VADiRSYRem: VANET-Based Diagnosis and Response System for Remote Locality

Abstract

The advancement in telemedicine has been helpful in the development of smart healthcare in urban and rural areas. In India, however, this solution often falls short of connecting the remote villages with modern healthcare due to the difficulty in accessibility. In a developing country like India, consisting of geographically remote settlements, it is not feasible to provide infrastructure-based health support to all. To conquer this difficulty, an alternative solution is required in which the existing network infrastructure is not essential for the communication purposes. The main objective of this proposal is to provide an alternative approach for the transmission of health care related messages. It is also important to minimize the cost of this type of healthcare services by avoiding the establishment of communication network across the country. Here, we propose the framework of an IoT-based system to work as an aid for quicker diagnosis and health support for the remote villagers. We have used vehicular ad hoc network for cost-effective and fast data transportation, besides other advantages.

Introduction

The remote healthcare system has experienced an incredible development in the ongoing era of IoT. This can be implemented using the already available resources of patient such as smartphone cameras and wearable biosensors. It reduces travel expenses, medical cost, time required for interaction, and provides easier access for the common man to specialist doctors without disrupting their daily responsibilities.

The above-discussed scenario is not always true in a developing country like India. India has a large population (approx. 121 crores [1]) in which 68.84% live in a rural area [1]. Another alarming fact is the current available doctor population in India is only 0.62:1000 [2] where the recommended doctor population ratio by WHO is 1:1000. In reality, very small percentage of population (only 30%) has access to the resources necessary for this type of healthcare service [2]. This is an indication of the fact that the recent advancements have been useful in providing quick medical help in urban and ‘not-so remote’ rural areas which has the basic infrastructural support. Unfortunately, the situation is not so promising in the case of remotest of the remote places in India. This is due to the challenges in implementing infrastructures necessary to telemedicine. The setting up of the telecommunication infrastructures is often too expensive making this almost infeasible due to the remoteness of the villages. In developing countries like ours, reaching the remotest of the remote corners with medical help remains a challenging task. It becomes impossible to manage any critical situations like epidemic break down. In most of the cases, proper road connectivity with nearest highway is not present. In case of any medical emergency, the two most important challenges are (a) establishment of route for transmission of data packets to the health center and (b) securing the health data and response packets as the carrier is unknown third party.

The main motivation behind this research proposal is to give a technically and financially feasible solution for providing medical help in the quickest possible way to the remote villagers.

With a powerful on-board unit, electronic control units and sensor, vehicles can be treated as a mobile resource for many services such as sensing, data relaying and storage, computing, cloud, infotainment and localization. The concept of Vehicle as a resource has focused on vehicular potential. For vehicular fatality reduction and for enhancement of ITS application and services, today’s vehicles are equipped with components which categorize them as ‘intelligent vehicles’. Sensors and actuators are deployed in these types of vehicles for intra-vehicle communication. These vehicles are included with electronic control units (ECUs) for processing and operation control. Using VANET, three types of communications are possible, such as Vehicle-to-vehicle (V2V), Vehicle-to-infrastructure (V2I) and Vehicle-to neighboring object (V2X). The ease of message transmission with very little delay involved makes VANET a good candidate for message transmission across the highways up to the health centers. The requirement for infrastructural support is also minimal.

In this paper, we have proposed a multilayer framework to establish a communication link between remote villages and health center. For this purpose, we have built a heterogeneous network comprising wireless sensor nodes and smart vehicles. An efficient framework for data transfer in this IoT environment is described in detail.

The rest of the paper is organized as follows. In the next section, we have presented a detail study of commonly used methodologies for mitigating the above two requirements. A novel framework is proposed in the third section for establishing the IoT-based communication system. For efficient communication, the framework is distributed in different layers. The fourth section presents a case study for proper explanation of the proposed framework. Simulation-based performance analysis is done in the fifth section before concluding with final remarks.

Related Works

In today’s world, remote healthcare is an obvious solution to manage the enormous requirement and researchers have proposed several solutions in this domain. In this section, we provide a brief analysis of existing works. Two types of approaches are popular in case of remote healthcare as: Wireless Body Area Networks and Telemedicine-based solutions.

In BSN_CARE [3], the authors have presented a BSN architecture which consists of different wearable and implantable sensors. Sensor nodes are integrated with biosensors such as electrocardiogram (ECG), electromyography (EMG), electroencephalography (EEG), and blood pressure (BP). The information collected by these sensors are forwarded to the Local Processing Unit (LPU). LPU can be any portable devices such as PDA and smart phone. It communicates with BSN_CARE server using any communication medium. Whenever any abnormalities are detected by LPU, it sends an alert to the person wearing biosensors. After receiving data from LPU, BSN_CARE stores the data in its database for further analysis. Depending on the degree of abnormality, it takes actions as: interact with family members, local physician, and emergency unit of a nearby healthcare system.

In [4], the authors proposed a remote-monitoring system to monitor children activity using a wearable vest. The focus is on the different chronical diseases of children. Sensors are used for data acquisition. In [4], the authors have used accelerometer, pulse oximeter and a temperature sensor. For data transmission purpose, they have used XBee transmitter module which is embedded in the garment. XBee receiver receives these transmitted data. This module relates to Raspberry Pi module which is a credit card-sized programmable computer using Raspbian software. The classification of the received data is done using this module. Feature extraction, dictionary generation and classification are the three main steps performed by this module. All the classified levels (oxygen saturation level, heartbeat and body temperature) transferred and the corresponding records are stored in the cloud server. Using a customized mobile application, these data are accessed by the parents or caregivers.

An investigation is done on telecardiology system in [5] by the authors. In this proposal, ECG devices are used for monitoring remote patients and the collected information are transmitted through the wireless medium. There is a diagnosis unit which provides the prescriptions to the concerned patient. Once any abnormality is identified, the patient lifesaving activities have started.

Another IoT-based approach is proposed in [6]. Taking consideration of the power consumption, the authors have used flexible solar energy harvester with MPPT as the power source. For data collection, pulse sensor and temperature sensor have been used. BLE module (HM-10) is used to transmit the sensed data to a smartphone which is act as an IoT gateway for sensed data visualization and emergency notification.

Due to the advancement of microelectromechanical devices, several small-sized sensor nodes can be used to observe patients’ biomedical parameters, fall detection and patient location [7,8,9]. Estimate of patient position in indoor location is a challenging task. Various noise and fading effects cause difficulties in distance estimation accuracy. These result to localization error [10].

Telemonitoring is another popular approach, using for remote patient monitoring. In this paper, we have analyzed several existing telemonitoring-based solutions. In [11], a synthesis of telemonitoring has been presented by Agency for Healthcare Research and Quality (AHRQ). A new review has been presented based on chronical diseases [12] where total 2611 patients were involved. [13] involved a trial with 3230 patients from three regions of England and it follows them for 12 months. A comparison is made between two groups of patients with and without telemonitoring and identifies the needs of emergency services required for remote patients. Telemonitoring trials were also performed in France and demonstrate good acceptability on the part of patients [14, 15].

It is observed that most of the approaches are concentrating on a single type of disease. In [16,17,18,19,20,21,22,23,24], the safety and efficacy of telemonitoring of rhythmic prostheses are demonstrated. American Heart Association/American College of Cardiology [AHA/ACC], European Society of Cardiology [ESC] and French Society of Cardiology have contributed in [25] telemonitoring of prostheses for arrhythmia.

Wearable sensors have been the main entity in the WBAN-based approaches as they collect vital information from human body for further analysis. The sensors attached to the body can cause inconvenience with regards to mobility and daily activity of the patients. Patients are aware about these devices that may influence the nature of collected sensitive physiological information. This necessitates the need for collecting physiological data as non-invasively as possible. This results in contactless data collection methods.

Another noticeable aspect is that specific set of sensors are required for every patient. A specific sensor measures specific health parameter such as temperature sensor sensed only body temperature and pulse sensor obtained only pulse rate. For developing countries, it is not financially feasible to arrange multiple sensors for one remote patient which leads to a huge cost.

In this paper, we have analyzed telemedicine-based approaches and following observations have identified. It is noted that telemedicine is involved for most of the applications. One telemonitoring system is built for aiding in detection of a specific disease, such as heart disease and kidney problem. As telemedicine is not the generic solutions, they fail to provide responses for other types of diseases.

Network connectivity is another key constraint for the existing proposals. Internet connection with uninterrupted electricity throughout the day is a mandatory prerequisite for this type of system which leads to a huge infrastructural cost. In developing countries, telemonitoring system has problem to sustain their activities for a long period due to the lack of regular funding sources.

Based on the above observations, we have identified a proposal to establish a data transportation system with minimal infrastructural cost. With the help of the proposed network, the medical request will be transported to the nearest health center and the required solution will return in similar manner.

Most of the proposed systems are atomic in nature and have planned to perform the whole work as a one. It generates an excessive computational complexity and adds more latency to the system. This results in increased delay in execution, i.e., the system will take more time for diagnosing the ailment, thus leading to increased response time. Based on the above discussion the required solution would consist of the following criterion:

  • The health information should be collected in non-invasive manner.

  • Will not need pre-established physical infrastructure.

  • The solution should be generic, i.e., not disease specific.

  • The delay should be minimized as far as possible.

  • The solution should be scalable, thus, allowing more villages to be connected without a change in the basic communication network.

For achieving above-mentioned objectives, in the next section, we have proposed a novel VANET-Based Diagnosis and Response System for Remote locality. In this solution, a multilayered framework is proposed for reducing complexity of total system as the functions of each layers and communications between them are well defined. With help of its modular nature, it provides flexibility to add new services. Due to segmentation, it is possible to break complex problems into smaller and more manageable sub-problems. This reduces the response time and makes the solution more scalable.

Description of Proposed Technique

In this section, we present a novel IoT framework comprising of VANET-Based Diagnosis and Response System for Remote locality (VADiRSYRem). Before delving into the technical details, initially we give an overview of the proposed layers used for data traversal between remote villages and health center. Figure 1 identifies a brief introduction of the six layers and their functionalities. The functionality is a set of actions that a layer offers to another (higher) layer.

Fig. 1
figure1

The layers and their functionalities

Registration

In this layer, we have concentrated on collecting basic health-related information from the villagers. This module is used for registration of individual villager into the system. A unique identifier is generated for every registered villager.

RAP (Rural Access Point) is used specifically for this purpose. RAP belongs to dumb network with specific functionality. One RAP node is assigned to each village which act as the gateway. These nodes are used for collecting and forwarding packets. They do not have any storage facility.

Initially, a registration request message (Reg_Req) containing the Aadhar number of each villager is collected by RAP. System generates a unique U_Id for every villager, where \({\left(U\_Id(Aadhar no)\mathrm{ x }(Z\right)}^{+}\), where Z+ is a set of positive numbers. Each villager registers with BHI[i,j[]], where i = 1 and j = 6. Each i represent U_Id of the villager and j = {height, weight, blood_group, blood_pressure, blood_sugar}.

This is forwarded to the RSU (Roadside units) those are located beside highway. One RSU can connect with multiple RAPs through multiple dumb networks. RSU receives Init_Req from different RAPs, maintain repository of medical requests, calculate priority of each requirement, build final request message and broadcast the messages according to priority.

The information remains stored in RSU until it is able to broadcast to a passing vehicle. This message is then sent to the nearest health center using VANET. The received information is used to update the Basic_Repository {U_Id, V_Srl, height, weight, blood_group, blood_pressure, blood_sugar}.

figurea

Dumb Networks

In this layer, we mostly focused on the communication network formation between RAP and RSU.

This layer of architecture is represented by a network which is a collection of nodes those can receive and transmit network messages. The inexpensive ZigBee can provide low-power, short-range wireless communication. For this reason, it can be chosen as an efficient and feasible option for this scenario. After studying the characteristics of the different topologies supported by ZigBee framework, we have found mess topology as most suitable for formation of our proposed dumb network. The primary responsibility of this layer is to gather information regarding health problems in remote villages. This layer is also responsible for the transformation and transmission of perceived information to the next layer.

Two types of ZigBee nodes are available in the market as; full-function device (FFD) and reduce function device (RFD). With respect to these functionalities, ZigBee devices are classified as Coordinator, Router and End Devices.

For receiving and forwarding health-related information, we require a ZigBee gateway node which can receive and transmit information. They are RFD and do not require any storage facility. In our proposed framework, we denoted this as RAP. For every remote village, we have a dedicated RAP which is an end user node of the proposed dumb network. In Fig. 2, we have shown five RAPs and each of them is connected with different villages. After receiving health-related information from village side, RAP transmits it to coordinator node. This is a FFD and there can be only one coordinator node in the dumb network. The coordinator is responsible for coordinating the movement of packets between the basic nodes and the higher layer. It forms Init_Req {Req_ID, V_Srl, E_Desc, Emg_L, ToV, TTL} message and forwarded the same through router nodes which is another FFD. This information is forwarded towards the highway. In this point, we have Highway Access Point (RSU).

Fig. 2
figure2

Dumb network formation

Dumb network is used for transmitting Init_Req message to its nearest RSU. Though all RSUs are static in nature, we have used Distance Vector Routing algorithm for finding shortest path between gateway node and RSU.

All RSUs are connected with multiple dumb networks corresponding to different remote villages. The corresponding gateway nodes communicate with RSU. A unique sequence number is assigned to every gateway node for representing the network. For a RSU connected with n numbers of villages, this sequence number can be represented by N[i], where i = 0, 1, 2, 3,n. Inside the network, each sensor nodes also have a unique identification number. For a network with m number of sensor nodes can be represented as, S[j], where j = 0, 1, 2, 3,…m.

The nodes present in dumb network, RAPs, RSUs are all static in nature. All these entities are the internal member of the system and assumed as trusted components. Though the network and as well as the sender and receiver all are trustworthy, no additional security measure is need not to be taken for Init_Req message.

Transformation Layer

The proposed system consists of highly mobile vehicle nodes and static sensor nodes. Though VANETs and WSNs have common features, such as self-deployment organization of nodes, they also have some crucial dissimilarities. For this purpose, we required a mapping between this ZigBee-enabled WSN nodes and VANETs node. The main issue to be addressed in this layer is to establish a mechanism for mapping between IEEE 802.15.4 and IEEE 802.11.p messages.

We have already incorporated gateway node in our proposed dumb network. This node is responsible for connecting the dumb network with other networks. As previously discussed, our dumb network is built using ZigBee-enabled wireless nodes those use IEEE 802.15.4 for communication purpose. This node communicates with the RSU. The hybrid WSN-VANET communication architecture assumes that RSUs are fitted with two wireless network interfaces, namely IEEE 802.11p and IEEE 802.15.4. The WSN nodes are communicating with RSUs via gateway nodes using IEEE 802.15.4 interface. Data from WSN are forwarded into the VANET by RSU. These RSUs are playing a major role in this architecture as it acts as a gateway between the WSN and the VANET.

The Link Aggregation Group (LAG) [26] methodology is used for receiving and forwarding network packets from multiple sources. It is a mechanism for connecting two switches with multiple links or establish multiple connectivity links between a switch and a server. Thus, a virtual link is created from multiple links which enables the transfer and receipt of data using a higher bandwidth. This is not only an inexpensive way of bandwidth upgradation whereas hardware upgradation is not required, it also allows incremental growth of interconnect bandwidth between two links. In our proposed system, we are going to follow DOCSIS 3.1[26] for link aggregation purpose.

For RSU, we have proposed Zig-Fi system which is built on the top of ZigBee and Wi-Fi. In this way, it becomes transparent and independent of these standards. The proposed architecture is shown in Fig. 3.

Fig. 3
figure3

Framework for transformation layer

The packet receiver component receives data packets from both interfaces using DOCSIS 3.1. The forwarded messages from dumb network are received through ZigBee interface and stored in a buffering queue named WSN Packet buffer. The message received from the RSU (i.e., those came from Wi-Fi interface) are stored in a separate buffer maintained for VANET packets. Separate data packet extractor is used for decapsulation of VANET and WSN packets. After decapsulation, the extracted data are going to store in data packets buffer. RSU received request messages (Init_Req message) through ZigBee interface in IEEE 802.15.4 format. It decapsulate it and store in data packets buffer. Using information of this message Fin_Req {Req_ID, RSU_ID, V_Srl, E_Desc, Emg_L, ToV, TTL} message is constructed in IEEE 802.11p format and stored in VANET packet buffer. For every request, packet received RSU maintains a Request_Repository {Req_ID, V_Srl}.

Processing Layer

At the end of the transformation layer, Fin_Req messages are ready to communicate through VANET. Single RSU can relate to multiple villages through WSN. From those villages, multiple requests can be generated at the same time. For this reason, every message should follow an order for transmission. This order is going to be decided by this layer. For assigning order of communication to every Fin_Req, following issues are to be solved.

  1. (A)

    Processing.

  2. (B)

    Prioritization.

  3. (C))

    Scheduling.

The system model for this layer is depicted by Fig. 4.

Fig. 4
figure4

The system model of processing layer

A. Processing

Processing is the first module in this layer. The Fin_Req messages are taken as input in this module. This request message basically contains health-related requirements. Depending on the type of medical emergency, we have categorized these messages and an emergency level is assigned to them. For an example, medical requirement generated due to animal bite differ from a medical emergency generated due to an epidemic. Both are different in nature and require different treatment.

Depending on the degree of seriousness, we have assigned four emergency levels (E) for guarantying prompt response, and it is represented as follows:

  • High: represented by 11, i.e., extreme emergency.

  • Medium: represented by 01, i.e., medium emergency.

  • Low: represented by 10, i.e., low emergency.

  • No emergency: represented by 00, i.e., there is no emergency at all.

B. Prioritization

Prioritization is the activity that arranges items or activities in order of importance relative to each other. In the previous module, we have already categorized the requests based on medical emergency. In this module, we are going to measure priority of requests on basis of the following metrics.

  1. 1.

    Emergency level: The possible values of this metrics are defined in the previous module only.

  2. 2.

    Time spend after request generation: TTL is associated with every message. From this, the amount of time already spent after request generation can be easily calculated.

  3. 3.

    Number of affected persons: The ToV parameter is useful in giving the total number of patients.

Depending on the above parameter, the priority (ρ) value for each message is calculated as follows:

$$\rho = \left( {\left( {w1*E} \right)*\left( {w2*ToV} \right)} \right) + \left( {{\text{seconds}}\_{\text{between}}\left( {SYSDATE - TTL} \right)} \right),$$
(1)

where w1 and w2 are weight factors. We have assigned w1 = 0.75 and w2 = 0.25. These values are decided depending on the fact that serious patient requires quicker response in comparison to patients suffering from less serious diseases. For calculation purpose of Emergency level (E), we are going to use decimal values of the corresponding field.

C. Scheduling

Depending on the computed ρ, the Fin_Req messages are scheduled. We are using preemptive priority scheduling algorithm for this purpose. The highest priority message is sent first. According to Eq. 1, difference between current time and TTL is a parameter for priority calculation. A low-priority Fin_Req becomes high priority with time because the difference between current time and TTL increases with time. It ensures this proposal is starvation free.

Smart Network Layer

This layer is responsible for network establishment-related issues between RSU and smart vehicles. In our proposed protocol, smart vehicles play a very important role. These vehicles are equipped with OBUs and are enabled to communicate with each other through IEEE 802.11p protocol. In the previous modules, we have already discussed about receipt of Init_Req message by RSU. RSU computes the final request message from Init_Req message as the Fin_Req. RSU starts to broadcast this message according to its ρ using VANETs.

Vehicular ad hoc networks are characterized with high mobile nodes (vehicles) and allows inter-vehicle communications (IVC) along with the roadside unit to vehicle communications (RVC). Each vehicle contains a wireless on-board computing unit (OBU). Vehicles can correspond with nearby vehicles known as a vehicle-to-vehicle (V2V) communication and with roadside infrastructure which is known as vehicle-to-Infrastructure (V2I). IVC and RVC can be performed for safety-related and information-related application.

After receiving one frame of Fin_Req message, vehicle starts to broadcast it to the neighboring vehicles, RSUs and nearest health center which falls within its transmission range. Each packet reaches their destination through forwarding of Fin_Req message to the nearest health center by a smart vehicle. After receiving every frames of Fin_Req message, the health center sends an Init_resp {Resp_ID, Req_ID, RSU_ID, H_C_I} to the corresponding RSU using VANET. After receiving the response message, RSU fetches the Req_ID from it and search it in its Request_Repository. After the match is found, it takes the corresponding V_Srl and forwards the message through the WSN.

Application Layer

In the above layers, we have discussed about different aspects of the proposed mechanism. In this layer, our aim is to integrate these different aspects as a whole and discuss about deployment issues of the above-discussed healthcare system in remote villages. The framework of this layer is represented by Fig. 5.

Fig. 5
figure5

Framework for application layer: the integrated system

Case Study

In Table 1 we have explained the symbols that we have used.

Table 1 Data dictionary

Assume, V01,V02,…,V0n are set of remote villages connected with RSU0 through wireless sensor networks. A medical emergency arises at V01. S0 is a gateway node of that WSN and it receives the request message from V01. Suppose S0 sense a request message Req0 (Table 2) sent by V01. Initially, the message will consist of the following information.

Table 2 Init_Req sent by S0

After receiving the message, S0 forwards it to RSU0 using Distance Vector Routing algorithm [27]. It extracts information from the message and stores it into data packets buffer. It will maintain the following information for this request (Table 3).

Table 3 Entry in Request_Repository

An analyzer works on data packet buffer. Depending on emergency_level (E), TTL and ToV (associated with the message) a priority (ρ) is assigned to Req0. ρ of Req0 is estimated as:

ρReq0 = (0.75*3)*(0.25*50) + 180 = 208.

The messages are sorted based on the priority. The RSU is going to forward the highest priority message to the next passing vehicle. The message is then modified as follows (Table 4).

Table 4 Fin_Req broadcast by RSU0

RSU0 starts to broadcast the modified Req0. A smart vehicle SV0, passing by the RSU, receives the Req0. SV0 then starts to broadcast Req0. Other nearby smart vehicles also receive Req0 and start to broadcast. In this way, Req0 is routed towards the nearest health center. Once the message is received by the health center, a response message will be sent for immediate action. In between, the basic repository at the health center is updated as well. The response message consists of the following information (Table 5).

Table 5 Init_resp sent by H_C_ID

Performance Analysis

The proposed scheme is implemented using NS2 simulator. We run the simulations in an area of 10000 m × 6000 m for 12 h, where road intersections are located at every 5 km. Other simulation parameters are summarized in Table 6.

Table 6 Simulation parameter

We have chosen performance metrics with respect to our basic objectives. Our aim is to propose a healthcare system for remote areas. Measurement of QoS in healthcare system can be done by achieving a cumulative measure of reliability, timeliness, robustness, availability and security. For measuring the above-mentioned performance indices, the followings are particularly significant.

Throughput: Throughput can be measured by the effective amount of data transmitted in a specific unit of time. This can be defined as:

$${\text{Throughout}} = \mathop \sum \limits_{i = 0}^{n} P_{{\text{s}}} L_{{\text{p}}} ,$$
(2)

where Ps is the total number of messages successfully received at the destination, Lp is the length (in bits) of the payload for each node and i is the transmission time.

We have compared throughput of the proposed algorithm with RCARE [23]. The comparison is shown in the above figure. Figure 6 is showing that the throughput of VADiRSYRem is little better than RCARE. In VADiRSYRem, we assume that every vehicle passing from the highway has taken part in message communication. Due to availability of a large numbers of communication resources, Ps can be maximizes which results to a higher throughout.

Fig. 6
figure6

Throughput vs. simulation time

End-to-end delay This is defined as the average time taken by a data packet to arrive in the destination. It is calculated based on the delay, caused by route discovery process (Processing delay), the queue in data packet transmission (Queuing delay), the delay required to push all the bits in a packet on the transmission medium in use (Transmission Delay) and the delay required for the bit to propagate to the end of its physical trajectory (Propagation delay). End-to-end delay is calculated as:

$${\text{EE}}_{{{\text{delay}}}} = {\mathop \sum \limits_{Trans\_no = 0}^{x} \left[ {\left( {\mathop \sum \limits_{c = 0}^{m} \left( {h\_size + \frac{bit\_app}{{Pack\_No}}} \right){\text{/cap + }}\frac{p}{Pack\_arr} + \frac{Pack\_s}{{bit\_app}} + \frac{dist}{{Prop\_vel}}} \right)/Trans\_no} \right]}$$
(3)

We have made a comparison between VADiRSYRem and RCARE in respect of End-to-end delay. The comparison is shown in Fig. 7. The figure is showing that the Delay of VADiRSYRem is less than RCARE [23]. In VADiRSYRem approach, the request message is broadcasted periodically to every passing away vehicle. No route is there to follow. Message can traverse in any direction. This results to very less (tends to zero) amount of route discovery delay. Due to the layered architecture, all the required preprocessing activities (such as message formation and preparation of queues) already completed much prior of the actual communication taken placed. This leads to minimized end-to-end delay and ensures the timeliness property of the system.

Fig. 7
figure7

End-to-end delay vs. no of packet delivered

Figure 8 depicts the average end-to-end delay with 12 h of simulation having variation in number of nodes (N). We can observe that by increasing N, we can reduce average delay. The average delay varies between 45 and 90 min having N = 100 and it varies from 55 to 100 min having N = 50. This is because with more numbers of vehicles (nodes) involved in the system the communication can be faster. Vehicles are the medium of communication in the proposed system. The more vehicles we have, the more communication takes place. It results to reduction of average end-to-end delay.

Fig. 8
figure8

Average end-to-end delay

Lemma 1

Communication delay gets minimized with increased number of vehicles beyond Thrv (threshold count for vehicles within an area).

Proof Figure 8 shows more numbers of vehicles are available for communication results to more numbers of communications initiated. It leads to the minimization of end-to-end delay in message passing.

Packet delivery ratio is the percentage of data packets delivered from source to destination. It can be used to represent the probability of packets being received. A packet may be lost due to, e.g., congestion, bit error, or bad connectivity.

Figure 9 shows the impact of number of vehicles on the packet delivery ratio. According to the above figure, we can observe, by increasing vehicles, we can achieve a better packet delivery ratio. The more numbers of vehicles involved in communication obviously increase the probability of more numbers of packets to be delivered. If the number of vehicles for message communication would increase, our algorithm provides a better result (Fig. 10).

Fig. 9
figure9

Packet delivery ratio

Fig. 10
figure10

Number of packet delivered vs. reliability

Lemma 2

As the number of packets to be transmitted is increased beyond Thr (Threshold value), reliability of the system increases.

Proof

This is attributed to the fact that packet delivery rate increases as the number of vehicles within the VANET is increased.

Case Study

Table 7 is depicting two scenarios. Five vehicles are involved in scenario 1 with reliability value of 0.892. Four vehicles are involved in scenario 2 with 0.78. This situation explains Lemma 2.

Table 7 Case study for Scenario 1 and Scenario 2

Reliability (R(n)) of the system is computed by measuring the number of packets sensed (S(n)), number of packets transmitted (T(n)) and number of packets generated (G(n)) by the system. The relation among them can be formulated using the following equation:

$$T\left( n \right) = \mathop \sum \limits_{v = 0}^{m} \frac{T\left( n \right)}{{S\left( n \right) + G\left( n \right)}},$$
(4)

where v is the number of nodes involved in the system.

We have made a comparison between VADiRSYRem and RCARE in respect of Reliability. The comparison has shown that VADiRSYRem has higher reliability than RCARE. In VADiRSYRem, message generation is done by only RAP and RSU. All other intermediates nodes only sense the messages received by them and without performing any modification transmit it. This results in maximizing T(n) and minimize G(n). It leads to increase in reliability with respect to number of packets delivered. Multiple communication paths are available between RAP to RSU and RSU to health center. System availability will not be affected due to failure of any communication link. It makes the system more available and robust. At any point of time, new villagers can be added in the proposed system. This increases the scalability of the proposed system.

Conclusion

Remote healthcare-monitoring system is a very emerging research domain nowadays. In this paper, we have proposed the framework of VADiRSYRem, an IoT environment comprising WSN and VANET to provide remote healthcare system in developing countries like India. With the help of the proposed solution, we can establish a virtual communication network throughout a large area with very less amount of infrastructural cost. We have taken help of VANET technology for communication purpose. The performance analysis though simulation has shown that this proposal gives higher throughput, packet delivery ratio, reliability and lesser end-to-end delay than RCARE system. The performance analysis also shows that if greater number of vehicles participated in the system, improved efficiency is expected.

References

  1. 1.

    2011. https://www.censusindia.gov.in/2011-prov-results/prov_results_paper1_india.html.

  2. 2.

    2017. https://indianexpress.com/article/india/less-than-one-doctor-for-1000-population-in-india-government-tells-lok-sabha-4760892/.

  3. 3.

    Gope P, Hwang T. BSN-care: a secure IoT-based modern healthcare system using body sensor network. IEEE Sens J. 2016;16:5. https://doi.org/10.1109/JSEN.2015.2502401.

    Article  Google Scholar 

  4. 4.

    Jansi R, Amutha R, Radha S. Remote monitoring of children with chronic illness using wearable vest, Telemedicine Technologies. In: book: Telemedicine Technologies, pp 121–137. 2019. https://doi.org/10.1016/B978-0-12-816948-3.00008-8.

  5. 5.

    Gopu G, Anitha T, Nagarajapandian M, Prabhakaran M. Investigation of telecardiology system to detect cardiac abnormalities. Telemed Technol. 2019;2019:43–55.

    Article  Google Scholar 

  6. 6.

    Wu T, Redouté J, Yuce M. An autonomous wireless body area network implementation towards IoT connected healthcare applications. IEEE Access. 2017;2017:5. https://doi.org/10.1109/ACCESS.2017.2716344.

    Article  Google Scholar 

  7. 7.

    Naranjo PGV, Shojafar M, Mostafaei H, Pooranian Z, Baccarelli E. P-SEP: a prolong stable election routing algorithm for energy-limited heterogeneous fog-supported wireless sensor networks. J Supercomput. 2017;73:733–55.

    Article  Google Scholar 

  8. 8.

    Saleh N, Kassem A, Haidar AM. Energy-efficient architecture for wireless sensor networks in healthcare applications. IEEE Access. 2018;6:6478–86.

    Article  Google Scholar 

  9. 9.

    Zhang H, Zhang H, Pirbhulal S, Wu W, Albuquerque VHCD. Active balancing mechanism for imbalanced medical data in deep learning-based classification models. ACM Trans Multimedia Comput Commun Appl. 2020;16:39–54.

    Google Scholar 

  10. 10.

    Pirbhulal S, Samuel OW, Wu W, Sangaiah AK, Li G. A joint resource-aware and medical data security framework for wearable healthcare systems. Future Gener Comput Syst. 2019;95:382–91.

    Article  Google Scholar 

  11. 11.

    Am T, Dm W, Kb E, Ms M, Jc GGS. Telehealth: mapping the evidence for patient outcomes from systematic reviews. In: Technical Brief No. 26, AHRQ Publication No. 16-EHC034-EF, Rockville, MD: Agency for Healthcare Research and Quality; 2016.

  12. 12.

    McFarland S, Coufopolous A, Lycett D. The effect of telehealth versus usual care for home-care patients with long-term conditions: a systematic review, meta-analysis and qualitative synthesis. J Telemed Telecare. 2019. https://doi.org/10.1177/1357633X19862956.

    Article  Google Scholar 

  13. 13.

    Steventon A, Bardsley M, Billings J, Dixon J, Doll H, Hirani S. Effect of telehealth on use of secondary care and mortality: findings from the Whole System Demonstrator cluster randomized trial. BMJ. 2012;2012:344. https://doi.org/10.1136/bmj.e3874.

    Article  Google Scholar 

  14. 14.

    Versteeg H, Timmermans I, Widdershoven J, Kimman GJ, Prevot S, Rauwolf T. Effect of remote monitoring on patient- reported outcomes in european heart failure patients with an implantable cardioverter-defibrillator: primary results of the REMOTE-CIED randomized trial. Eurospace. 2019;21(9):1360–8. https://doi.org/10.1093/europace/euz140.

    Article  Google Scholar 

  15. 15.

    Guedon-Moreau L, Lacroix D, Sadout N, Clementy J, Kouakam C, Hermida JS. Costs of remote monitoring vs. ambulatory follow-ups of implanted cardioverter defibrillators in the randomized ECOST study. Europace. 2014;16(8):1181–8. https://doi.org/10.1093/europace/euu012.

    Article  Google Scholar 

  16. 16.

    HAS, Remote monitoring systems for automatic implantable cardiac defibrillators. 2017. https://www.has-sante.far//jcms/c.2790581/fr/systems-de-telesurveillance-pour-defibrillateurs-cardiaques-automatiques-implantables.

  17. 17.

    Hindricks G, Taborsky M, Glikson M, Heinrich U, Schumacher B, Katz A, et al. Implant-based multiparameter telemonitoring of patient with heart failure (in time); a randomized controlled trial. Lancet. 2014;384(9943):583–90. https://doi.org/10.1016/S0140-6736(14)61176-4.

    Article  Google Scholar 

  18. 18.

    Crossley GH, Boyle A, Vitense H, Chang Y, Mead RH. The CONNECT (Clinical Evaluation of Remote Notification to Reduce Time to Clinical Decision) trial: the value of wireless remote monitoring with automatic clinical alerts. J Am Coll Cardiol. 2011;57(10):1181–9. https://doi.org/10.1016/j.jacc.2010.12.012.

    Article  Google Scholar 

  19. 19.

    Guedon-Moreau L, Lacroix D, Sadoul N, Clementy J, Kouakam C, Hermids JS, et al. A randomized study of remote follow-up of implantable cardioverter defibrillators: safety and efficacy report of the ECOST trial. Eur Heart J. 2013;34(8):605–14. https://doi.org/10.1093/eurheartj/ehs425.

    Article  Google Scholar 

  20. 20.

    Crossley GH, Chen J, Choucair W, Cohen TJ, Gohn DC, Johnson WB, et al. Clinical benefit of remote versus trans telephonic monitoring of implanted pacemakers. J Am Coll Cardiol. 2009;54(22):2012–9. https://doi.org/10.1016/j.jacc.2009.10.001.

    Article  Google Scholar 

  21. 21.

    Varma N, Epstein AE, Irimpen A, Schweikert R, Love C. Efficacy and safety of automatic remote monitoring for implantable cardioverter-defibrillator follow-up: the Lumos-T Safely Reduces, Routine Office Device Follow-up (TRUST) trial. Circulation. 2010;122(4):325–32. https://doi.org/10.1161/CIRCULATIONAHA.110.937409.

    Article  Google Scholar 

  22. 22.

    Mabo P, Victor F, Bazin P, Ahres S, Baduty D, Da Costa A, et al. A randomized trial of long-term remote monitoring of pacemaker recipients (The COMPAS trial). Eur Heart J. 2012;33(9):1105–11. https://doi.org/10.1093/eurheartj/ehr419.

    Article  Google Scholar 

  23. 23.

    Amara W, Montagnier C, Cheggour S, Boursier M, Gully C, Barnay C, et al. Early detection and treatment of atrial arrhythmias alleviates the arrhythmic burden in paced patients: the SETAM study. Pacing Clin Electrophysiol. 2017;40(5):527–36. https://doi.org/10.1111/pace.13062.

    Article  Google Scholar 

  24. 24.

    Boriani G, Da Costa A, Ricci RP, Quesada A, Favale S, Iacopino S, et al. The MOnitoring Resynchronization dEvices and CARdiacpatiEnts (MORE-CARE) randomized controlled trial: phase 1 results on dynamics of early intervention with remote monitoring. J Med Internet Res. 2013;15(8):e167. https://doi.org/10.2196/jmir.2608.

    Article  Google Scholar 

  25. 25.

    Slotwiner D, Varma N, Akar JG, Annas G, Beardsall M, Fogel Rl, et al. HRS Expert Consensus Statement on remote interrogation and monitoring for cardiovascular implantable electronic devices. Heart Rhythm. 2015;12(7):e69-100. https://doi.org/10.1016/j.hrthm.2015.05.008 (Epub 2015 May 14).

    Article  Google Scholar 

  26. 26.

    White G. Active queue management in DOCSIS® 31 networks. IEEE Commun Mag. 2015;2015:0163-6804/15.

    Google Scholar 

  27. 27.

    Khan I, Qureshi I, Shah SB, Akhlaq M, Safi E, et al. Comparative analysis and route optimization of state-of-the-art routing protocols. EAI Endorsed Trans Mobile Commun Appl. 2018;3:13. https://doi.org/10.4108/eai.22-3-2018.154385.

    Article  Google Scholar 

  28. 28.

    Barua M, Liang X, Lu R, et al. RCare: extending secure health care to rural area using VANETs. Mobile Netw Appl. 2014;19:318–30. https://doi.org/10.1007/s11036-013-0446-y.

    Article  Google Scholar 

Download references

Author information

Affiliations

Authors

Corresponding author

Correspondence to Suparna DasGupta.

Ethics declarations

Ethical Approval

This article does not contain any studies with human participants or animals performed by any of the authors.

Conflict of Interest

The authors declare that they have no conflict of interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

This article is part of the topical collection “Applications of Software Engineering and Tool Support” guest edited by Nabendu Chaki, Agostino Cortesi and Anirban Sarkar”.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

DasGupta, S., Choudhury, S. & Chaki, R. VADiRSYRem: VANET-Based Diagnosis and Response System for Remote Locality. SN COMPUT. SCI. 2, 41 (2021). https://doi.org/10.1007/s42979-020-00430-6

Download citation

Keywords

  • Remote healthcare system
  • IoT based
  • Smart vehicles
  • Vehicular ad hoc networks
  • ZigBee-enabled wireless sensor nodes