Introduction

Wireless body area networks (WBANs) include a collection of body sensor nodes that are low power, invasive or non-invasive, light-weight devices which are either worn on the body or implanted inside the body. The architecture of WBAN [1] can be considered as three different tiers, namely: Tier-1 as Intra-BAN, Tier-2 as Inter-BAN and Tier-3 as Extra-BAN as shown in Fig. 1. In Tier-1, the body sensor nodes collect the data and send to the coordinator. In Tier-2, the coordinator processes the received data and sends the information towards the sink node. From the sink node the packets are transmitted to the corresponding health-center through internet or other communication techniques.

Fig. 1
figure 1

Wireless body area network architecture

Full size image

According to the survey conducted by the Economic and Social Affairs Department of United Secretariat [2], after 10 years people with age above 65 will be approximately 15% of total world population. As, older people are more exposed to medical health issues, the need for low cost health monitoring devices [3] becomes a major part human life.

WBAN is actually a subset of conventional wireless sensor networks (WSNs), which can be used for early detection of various diseases, for real-time patient monitoring of elderly people, etc. The body sensor node is either worn on the body surface or implanted inside the body. The sensed data is sent to the Base Station and is then forwarded to the sink node. The sink node is responsible for sending the information to its corresponding healthcare center.

The reliability and efficiency of WBAN depend on how the system responds quickly and accurately, to send and receive the data between the nodes, which eventually depends on the selected routing protocols or algorithms. The process of sending information from either an in-body or an on-body sensor node includes the radiation emitted from wireless transceivers which is similar to WSNs.

Although the routing protocols that are used in WSNs have been under study for past few years, these protocols cannot be used for WBANs due to its stringent requirements. For WSNs, the main focus is on minimal routing overhead and maximal throughput than reduced energy consumption [4]. Also, WSNs are mostly homogenous networks, the WBANs are heterogeneous too [5].

In this article, a comprehensive review of the existing recent routing protocols/algorithms is discussed. “The required evaluation metrics for healthcare applications” section discusses the performance metrics that should be considered for WBANs. The classification of the existing routing protocols is explained in “Classification of routing protocols” and “Future challenges and comparative analysis of routing protocols” sections gives the future challenges and a comparative analysis of different selected protocols. Finally, “Conclusions” concludes the survey.

The required evaluation metrics for healthcare applications

In order to identify the important metrics that have to be considered in WBANs during the design process, a general overview about the routing challenges in WBANs should be studied. The certain routing issues and challenges include network topology, postural body movements, limited resources, quality of service metrics, radiation and interference, global network lifetime, heterogeneous environment, etc. By analyzing all these factors we can conclude and list the important performance metrics to be considered while implementing the whole WBAN. The following section defines the metrics:

  • Network lifetime defines the total operation time of the network until the last node is dead.

  • Path loss is the difference between the transmitted power at the source node and received power at sink node.

  • Stability period is the time before first node die.

  • Residual energy is the difference between initial energy and used energy during the operation of the network.

  • End-to-end delay is the average time taken by a data packet to reach the sink from the source node.

  • Packet delivery ratio is determined by number packets obtained at the sink divided by the number of packets send from the source.

Classification of routing protocols

The classification of routing protocols can be done in different categories that correlate with the routing challenges of WBAN. The following section gives an overview about the existing protocols, which can be categorized as Cluster-based, Cross-layered, Postural movement based, quality of services (QoS) aware and Temperature-aware routing algorithms [6] as shown in Fig. 2.

Fig. 2
figure 2

Classification of routing protocols

Full size image

Cluster-based routing protocols

In both WSNs and WBANs, the limited energy source is the main constraint to be analyzed. Hence, several efficient cluster based schemes are proposed for both networks to minimize the power consumption and maximize the network lifetime [7]. While comparing hybrid indirect transmission [8] to power-efficient gathering in sensor information systems [9] and low-energy adaptive clustering hierarchy (LEACH) [10], it consumes less amount of energy if the number of nodes are small. However, AnyBody [11] protocol is better than LEACH, as the numbers of clusters remain constant with an increase in the number of nodes, but LEACH does not. Also, the installation cost is also less with AnyBody. The Table 1 shows an overview of the existing cluster-based protocols.

Table 1 clustered routing protocols in body area networks
Full size table

Cross-layered routing protocols

These protocols use the concept of cross layering [13] which is already addressed in WSNs, where each layer (adjacent or non-adjacent) in the protocol stack shares their information unlike in the strict layered model. In WBANs, we can utilize the cross layering concept between network and medium access control (MAC) layers for routing and thereby can upgrade the overall network performance. Table 2 shows the different cross-layered routing protocols. Considering the detailed analysis of cross layered protocols, Cascading Information retrieval by controlling access with distributed slot assignment (CICADA) [14] and time zone coordinated sleep scheduling [15] have less power consumption. Wireless autonomous spanning tree protocol (WASP) [16] has better packet delivery ratio and CICADA has less delay.

Table 2 Cross layered protocols in body area networks
Full size table

Postural movement based routing protocols

The body postural movements affect the network topology of the network, which results in link disconnection. The researchers introduced a cost function that is periodically updated for choosing the best route to forward packets to the sink. The protocols listed in Table 3 are the existing postural movement based routing protocols. Among the discussed protocols, on-body store and flood routing (OBSFR) [18] has better performance in reducing the packet delivery delay, but opportunistic postural movement based routing protocol has lower energy consumption than others.

Table 3 Postural routing protocols in body area networks
Full size table

QoS aware routing protocols

Presently, there are a number of diverse QoS aware protocols available in WSNs, which cannot be as such implemented in WBANs, but by considering its unique curbs it can. In WBANs different data types require different QoS [24]. Hence the proposed protocols should be aware of different types of QoS metrics for various types of data. The various QoS aware routing protocols are shortlisted as given in Table 4.

Table 4 QoS aware routing protocols in body area networks
Full size table

The comparative analysis has shown that energy-aware peering routing protocol (EPR) [45], QoS-aware peering routing protocol for delay sensitive data (QPRD) [41] and QoS aware peering routing protocol for reliability sensitive data (QPRR) [38] have less power consumption when compared to other protocols. Some protocols do not consider energy consumption, while others. One of the most used QoS aware protocols is data-centric multi objective QoS-aware routing protocol (DMQoS) [50] because it can decrease the delay for delay-sensitive information, and similarly, it can provide reliable routing for reliable-sensitive information. The other QoS-aware protocols are used or selected for a particular network, according to the data type and its QoS requirements.

Temperature-aware routing protocols

The antenna radiation, its absorption and interference are the major challenges to be considered while designing a body sensor network, since the radiated fields cause a temperature rise of node’s electronic circuitry. The field of radiation also has a strong impact on the human body [57] that may damage the human tissues due to its continuous exposure. The goal of all temperature-aware protocols is to decrease the temperature rise of in-body sensor nodes by avoiding routing through hotspots. Table 5 discusses the existing temperature-aware protocols. From the comparison of different protocols, it is seen that LTRT [58] performs much better than others while TARA [59] performs worst. In terms of temperature rise, HPR [60] shows less temperature rise in comparison with others. Finally, the latest M2E2 protocol [61] has proved that it is the best one among all thermal aware protocols, suitable for heterogeneous, multimode, energy efficient body sensor networks.

Table 5 Temperature aware routing protocols in body area networks
Full size table

Future challenges and comparative analysis of routing protocols

The scope of this article is to open up new research areas in WBAN domain for routing protocol designs. Among all the routing protocols, the cluster based protocol HIT [8] aims at maximizing the network lifetime, but it does not consider the packet delivery ratio which is an important QoS metric. The second one, AnyBody [11] protocol considers the delivery ratio, but does not consider average delay, mobility and energy consumption. It leads to poor security measures. In order to optimize the performance of sensor networks for some specific applications, it is necessary to include the aforementioned metrics also in the design considerations.

The cross layered concept is attaining great significance and interest among researchers due to its flexibility and effectiveness in sensor networks. The future research work aims at improving the reliability of CICADA [14] which performs better in terms of energy efficiency and average delay when compared to other cross layered protocols. When IEEE 802.15.4 standard networks are used, it will be good if the TICOSS [15] protocol is redesigned for reducing the average delay which is not considered in the existing one. If the entire network performance is to be optimized, the choice will be the Biocomm protocols along with new techniques to reduce the node temperature. Hence, the scope for research in this area is very much wide enough to work with. The comparative analysis of postural movement protocols has shown that, none of them considered the thermal effects of nodes and QoS issues together. Therefore, the future protocols could be proposed in such a way that, it could achieve better QoS parameters along with techniques to reduce the node temperature rise and methods to counter security attacks. The survey on QoS aware protocols unveils various research areas for future work because of its importance. Every new protocol, which has been designed, is meant for addressing the limitation of the previous one. For example, in routing service framework [56] and reinforcement learning based routing with QoS support [55] the energy consumption is not considered, but it is taken into account in the remaining protocols.

In almost, all the existing QoS aware protocols, only the QoS metrics are examined, without concentrating on the human body movements and temperature rise of implanted devices. The proposed temperature aware protocols perform better by reducing the temperature rise due to radiation from antenna and other node circuitry. Along with the thermal issue and power consumption, it will be better if these protocols can also address the routing issues like shortest path as in QoS aware.

Table 6 summarizes the comparison between some of the routing protocols used in WBAN. From the analysis, it is seen that almost all the protocols have considered different QoS metrics for their performance analysis. Hence, choosing the protocol for a WBAN system depends on the particular application of the system; whether it should be energy efficient, good reliable one or it should reduce the temperature of the node circuitry. Table 7 lists the pros and cons and the application domain of each protocols used in body area networks. This table helps to choose a particular protocol based on the QoS requirements. For example, if the application of the proposed system is patient monitoring in hospitals, then the protocols like WASP [16] or TICOSS [15] can be selected due to its high packet delivery ratio and low average delay. If the sensors are implanted within the body, then the protocols like Co-LEEBA [33], TARA [59], or routing algorithm for network of homogeneous and Id-less biomedical sensor nodes (RAIN) [66] can be chosen. If the network is heterogeneous one, then M-ATTEMPT or M2E2 can be used.

Table 6 Comparative analysis of routing protocols
Full size table
Table 7 Pros and cons and the application domain of routing protocols
Full size table

Conclusions

Wireless body area network is a part of wireless sensor network, with a number of nodes deployed within and on the surface of human body to measure different biological parameters for a particular application. In this survey article, various existing routing protocols that are used in WBANs are categorized and briefly analyzed from the available articles between the years 2002–2016. It is seen that the routing protocol plays a vital role in the design process of every efficient, reliable, low cost wireless body sensor networks. Based on the structure and nature of networks, the routing protocols for WBANs are categorized as cluster-based, cross-layered, postural movement based QoS aware and temperature-aware protocols. It is observed that there is no strict classification of protocols is possible since most of them aims or results in achieving the challenges of sensor networks. It is also concluded that each protocol is application dependent, i.e., the protocols used for daily monitoring and the critical medical cases are different. The future directions for each group of protocols are also presented which helps the researchers to focus on their interested area. Also, a comparative study of different protocols has been examined so that an appropriate protocol can be selected according to the targeted application. This survey will benefit the researchers to study the existing routing protocols for WBANs in the field of healthcare systems.

The Future work includes the design and implementation of a body sensor prototype with a newly designed routing protocol, which will be highly energy efficient and reliable one for rehabilitation of old age people using a microcontroller based system with suitable sensors.