29-12-2015, 04:48 PM
Abstract
The technology of sensor, pervasive computing, and intelligent information processing is widely used in Body Sensor Networks (BSNs), which are a branch of wireless sensor networks (WSNs). BSNs are playing an increasingly important role in the fields of medical treatment, social welfare and sports, and are changing the way humans use computers. Existing surveys have placed emphasis on the concept and architecture of BSNs, signal acquisition, context-aware sensing, and system technology, while this paper will focus on sensor, data fusion, and network communication. And we will introduce the research status of BSNs, the analysis of hotspots, and future development trends, the discussion of major challenges and technical problems facing currently. The typical research projects and practical application of BSNs are introduced as well. BSNs are progressing along the direction of multi-technology integration and intelligence. Although there are still many problems, the future of BSNs is fundamentally promising, profoundly changing the human-machine relationships and improving the quality of people’s lives.
1. Introduction
As important public network applications, BSN applications are in great demand in medical care [1–3], sports and entertainment [4–6], the military-industrial sector [7], and the social public field [8–10], and BSNs have gradually become a research hotspot. BSNs are a kind of WSN which is formed by physiological parameter sensors placed in the human body, on the body surface or around the body. The main techniques it covers are sensors, data fusion, and network communication. It is not only a new type of universal health care, disease monitoring, and prevention solution, but also an important component of the so-called Internet of Things. Its main purpose is to provide an integrated ubiquitous computing hardware, software, and wireless communication technology platform, and an essential condition for the future development of ubiquitous health care monitoring systems [11].
BSNs originated from WSNs, so there are many similarities between them. However, the characteristics are correspondingly different because of their different application purposes. Firstly, considering network deployment, WSNs can be deployed to inaccessible environment, such as forests, swamps or mountains. Many redundant nodes are placed in the environments mentioned above to solve the problem of node failures, so node density is higher, whereas BSN nodes are deployed in, on or around the human body, so the total number of nodes is generally up to a few dozens. Each node ensures the accuracy of monitoring results by its robustness [12]. Secondly, considering attributes, nodes in WSNs perform the same functions, and have the same properties. The size of nodes is not very critical. Once the node is deployed, it will probably no longer need to be moved. According to the different physical signals collected, BSN applications have different sensor types [13]. Moreover, the requirements of BSN node design are relatively high. The node size must be small enough, and the nodes need to have high wearability and high biocompatibility [14]. Due to the locations the nodes are deployed, they will move as the human body moves. Thirdly, considering energy supply, WSNs and BSNs can be battery-powered. The former, deployed outdoors, can also be powered by wind energy or solar energy, while the latter can also be powered by kinetic energy and heat [15,16]. Finally, considering data transmission, the transfer rates of WSNs are almost the same, but those of BSNs are different, as the data type and channel assignment are different among nodes on the body surface and in the body [13]. Additionally, BSNs deployed in the human body are for monitoring human physiological data, which are subject to user’s personal safety and privacy protection issues. Therefore, QoS and the real-time prosperity of data transmission must be considered [17,18].
The general architecture of a BSN is shown in Figure 1. Sensor nodes which are placed in the body collect physical data and perform preliminary processing. The data are gathered by a sink node and then transmitted to a base station in order to share over the Internet, which is the basis of many applications, including health care systems, social welfare, patients and immediate service, diagnosis services by doctors and medical experts, and emergency treatment systems, etc.
Nowadays BSN research still faces many key technical challenges. Figure 2 summarizes the main research areas of BSNs. The research work on sensor design and use mainly focuses on the wearability of sensor nodes [19], the capacity of fault diagnosis and fault treatment [20], energy consumption [21] and sensor deployment [22], etc. In the aspect of data fusion, research mainly includes the design and implementation of denoising [23], feature extraction [24], data classification [25], data compression [26], and other key technologies. A growing number of scholars combine situational awareness and data fusion technologies in the activity recognition area [27,28]. In the aspects of network communication, research mainly focuses on the problems of network topology design [29], channel characterization [30], channel access control [31,32], routing algorithm design [33], and lightweight communication protocols design, etc. These key technologies must be considered when building a complete BSN system. They are not only of great research value, but also of important practical value.
Existing surveys have made detailed investigations on BSN architecture, signal processing, communication protocols, context awareness and QoS. Reference [12] mainly focused on propagation of information, communication protocols, QoS and security. Moreover, it gave some examples of medical monitoring based on BSN. In Reference [13] the authors provided a detailed research on sensor devices, physical layer, MAC layer and radio technologies, then made a classification on body sensor projects and summarized some open research issues on multi-aspects. In Reference [34], researchers presented an overview of BSN in enabling pervasive healthcare and assistive environments, listing many applications of sensor devices and communication protocols. The authors also analyzed current obstacles and future open issues from an overall angle. Reference [35] placed emphasis on network communication, discussing the issues of physical layer, MAC layer, network layer and routing protocols in detail. Different from the above papers, we set forth our views from the perspective of system design, with a detailed explanation of the research status of three aspects: sensor, data fusion, and network communication. In each section, we summarize the issues which need to be considered in design phase, listing some typical related cases. We also present the trends of development and technical challenges. In addition, we list some key research achievements of BSNs in various fields, such as medicine, social welfare, sports, and man-machine interfaces. Moreover, we have provided a detailed analysis of trends and challenges of BSNs from a systemic perspective.
The remainder of this paper is organized as follows: Section 2 describes the various types of sensors and hotspots in BSNs. Section 3 explores the development of data fusion. Section 4 discusses BSN network topology and communication between layers. In Section 5, we introduce the typical applications in various fields and the system design problems to be considered in BSNs. Section 6 briefly concludes this paper.
2. Sensors
Sensors are the key components of BSN, as they connect the physical world with electronic systems. They are mainly used to collect the information about physiology and the surrounding environment. Sensor nodes, which have a sensor as their main part, are responsible for processing information by format conversion, logical computing, data storage, and transmitting. One sensor node generally comprises a sensor module, processor module, wireless communication module, and power supply module [36]. The sensor module is responsible for collecting the status of measurands and converting data from physical quantities to electrical signals. The processor module is responsible for controlling the sensor nodes. The wireless communication module, consisting of network layer, MAC layer and wireless transceiver in the physical layer, is responsible for communication among sensors and computers. The power supply module is responsible for providing energy for entire the sensor node.
3. Data Fusion
Data fusion in BSNs is a procedure for processing data or information coming from multiple sensors with multi-level, multifaceted processing to make data more effective and meet users’ needs better. BSNs produce a large amount of physiological data according to the application purpose. The way in which these data are manipulated is a fundamental issue faced by sensor node designers. Data fusion techniques combine data from multiple sensors and related information, which can achieve more accurate inferences compared to a single, independent sensor. It also can filter noise effectively, making predictions and inferences from monitored actions or phenomena [56]. At the same time, data fusion techniques can reduce data redundancy, and consequently reduce the load and energy consumption of BSNs, with the advantage of prolonging the network lifetime [57]. Not all data in BSNs needs to be fused, such as blood pressure, body temperature, and heart rate. Data of these types are measured by single sensors, and can reflect physiological performance directly. Other types of data which are measured from multiple sensor in BSNs, like heart waves, pulse waves, and data to identify people’s motion, cannot be used directly. In these cases, data fusion techniques are necessary.