Keywords

1 Introduction

The fact is unanimous that continuous health monitoring is essential but these often become infeasible due to scarcity of doctors and hospitals. In extension, the remote location and poor people cannot access round-the-clock health facility directly. Some chronic diseases like cardiovascular diseases, diabetes etc., are gradually emerging as alarming factor in human mortality. Prediction of risks factors of these diseases and continuous monitoring are very much needed. In one of our published papers, a prediction of risk of cardiovascular diseases considering hereditary factors [1] has been proposed. Continuous monitoring of patient of this category is highly required. In the present day context, it is a big challenge to provide health care with limited financial and human resource. Wireless body area network is the answer to this problem. IEEE 802.15.6 is the wireless body area network standard. It is a standard for supporting vast range of data for human body area network. It also provides confidentiality, authentication, integrity, privacy protection and replay defense (Fig. 1).

Fig. 1
figure 1

Layered architecture of wireless body area network

Full size image

WBAN is also associated with wireless technologies like ZigBee, WSNs, Bluetooth, cellular networks [2].

Several hospitals are adopting these services. The wireless body area network is now a reality because of advancement of wireless sensor network which is a collection of spatially distributed autonomous devices to monitor different physical and environmental aspects. Body area network is a wireless network of wearable computing devices that can monitor different body responses of human being like EEG, ECG, blood pressure, pulse rate, etc. Routing algorithm plays a very important role in optimum response of BAN.

Along with the human body network, intra-body communication is also a challenging research domain using the redaction device. Red Tacton uses the human body surface as communication path [3]. Human body can be used to send data to several sensors with the help of this technology also.

Several routing algorithm are being proposed in this field. MANET [4] is an energy- efficient routing algorithm for human body area network where it has two phases, in initialization phase all nodes broadcast hello message containing neighbour information and distance to sink node and in the routing phase routes the data with fewer hop count. Dynamic Duty Cycle MAC Algorithm [5] proposed by Jinhyuk Kim et al. is a priority-based energy-efficient routing algorithm that guaranteed low latency. DSDV (Destination Sequence Distance Vector) [6], and reactive protocol DSR (Dynamic Source Routing) [7,8,9,10] are two proactive and reactive routing algorithm used in this domain. DSR is based on source routing and DSDV is based on distance vector routing. Okundu et al. has proposed an algorithm containing three processes as link establishment, wakeup service and alarm process. This algorithm is energy efficient and capable of avoiding collisions [11]. TDMA-Based MAC protocol for WBANs called Med MAC [12] proposed by N. F. Timmons et al. consists of two schemes for the power saving: Adaptive Guard Band Algorithm (AGBA) and with Drift Adjustment Factor (DAF). Low Duty Cycle MAC protocol for WBANs [13] can perform analog to digital conversions. Three bandwidth management schemes: Burst, Periodic and Adjust bandwidth is used by B-MAC [14] that reduces the used bandwidth. Time-out MAC (T-MAC) for WBASNs uses flexible duty cycles for increasing energy efficiency [15]. H_MAC is another very interesting algorithm that uses heart beats for synchronization [16]. Reservation-based dynamic TDMA (DTDMA) protocol [17] provides more dependability in terms of lower packet dropping rate and low energy consumption especially for an end device of WBAN. Wise MAC is another MAC algorithm that is scalable and adaptive to traffic load [18]. PACT is an algorithm that is suitable for low delay application [19]. Another dynamic clustering algorithm is LEACH which is a distributed approach [20]. FLAMA is another energy-efficient algorithm that is good for normal traffic [21]. HEED which uses TDMA clustering provides prolonged network lifetime [22].

2 Proposed Method

Through wireless body area networks vast amount of crucial information are transmitted over the networks through many number of nodes. Energy is a very crucial factor for this type of routing algorithm. Very crucial set of data are being transmitted through the nodes that are part of the network. It is a always a better approach not to transmit redundant information again and again, as sending same set of information again and again does not add any effect but and create congestions.

We proposed a routing algorithm for the body area network that is energy efficient. Our algorithm considered three major factors when searching for the path to destination, these are:

  1. i.

    No. of hops to the destination through the node (HC)

  2. ii.

    Energy level of the node and (E)

  3. iii.

    Total no. of packet traversed through the node on last t intervals noted as queue length.

Every node broadcasts their neighbour information along with energy level and queue size in every t time interval as control information along with the data. Every node transmits set of data in next interval only if there are drastically changes in data. This feature will help to reduce congestion over the network. The threshold value is selected depending on the type of information the nodes are sharing.

2.1 Control Information

  1. i.

    Hop count control information is being shared if there is any changes:

    $${\text{HC}}\left( {{\text{N}}_{\text{i}} ,{\text{T}}_{\text{i}} } \right) - {\text{HC}}({\text{N}}_{\text{i}} ,{\text{T}}_{{{\text{i}} + 1}} ) \ne 0$$
    (1)
  2. ii.

    Energy level of any node is being shared if there is any changes in the it

    $${\text{E}}\left( {{\text{N}}_{\text{i}} ,{\text{ T}}_{\text{i}} } \right) - {\text{E}}({\text{N}}_{\text{i}} ,{\text{T}}_{{{\text{i}} + 1}} ) \ne 0$$
    (2)
  3. iii.

    Queue length information is also being shared based on previous value:

    $${\text{QL}}\left( {{\text{N}}_{\text{i}} ,{\text{ T}}_{\text{i}} } \right) - {\text{QL}}({\text{N}}_{\text{i}} ,{\text{T}}_{{{\text{i}} + 1}} ) \ne 0$$
    (3)

2.2 Actual Data

Actual data is the most crucial part of the human body area network as based on that data the patients will be taken care.

In actual data field the nodes send human health information: heart beat, blood pressure, pulse rate, body temperature and ECG. All the set of information again need not be sent in every interval as this may increase congestion on the network. If there is any change from the previous interval then only the information is sent again. The nodes for sending actual data are selected sequentially one after another in every time interval, one consideration we have followed is that if any node has the same information as previous interval is will forward to the next node.

$${\text{Select}}\left( {{\text{N}}_{\text{i}} ,{\text{T}}_{\text{i}} } \right) = \{ {\text{N}}_{\text{i}} ,{\text{T}}_{\text{i}} \left| {} \right.{\text{Data}}\left( {{\text{N}}_{\text{i}} ,{\text{T}}_{\text{i}} } \right) \ne {\text{Data}}\left( {{\text{N}}_{\text{i}} ,{\text{T}}_{{{\text{i}} + 1}} } \right)$$
(4)

2.3 Selection of Node for Sending Information Through

We formulate an equation to select the node for transferring data through by considering the weighting factors of the three conditions. We assign weight 0.4, 0.4 and 0.2 for no. of hops, energy level and queue length respectively. This is to be observed that the first condition that is no. of hops to destination through the node is inversely proportional for the selection of node, the second condition that is energy level is also proportional to the selection of the node but the last condition that is queue length is inversely proportional to the selection process as is large numbers of packet traversed through a particular node there is probability of congestion.

Perform the following calculation for choosing nodes that will send data:

$${\text{W1}}*\left( { 1/{\text{HC}}} \right) + {\text{W2}}\,*\,{\text{E}} + {\text{W3}}\,*\, \left( { 1/{\text{QL}}} \right)$$
(5)

W1, W2 and W3 are the weight assigned for all above-defined three distinct features those are being considered in our propose method according to their influence in node selection.

And we choose the weight as follows:

$$\begin{aligned} & {\text{W1}} = 0. 4\\ & {\text{W2}} = 0. 4\\ & {\text{W3}} = 0. 2\\ \end{aligned}$$
  1. 1.1

    Algorithm:

    Our proposed algorithm has following three subparts.

    1. 1.1.1

      Sharing Control Information

      IF (HC(Ni, Ti)-HC(Ni,Ti+1) ≠ 0 ) {Share Hop Count information with neighbours} IF (E(Ni, Ti) - E(Ni,Ti+1) ≠ 0 ) {Share Energy Level information with neighbours} IF (QL (Ni, Ti) QL(Ni,Ti+1) ≠ 0 ) {Share Queue Length information with neighbours}

    2. 1.1.2

      Selection of Nodes Through Which Data to be Transmitted Depending Upon the Control Information

      FOR (i = 0 to N) IF (MAX{(Ni,, Ti), W1*(1/HC) + W2*E + W3*(1/QL)}) {Select Node Ni } ELSE {Continue}

    3. 1.1.3

      Selection of Node Sending Actual Data

      FOR (i = 0 to N) IF(Data(Ni,Ti) ≠ Data(Ni,Ti+1) Send Data Break} ELSE Continue}

3 Test Result

A sample result of selection of nodes depending on the control information is presented in the following Fig. 2.

Fig. 2
figure 2

Test result1

Full size image

The comparison between three factors, viz. energy level, hop count and queue length and thereby estimated value has been presented in the following Fig. 3.

Fig. 3
figure 3

Comparative selection of nodes depending on three considered values

Full size image

4 Conclusion

Body area network is an emerging area of research. This has a great contribution to the field of healthcare application and that is possible by the wireless sensors network. As body area network is being used in the remote health monitoring, sending proper data at appropriate time is main focus. And at the same time the life time of the nodes and sensors are also very important for the communication. In the paper, we proposed a method of routing that considers the hop count, energy level and congestion of the nodes also.

So, we can conclude the following from our proposed method:

Node Selection Function:

$$\partial\infty \, 1/{\text{Hope}}\;{\text{Count}}$$
(6)
$$\partial\infty \, {\text{Measure}}\;{\text{Energy}}$$
(7)
$$\partial\infty \, 1/{\text{Queue}}\;{\text{Length}}$$
(8)