body Sensor Networks which are used for flexible health care applications. The e-health involves the patient's body in which sensor nodes are implanted require a secure network for communication due to privacy concern. Fuzzy commitment technique is used in key management schemes. The key management schemes used for security consume lot of energy which is undesirable because sensors have limited energy resources due to their small battery size. In this review paper, we have reviewed different key management schemes and compared their performance parameters in terms of energy consumption.
Keywords: Body Sensor network, biometrics, Fuzzy commitment.
I. INTRODUCTION

The period of evolutionary advancements in wireless technology, material science, miniaturization of devices and the Internet allow us to change the way in which the sensors used for monitoring and controlling the in accessible environments. There are many applications of sensor networks such as military, environmental, healthcare and other commercial applications. The most important and critical is the medical field and the utmost subject is the way in which health care services are delivered in normal and emergency situations. The health care market has become a four trillion dollar industry worldwide. Predictions estimate a doubling within ten years. With this enormous growth there comes a need for new information technologies to achieve two critical goals: cutting costs while improving the quality of care. According to current census, the average life of a person has significantly increased from 58 years to 78 years. This trend is global, so the worldwide population over age 65 is expected to be more than double arises from 357 million in 1990 to 761 million in 2025. These statistics requires the most attention towards more flexible, comfortable and affordable health care solutions. The developments in miniaturization and low powered electronics have lead to the development of health monitoring systems called body sensor networks (BSNs). BAN has revolutionized the patient monitoring system. Remote healthcare monitoring has the advantages of reduced medical costs, increased medical quality, continuous and timely patient monitoring, and timely presenting the correct adaptive remedy. A Body Area Network is formally defined by IEEE 802.15 as, "a communication standard optimized for low power devices and operation on, in or around the human body (but not limited to humans) to serve a variety of applications including medical, consumer electronics / personal entertainment and other" [1]. A Wireless Body Area Networks (WBANs) consist of sensor nodes which may be wearable or implanted inside the patient's body to monitor the body and to collect the vital signals (heart rate, blood pressure, oxygen saturation, muscle activity) continuously and send them to the medical server installed at the hospitals through Internet. BSNs are composed of some biosensor nodes which are micro scale electronic equipments integrated with biosensors and wireless transceivers [2]. In more common terms, a Body Area Network is a simple-star network topology which is formed by the group of sensor nodes and a central processing unit which collects all the vital signals from these nodes, process them and send them to the personal server which may be a cell phone through Bluetooth (802.15.1) or Zigbee(802.15.4). The personal server sends this information to medical server installed at the hospital through Internet or other techniques like GPRS (3G, 2G) etc. [3]. The BSNs are connected in various topologies according to its applications. The main two topologies are star and mesh topology. In star topology, there is central control unit like hub usually called Personal Digital assistant in BSNs. It is similar to local treatment unit which collects the different information from all the sensor nodes and transfer it to the hospital server. In mesh topology, the communication is peer to peer like in ad-hoc networks and nodes communicate with each other intelligently without any central controlling unit and send data to cell phone via Bluetooth to hospital server unit [4].

Thus, BSNs provide us applications in various fields which includes (1) continuous and automated monitoring of human body which needs frequent care (2) Intelligent Treatment, such as automatically drug delivery at specific time to a specific part of body (3) Flexible health monitoring by overcoming the wired communication used for inner body signal measurements, (4) also help to protect those exposed to potentially life-threatening environments, such as soldiers and deep-sea and space explorers.
Practically, wireless sensor networks along with numerous applications have a lot of challenges. These applications have many challenges [5]:
1) Security: Considerable effort would be required to make BSN transmission secure. It would have to be made sure that the patient's data is only derived from its own sensors and is not mixed up with data derived from the sensors of other patients.
2) Limited Energy Sources: Sensors use batteries for power supply. Due to miniaturization of sensor devices and their flexible implantation in human body, batteries should be made small in size. Therefore sensors get easily depleted of energy and have limited energy for long term usage.
3) Flexibility: The sensors used in BSNs would have to be less complex, small, light weighted, energy efficient, easy to use and scalable. The storage capacity required to store the patient's data should less as this will affect the sensor size.
4) Interference: The movement of patients with implanted or wearable nodes would cause a problem of interference with other patient's BSN. The confidential data may be lost or interchange at the server, which cause a serious problem to patient's safety and privacy.

The main challenges are security and minimizing the power consumption. Because data exchanged in e-health contain a great deal of sensitive health information, the Health Insurance Portability and Accountability Act (HIPAA) mandates that e-health data must be protected from unauthorized access and tampering. For example, in Black Hat Conference in 2012 the researchers from McAfee have revealed that they can take control of insulin pumps implanted inside diabetes patients, while scientists at the University of Massachusetts have shown that they can use radio attacks to turn off defibrillators inside heart patients. The problem is that the security on the radio link is breakable, and the implants' operation can be remotely over-ridden [6]. A secure communication between the sensor nodes has become a mandatory factor. But along with this the other factor of concern is energy consumption. The security mechanism required for sensor nodes consumes a lot of energy which is undesirable as the sensors have limited battery because of their small size. Key management is the set of techniques and procedures supporting the establishment and maintenance of keying relationships between authorized party. A keying relationship is the state where the communicating entities share keying material to facilitate cryptography techniques. Most attacks aimed at key management level rather than cryptographic algorithm itself [7]. The key management scheme must be efficient and secure enough to overcome all the risks. The other main factor for nowadays concern is power consumption. Sensors have limited bandwidth and computational resources. However, key management protocols require a great amount of energy consumption, particularly in the transmission of initial key negotiation messages. The key management scheme is such that it can consume minimum power resulting in increasing the lifetime of the network. There are mainly six stages in key management-
1) key generation
2) key storage
3) key establishment
4) key usage
5) key change or refreshment
6) Key destruction.
Each stage requires a significant amount of energy. This significant energy consumption at each stage lead to decrease in lifetime of the sensor node and there is a need of frequent change in the sensor nodes during the observation period of a patient which further increases the cost as well increase the inconvenience for the patient.
As per the organization of the paper, we have reviewed various key management schemes and their various aspects in terms of energy consumption. In addition, we discuss the various problems related to the techniques and their future scope. The remainder of the paper is organized as follows: In Section 2, we present and analyze related work of key management schemes for BSNs. In Section 3, we discuss the research gap and various problems related to existing techniques. This is followed by conclusion and future scope in Sections 4 and 5.
II. RELATED WORK

BSNs are primarily used for two main e-health applications: to collect the medical data from patient's body at regular intervals and monitor it from remote distance and to give automatically drug dose or to handle the emergency conditions easily from far distance [8].

Cherukuri [9] proposed a scheme for security in BSNs which enhance the inexpensive mechanism for secure key distribution rather than costly operations such as asymmetric encryption algorithms. They discussed the basic need for cryptography in sensor networks because as sensors are implanted in patient's body at very delicate organs and their safety is necessary. They discussed the constraints of biosensor networks such as energy, bandwidth, limited memory, and limited computational capabilities and conclude that asymmetric cryptographic techniques require much communication and computational factors as compared to symmetric key distribution. The biosensors are placed inside the human body and the biometrics derived from human body is used for key distribution. The security mechanism discussed in this paper used a Fuzzy Commitment technique for distributing keys. In fuzzy commitment technique sender get committed to a chosen value while keeping it hidden from others and make the receiver to reveal the committed value later. It consists of two phases i.e. commit phase and reveal phase. In commit phase the sender calculates value and protects it and then in reveal phase the receiver calculates the value and validates it. They try to minimize the communication and computation by biosensors due to their power constraints and whole of data aggregation and dissemination would be done by control node and base station as they have enough energy. The main focus of this paper is towards the secure communication between the sensor nodes. After pre-deployment of keys in sensor nodes, there was a need of re-keying after sometime for the refreshment of keys for security purpose. They use biometric values for key calculation which are automatically changed with time. A random binary number is generated from the biometric value and is xored with the data tobe send using hash function. The receiver calculates a different biometric value at that time and xored it with the received value to get the similar biometric value. If the value matches then the message is decrypted by the receiver. If there is any error in the calculation the hamming distance is used to detect the error which can tolerate up to 10% of error. They eliminate the computational cost and reduce the unnecessary communication on between the nodes under constrained conditions as compared to traditional asymmetric key establishment techniques for generic sensors networks. They make use of secure key distribution without using expensive computations like exponentiation and more communication rounds. This would lead to an appropriate use of energy and bandwidth. An error correcting method must be used to alleviate the problem of error occurred during the calculation of biometric values due to their randomness. The entropy of some biometrics like heart rate is not satisfactory, so the entropy could increase to satisfactorily level by multiple readings of signal at same instant which will lead to accuracy. This algorithm is less complex but they do not consider time synchronization between two nodes for establishing the shared key. This scheme is well aware of huge energy consumption during communication but ignored the relation between transmission distance and energy consumption in nodes.

Muhammad [10] proposed a key management scheme named BARI+ for data integrity and protecting patient's confidentiality from eavesdropping and unauthorized access. The crucial application fields highlight the use of security in BSNs. As per the claim of this paper, previous key management schemes like LEAP+, MUQAMI+, and SHELL+ are energy efficient schemes but they are designed for large sensor networks in which they assumed that all nodes are not in communication range of each other. BARI+ is a distributed key management scheme in which keys are refreshed at regular intervals and distributed by nodes in timely order. They assume that PS is preloaded with node identities and all nodes are preloaded with relevant keys before their deployment. This scheme used four types of keys: communication key which is used to transfer data through network, administrative key used to refresh the communication key, basic key is used for communication between node and PS and a secret key shared between sensor node and the medical server. This scheme manages the distribution by refreshing the keys at regular intervals by different nodes as the time slots issued by PS. This scheme is compared with previous schemes like MUQAMI+ and LEAP+ for energy consumption with respect to different phases. The PS and nodes in BARI+ scheme consumes less energy in initial deployment, admin refreshment and communication refreshment phase as compared to other schemes. If shared distribution and refreshment of keys is done continuously and each node in turn participate for refreshing the key, then the networks can be prevented from various attacks like flooding, spoofing etc. and also their energy consumption should be minimized. The schemes designed for WSN are not applicable in BSNs because of their different constraints and operating environment. However, in this scheme, PS needs to broadcast and receive the key management messages in continuous manner, which consumed a lot of energy. Hence, this scheme is not a good solution to lower the energy consumption during key management for BSNs, although it is better than LEAP+ and MUQAMI+ which are designed for WSNs. This is an improved fuzzy commitment technique in which more than two nodes negotiate a shared key but it does not provide time synchronization among the multiple sensor nodes sharing the same biometrics.

Lu [8] proposed a privacy preserving scheme SPOC in which opportunistic environment is used for medical help in emergency conditions. However, usually smart phones are used for sending data to the medical centre through wireless communication. As smart phones are also used for other purposes also like calling, internet surfing and they do not have enough energy to be used during emergency conditions as in such conditions data has to be send every 10s. As per the algorithm of this paper opportunistic conditions are used, for example to get help from the neighboring nodes during high traffic. This lead to another problem related to security because the personal health information of the patient is revealed to other users also which is a serious attack on privacy. In this paper, they proposed a new opportunistic framework to address this challenge. The authors had proposed a framework in which only that user is eligible to participate in opportunistic computing who satisfies certain conditions. The effective attribute based access control allows the user to collect the information about the other medical users with appropriate resources and PPSC (privacy preserving scalar product computation) allow only those medical users to participate in opportunistic computation who have similar symptoms without revealing patient's symptoms. The proposed scheme has three phases:
1) system initialization
2) user-centric privacy access control
3) Analysis of opportunistic computing in emergency.
Then according to their symptoms they are categorized and participate in opportunistic computation. When an emergency occurs, the patient's network contact with the nearby users for additional resources. If they have enough energy, then their symptoms are checked. If they matched, then they participate in computation .The performance metrics used in evaluation of SPOC are the average number of helpers that participate in opportunistic computing and the ratio of total resources used by patient in emergency to the total resources used in computing the personal healthcare information in a specific time period. If the number of users and their arrival at a particular location increases then the average number of helpers also increases. This scheme has minimum computation cost and communication cost as compared to traditional scheme. By taking the help of nearby users the energy resources are utilized more efficiently and computation cost is reduced. Also this will lead to handle the critical emergency situations by consuming less energy. This algorithm is very complex and consists of many recursions and, no work has been done on the security issues related to the attacks within the network. The key storage requirement for the trusted authority in system initialization phase also increases with increase in number of nodes which affects their scalability and flexibility of the network.

Zhao [11] proposed a scheme for efficient key management to address the two pivotal problems of key storage and power consumption during cryptography process. In the previous proposed schemes no satisfactory work has been done on the energy consumption prospect. All the previous schemes ignored the energy consumption during the communication phase but this phase contributes a lot in computational cost as compared to the overall cost. Time synchronization is another fact that must be taken into account when biometrics derived from human body is used to secure the keying material. In BSN physiological values (heart rate, muscle activity etc.) are used for secure communication as they are time variant and difficult to interpret. Thus sender and receiver need accurate time synchronization to get tuned. They proposed a fuzzy commitment technique for energy efficient key management for BSNs which give the solution to the energy problem of accurate time synchronization. A hybrid (star and mesh) multihop network is used which include a Personal Digital Assistant, wearable and implanted nodes. Wearable nodes are specifically used as cluster heads as they are large in size and have an ample amount of energy for communication with medical server and are also easily replaceable. Before deployment of nodes, two keys K1 and K2 and a hash function H (.) is loaded into nodes for security of system. The keys required for communication are calculated with the help of this these keys. Some biometrics such as heart rate is not good choice because their level of variation is unsatisfactory. To address this problem they use a method that when a sensor node requires a key it collect the biometric signal and transfer it into binary value b. Then it calculates k=h (K2,b) which becomes a pseudorandom value as K2 is unknown to the intruder and hence k also. As mentioned above, time synchronization fuzzy commitment technique requires a significant amount of energy. Therefore, weak time synchronization is used in which a time window is used to allocate a fixed space to each node to store medical data and when the next time window comes the new data will replace the old data. This reduces the energy consumption during the negotiation of session keys. The results of this work were that even if there is increase in number of biosensor nodes the key storage requirement remains low and constant. However, in traditional techniques the key storage space increases as increases the number of nodes. In terms of energy consumption, the energy remains low in this key management scheme during the session key negotiation as number of secret attempts increases for weak time synchronization as compared to conventional schemes. As hybrid network is used with multihop communication rather than single hop this reduces the energy consumption across all transmission paths and weak time synchronization can reduce the energy consumption during the key negotiation phase. The algorithm is less complex and consumes less time because of synchronization. In this paper, the detailed analysis of energy consumption at each stage of cryptographic process has not been done. In analysis of this algorithm it was found that if we analyze the energy consumption at each and every step we are able to minimize the consumption as compared to traditional methods and hence increase the lifetime of the nodes as well as reduce the cost of installation of BSN which is very important factor as individual point of view.

TABLE I . COMPARISON BETWEEN KEY MANAGEMENT SCHEMES

Performance Parameters

Scheme Use of Biometrics Energy consumption during refreshment schedule Opportunistic Computation Security
Memory for Key storage Energy Efficiency
BIOSEC Yes ' ' Yes ' Low
BARI+ Yes Yes ' Yes More Limited
SPOC Yes ' Yes Yes More Average
EFFICIENT KEY MANAGEMENT Yes ' ' Yes Less Good


III. RESEARCH GAP

In the previous research, limited work has been done to check the following factors:
' Energy consumption during communication vs. number of nodes.
' Energy consumption during encryption of keys vs. number of nodes.
' Energy consumption during decryption of keys vs. number of nodes.
' Energy consumption during generation of keys.
' Energy consumption during signature generation.
' Energy consumption during initialization of keys.
' Energy consumption during broadcast of keys.
If we analyze the energy consumption at each and every step of encryption and decryption process we are able to minimize the consumption as compared to traditional methods and hence increase the lifetime of the nodes as well as reduce the cost of installation of BSN which is very important factor as individual point of view.
IV. CONCLUSION
Security and data confidentiality is an important aspect in e-health. To overcome this problem, cryptographic scheme is used. But, due to limited energy resources the key management scheme used must be energy efficient. This is the main problem discussed in this paper is about the energy consumption related to various stages in cryptographic process and time synchronization in nodes to measure biometrics. If we create a mesh network of Body Sensor nodes and simulate the network according to following steps. Then we may able to achieve a technique which calculates energy at every step of cryptographic process.
' Develop an algorithm for ultra low energy consuming routing based on fuzzy commitment technique and weak time synchronization as in the previous work.
' Based on the limitation of fuzzy commitment technique, enhance the proposed algorithm in terms of energy consumption w.r.t. generation of keys, encryption, broadcast and decryption of keys.
' Evaluate the performance of both algorithms for better results.
V. FUTURE SCOPE
The problem of concern in sensor networks is the energy consumption as they have limited sources of energy due to their size and their implementation in harsh and inaccessible locations. Due to their application in e-health, security is another main factor which must be consider as sensors are installed within the patient's body thus and communication must be secure. For the security and energy consumption point of view, a detailed energy analysis of key management schemes has to be done to minimize the computational cost. For this, a body sensor network mesh structure is developed in which the nodes can communicate with each other like an ad-hoc network without any Personal Digital Assistant. Develop an algorithm for ultra low energy consuming technique for communication. The commonly used technique is Fuzzy Commitment technique in which physiological values are used as keys for security and weak time synchronization is needed for synchronized communication between the sensor nodes. Then based on the limitations of fuzzy commitment technique, we develop an enhanced algorithm in terms of energy consumption w.r.t. generation of keys, encryption, broadcast and decryption of keys and then evaluate the performance of both algorithms. Then an ultra low energy algorithm has been developed then we are able to analyze the energy consumption at each and every step of encryption and decryption process and energy consumption is minimized as compared to traditional methods and hence this increases the lifetime of the nodes as well as reduce the cost of installation of BSN which is very important factor as individual point of view.

Source: Essay UK - http://www.essay.uk.com/free-essays/information-technology/body-sensor-networks.php



About this resource

This Information Technology essay was submitted to us by a student in order to help you with your studies.


Search our content:


  • Download this page
  • Print this page
  • Search again

  • Word count:

    This page has approximately words.


    Share:


    Cite:

    If you use part of this page in your own work, you need to provide a citation, as follows:

    Essay UK, Body Sensor Networks. Available from: <https://www.essay.uk.com/free-essays/information-technology/body-sensor-networks.php> [05-06-20].


    More information:

    If you are the original author of this content and no longer wish to have it published on our website then please click on the link below to request removal: