Wireless Body Area Sensor Networks (WBASNs)

ABSTRACT

One leading solution to offer proactive and reasonable healthcare provisioning to overcome the growing population and medical professionals shortage is handled through efficient monitoring of sensor nodes capable of prior disease prediction by real-time data monitoring. It leads to appropriate enhancements in the Quality of Service (QoS) in every human life. Wireless Body Area Sensor Networks (WBASNs) are specifically designed for providing a promising solution with superior node functionality and mobility of wireless network over the conventional wired medical system with the adoption of miniaturized size sensors inside, low-power devices for the patients’ body and engaged in the constant monitoring of physiological signals. Moreover, these data are highly prone to threats or an attack, which causes data loss and results in the reduced delay in providing services to the end-users. Therefore, this issue can be resolved via the appropriate compression and encryption of medical data intended for other transmission processes. Some open research challenges need to be resolved concerning compression and encryption. This investigation concentrates on Huffman’s appropriate analysis for the compression and Secure Symmetric Force Encryption (SSFE) algorithm for the encryption process. These analyses’ ultimate target is to provide an efficient WBASN system for transmitting reliable data without any delay. Also, motivated by the increasing demands of improved healthcare and remote systems, decreased device cost, improved healthcare management, and reduced network infra-structure designing cost to fulfill data rate constraints. WBASNs still provides a stronger growing research field where researchers and engineers pretend to investigate to supplement newer solutions with the enhancing opportunities in WBASNs.  The significant contributions of this investigation are given below.

 

1) Firstly, a framework is anticipated to select an appropriate dictionary for EEG signals, which is known as the Discrete Cosine Transform (DCT). It is adopted for the reconstruction of EEG signals effectually and makes the transmission over the network in an efficient manner. After signal reconstruction, the compression process needs to be performed to make communication easier. Here, Huffman coding is employed for attaining a higher compression ratio (CR). The CR of the anticipated model is 11.27, Mean Square Error (MSE) is 0.2, Peak Signal to Noise Ratio (PSNR) is 37.89, and Percentage of Root-Mean Square Difference (PRD) is 2.54 respectively. The execution time for computing CR and PRD is 4.18 and 2.54, respectively. The construction time of Huffman Coding based DCT is 0.86 seconds.

2) The secondary target of this research aims at analyzing the data for the encryption process. Therefore, the medical data can be provided securely and privately. Followed by the efficient execution of compression sensing over the EEG signal, the data needs to be encrypted and decrypted with reduced computational complexity. Thus, the Secure Symmetric Force Encryption algorithm is employed to make this process simpler and attain better results than various prevailing approaches. The power consumption during encryption (mA) is 0.13, voltage drain during encryption (V) is 5.02, and the time taken for execution is 0.000285 seconds, respectively. Similarly, the code size used in this work is 10000 bytes, the input size is 16 bytes, energy consumption is 0.0001102 Joules, and Throughput is 87500.55 bytes per second. This method gives better results when compared to prevailing methods like AES and LEA, respectively.

During the EEG signal transmission process, the energy consumption over WBASN is also analyzed with the exploration of proposed model efficiency. The investigations are performed by numerical analysis for realizing the computational complexity and implementation of compression and encryption techniques over the WBASN. The anticipated model facilitates better compression and encryption of data with reduced computational and time complexities. Therefore, it gives substantial effort to save the node’s battery lifetime and improves the overall performance. The simulation is carried out in a MATLAB environment to make the simulation easier and efficient.

 

 

 

 

 

 

 

 

CHAPTER 1

INTRODUCTION

• Prologue

The necessities of health care resources are increasing due to rapid employment in healthcare applications with diverse technologies. In various countries, it gives the healthcare industry a growing cost of healthcare with numerous populations (M. Maloney et al., 2004). Recently, information and communication development in clinical practices are available readily with various mobile computing devices and Personal Digital Assistants (PDAs). In the current scenario, 50% of healthcare institutions possess Wireless Local Area Networks (WLANs), Wi-Max, and Wi-Fi devices. Therefore, remote monitoring facilitates patients’ monitoring from the outside environment. Diagnosis of disease over remote monitoring can enhance every individual life.

The entire devices are connected through WLAN, facilitating clinicians to observe medical data with prescribed medications and electronic formats instantly. The users are benefited from the advanced smart home monitoring systems, personalized care, teleconsultations, and individual treatments. Originally, medical procedures are connected to patients’ through wires (B.J. Nelson et al., 2010). These installations are beneficial from the advancements of electronic devices. The wired system’s drawbacks towards the patient’s treatment are done with certain coverage regions for a specific time. Thus, wireless networks have been introduced to overcome this crisis.

1.2 Wireless Sensor Networks (WSNs)

WSNs are a self-configured device with a huge amount of low-cost tiny devices of sensor nodes that have the competency to compute sense and communicate with each other. These devices are utilized to examine environmental conditions like sound, vibration, temperature, motion, pressure, or pollutants.  It also assists in improving the essential data from the system and collecting the outcomes from the sink. The sensor node communication is performed with radio signals via power components and transceivers. In WSN, every individual node possesses certain constraints like storage capacity, processing storage, communication bandwidth, and speed.

The sensor nodes function like transceivers that routes and collect data from gateway/sink to end-users (M. R. Yuce et al., 2009). They carry out their unique functionalities for computations to partially transmit essential data. The sink node communicates with the task manager or ends users via satellite, internet, and wireless networks such as mesh networks, Wi-Fi, Wi-Max, and cellular systems. Moreover, in various cases, the sink nodes are connected directly with multiple end-users and multiple gateways/sinks. The WSN components are given in Fig. 1.1.

Fig 1.1 WSN components

Similarly, the power unit is considered a critical element for node energy supply. The power is stored in capacitors or rechargeable batteries. Some natural resources like solar power in cells, or photovoltaic forms, kinetic energy, wind power with turbines from water, and so on are utilized.

The sensor nodes are essential components of wireless nodes that partitions themselves from another embedded system with various communication facilities (M.R. Yuce et al., 2011). It is accountable for gathering information like humidity, temperature, and light and is composed of two sub-units: an Analog to Digital Converter (ADC) and sensors. These sensor units generated signals based on the diverse process. These signals are transformed from ADC and fed into the processing unit.

The benefit of WSNs towards health monitoring systems is tracking individuals’ health conditions to eliminate emergency conditions in the expenditures of low-cost devices. It is accessed by integrating sensors in patients’ bodies to observe their characteristics. The physicians are aware of every individual condition to offer services during poor health conditions immediately. The advancements of implanted biomedical devices are integrated with sensors to assist bio-medical applications.

• Wireless Body Area Sensor Networks (WBASNs)

WBASNs are imminent to health care, disease, and diagnostic monitoring and associated medical processes and offer superior reliability over communication for medical devices, specifically through implantation over the human body. WBASNs are modeled with a minute and smaller sensors to observe the health conditions of every individual securely. It is performed using sensor node placement over the human body as intellectual patches over the skin are deeply implanted over body tissues and integrated into clothing. With this technology, various essential healthcare solutions are offered to enhance human life quality (Maskooki et al., 2011). These nodes help monitor the patient’s body signs like Electro Encephalogram (EEG), Electro Cardiogram (ECG), blood pressure, and other essential environmental factors such as humidity and temperature without any failure. The collected data from every patient from placed WBASNs is transmitted to centralized healthcare systems, ultimately holding permanent records. The data are accessed remotely using physicians to compute the state of the patient’s health condition. Moreover, the patients are alerted with remainder messages and alert systems.

• Architecture

The typical WBASN architecture is shown in Fig 1.2. These nodes are placed in patients’ to gather biological data and carry out essential functions. While performing WSN functionality, the collected data are communicated from the sink node for transmission to BS via the internet. It is executed in various applications like social welfare, health care, and immediate prediction services by emergency treatment systems and physicians/medical experts.

Fig 1.2 A typical WBASNs architecture

• WBASN Sensors

The sensor nodes in WBASNs are determined as a key element that connects the electronic systems with the physical world using information collected from the surrounding and working environment. The sensor node processes the information using logical computing, format conversion, data transmission, and storage (N.A. Khan et al., 2012). Every sensor node comprises the processor module, sensor module, power supply module, and wireless communication process, where the filtering process and compression for given healthcare applications.

The WBASNs module computes and gathers the data for converting the electrical signals. The processor module controls the nodes. The successive communication module transmits sensors with Media Access Control (MAC), network, and physical layers. The sensor nodes are supported by the power supply module to acquire energy (Abdellatif et al., 2018). The system over BSN predicts human behavior with various types of sensors, as shown in Fig 1.3. There are diverse kinds of BSN sensors that are executed with the diverse application for certain requirements. Based on the measured signal types, BSN sensors are partitioned into two types. They are given below:

• The first sensor type gathers consistent time-varying signals with the use of gyroscopes, Electro-Encephalograph (EEG) sensors, Electro Cardio Gram (ECG) sensors, visual sensors, auditory sensors, and Electro-Myography (EMG) sensors. These signals are collected using implanted body sensors and utilized to measure pressure over huge real-time alerts with an enormous amount of power consumption and data transmission.
• The next sensor type includes temperature, glucose, blood pressure, humidity, blood oxygen saturation by monitoring sensors. These types of sensors collect discrete time-varying physiological signals when enormous data transmission is lesser.

 

 

Fig 1.3 Various types of WBASNs based sensors in healthcare applications

With these data transmission media, the generally used BSNs sensors are partitioned into these categories:

• Wireless sensors utilize wireless communication technologies like Zigbee or Bluetooth, Radio Frequency Identification Devices (RFID), and Ultra-Wideband (UWB) to communicate with other devices or sensors. Various real-time applications utilize multiple sensors to enhance the wearing capability and diminish the sensor interferences with general activities.
• Wireless sensors are utilized as an alternative for diverse wired sensors which use the wired communication technologies. The transmission mode is more constant, and the process of eliminating wire over estimated BSN trends.

Human Body Communication (HBC) sensors consider patients’ bodies as a transmission medium. These sensors adopt sub-frequencies with no antennae. HBS diminishes power consumption with sensor nodes size (Bayilmis et al., 2012). Fundamentally, these sensor types are integrated with body-worn devices with communication speed reduction compared to the normal wireless sensor. IEEE 802.15.6 standard assists low power, reliable wireless communication systems. Based on implant sensor nodes positions, BSN sensors are divided into three categories:

1. Wearable sensors are pressure sensors, temperature sensors, and accelerometers. Here, the model design considers the weight and size to protect it from user activity interference.
2. The implanted sensors are placed in the human body as a camera pill. Consider the design process with weight and size to protect from interference with general user activities. These sensor types are minute and bio-compatible, and non-corrosive material.
3. The visual sensors are placed on recognizing and collecting the behavior information regarding the surroundings.

Similarly, based on automatic adjustment ability, BSN sensors are partitioned into two categories.

1. As the name specifies the self-adapting and adjusts the processing techniques, boundary constraints or conditions, parameters, and settles it towards structural characteristics and statistical distribution of data measured to acquire a special treatment process.
2. Similarly, non-adaptive sensors have a simple design without self-adjusting functionality. The requirements towards accuracy and complexity for enhancing this self-adapting technique are applied regularly with sensor design.

 

• WBASN applications

This section discusses the WBASNs application that is classified into two phases: non-medical and medical depends on its application function.

• Medical Applications

In medical applications, WBASN facilitates the constant monitoring of physiological factors like blood pressure, body temperature, and heartbeat (R.M. Shubair et al., 2015). However, the various bio-medical sensor nodes are deployed for monitoring the vital signs of patients for identifying abnormalities over vital signs and then transmit data to gateway like mobile devices.

• Wearable Applications

Some wearable medical devices are placed or worn in healthcare sectors placed near or over the skin of patients. These nodes can measure and sensor various healths based body signs and then transmits the medical data to sink nodes. Glucose level monitoring, temperature monitoring, SpO₂, Electromyography (EMG), Electrocardiogram (ECG), Electroencephalogram (EEG), and blood pressure monitoring are samples of wearable medical applications.

• Implant Applications

The devices are placed or implanted over patients’ bodies with bloodstream for diabetes control systems, cancer prediction, and cardiovascular diseases.

• WBASN types

Wireless Sensor nodes compute various metrics that can collect and respond to the collected data using physical stimuli, if necessary, process the data, then report the information wirelessly, as shown in Fig 1.4. However, the actuator node specifies a wireless device that interacts with the user based on sensors’ data. As well, the sensor components are alike the actuator. The sensor nodes types are available for individual patients,’ as explained below:

 

 

• Electrocardiogram (ECG)

ECG nodes are utilized to monitor ECG signals; it is used to diagnose heart disease in the health care systems (M. Pacelli et al., 2006). As well, ECG nodes are utilized for monitoring different functionality of diverse heart similar locations. The electrodes are attached over the skin of individuals like limbs or chest of the body.

 

Fig 1.4 WBASN for Healthcare application

• Blood pressure

The blood pressure sensor node utilizes oscillometric approaches that are modeled to compute diastolic and systolic human blood pressure.

• Electroencephalogram (EEG)

EEG sensor predicts the brains’ electrical activities via the electrodes attached to scalps’ at diverse locations. Electrodes sense the brains’ electrical activities, and it transmits to an amplifier that produces tracing patterns.

 

• Electromyography (EMG)

It computes muscles produce electrical signals via contractions or rest. In the human body, the impulses specify an electrical signal, which leads to muscle reaction in diverse ways; both muscle and nerve lead to this response.

• Blood glucose

Diabetes is a long-term problem that needs constant monitoring of the blood sugar level of patients’. Blood glucose is to examine the sugar level (glucose) circulating over the blood. As well, it is known as glucose monitoring or blood sugar monitoring. The glucose measurements are performed with conventional techniques like optical meter (glucose meter) and other recent techniques like monitoring infra-red technology and optical sensing. Some other examples of monitoring glucose levels are the Gluco-watch that specifies a set of wearable systems.

Fig 1.5 Generic view of WBASN based sensor components

• Temperature sensors

Temperature sensors are used for measuring temperature during health monitoring via collected signal information from the human body. Moreover, when there are certain amounts of variations are identified, an alarm signal is issued. As well, the humidity sensor computes the humidity of the environment around the individual patients.

Some other devices are used for finding the health-based application area like coordinator nodes, relay nodes, and personal devices (wireless). When considering personal devices (wireless), these devices are accountable for collecting information from sensors and actuators that perform interactions with other users like nurses, patients, or GP via external gateways or device displays (Adel et al., 2018). This device is also termed as sink nodes, body gateway, personal digital assistant (PDA), Body Control Unit (BCU) for some applications. The coordinate nodes are analogous towards the outside world’s gateway, successive WBASN, trust center, and access coordinator. All WBASN nodes are used for communication through parent nodes, an intermediate node, relay nodes, and child nodes (Al-Sa’D MF et al., 2018). The ultimate functionality of these relay nodes is to forward data obtained from other nodes until it reaches PDA. Fig 1.5 depicts the sensor components in WBASNs.

• Security Requirements

More specifically, the use of wireless technology in healthcare is related to privacy and security concerns. Fig 1.6 shows the energy efficiency based technique generation involved in WBASNs. The systems’ security methods offer various services over certain bio-medical data for some applications (Dao et al., 2018). Those requirements are listed below:

Encryption- The data which is encrypted is disclosed during transmission. It offers confidentiality towards eavesdropping or from malicious attackers.

Integrity – It comprises data authentication and integrity. It is essential while altering or modifying the data that is transmitted over the insecure channel. The appropriate data integrity methods are performed to fulfill whether the information is changed or not.

Confidentiality- the data over the system should not be exposed, as it needs data confidentiality. Some information causes overhead or leads to vulnerability or eavesdropping during the transmission process. Therefore, the data is shared or encrypted with a secret key via a secured channel to attain confidentiality.

Fig 1.6 Flow diagram of wireless sensor-based functionalities

Protection- Data freshness fulfills that data frames are in order and not changed during the re-transmission or transmission process. This technique is essential for providing integrity and confidentiality.

Authentication- In WBASN, non-medical and medical applications require data authentication. This secure technique is utilized for achieving data authentication. This method shares a secret key for evaluating the Message Authentication Code for all transmission data.

Availability – it fulfills the accessibility of the patient’s data from the physician. It can be eliminated by disabling the ECG model and causes a necessary condition like losing a life. During availability loss, some techniques have to maintain switching operations or change to other network models.

Dependability – The system has to be more dependable or reliable. During data retrieval, some failure is specified as a crucial concern for WBASNs as it is measured as a life-threatening issue for patients. Error-correction based coding approaches are used to resolve this crisis.

Secure Localization – Some WBASN applications also require an appropriate estimation of patients’ locations. The lack of various tracking approaches leads attackers to acquire data like replaying towards fake signals while knowing the patient’s location.

• Problem statement

The reliability of physiological data transmission is still a huge challenge in WBASN and healthcare monitoring systems. It has to be given higher significance. As well, various existing investigations do not concentrate on reliability over WBASNs bottleneck zone. Some studies fail in proving the reliability process to resolve the loss in bio-medical data packets.

Data reliability is the most crucial factor while considering WBASNs, and specifies a key component in WBSN health care applications. Diverse issues are investigating the transmission reliability for health care data in the WBSN environment to acquire higher success probability in the reception of packets without loss in the sink node. Various coding techniques are used over bio-medical data to enhance reliability by reducing loss (T. Padma Priya et al., 2015). Thus, the investigations towards the network coding approaches are highly recommended, and it should resolve the drawbacks encountered in XOR and random linear network coding.

• Challenges in WBASNs

WBASN is a special kind of WSN which inherits various challenges. Moreover, different new challenges and unique WBASN characteristics lead to a huge problem, and it needs better solutions. Thus, multiple investigators concentrate on predicting the difference among the WSN and WBASN, where the practical adoption of WBASN cannot be achieved without any proper ethical, technical, and social challenges for tackling them with diverse kind’s network faces. The essential factor is to acquire reliable network transmission of data with reduced delay and maximal Throughput when considering power consumption. Similarly, it has to minimize unnecessary communication like idle listening, control frame overhead, and frame collisions. The users’ needs, like safety, privacy, security, ease of use, and compatibility, are also considered essential (Zeng et al., 2016). The most challenging factors related to WBASN are the reliability of data during transmission. This work shows a great deal in data reliability with an effectual compression technique for improving the system functionality.

• Research Motivation

The main motivation behind this investigation is to offer a reliable data transmission of medial data without any loss in the WBASNs environment. When the patient is under risk, WBASNs carry medical packets like heart rate to EEG that has to be reliably delivered to the physicians. Hence, some loss in those bio-medical packets causes risk in a patient’s life and leads to misinterpretations due to incomplete packet reception. However, various investigations concentrate on the reliability process and to resolve the issues associated with this. Reliable data transmission is more challenging in medical monitoring systems like an emergency event. Therefore, further research is needed for the productive development of compression techniques to avoid packet loss.

However, WBASNs are not modeled for reliable medical data transmission. Hence, it paves the way for using network coding techniques like Huffman coding to acquire superior data transmission reliability in health care applications. Motivated by the characteristics of these network coding techniques, the data reliability performance is enhanced in WBASNs.

In WBASNs, reliable transmission and reduced packet drop over bio-sensor nodes are essential with body area networks. It shows a key requirement in WBASN applications specifically for health care systems. The ultimate motivation towards this problem is to reduce the packet loss by implying compression techniques and ensure data transmission reliability to acquire effectual delivery of medical data transmission.

• Research Objectives

The ultimate objectives of this investigation are given below:

• To improve the lossless EEG signal decomposition performance.
• To compute the protection and information security over the Internet of Things (IoT) applications.
• To improve maximal compression rate towards input signals devoid of any information loss.
• To enhance privacy with the SSFE algorithm using EEG signals.
• To reduce the energy utilization of the WSBAN devices placed over the patients.

 

• Research question

EEG signal processing needs a prolonged period for recording and its outcomes in a huge amount of data. It leads the basic requirements like communication bandwidth and huge storage over WBASN. However, the storage can be reduced by compressing the huge amount of EEG data for performing transmission over WBASN. Thus, an efficient compression technique and signal reconstruction algorithm is required. In earlier researches, CS is used for performing communication and reduces the data rate. Thereby, the energy consumed by the nodes is also reduced. But, the essential factor is that the domain for signal compression should be sparse. It is known that EEG signals are neither sparse in the frequency domain nor time domain. The compression scheme’s primary challenge is determining the field as the dictionary where the EEG signal is considered sparser. Generally, wavelet-based techniques are used for compression as the characteristics are not completely analyzed. Thus, the research question rises with signal reconstruction’s need to perform appropriate compression for different transmission purposes.

• Research gap

Various investigators do not analyze the dictionary properties, which make the appropriate EEG signal selection. Due to the lack in this path, the compression sensing and signal reconstruction are not provided efficiently. Subsequently, it is observed that Discrete Cosine Transform (DCT) is used for handling the complex EEG signals effectually and sparsely. These two factors are considered as the major research area to attain wider knowledge construction. Various investigators have not focused on handling encryption in EEG signals and concentrate on three essential factors: signal reconstruction, compression, and the encryption process.

• Research Methodology

The quantitative scientific research methodology adopted in this work is based on theories, experiments, and hypothetical conditions based on data generated from the WBASN environment. The significant phases of this research methodology are explained in Fig 1.7. This flow includes the successive stages of research methodology: analyzing the previous research works, predicting the problems related to WBASNs, examining the situation in the mathematical or algorithmic model, processing the data, implementing the data, and computing the results. This implementation of the proposed model assists in acquiring the aim and objective of this research with a network coding technique known as Huffman Coding. The research methodology is shown in Fig 1.7.

Fig 1.7 Flow diagram of Research Methodology

• Research scope

This research’s ultimate scope is modeling efficient compression and encryption techniques for EEG signal processing over WBASN and performs appropriate data reconstruction at the receiver side. Similarly, this work attempts to fulfill EEG’s constant monitoring and acquires better communication from the patients’ to physicians.

• Research contributions

Based on the research gap discussed above, this research’s ultimate target is divided into two major parts. They are 1) Compression and 2) Encryption.

1) In this dissertation, a framework is anticipated for selecting an appropriate dictionary for EEG signals, which is known as the Discrete Cosine Transform (DCT). It is adopted for the reconstruction of EEG signals effectually and makes the transmission over the network in an efficient manner. After signal reconstruction, the compression process needs to be performed to make communication easier. Here, Huffman coding is employed for attaining a higher compression ratio.

2) The secondary target of this research aims at analyzing the data for the encryption process. Therefore, the medical data can be provided securely and privately. Followed by the efficient execution of compression sensing over the EEG signal, the data needs to be encrypted and decrypted with reduced computational complexity. Thus, the Secure Symmetric Force Encryption algorithm is employed to make this process simpler and attain better results than various prevailing approaches.

During the EEG signal transmission process, the energy consumption over WBASN is also analyzed with the exploration of proposed model efficiency. The investigations are performed by numerical analysis for realizing the computational complexity and implementation of compression and encryption techniques over the WBASN. The anticipated model facilitates better compression and encryption of data with reduced computational and time complexities. Therefore, it gives substantial effort to save the node’s battery lifetime and improves the overall performance.

• Thesis organization

The thesis is structured as follows.

Chapter 1 discusses the basic ideas regarding the EEG signal processing, advantages, limitations, research gap, research scope, objectives of the proposed model, etc.

Chapter 2 depicts the background analysis over the EEG signal processing, reviews on energy-efficient routing, interference mitigation, mobility management, security policies, compression algorithms, security threats, challenges, and future recommendations.

Chapter 3 provides the EEG signal transmission over WBAN with background ideas regarding EEG signal, generation, recording system, and applications. The chapter discusses the compression techniques, DCT, and IDCT for signal reconstruction.

Chapter 4 depicts the feasibility of the anticipated framework to enable encryption from a transmission point of view. Here, Secure Symmetric Force Encryption is used for compressing the EEG data by generating a secret key to ensure privacy and security over the network.

Finally, Chapter 5 concludes this dissertation by summarizing the research contributions and suggesting the research’s potential extension.