Sampling Rate Prediction of Biosensors in Wireless Body Area Networks using Deep-Learning Methods

https://doi.org/10.1016/j.simpat.2020.102101Get rights and content

Highlights

  • Adaptive sampling by presenting modified Fisher test and developing Spline function.

  • An adaptive method for determining activation and sleep time slices for biosensors.

  • Developing ANFIS and LSTM for predicting sampling rate of biosensors in future.

  • Forecasting energy expenditure of biosensors and patient's status in the future.

Abstract

In this paper, we propose a scheme which aims at determining and forecasting sampling rate of active biosensors in Wireless Body Area Networks (WBANs). In this regard, from the first round until a certain round, the sampling rate of biosensors would be determined. Accordingly, we introduce our modified Fisher test, develop Spline interpolation method, introduce three main parameters namely information of patient's activity, patient's risk and pivot biosensor's value. Then, by employing these parameters plus introduced statistical and mathematical based strategies, the sampling rate of the active biosensors in the next round would be determined at the end of each entire round. After reaching a pre-denoted round the sampling rate of biosensors would be predicted through forecasting methods. In this regard, we develop two machine learning based techniques namely Adaptive Neuro Fuzzy Inference System (ANFIS) and Long Short Term Memory (LSTM) and compare them with four famous similar techniques. In addition to using forecasted sampling frequencies of the biosensors for controlling their energy expenditure, these forecasted values would also be used to forecast patient's status in the future. This is the first work in this domain that uses current information of the patient to determine adaptive sampling frequency and then employs the time series of determined sampling frequencies to forecast the patient's status and biosensors energy expenditure in the future. For estimating our schemes, we simulated them in MATLAB R2018b software and compared the results with a number of similar schemes. Based on the simulation results, the proposed schemes are capable to reduce data traffic by 81%, decrease energy consumption of the network by 73% while having the capability of predicting sampling rate of biosensors with 97% accuracy.

Introduction

Wireless networks are employed in various medical fields, including patient monitoring, emergency systems, fitness programs, chronic diseases and nursing care. These cases include heart rate measurement, blood pressure and blood oxygen level measurement systems, health monitoring systems, artificial pacemakers, and hearing aids. In more advanced cases, devices also monitor the therapeutic course and medications and also their needed amounts. Furthermore, there are also programs designed for the physician to monitor his patient after being discharged from hospital. Many wireless sensor networks (WSNs) technologies are talented to enhance different parts of human lives from travelling to medicine aspects. These technologies facilitate the daily life problems [1]. Because of the growing trend of the world population, especially elder population, some sort of intelligent cares are needed to monitor their health status. Continuous monitoring operations are the result of new presented technologies of heath monitoring networks [2]. At homes that are equipped with health monitoring networks, different controlling strategies for different persons can be presented to check their vital signs and make consequent decisions [[3], [4]]. Continuous monitoring of patients or those people who are susceptible to risky situations helps detect their emergency conditions and reveals related alarms for medical care. Aggregation, fusion and forwarding the captured data by biosensors to any destination is performed by coordinator which is programmed to organize and synchronize the network based on the proposed algorithms [5].

Wireless Body Area Networks (WBANs) contain a number of biosensors placed on the patient's body to receive vital body messages including blood pressure, oxygen level, respiration rate, etc.‌‌‌‌ The biosensors can connect to each other wirelessly to form a wireless body biosensor network. Fig. 1 illustrates an example scenario of WBAN. As shown in Fig. 1, the biosensors of WBAN capture the vital sign data of the patient and communicate them to a decision making center, a remote practitioner, etc. The biosensors may be placed either inside or on the body. Also, wearable technology is another way of connecting the biosensors to the body in these networks. Several communication technologies like Radio Frequency IDentification (RFID), Bluetooth Low Energy (BLE), Ultra Wide Band (UWB), ZigBee (IEEE 802.15.4), Insteon, Z-Wave, RuBee and other related technologies enhance WBAN functionality [6]. RFID uses electromagnetic fields to automatically identify and track tags attached to objects. An RFID tag consists of a tiny radio transponder; a radio receiver and transmitter. RFID is capable of providing object level automatic identification. In parallel with speeding up the development of RFID technology, it is widely used in different fields, such as WSN and WBAN [7]. BLE is a low power consuming Wireless Personal Area Network (WPAN)  technology introduced by Bluetooth Special Interest Group (Bluetooth SIG) to focus on novel applications in the healthcare, fitness, beacons, security, and home entertainment industries. BLE can support forty channels including three advertisement channels and many data channels [8]. UWB is one of the enabling technologies for WSN and related networks like WBAN which provides high transmission data rate, low power consumption and complexity, precise locationing and tracking ability and little interference to other systems [9]. ZigBee which works based upon IEEE 802.15.4, provides high-level communication aiming at creating personal area networks with small, low-power digital radios, like home automation, medical device data collection and other low-power low-bandwidth needs, designed for small scale projects which need wireless connection. Hence, ZigBee is a low-power, low data rate and close proximity wireless network [10]. Z-Wave is a wireless communications protocol used primarily for home automation. It is a mesh network using low-energy radio waves to communicate from appliance to appliance, allowing for wireless control of residential appliances and other devices, such as lighting control, security systems, thermostats, windows, locks, swimming pools and garage door openers [11]. RuBee executes according to two-way wireless protocol by employing long wave magnetic signals aiming at communicating measured data [12]. By considering the fact that Radio Frequency (RF) technology is sometimes highly involved with congestions in communicated channels, one can say that Optical Wireless Communication (OWC) based phenomena like Visible Light Communication (VLC) can significantly solve the mentioned problem. VLC employs Light Emitting Diodes (LEDs) for data communications between its nodes. Employing LEDs in VLC has been reviewed in [13] which proofs the successfulness of this technique compared with other communication methods in the cases of energy efficiency, cheap equipment, high security and also high bandwidth. Some of the OWC advantages are large bandwidth capacity, unregulated spectrum, high degree of spatial confinement and high reuse factor, inherent security and robustness to ElectroMagnetic Interference (EMI). The set of biosensors is employed to gather patient's information and send them to the central coordinator to check patient's status instantaneously [14].

The biosensors of the WBANs can be placed inside the patient's body or put on his skin. In the other hand, wearable technology is another way to connect the biosensors to the patient, as well. In addition, the biosensors may be equipped with devices carried by humans in various situations in clothes, hands or bags. In addition, some new and emerging communication technologies are 5G cellular, Vehicle-to-everything (V2X) wireless, Long-range wireless power, Low-power wide-area (LPWA) networks, Wireless sensing, Enhanced wireless location tracking and Millimeter-wave wireless. Although 5G cellular systems are starting to be deployed in 2019 and 2020 but a complete rollout with take five to eight years. This technology may supplement Wi-Fi as a more cost effective option for high-speed data networking in large sites. V2X wireless systems will enable conventional and self-driving cards to communicate with each other and with the road infrastructure. In addition to exchanging information and status data, V2X can provide a wide range of services, including safety capabilities, navigation support, driver information, and fuel saving.

First-generation wireless power systems have not delivered the user experience that manufacturers expected. The need to place devices on a specific charger point is only slightly better than charging via cable, although there are several new technologies that can charge devices at ranges of up to 1 m or over a table or desk surface. The expectation is that long-range wireless power could eliminate power cables from desktop devices.

LPWA networks provide power-efficient and low-bandwidth connectivity for IoT applications for long battery life. Current LPWA technologies include Narrowband IoT (NB-IoT), Long Term Evolution for Machines (LTE-M), LoRa, and Sigfox and typically support very large areas such as cities or countries. Wireless-sensing technology can be used in a variety of applications from medical diagnostics to smart homes. Wireless signals can be used for sensing purposes in applications such as an indoor radar system for robots and drones or virtual assistants to improve performance. A key trend is for wireless communication systems to sense the locations of devices connected to them. High-precision tracking to about 1-m accuracy will be enabled by the upcoming IEEE 802.11az standard and is expected to be a feature of future 5G standards. Millimeter-wave wireless technology operates at frequencies in the range of 30 to 300 GHz, with wavelengths in the range of 1 to 10 mm. The technology can be used by wireless systems such as Wi-Fi and 5G for short-range, high-bandwidth communications. Adaptive sampling is a technique in which the samples are selected according to the values of the characteristic under study. The motivation of adaptive sampling benefits managing system resources while achieving at high level data acquisition.

In fact, adaptive data sampling can be continued until some conditions are satisfied. Sampling rate adaptation has been performed in WSN by some previous works [[14], [15] and [16]]. The authors of these works tried to propose useful techniques to sense and send the network data to a controller, adaptively. They considered energy consumption of network sensors to perform adaptive sampling. Adaptive sampling methods proposed by authors in [[20], [21] and [22]] paid attention to different condition of the focused environment like physical context parameters. Also, the authors in [[24], [25], [26] and [30]] performed adaptive sampling aiming at enhancing efficient data collection and management in WSN and WBAN. They considered various conditions of periodic networks to execute adaptive sampling.

Limitation of energy consumption in biosensors of these networks should not be neglected, thus the presented sampling strategies must pay attention to optimality in energy consumption to acquire vital sign data. This means determining sampling rate adaptively and according to the patient's statuses prevents over energy consumption in periodic data communication. Despite energy limitation, creating redundant overhead data is another issue in WBANs, therefore, producing and communicating unnecessary data leads decreasing network functionality in either energy consumption or overhead data. In WBANs, the regular biosensors suffer from processing restrictions. Also, they are equipped by non-rechargeable power supplies which lead energy consumption issue. But, the coordinator almost does not have any restrictions in energy consumption and processing capabilities [15]. The framework of the proposed system is shown in Fig. 2. First the vital sign data of the monitored patient are measured; then, they would be preprocessed to determine sampling frequency of the biosensor. At the rest, the proposed sampling frequency forecasting method is utilized to forecast the sampling frequency time series of the biosensor in the later rounds.

In this paper, we assume the most important biosensor as the pivot biosensor. This node contributes in sampling rate determination. Also, to roughly recognize patient's status we assume patients’ activity. This means, the patient's activity acts as one of the most important components in sampling rate determination. The proposed determining and forecasting sampling rate method performs based on three main component including patient's risk, patient's activity and the values of the pivot biosensor.‌‌‌‌‌The proposed method observes variances of the captured vital sign data and also an interpolation function to achieve at optimum sampling rate. For this reason, first a statistical test is executed to denote whether the hypothesis is accepted or rejected. Then, at the second step, the interpolation function determines the optimum sampling rate. Also, our approach employs ANFIS and LSTM to predict sampling rate for network biosensors. More details would be described at the rest of the paper. This paper is an extension of our earlier work [31]. Several estimations and comparisons with a number of similar methods are presented in this paper. Some of motivations of our method are:

  • Determining activation and sleep time slices for biosensors.

  • Presenting an energy efficient adaptive approach to employ information of patient's activity aiming at biosensor activation and deactivation.

  • Presenting Modified Fisher test as a development of ANOVA model to analyze variance of weighted captured vital sign data.

  • Applying cubic Spline interpolation function as the behavior function to determine optimum sampling rate.

  • Defining pivot biosensor and employing its values in addition to values of patient's risk as two main controlling points in the process of optimum adaptive sampling rate determination for active biosensors.

  • Developing Adaptive Neuro Fuzzy Inference System (ANFIS) and Long Short Term Memory (LSTM) for predicting sampling rate of biosensors in futures round(s).

  • This is the first work that focuses on utilizing the current information of the patient to determine adaptive sampling frequency and then employing the time series of determined sampling frequencies to forecast the patient's status and biosensors energy expenditure in the future.

In Section II we briefly mention a number of recent related works. In Section III the details of NEWS based Local Emergency Detection are reviewed. Our approaches for adaptive sampling rate determination including introduced modified fisher test, employed main parameters, developed interpolation function and also proposed algorithms are described in Section IV. Also, the details of our sampling frequency forecasting scheme are explained in Section V. The performance evaluation of our methods and comparison with similar works are explained in Section VI while Section VII concludes the paper.

Section snippets

Related Works

In this section, some of related works are mentioned. Yoon et al. [13] proposed a strategy to adaptively manage compression rate ‌and sensing rate of energy harvesting WSNs. Their proposed approach can improve data resolution in parallel with decreasing blackout nodes’ number. In addition, their approach is capable to adaptively control sampling rate for sensors. Enhancing packet delivery ratio plus decreasing energy consumption of the sensor nodes are the results of their approach. Lee et al.

NEWS Based Emergency Detection

In the common WBSNs the network nodes send the sensed data periodically with the maximum sampling rate. This means the sensor nodes measure all of the patient's vital signs and send them to the coordinator or any destination for making appropriate decisions. In this trend which is introduced as Local Emergency Detection (LED*) [39], energy consumption rises up rapidly which leads to reduction of network overall useful lifetime. This is shown in algorithm 1. As another result, communicating data

Developing Fisher Test

A statistical test based upon Fisher test is introduced in this section by which the sensor nodes inform about the changes of their corresponding features. So the introduced test should calculate the variance of the measurements between current period and the previous ones. This estimation helps to calculate accurate sampling rate for network nodes. Hence, we are interested in developing Fisher test by considering weights for observations during different periods. This means modified Fisher

Sampling Rate Prediction

To forecast sampling rate of network biosensors, two affective tools entitled ANFIS and LSTM are used. Hence, first we point to ANFIS and its operations for sampling rate forecasting and then pay attention to LSTM and introduce the way it can predict sampling rate.

Performance Evaluation

This section indicates the results of performance evaluation of the proposed methods and compares them with previous works. For evaluating the functionality of the proposed approach and its effects on the network we simulated it by using

MATLAB R2018b simulator. The results of simulations are compared with the simulations results of pervious sampling determination methods. We used a dual core lap top Dell Vostro 1310 with Intel Core 2 Duo T9300 / 2.5 GHz, 6 MB L2 Cache, 4 GB RAM and 800 MHz Data

Conclusion

The proposed schemes in this paper determine and predict sampling rate of active biosensors in WBANs. Our approach divides the network execution time of the network into two parts: first part in which the sampling rate of the biosensors in the next round is determined and the second part in which the sampling rate is forecasted. The presented sampling rate determination scheme was performed by introducing and using modified Fisher test, developing and using Spline interpolation method and

Ethical Considerations

This study was also approved by the Ethics Committee of Islamic Azad University, Dezful Branch and written informed consent was obtained from all participants.

Declaration of Competing Interest

The authors declare no conflict of interest.

Acknowledgments

The authors acknowledge the Islamic Azad University, Dezful Branch, who participated in this research paper.

Funding/Support

Islamic Azad University, Dezful Branch, Dezful, Iran.

References (44)

  • W. Ji et al.

    Design and Implementation of a RFID Reader/Router in RFID-WSN Hybrid System

    Future Internet

    (2018)
  • N. Mirza et al.

    Bluetooth low energy based communication framework for intra vehicle wireless sensor networks

  • X. Wu, Y. Cao, X. Yang, and C. Luo, “Performance Analysis of UWB Based Wireless Sensor Networks in Indoor Office LOS...
  • L. Xiaoman et al.

    Design of a ZigBee wireless sensor network node for aquaculture monitoring

  • J.D. Fuller et al.

    Rogue Z-Wave controllers: A persistent attack channel

  • R.A. Khan et al.

    The state-of-the-art wireless body area sensor networks: A survey

    International Journal of Distributed Sensor Networks

    (2018)
  • J.J. George et al.

    A survey on visible light communication

    International Journal Of Engineering And Computer Science

    (2014)
  • A. Dohr et al.

    The Internet of Things for ambient assisted living

  • I. Yoon et al.

    Adaptive sensing and compression rate selection scheme for energy-harvesting wireless sensor networks

    International Journal of Distributed Sensor Networks

    (2017)
  • C. Lee et al.

    Harvesting and Energy aware Adaptive Sampling Algorithm for guaranteeing self-sustainability in Wireless Sensor Networks

    International Conference on Information Networking (ICOIN)

    (2017)
  • I. Scarabottolo et al.

    A spectrum-based adaptive sampling algorithm for smart sensing

  • W. Zhang et al.

    Medical data fusion algorithm based on Internet of things

    Personal and Ubiquitous Computing

    (2018)
  • Cited by (22)

    • General probability distribution model for wireless body sensors in the medical monitoring system

      2022, Biomedical Signal Processing and Control
      Citation Excerpt :

      This network came from the idea of the Wireless Sensor Network (WSN), and WBAN focuses on sensors placed on/in a human body [14–16]. WBAN is a collection of several low-powered and lightweight biosensors that can monitor and detect some functions of the body [17]. The detected functions are applied to the processing stations for abnormal characterization.

    View all citing articles on Scopus
    View full text