Abstract

Wireless Sensor Network (WSN) is one of the emerging technologies in the 21st century due to its growing demand in automation. WSNs are organized in a large environmental area and there are more chances for the sensor nodes to get affected because of external temperature. As the environmental temperature rises, the lifetime, quality of service and temperature of sensor nodes are easily influenced. Thus Environmental Aware Thermal (EAT) routing protocol is introduced to minimize the issue. In this protocol, the incoming data signals are assigned with normal, abnormal and critical priority levels. It consists of three potential fields such as environment, energy and quality of service. The routing path is chosen in such a way that the critical data reaches its destination with minimum delay. Therefore, the path is selected depending on surrounding temperature, threshold level and residual energy. The network performance was analyzed in three different cases: 1, 2 and 3. The total amount of power consumption, temperature variation, delay and lifetime of sensor node in all three cases are inferred.

Keywords: Environmental temperature, multipath routing protocol, wireless sensor network, quality of service

6.1 Introduction

Generally, a Wireless Sensor Network (WSN) is a distributed network with many sensors. It is based on wireless technology and is used to collect information from external environments like forests, flooded regions, agricultural land, battlefields, etc. A WSN comprises of numerous tiny sensors to monitor the area where it is being located. The collected signals are forwarded to the destination through intermediate nodes. A path is established to transmit the data from source to destination. This path is known as a routing path and the protocols designed to carry out this function are referred to as routing protocol. A routing protocol uses a predefined set of rules and regulations to choose a shortest path to destination from multiple available paths. An efficient routing protocol will increase the efficiency of a system and therefore it is considered as the heart of the communication networking system. The common protocols used in a WSN are given in Fig. 6.1.

1. Node centric routing: In this type of protocols the destination is identified as numeric.
2. Data centric routing: In this routing, the information obtained from the attributes are transmitted rather than receiving information from other nodes.
3. Source initiated routing protocols: The source node advertises that it has data to send and routing path is initiated from source.
4. Destination initiated routing protocols: In these protocols, destination initiates for the routing path.

The categories of routing protocol are single and multipath routing protocol. Nowadays, multipath routing protocol is incorporated in wireless sensor networks to obtain good quality in data transmission. In the case of single node routing protocol, data loss between source and destination occurs if there is a fault in sensor node. Also, energy consumption and data failure rate are noticed as high in single path routing protocol.

6.1.1 Single Path Routing Protocol

In single path routing protocol, the connection establishment between source and destination are designed using a single path. This protocol estimate link quality and these links are used to determine the best optimum path in WSNs. Basically, this is one of the most supportive techniques utilized in single path routing to provide reliability [1]. All the nodes are connected to the node head as illustrated in Fig. 6.2. If a node wants to transmit the data to base station it first sends the data to the node head and from the node head it reaches the base station.

Due to continuous data transmission of node head it generates more heat and the communication path established to base station is disconnected due to node head failure as shown in Fig. 6.3. Once the link gets failed then there is a data loss and it cannot reach the base station. It becomes a life-threatening problem in case of critical data transmission.

In a single path routing approach, a route discovery can be carried out with minimum resource utilization and computational complexity but it results in reduced throughput [2]. Additionally, reduced flexibility obtained as a result of this approach may significantly degrade the performance of the network in critical situations. Due to limited power supply, physical damages and high dynamics in wireless links causes failure in active path link. The data packet is not forwarded and thus an alternative routing path is found to transmit the data continuously resulting in increased delay in data delivery and maximum overhead. Hence due to unreliability and resource constraint of wireless links, a single path routing protocol is not widely used in various applications [3] as it cannot meet different criteria performance requirement in WSNs.

6.1.2 Multipath Routing Protocol

The multipath routing protocols are used in different applications. They provide an alternative routing path if there a link failure in established multiple path connection between the source and the sink. The link path is established using hop count. Furthermore, in multihop WSNs the environmental factors, orientation, antenna shape, distance and radio interference vary during the entire lifetime of wireless sensor networks. All these factors affect the link quality between the sensor nodes [1]. Consider a network with multiple routing path to reach the destination. A routing path established to transmit data from source to destination is as shown in Fig. 6.4. Due to continuous data transmission, there is a node failure due to excess heat generation in the established routing path as depicted in Fig. 6.5. Therefore, the communication between the source and destination is disconnected and data cannot reach the destination [4].

In multipath routing protocol the data are sent to the source and the source establishes a new routing path to the destination. Fig. 6.6 shows the reestablishment of connection to the destination through the next shortest route. The main advantage of utilizing multipath routing protocol is to maintain uniform traffic within the network, where the data are divided equally among all multiple paths. As a result, the energy consumption is also balanced. Moreover, it increases the reliability of the system by creating multiple copies of data packets and transmits to destination.

6.1.3 Environmental Influence on WSN

In wildlife monitoring applications, the performance of the network was observed at different days for night and daytime in different climatic conditions like summer and winter, especially in an outdoor environment. Thelen et al. [5] discussed the radio propagation through high humidity in potato deployment field. The path loss exponent value was 4 irrespective of different growing seasons. The radio range diminishes to 10 m as the potato crop starts flowering. Thus, it is necessary to deploy sensor nodes at a distance of at most 10 m in precision agriculture applications and a microclimate is sensed during the entire growing season. The influence of the potato foliage is found to be 17 dB, as nodes are placed at a distance of 15 m. G. Anastasi et al. [6] suggested that rain and fog affects the performance of WSN especially in data transmission range and reception. The data transmission range of mica2/mica2dot sensor nodes is poor in the presence of rain or fog. Carlo Alberto Boano et al. [7] looked into the variations of link quality and data delivery performance at ambient temperature influence in low-power radio communications. The experimental result highlights that the communication between sensor nodes gets affected due to temperature and thus minimum transmission power is required at low temperature.

6.2 Motivation Behind the Work

A wireless sensor network plays a vital role in many applications such as health care, precision agriculture, environmental surveillance military, etc. A WSN must support a certain degree of reliability, energy and delay bound for data transportation to be utilized in these applications. Therefore, it is necessary to design and develop an energy-efficient protocol. Apart from these factors, environmental awareness is also an important factor that should be considered in multipath routing protocol design. The cost of sensor nodes is less and is deployed in large scale. These are easily influenced by environmental factors like electromagnetic interference, vibration, temperature and humidity. Once the surrounding temperature increases, it degrades the performance of sensor nodes and excessive rise in temperature may damage the sensor nodes. An extreme high humidity environmental condition minimizes the link quality and raises the probability of short-circuitry in sensor nodes. Similarly, a Strong electromagnetic interference increases the data loss rate.

Thus the sensor nodes utilized in health care applications must withstand the environmental characteristics and fluctuating channels. Besides, a communication protocol must be designed to maintain a bounded packet delivery rate (during critical stage of human) though there is a drop in established link. The sensor nodes deployed at the outdoor environment usually experience high fluctuation due to the variation in weather conditions. Thus the designed protocol must withstand variations in environmental conditions, channel fluctuations and successful data delivery. If the designed routing protocol does not withstand the environmental changes and the data packets routed through a sensor node are affected by temperature once it crosses a heat zone, data delivery through this particular path is terminated. In case of environment-aware routing, if the routing path senses extreme temperature it adjusts to an alternate routing path to prevent data loss.

6.3 Novelty of This Work

Many researchers tried to incorporate the impact of environmental influence into the network’s performance. But only a few of them could find an appropriate and reliable result for the influences of different environmental conditions. However, the main parameter degrading the network lifetime and quality of service is very scanty in literature and only very few environmental parameters like fog, moisture, humidity, and reflecting angle were considered. This work aims to develop an invulnerable routing protocol which resists an environmental impact. Thus an Environmental Aware Thermal (EAT) Routing Protocol has been developed. It consists of potential fields like energy, environment and quality of service. The energy field ensures that the sensor nodes select neighbor nodes with more energy as relay nodes. The environmental field makes sure that the estimated routing path finds an alternative routing path as the sensor node temperature increases beyond the threshold limit. The quality of service field makes the data reach the destination successfully from the source. The routing path is estimated once the above-mentioned potential fields are satisfied. The major contributions of EAT protocol are summarized as follows.

1. Improved routing possibilities under critical temperature zone: Based on the acquired data from environment, the EAT routing protocols can identify an additional routing path to avoid the critical temperature zone.
2. QoS field: To improve the quality of service, the data are assigned with three different priorities (normal, abnormal and critical). This protocol ensures that the critical data will reach the destination node without any delay.
3. Energy field: This protocol measures the available remaining energy within the network. If a node wants to choose an alternative path due to high temperature zone, then routing path with high energy nodes are selected. Similarly, the relay node with high energy is selected for long-distance data transmission.

The remainder of this chapter is organized as follows: conventional protocols on node disjoint, partially disjoint protocols and temperature influence on different applications are reviewed in Section 6.4. In Section 6.5, the implementation of Environmental Aware Thermal (EAT) Routing Protocol, assumption and flow chart is illustrated. Section 6.6 highlights the simulation parameters utilized. Section 6.7 discusses the results obtained from simulation and analyzes the environmental impact on a WSN. Finally in Section 6.8 the conclusion is presented.

6.4 Related Works

The multipath routing protocols are designed to provide reliability and energy efficiency in a WSN. The protocols are classified into two types based on node path disjointness such as node disjoint and partially disjoint.

In node disjoint protocol, there is no single common node in any discovered routing paths. In case of node failure, data transmission through that particular route is interrupted. This protocol guarantees that other constructed paths are not affected. The different node disjoint protocols are as follows: In N-to-1 multipath protocol [8] the routes are updated periodically at the end of discovery process or based on the demand from base station. A hybrid multipath approach is introduced for reliable and secured data collection. The information at the source is split into multiple data using secret sharing scheme. The divided data travels along the multiple path for concurrent delivery. The reliability of packet is increased due to an alternate path packet salvaging strategy. This protocol is resistant to collusive attack and link failure of nodes. HSPREAD [9] is an extension of N-to-1 multipath routing protocol. It is used to find the nodes being disjointed from BS in a single route discovery process, following which a hybrid multipath data collection approach was proposed. In this method an alternative routing path is determined for every individual packet and is combined with concurrent multipath dispersion to obtain concurrent route for end-to-end data collection. Additionally, this scheme improves the security of end-to-end data delivery by combining multipath data dispersion and secret sharing mechanism. The multipath route discovery operation is similar to N-to-1 multipath routing protocol. Hence energy efficiency is found to be a major drawback in this protocol. The authors in [10] proposed DCHT protocol based on node disjoint multipath routing protocol. In this scheme, multipath routing path is established by direct diffusion process. This process strengthens the multiple path by providing minimum latency and high quality link. The quality of any established path is judged by interference strength and data transmission latency. As the interference strength is dynamic in WSN, more network resources are required for routing in DCHT. Thus the disjoint paths with less interference and path cost are selected. In Efficient and collision aware (EECA) node-disjoint multipath routing algorithm (EECA) [11], two collision-free routing paths are estimated based on node position information. These two paths are established using power and constrained adjusted flooding mechanism. The data are transmitted with minimum power. The EECA protocol is limited within the neighbor nodes in the discovered route. Additionally, collision between the established two routing paths is achieved utilizing the broadcast nature of wireless communication. However, the multipath interference is reduced in routing protocol. In Geographic node-disjoint path routing protocol (GNPR) [12], two routing schemes based on direction and distance are established. These metric schemes are incorporated in greedy routing (GR) and compass routing (CR). The data packets are forwarded to the neighboring nodes with a smallest angle to reach the destination in CR. Similarly, the node transmits data packet to neighbor present near the destination in the space in GR. It performs better in terms of delay. In Pairwise directional geographical routing protocol (PWDGR) [13] the pairwise nodes are selected which are 360° around the sink. The routes are established in the following manner: source-pairwise-sink. This connection provides a balanced traffic in the network and avoids hot spot issue by uniformly selecting the nodes for routing path. The GPS module can be integrated into sensor nodes in PWDGR to find the location but the cost becomes high in large-scale deployment. In Minimum Energy Cost Aggregation Tree (MCEAT) algorithm [14], multipath node disjoint problem is considered as Steiner tree problem and the solution is determined through genetic algorithm. The main objectives of these optimization algorithms are reliability, transmission delay and energy. In this algorithm two factors are considered, one with relay node and other without relay node. The solution for without relay node problem is obtained using 2 approximation algorithm and for networks with relay node is determined using O(1) approximation algorithm. Since the Steiner tree problem is NP hard, this approach is efficient only for small-scale deployment areas.

The node disjoint routing protocol provides several advantages such as reliability. This algorithm finds it difficult to estimate several paths between source and sink in case of sparse deployment. Besides, this protocol requires frequent updating of information about the neighboring nodes, resulting in larger routing overhead. Thus partially disjoint multipath routing protocols are formed which is similar to node disjoint protocol, the partially disjoint multipath routing can also incorporate multiple shared nodes and a single node failure interrupts all the other alternate paths including the failure node. The various partially disjoint multipath routing protocol are described below.

Security Aware Ad hoc Routing protocol (SAR) [15] is a first partially disjoint multipath routing protocol. In this protocol the routing decisions are made by considering the following factors, namely QoS parameters, priority of data packets and energy conservation. This protocol utilizes a table driven multipath approach to provide fault to tolerance, energy consumption and QoS parameters. SAR provides quality of protection to all the data packets flowing through this protocol. Thus the routing overhead maintenance is overwhelming. Reliable Information Forwarding (ReInForM) routing protocol [16] transmits multiple copies of data packets through multiple paths from to source to sink with the desired reliability. Dynamic packet is created to minimize the number of paths required for reliability. This is done through topology and channel error rates. ReInForM utilizes all desired path with uniform and efficient load balancing. The routing mechanism implemented in this protocol is costly due to frequent information exchange of neighboring nodes. In State Free Gradient-Based Forwarding Protocol (SGF) [17], the sensor nodes do not maintain routing table in which the information about neighboring nodes or network topology is not maintained. Hence this protocol remains suitable for large networks. Instead of routing table, SFG constructs a cost field called gradient. This gradient directs each data packet to proper routing path. The entire gradient mechanism is maintained by data transmission with little overhead. To adapt to topology variations, the forwarder node is selected through distributed contention process from multiple nodes. This protocol provides less delay with increased packet delivery. Energy-Balanced Routing Protocol (EBRP) [18] approach is constructed by combining virtual potential field and the concepts of potential in physics. The virtual potential field consists of depth energy and residual energy. In this protocol the data are forwarded through the nodes with high residual energy. The routing loop problems are eliminated by using loop elimination algorithm and basic algorithm. This algorithm improves energy balance, increased network lifetime and throughput.

In addition to the above routing protocol a few researchers have analyzed the impact of surrounding temperature in an environment with the following results. The area of monitoring on off-site region depends on the position of electronic nose which is a part of the WSN system [19]. The node located beyond the landfill region does not monitor continuously, but it acts as a sensor when activated at particular conditions, both inside and outside the landfill are obtained. Additionally, a WSN is organized based on the energy aware approach to increase the lifetime of entire system with benefits in terms of cost and better advancements in monitoring structure. In this work [20] a heuristic algorithm is designed and reference architecture that aids the decision of anomaly detection depends on the demands of agricultural environments are utilized. The author had performed a preliminary evaluation and analyzed different anomaly detection algorithms in terms of scalability metrics, execution time and accuracy. From the obtained results it was inferred that the power consumption is reduced by 18.59% and lessens the temperature of the device by 15.94%. The obtained values are completely dependent on edge device characteristics and the application workload. The sensors are placed in different environments to collect various data such as humidity, light, temperature, etc. [21]. Though it is useful to collect different data, it is still a prominent issue to infer the impact of environmental conditions on data collection in terms of accuracy and prolonged network lifetime. Hence an optimized dictionary updating learning-based compressed data collection algorithm (ODUL-CDC) is developed to degrade the influence of environmental noise on the accuracy of WSNs data collection and to increase the life time of sensor node. The main purpose of using the dictionary learning method is to get a sparse dictionary, which is obtained by learning from the training data. Henceforth the main purpose of introducing the self-coherence penalty term is to reduce the over fitting of the training data during the dictionary updating process. Before installation of sensor nodes, it is important to determine the total cost required to complete the entire set up [22]. A sensor network is designed with an operating frequency of 920 MHz band to measure the quantities like atmospheric pressure, dust, temperature and relative humidity, etc. The system is developed based on LoRa networks and the above-mentioned parameters are measured in the actual environment of Kamihama campus at Mie University. From the results it is observed that the temporal and spatial characteristics of measured quantity are determined for proper positioning of end devices in LPWAN-based WSN.

Thus from the above discussion it is clear that only few researches were carried out by considering the impact of temperature on sensor nodes. But in routing protocol design the influence of temperature variations in the environment is not included. Hence due to low-cost implementation and large-scale deployment the effect of environmental factors on WSN cannot be neglected practically. Thus EAT routing protocols are designed to consider the influence of environmental temperature on performance of sensor nodes.

6.5 Proposed Environmental Aware Thermal (EAT) Routing Protocol

In EAT routing protocol, the environmental influence on a particular network and its effects are estimated. The effects are observed for lifetime, data delivery delay, device performance, and network efficiency during critical periods. Fig. 6.7 shows the operation of EAT protocol.

• In the initialization phase, source initiates the broadcast to gather information from intermediate nodes like hop distance, temperature and energy from source to destination.
• The neighboring node starts calculating the temperature and remaining energy. If the temperature of the node is found to be less than the threshold value (Node_temp<Th _min), the packet is passed to the next intermediate node for further processing or else the data packets are discarded. Likewise, the remaining available energy is also calculated. If the node’s energy is high, then the packet is forwarded to the neighboring node.
• If both temperature and energy conditions are satisfied, the node calculates the distance between source and destination. A connection is established through a path with minimum hop count. If the distance is too long, then the node will choose a relay node to reach the destination.
• Once the connection is established, the sensor nodes are ready to transmit the packets to the destination. Before data transmission, the packets are categorized into normal, abnormal and critical priority levels.
• After assigning priority, the protocol checks the surrounding temperature. If it is above the threshold value (Node_ Etemp>Th_max), then the packets are retransmitted to source to choose an alternative path. Next, the node’s temperature is calculated. If temperature of sensor node is greater than the threshold value (Node_temp>Th_max), the sensor node forwards only critical data signals; otherwise all priority signals are transmitted.

6.5.1 Sensor Node Environmental Modeling and Analysis

The influence of temperature on sensor node and its effects on the data transmission, delay and energy consumption is observed. However, in the atmosphere there are many environmental factors like humidity, moisture, electromagnetic interference and temperatures that influence the sensor node’s performance. From the above-mentioned parameter, temperature is one of the most influencing factors which degrades the performance of the sensor node. In this paper, environmental temperature influence on sensor node is focused. Thus, single node environmental influence and multi-node environmental influence is developed to analyze the influence of temperature on sensor nodes.

6.5.2 Single Node Environmental Influence Modeling

The threshold temperature is fixed for each sensor node to identify the surrounding temperature around the single node. The threshold value is fixed based on surrounding temperature for best operation. Each node continuously senses the surrounding temperature. If the surrounding temperature is minimum and below the threshold value (-10 °C to 10 °C) the values are calculated using Eq. 6.1. If the node is deployed in normal environmental temperature field of 10 °C to 80 °C, the influence on external environmental influence is set to be 1 as given in Eq. 6.2. At this point, the temperature influence is considered as negligible. If the temperature exceeds maximum threshold value, then Eq. 6.3 is used to calculate the field temperature.

(6.1)

(6.2)

(6.3)

Where is a single node surrounding environmental temperature field are defined as the sensor normal operation at k environmental factor. is the sensor node operating threshold set point. T^k(n) is temperature of individual node at k environmental factor. for low environmental temperature field and  is defined for high environmental temperature field. If the data is transmitted through a node (n) and T^k(n) changes to the state 1. It indicates that the temperature of a particular node increases. At this stage the protocol verifies the value. If the condition is satisfied, data transmission is terminated through that particular node.

6.5.3 Multiple Node Modeling

The data packets are transmitted through multiple nodes to reach their destination. Therefore, multiple node temperature modeling is essential to understand the complete influence of environmental effect on sensor nodes. Due to the environmental factor, the lifetime of sensor node, energy of particular node and data losses of the node are being affected. To analyze the environmental temperature T_m(n) influence on sensor node the following Eq. 6.4 is used.

(6.4)

Where T_m(n) is a single node surrounding temperature created by node (n) at k factor. In a real-time environment, multiple factors like humidity, moisture and electromagnetic interference, etc., influence the node performance. In this study only temperature is taken into consideration. The path selection is done based on single node surrounding temperature value. To ensure continuous working of sensor nodes two threshold values, T_min and T_max are fixed. If the temperature of sensor node increases beyond T_max, that specified node area is called as “unsafe zone” and this node is not selected for further communication purpose. This “unsafe zone” data is collected by neighboring node and the same node is continuously monitored until it returns to normal temperature. It is given in Eq. 6.5.

(6.5)

Where T_e (n, p) is the neighboring field temperature potential of node (n) and node (p).

6.5.4 Sensor Node Surrounding Temperature Field

The total environmental temperature of a particular sensor node (n) is defined by combining the multiple node environment T_m(n) and neighboring field environment T_e (n, p). It is given in Eq. 6.6.

(6.6)

Where T_evn (n) is environment of particular sensor node, k_{(n, p)} environmental factor at node (n) and node (p).

6.5.5 Sensor Node Remaining Energy Calculation

To ensure continuous operation of sensor node, remaining energy calculation is very important. The remaining energy (E_ng(n)) of particular node is calculated using Eq. 6.7.

(6.7)

Where E_r is the remaining available energy of node (n) at the time of (t), E_i(n) is the initial energy available while deploying the sensor node (n). Utilizing Eq. 6.7 the remaining energy of a particular node at time (t) can be determined. But the required energy for sensor operation is calculated using Eq. 6.8.

(6.8)

Where P_rx denotes the node receiving operation, P_tx performs data transmission, P_idl indicates node in idle state, P_slp is the nodes in sleep state, P_{Rfrequency_startup} is the radio frequency startup power during transmission, P_t and T_r is the data and transmission rate of packets. In a network, all nodes perform many operations like sensing, transmission, receiving, sleep and idle stage. Each stage of sensor operation consumes a different energy level from the battery.

6.5.6 Delay Modeling

The data being transferred from source to destination will undergo different delays along its desired path. The types of delay include processing delay, queuing delay and sensing delay. The processing delay arises during data transmission from one node to another node. Queuing delay is due to the nodes transmitting previous data packets. The sensing delay is with the initialization of nodes for data transmission. Moreover, transmission delay occurs while sending the data, and reception delay is during data reception at each node.

(6.9)

Where D_n is sum of delay, D_sn is sensing process delay, D_prs is process delay, D_qu is queuing delay, D_tx is transmission delay, D_rx is receiving delay and is relay node processing delay.

6.6 Simulation Parameters

The EAT protocol are simulated using MATLAB. The sensors are assumed to be deployed within an area of 250 × 250 m. The total number of sensor nodes used is around 100 and the range is set to 50 m from source to destination. The ambient temperature is kept at 40°C. The minimum and maximum operating temperature is around 10°C and 80°C. In the simulation model, the node’s position is fixed and has the same transmission range. The specific heat of the node is fixed as constant value. The node gets cooled down at the rest state of the sensor node. The multi-hop network model is prepared. During installation, all nodes are placed at uniform distance with equal energy. The simulation parameters used are shown in Table 6.1.

6.7 Results and Discussion

In this section, the influence of temperature on networks, amount of power consumed, sensor network lifetime at three different cases and variation of delay at different temperature are discussed.

6.7.1 Temperature Influence on Network

The sensor network performance degrades and its malfunction probability also increases sharply at low and high temperature. If the node operates at normal environmental temperature, then the effect caused due to surrounding temperature on the network can be taken as negligible. Fig. 6.8 shows the temperature influence on sensor nodes at three different cases. In case 1, normal operation (no temperature influence) is considered. Here, the sensor node operates at nominal temperature interval and does not consider the influence of temperature on sensor performance. In case 2, the factors influencing the sensor node at different environment field are considered. As the surrounding environmental temperature increases, the sensor node temperature rises linearly at a time interval t. In case 3, the temperature variation of sensor nodes along the routing path due to continuous variation in environmental temperature is analyzed.

6.7.2 Power Consumption

The total amount of power consumed is determined by taking the difference between initial energy and the remaining energy. The environmental field ensures that the constructed multipath does not utilize the sensor node whose temperature is beyond the maximum threshold limit. The QoS field takes care of successful delivery of packets to sink. The energy field helps to select an intermediate with high residual energy to involve in the next hop of data transmission. Fig. 6.9 illustrates the average power consumption of different data rate for case 1 to 3. From the obtained results, it is observed that the amount of power consumed is less in case 1 and case 2. In case 3 energy increases with data rate. Thus a large amount of power is consumed in case 3, thereby reducing the node lifetime considerably. Furthermore, the routing decisions get affected due to residual energy and the data avoid passing through the node with lesser energy.

6.7.3 Lifetime Analysis

Fig. 6.10 shows the lifetime analysis at different cases. In case 1, the sensor node works for longer duration compared to other two cases. In case 2 condition, the nodes are influenced by environmental temperature which causes fast discharging of available energy in the battery. If the discharging rate of battery power increases, then the total lifetime of the sensor node gets decreased. In case 3, due to rerouting process the sensor spends more energy for transmitting the data to long distance. As the transmission distance increases, the energy consumption will also remain high. Likewise, if the packet size increases then the energy consumption also increases. Thus the lifetime of sensor node gets reduced in case 3. Moreover, the improper energy calculation of sensor node during route node selection results in rapid death of sensor nodes within the network.

6.7.4 Delay Analysis

The delay modeling is performed at all the three cases and the corresponding result (delay vs. temperature) is shown in Fig. 6.11. The delay is measured based on the number of packet reaches within a specified time interval. From the graph, it is observed that the delay is minimum in case 1 as the transmitted data packets reach the destination through the shortest path. So all nodes perform data transmission with minimum delay. In case 2, the delay is high due to external temperature influence on a particular sensor node. This results in the limited operation of the node. At this condition, the data packet transmission will be stopped and the neighboring nodes will update the current temperature value of the affected node. Likewise, in case 3 the sender node will completely reroute to the next shortest path. As the transmission range increases, the delay gets increased during delivery of data to destination node.

6.8 Conclusion

WSNs are deployed in unattended areas and are provided with minimum energy for operation, which affects the network’s performance and lifetime. Thus Environmental Aware Thermal (EAT) routing protocol was proposed. This protocol mainly concentrated on the effect of surrounding environmental temperature and selects an optimum routing path accordingly. Temperature, delay, lifetime and power consumption of sensor nodes at three different cases are analyzed. From the obtained results, it was inferred that case 1 results have efficient QoS and increased network lifetime at normal environmental temperature. In case 2, as the temperature increases, the delay gets increased and network lifetime becomes minimum for a single sensor node. In case 3, a fully established sensor network was considered. In this case, the environmental temperature influence on the QoS, lifetime, and temperature of sensor nodes was observed. Therefore, the effect of environmental conditions on the performance of sensor node was analyzed. In future, other environmental factors like humidity, rain and moisture influence on EAT protocols need to be evaluated to analyze the effectiveness of the entire network operation. Also, the real-time implementation of sensors and its corresponding data can be examined.

References

1. 1. Macit, M., Gungor, V.C. and Tuna, G., 2014. Comparison of QoS-aware single-path vs. multi-path routing protocols for image transmission in wireless multimedia sensor networks. Ad Hoc Networks, 19, pp.132-141.
2. 2. Kang, J., Zhang, Y. and Nath, B., 2004, August. End-to-end channel capacity measurement for congestion control in sensor networks. In Proceedings of the 2nd International Workshop on Sensor and Actor Network Protocols and Applications (SANPA’04).
3. 3. Son, D., Krishnamachari, B. and Heidemann, J., 2006, October. Experimental study of concurrent transmission in wireless sensor networks. In Proceedings of the 4th International Conference on Embedded Networked Sensor Systems (pp. 237-250).
4. 4. Akan, O.B., 2007. Performance of transport protocols for multimedia communications in wireless sensor networks. IEEE Communications Letters, 11(10), pp. 826-828.
5. 5. Thelen, J., Goense, D. and Langendoen, K., 2005. Radio wave propagation in potato fields. In 1st workshop on wireless network measurement, Riva del Garda, Italy, April 2005 (pp. np-np).
6. 6. Anastasi, G., Falchi, A., Passarella, A., Conti, M. and Gregori, E., 2004, October. Performance measurements of motes sensor networks. In Proceedings of the 7th ACM International Symposium on Modeling, Analysis and Simulation of Wireless and Mobile Systems (pp. 174-181).
7. 7. Boano, C.A., Tsiftes, N., Voigt, T., Brown, J. and Roedig, U., 2009. The impact of temperature on outdoor industrial sensornet applications. IEEE Transactions on Industrial Informatics, 6(3), pp.451-459.
8. 8. W. Lou, An efficient N-to-1 multipath routing protocol in wireless sensor networks, in: IEEE International Conference on Mobile Adhoc and Sensor Systems, 2005, pp. 664–672.
9. 9. W. Lou, Y. Kwon, H-SPREAD: a hybrid multipath scheme for secure and reliable data collection in wireless sensor networks, IEEE Trans. Veh. Technol. 55 (4) (2006) 1320–1330.
10. 10. S. Li, R.K. Neelisetti, C. Liu, A. Lim, Efficient multi-path protocol for wireless sensor networks, Int. J. Wirel.Mob.Netw. 2 (1) (2010) 110–130.
11. 11. Z. Wang, E. Bulut, B.K. Szymanski, Energy efficient collision aware multipath routing for wireless sensor networks, in: IEEE International Conference on Communications, 2009, pp. 91–95.
12. 12. A. Kumar, S. Varma, Geographic node-disjoint path routing for wireless sensor networks, IEEE Sens. J. 10 (6) (2010) 1138–1139 .
13. 13. J. Wang, Y. Zhang, J. Wang, Y. Ma, M. Chen, Pwdgr: pair-wise directional geographical routing based on wireless sensor network, IEEE Internet Things J. 2 (1) (2015) 14–22.
14. 14. T.W. Kuo, M.J. Tsai, On the construction of data aggregation tree with minimum energy cost in wireless sensor networks: Np-completeness and approximation algorithms, in: IEEE International Conference on Computer Communications (INFO-COM), 2014, pp. 2591–2595.
15. 15. S. Yi, P. Naldurg, R. Kravets, Security-aware ad hoc routing for wireless networks, in: ACM International Symposium on Mobile Ad Hoc Networking & Computing, 2001, pp. 299–302.
16. 16. B. Deb, S. Bhatnagar, B. Nath, Reinform: reliable information forwarding using multiple paths in sensor networks, in: IEEE International Conference on Local Computer Networks, 2003. LCN ’03. Proceedings, 2003, pp. 406–415.
17. 17. P. Huang, H. Chen, G. Xing, Y. Tan,Sgf: a state-free gradient-based forwarding protocol for wireless sensor networks, ACM Trans. Sens. Netw. 5 (2) (2009) 14.
18. 18. F. Ren, J. Zhang, T. He, C. Lin, S.K.D. Ren,Ebrp: energy-balanced routing protocol for data gathering in wireless sensor networks, IEEE Trans. Parallel Distrib. 22 (12) (2011) 2108–2125.
19. 19. Campanile, L., Iacono, M., Lotito, R. and Mastroianni, M., 2020. A WSN Energy-aware Approach for Air Pollution Monitoring in Waste Treatment Facility Site: A Case Study for Landfill Monitoring Odour. In IoTBDS (pp. 526-532).
20. 20. de Souza, P.S.S., Rubin, F.P., Hohemberger, R., Ferreto, T.C., Lorenzon, A.F., Luizelli, M.C. and Rossi, F.D., 2020. Detecting abnormal sensors via machine learning: An IoT farming WSN-based architecture case study. Measurement, 164, p.108042.
21. 21. Chen, J., Zhou, F., Guo, Z. and Wan, J., 2020. Compressed Data Collection Method for Wireless Sensor Networks Based on Optimized Dictionary Updating Learning. IEEE Access, 8, pp. 205124-205135.
22. 22. Kaichi, A., Narieda, S., Fujii, T., Umebayashi, K. and Naruse, H., 2020, December. On Placement of End Devices in LPWAN Based WSN for Environmental Monitoring Applications. In 2020 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC) (pp. 1519-1522). IEEE.