Chapter 2
Wireless sensor networks applications to smart homes and cities
A. Belghith*
Abstract
Wireless Sensor Networks (WSNs) provide several types of applications providing comfortable and smart-economic life. Energy saving minimizing the rare sources of energy, noise and atmospheric monitoring reducing the pollution, and healthcare monitoring helping the health are examples of important applications in WSNs.
Any application requires communication between the sensors and the different kinds of servers. This communication can be performed using three main access technology architectures. These architectures profit from important wireless communication technologies such as IEEE 802.15.3 and IEEE 802.15.4 for Wireless Personal Area Network (WPAN), IEEE 802.11g and 802.11n for Wireless Local Area Network (WLAN), and High Speed Downlink Packet Access (HSDPA) and Long-Term Evolution (LTE) for Wireless Wide Area Network (WWAN).
In order to enhance application performance, the routing strategies discovering the best path to the destination, the energy-saving methods maximizing the life of sensors, and security protocols guarantying our privacy are the most important techniques used in WSNs.
Keywords
WSN
applications
access technologies
power saving
routing
security
1. Introduction
Wireless Sensor Networks (WSNs) [1] connect our world more than we dream up. Noise and atmospheric pollutions, garbage level sensor, road traffic monitoring, and smart parking are some of the many WSN applications to smart cities. In smart homes, it is now difficult to avoid using home video message, alarm to mobile phone, door, window, and light control applications while providing comfortable and smart-economic life.
To ensure this admirable way of life, new challenges of WSNs arise. Which access technologies can be used in telecommunication systems for smart cities and homes? Wireless protocols have to be used to communicate with thousands of nodes while managing a trade-off between data transfer rate, speed, and power consumption.
The huge number of sensors requires data aggregation mechanisms in order to prevent information redundancy and high energy consumption, storage capacity, and communication bandwidth. Obviously, these aspects cannot be met given the limitations of WSNs.
Moreover, it is interesting to define efficient network discovery and intelligent path determination to obtain reliable routing protocols taking into account the characteristics of sensors in the smart cities and homes.
Reliability is also guaranteed by security aspects such as encryption, access control, and secure data aggregation. Encryption becomes vital as data exchanged are related to personal and confidential data. Access control avoids privacy information discloser, especially for remote home monitoring.
In this chapter, we first give some examples of WSN applications in smart homes and cities. Then, we discuss the access technology to be used for the applications. Finally, we present some protocols useful to provide better applications’ performance such as the routing strategies, energy-saving methods, and security protocols.
2. WSN applications examples
In this section, we present essential applications of WSNs for smart cities and homes. There are three key applications: energy saving, noise and atmospheric monitoring, and healthcare monitoring.
2.1. Energy-saving applications
The huge energy consumption and the high fuel cost require an efficient use of the energy that becomes more and more rare. Fig. 2.1 shows the total consumption by End-Use sector in the United States [2].
Figure 2.1 Total Consumption by End-Use Sector in the United States, 1949–2014 (Quadrillion Btu)
Exact values of energy consumed in the electric power, residential, and transportation sectors in 1994 and 2014 are presented in Table 2.1. We note an increase of about 15% in the past 20 years and therefore it is essential to efficiently react to this growth.
Table 2.1
Total Energy Consumed in 1994 and 2014 by Different Sectors in the United States
| Total Energy Consumed by | In 2014 (Trillion Btu) | 1994 (Trillion Btu) | Increase in Past 20 Years (%) |
| The electric power sector | 38,520.472 | 32,398.714 | 15.9 |
| The residential sector | 21,530.939 | 18,111.572 | 15.88 |
| The transportation sector | 27,117.707 | 23,365.133 | 13.83 |
The increase in energy consumption is not related only to the USA, but concerns all countries in the word such as in Finland where lighting consumes over 30% of the total electricity used in households for appliances [3] and in Egypt where the consumption of energy in a house can reach 1500 kWh/month [4].
Energy-efficient buildings have to be designed in order to significantly reduce the energy use, especially for heating and cooling. The decrease of the energy use can be performed when reducing the demand for energy by avoiding waste and implementing energy-saving measures. Waste energy avoidance can be performed by having good insulation means, air tightness, and ventilation. In Ref. [5], a procedure for measuring and reporting commercial building is proposed. This procedure consists of two steps. The first step is the measurement system design that identifies the performance metrics, the physical location of each measurement, the frequency of measurements, and the measurement equipment. The second step is the data collection and analysis, which specifies how to monitor data, assemble data, and calculate monthly and annual metrics.
We note that the wasted energy reduction is not sufficient. It is also interesting that the buildings use sustainable and green sources of energy instead of finite fossil fuels. For example, a green building architecture is proposed in Ref. [6]. Monitoring the temperature, light intensity, and presence of persons provides information that is used by the control subsystem of the proposed architecture. The control subsystem can tune the energy to consume, inform the users about the status of the energy consumption periodically or when there are excessive energy consumption in real-time reports, and schedule flexible tasks. For example, when using smart meters in a building [7], the cost of energy varies depending on the period of day and therefore the control subsystem can schedule some house tasks when the price of energy becomes low. Finally, taking into account the atmospheric characteristics, we can program which type of energy can be used such as solar, wind, and geothermal energies.
Wireless sensors can also use green energy such as light, motion, and vibration to work. For example, GreenPark developed ultralow-power sensors that use environmental energy sources in order to reduce the energy consumption and extend the use of sensors in a building [8].
2.2. Noise and atmospheric monitoring
WSNs can be very helpful for the health of urban residents as they can monitor noise and atmospheric pollution. First experiences using WSNs for noise pollution monitoring are presented in Ref. [9]. These experiences are based on Tmote prototyping platform for collecting noise pollution data in both indoor and outdoor settings. Experimental results show the feasibility of using WSNs in noise monitoring. Therefore this kind of application can prevent expensive and complex tasks done especially by private entities.
The increase of noise pollution motivates researchers to continuously propose designs for noise pollution monitoring systems. For example, in Ref. [10], authors propose noise measurement instruments and techniques taking into account the transmitter, receiver, and atmosphere. Note that the noise follows a path depending on the atmosphere to reach the receiver. In Ref. [11], a design of an energy-harvesting noise-sensing WSN mote is proposed in order to mitigate and fight noise pollution. The proposed mote extension is able to detect noise levels in urban environments where there are multiple pulse loads. Experimental results show that the WSN mote provides an improvement of more than 300% in the analytically derived duty cycles.
Note that some commercial WSN devices for atmospheric pollution monitoring are presented in Ref. [12] such as Waspmote, Generic Ultraviolet Sensors Technologies and Observations (GUSTO), and CitiSense. Waspmote [13] can monitor several parameters to verify if the quality of air we breathe is healthy. These parameters consist of Nitrogen dioxide (NO2), Carbon dioxide (CO2), Methane (CH4), and Hydrocarbons (Ethanol, Propane, Butane, etc.). GUSTO [14], based on the Differential Ultraviolet Absorption Spectroscopy (DUVASTM) technology [15], can measure and transmit urban pollutants such as NO2, O3, and benzene in real time.
Figs. 2.2 and 2.3 show Waspmote and GUSTO devices, respectively. We note that these devices do not have display. They send air pollution measurements to a collector for further analysis and investigation.
Figure 2.2 Waspmote Devices
Figure 2.3 GUSTO Device
CitiSense [16], developed by a University of California-San Diego team, is a pollution monitoring system that can be integrated in smartphones. Therefore, it provides very smart useful service to users in order to prevent staying in highly polluted places. Fig. 2.4 shows that CitiSense finds out that air quality in the current place is moderate. Note that real-time information about air pollution can be obtained using other sophisticated tools such as over the Google Map. However, information is not displayed for public users (only for users having authorization) [17]. Many areas deploy this technology such as in Qatar [18].
Figure 2.4 Air Quality Monitoring by CitiSense
Like noise pollution monitoring, several system designs for air pollution were proposed. In Ref. [19], authors propose an air pollution system that monitors the air quality in real time while reducing the energy consumption of sensors using a power-saving strategy. The proposed power-saving strategy is described later. Sensors sense the pollution information and then compare the pollution level with defined standard reference values. If the pollution level sensed is high, then data are sent through transmitter and sensors wait 5 min before sensing the pollution information in the next time. Otherwise, when the sensed pollution level does not exceed the threshold values, sensors do not send data and wait 15 min before performing next measurements.
In Ref. [20], Kalaimani and Sakthivel propose a simple WSN-based air quality monitoring system (WSN-AQMS) for industrial areas. The proposed monitoring system controls and monitors the physical environment while reducing the energy consumption and the rate of data exchanged between sensors. It selects WSN components depending on the building’s purpose, the number of nodes needed, and the options of evaluation. The architecture of the monitoring system uses Gas sensors, humidity sensors, and Global System for Mobile communications (GSM) modules for cellular communications.
2.3. Healthcare monitoring
In addition to the noise and atmospheric monitoring, which are useful for keeping health, WSNs can be also used for smart healthcare of residents while keeping their comfort and privacy. In Ref. [21], authors propose system architecture for residents’ health monitoring. The proposed system integrates the existing medical technology in low-cost sensors in order to nurse elder and handicapped persons. The efficiency of this solution is based on the following proprieties: portability of the small devices (sensors), scalability of the number of devices used when decreasing the complexity of the functionalities, autoconfiguration capability especially when using Internet Protocol version 6 (IPv6) protocol [22], and the real-time response of the deployed sensors when measurements exceed specific thresholds.
Evidently, WSN monitoring is used in hospitals. Many projects are developed in this field [23] such as HealthGear (Microsoft project) [24], MobiHealth (European Commission project) [25], CodeBlue (Harvard University project) [26], and Wireless Sensor Network for Quality of Life (WSN4QoL) (Marie Curie project) [27].
HealthGear consists of a set of noninvasive physiological sensors that are connected via Bluetooth to cell phones. The sensors contain modules that measure many parameters such as the blood’s oxygen level of users as well as the user respiration and motion. The measurements are then transmitted using Bluetooth in order to analyze them and represent the results in an interface (see Fig. 2.5).
Figure 2.5 Interface of HealthGear
MobiHealth profits from available technologies such as mobile medical sensors, public wireless network, and new services of Universal Mobile Telecommunications System (UMTS). It provides medical follow-up and medical research. Note that MobiHealth deals with many fields such as remote monitoring of glucose levels in Europe [28] and pregnancy telemonitoring in community centers and clinics in Zambia [29]. As an example, the interface of Diabetes management is presented in Fig. 2.6.
Figure 2.6 Interface of MobiHealth for Glucose Measurements
CodeBlue consists of wireless infrastructure that is deployed in emergency medical care environments. It integrates low-power wireless sensors, Personal Digital Assistants (PDAs), and Personal Computers (PCs). The wireless sensors integrate Global Positioning System (GPS) to track persons, an Electromyography (EMG) module that checks the health of the muscles for motion capture, and a mote-based pulse oximeter for transmitting periodic packets containing heart rate.
WSN4QoL aims to provide real-life implementations on the pervasive healthcare applications. The applications targeted by this project are for pacemaker, blood pressure, cochlear implant, and electroencephalogram (EEG).
3. Access technologies
In this section, we present three architectures for access technologies. These architectures are summarized in Ref. [30].
3.1. First access technologies architecture
The first architecture is presented in Fig. 2.7. It can be deployed at home, at work, and in hospitals when monitoring healthcare system. The wireless sensors perform measurement and send results to the personal server using Wireless Personal Area Network (WPAN) technologies such as IEEE 802.15.1 [31], IEEE 802.15.3 [32], and IEEE 802.15.4 [33]. The personal server can be a PDA and needs a second wireless interface, in addition to the WPAN interface, in order to connect to the home server. The connection with the home server is performed using Wireless Local Area Network (WLAN) technologies such as IEEE 802.11a/b/g/n.
Figure 2.7 First Access Technologies Architecture: Use of WPAN and WLAN
The home server, called also the central server, has to be connected to the Internet in order to send data to the distant server. This connection has to be secured. Finally, the distant server gathers different information, analyzes it, and displays the final results. The final results can be obtained via web pages in order to facilitate the information access.
The well-known WPAN technology is Bluetooth, which is based on the IEEE 802.15.1 standard. This technology is designed for short range, low cost, and low energy consumption. It operates in the Industrial, Scientific, and Medical (ISM) frequency band of 2.4 GHz, but there are methods to reduce interference [1,34].
IEEE 802.15.3 represents the high rate of WPAN. It provides high throughput and therefore this technology is useful when the sensors need to send images and videos. Moreover, sensors using IEEE 802.15.3 can directly communicate between them and therefore this technology is useful when there is a need to create mesh networks [35]. However, this technology consumes more energy due to the overhead for communication links management [36].
IEEE 802.15.4 represents the low rate of WPAN. It provides low throughput, but enhances the reduction of the energy consumption. Moreover, this technology has sophisticated methods to prevent interference [37]. For example, the bad channels are not used for data transmission such as in Ref. [38] or all channels are used but with different probabilities depending on the channel quality such as in Refs. [1,39,40].
Next, we briefly present the main characteristics of WLAN technologies (see Table 2.2). IEEE 802.11a [1,41] uses orthogonal frequency division multiplexing (OFDM) technique and works in the 5-GHz frequency band. The throughput can reach 54 Mbits/s. The main drawback of this standard is that the signals cannot penetrate obstacles such as walls and solid objects [42,95].
Table 2.2
Main Characteristics of IEEE 802.11a/b/g/n
| Standard IEEE | 802.11a [41,95] | 802.11b [43,102] | 802.11g [44] | 802.11n [45] |
| Frequency (GHz) | 5 | 2.4 | 2.4 | 2.4 and 5 |
| Modulation | OFDM | DSSS | OFDM and DSSS | MIMO–OFDM |
| Throughput (Mbit/s) | 54 | 11 | 54 | 248 |
| Range (m) | 35 | 35 | 35 | 70 |
| Publication date | October 1999 | October 1999 | June 2003 | October 2009 |
| Advantages | Good throughput | Low cost | Good throughput and better obstacle penetration | High throughput and better range |
| Drawbacks | Signals cannot penetrate obstacles | Medium throughput and sensitive to interferences | Wireless congestion |
DSSS, Direct Sequence Spread Spectrum; MIMO, Multiple Input Multiple Output; OFDM, orthogonal frequency division multiplexing.
IEEE 802.11b uses the Direct Sequence Spread Spectrum (DSSS) technique and works in the 2.4-GHz frequency band. The throughput can reach 11 Mbits/s [46]. Note that the use of Carrier Sense Multiple Access/Collusion Avoidance (CSMA/CA) for physical channel access decreases the throughput [102]. Moreover, CSMA/CA is not fair [47]. The main advantage of IEEE802.11b is the low cost and so this technology is widely deployed. However, the throughput is not high and as the frequency band of 2.4 GHz is used, signals are very sensitive to interferences.
IEEE 802.11g uses both OFDM and DSSS techniques and works in the 2.4-GHz frequency band. The throughput can reach 54 Mbits/s. Like IEEE 802.11b, the use of CSMA/CA decreases the throughput. The main advantage of this technology is in its high throughput and the interoperability with the widely used standard IEEE 802.11b [48]. However, as the frequency band of 2.4 GHz is used by many types of equipment such as IEEE 802.11b wireless cards, phones, microwave ovens, and baby monitors, IEEE 802.11g suffers from wireless congestion [49].
IEEE 802.11n uses OFDM and Multiple Input Multiple Output (MIMO) techniques and works in the 2.4- and 5-GHz ISM frequency bands. MIMO enables the opportunity to spatially resolve multipath signals [50]. The throughput can reach 600 Mbits/s when using four antennas for transmission and four antennas for reception. Note that 4 × 4 is the maximum MIMO configuration allowed in IEEE 802.11n.
3.2. Second access technologies architecture
The second architecture is presented in Fig. 2.8. This architecture utilizes only WPAN technologies and therefore it does not need the personal server. Therefore, this technology is the cheapest as it reduces the number of devices needed. The wireless sensors send measurement results directly to the home server using WPAN technologies. However, the wireless sensors require more energy to access the home server as they have to increase Radio-Frequency (RF) output power. The increase of the RF power can also cause more collisions and hence more retransmissions. The increase of the retransmissions degrades the Quality of Service (QoS) and consumes more energy as the same data are transmitted several times.
Figure 2.8 Second Access Technologies Architecture: Use of WPAN Only
3.3. Third access technologies architecture
The third architecture is presented in Fig. 2.9. As in the first architecture, the wireless sensors send measurement results to the personal server. Then, the personal server gathers and forwards data to the home server. Unlike in the first architecture, the connection between the personal server and the home server is performed using Wireless Wide Area Network (WWAN) technologies such as 2G, 2.5G, 3G, and 4G. Finally, the home server, connected to the Internet, sends data to the distant server.
Figure 2.9 Third Access Technologies Architecture: Use of WPAN and WWAN
Now, we briefly describe some WWAN technologies. The second generation (2G) of mobile networks started to be deployed in the beginning of the 1990s. The main 2G mobile network, and the most successful, by far, is the GSM [51,102]. The services were limited to voice and Short Message Service (SMS).
The so-called two-and-half generation (2.5G or 2G+) such as General Packet Radio Service (GPRS) [52] and Enhanced Data Rates for GSM Evolution (EDGE) [53,102] added packet data services and increased data rate. This generation is mainly used for Internet-style access and email. The theoretical maximum rate in the GPRS system is 115 Kbps while the EDGE system provides better theoretical maximum rate (up to 384 Kbps) [102].
Nevertheless, users need wireless high-speed Internet access. Moreover, users want to be able to access the Internet from a large area. The 3G system can support multimedia, data, video, and other services along with voice. The main 3G systems are the UMTS [54] and CDMA2000 [55,102]. The first deployment of CDMA2000 and UMTS took place in 2000–01.
Yet, the 3G systems are still in evolution. The first data rates were in the magnitude of 1 Mbps. Nowadays much higher data rates are expected in both uplink and downlink with the High Speed Downlink Packet Access (HSDPA) and the High Speed Uplink Packet Access (HSUPA) evolutions [56] (see, eg, Release 7 of UMTS). Apart from the displayed physical data rates, application-level data rates are smaller. For example, in 2007, the HSUPA could reach 1 Mbps for a File Transfer Protocol (FTP) application [57]. More recent versions of HSUPA have higher figures. The next step after 3G is (evidently) 4G or what is also known as Beyond 3G (B3G).
Long-Term Evolution (LTE) defined by third Generation Partnership Project (3GPP) Release 8 in 2008 is a very promising technology providing a high peak data rate of 163 Mbps in a channel bandwidth of 10 MHz and a low latency of 15 ms [58]. The enhancement of LTE, called LTE-Advanced (LTE-A), aims to reach a peak data rate of 1 Gbps in order to have a fourth-generation (4G) access technology. This technology continues to evolve through Release 13 that is planned to be completed in March 2016 although some features will be added [59]. This release includes advanced features such as supporting Advanced three Band Carrier Aggregation (three in Downlink/one in Uplink).
The different access technologies can be evaluated using experimental tests or simulation tools. Note that a web-based simulation tool, proposed in Ref. [60], provides a simulation environment with Network Simulation 2 (NS-2) [61] and includes WSN and Bluetooth modules that can be used to practically evaluate different access technologies for WSN networks.
4. Routing strategies
Routing strategies applied to WSN applications for smart cities and homes have to be simple but efficient in order to enhance the QoS supported by these applications. Moreover, it is essential to not propose complex strategies due to the limit capacities of sensors. An intelligent routing strategy is proposed in Ref. [62]. The proposed routing strategy is based on an efficient network discovery. The network discovery uses a multihop routing tree constructed by the Spanning Tree depending on the metric used. The construction of the Spanning Tree is performed using the Prim’s approach [63]. After adding each sink node to the tree, tree edges are then iteratively selected in breadth-first search, based on the defined routing metric until all nodes are added. The proposed routing strategy uses messages for the discovery of the best parent, announcement of the potential parents (sending beacons periodically), association with the parent selected, and acknowledgment when the association is successful (ack is sent by the parent).
The best parent is determined based on the metric used. The choice of the metric depends on the application requirements. Metrics can be, for example, the number of hops, quality of signals, and residual energy. In Ref. [62], the defined metric combines between these three kinds of metrics using Fuzzy Logic (FL) decision approach. One example of the FL approach used in routing in WSN is presented in Ref. [64].
Experiments are run on the I3ASensorBed testbed [65]. This testbed, deployed in the Albacete Research Institute of Informatics, contains 47 nodes to emulate WSN applications in smart cities. Sensors can monitor temperature, humidity, CO2, presence, door and window state (open or close), and energy consumption sensors. The testbed is accessible using a WEB interface to select nodes that perform applications, configure the nodes, and schedule application running. Note that there are other testbeds in the literature such as Wireless Sensor Network Testbeds (Wisebed) [66], Realnet [67], Twist [68], FIRE [69], and Neteye [70].
Experimental results show that the proposed routing strategy outperforms Hop Algorithm (HA) and Received Signal Strength Indicator Algorithm (RSSIA) in terms of energy consumption per delivered data packet (see Fig. 2.10) and packet delivery ratio (see Fig. 2.11). HA and RSSIA favor routes that have minimum number of hops to the receiver and the maximum RSSI of nodes, respectively.
The experimental results of energy consumed per delivered data packet can be explained by the fact that HA takes into account only the number of hops and therefore the number of delivered packet decreases as nodes do not have the same residual energy and the same quality of channel. On the one hand, the Modulation and Coding Scheme (MCS) used depends on the channel quality. An efficient MCS (so more data transmitted) is used when the channel quality is better. On the other hand, when data is forwarded by nodes having low residual energy, these nodes can shut down and so the amount of data transmitted decreases. For routing algorithm RSSIA, when the RSSI is the single parameter in selecting routes, data are forwarded by long routes (having high number of hops) and therefore require more total energy consumed.
The experimental results of the packet delivery ratio can be explained by the fact that HA favors routes having the minimum hops and therefore these hops have poor channel quality. When the channel quality is poor, the MCS used has to be robust and therefore lower number of data packets is transmitted. As RSSIA chooses long routes and consumes more energy, the packet delivery is lower than that provided by the FL strategy that combines several metrics in order to select routes having high quality.
The FL decision approach can also be performed in order to reduce the energy consumed. For example, in Ref. [71], authors propose a routing protocol using the FL approach depending on three parameters: the degree of closeness of node to the shortest path, the degree of closeness of node to sink, and the degree of energy balance. Simulation results show that the proposed routing protocol can reduce the energy consumed by 75% when compared to Greedy Perimeter Stateless Routing (GPSR) [72] and minimum transmission energy (MTE) [73]. These two routing protocols (GPSR and MTE) select routes depending only on the location of neighbors and therefore do not balance energy between nodes.
In Ref. [74], authors proposed an adaptive routing strategy taking into account many individual and environmental criteria in order to reduce the traffic congestion and the environmental impact. For example, an individual can choose a trip that contains several touristic areas or provides a safe route avoiding high-criminality areas.
To reduce the traffic congestion and the environmental impact, the route selection has to consider the system state depending on the characteristics of system areas [75,76]. The pattern used in Ref. [74] and located in the city of Milan (Italy) consists of four layers: traffic, pollution, crimes, and events. The traffic layer contains data about the total number of calls and texts generated over a period of 2 months. The pollution layer is based on data obtained by seven sensors. These sensors perform measurements each hour over the course of the past 2 months. The third layer contains 1276 crimes happened during the past year. Finally, the events layer contains 100,000 geolocated tweets generated over a 1-month period. These different layers aim to combine individual level (eg, crimes layer) and global level (eg, pollution layer). Note that Geographic Information System (GIS) can be used to enhance pattern definition.
As the different individuals do not have the same constraints of route selection, variable coefficients of static and dynamic constraints can be defined. Static constraints correspond to restrictions that do not change over time or change over large temporal scales while dynamic constraints correspond to rapid changes within the system itself such as the traffic flow, weather, or accidents.
Simulation results show that the proposed routing strategy decreases time to reach destination. In addition to the efficient route selection, sensors monitor the state of the city in real time and therefore automatically identify areas that are experiencing a temporary congestion and give authorities the possibility to rapidly take action when crimes or accidents occur.
5. Power-saving methods
Energy management in smart homes and cities can be performed by different methods. In general, the visualization of power consumption reduces the energy consumption of cities and homes between 10% and 30% [77]. For example, optical reader can be used to read the power indicator of a utility revenue meter. In Ref. [78], Gagnon proposed a system that contains three components: a sensor that monitors the cyclical property of the indication and generates information depending on the energy consumption measurements, a transmitter that sends the information obtained, and a remote display that indicates the energy consumed when receiving information via the transmitter. The main drawback of this method is that the information collected is very limited as it depends only on the cyclical property of the indication.
Smart meters can replace optical reader in order to reduce the energy consumption [79]. The smart meters can access the information in real time and transmit data using wired or wireless networks. Evidently, for flexible use of the smart meter, it is recommended that smart meters enable wireless transmitter. However, the main drawback of smart meters is the low data refreshing rate (in general each 15 min) due to the communication constraints.
Traditional electric current and voltage probes can be installed inside the electrical panel of customers [80]. This method can collect high-quality information for the analysis of nonintrusive load monitoring. However, the installation of the measuring probes is difficult as it requires licensed professionals.
In Ref. [81], the authors propose the use of a new energy-saving method based on the magnetic sensor array technique. The measurement device for monitoring home energy use is presented in Fig. 2.12. It is composed of an array of magnetic sensors, an electric panel, and a conduit (containing conductor currents).
Figure 2.12 Measurement Device Based on Magnetic Sensor Array
The main advantage of this method is that the installation is easy as there is no electrical contact to the conductors so a homeowner can easily install it. Moreover, data refreshing rate is high (each 1 s). Once the sensor is calibrated using the Power Line Communication (PLC) scheme [82], it collects the current information and then transmits the information collected to a receiver using 433-MHz low-power RF communication. Note that the PLC technology provides a self-contained system. The receiver is connected to a remote Internet server using Local Area Network (Ethernet) where information is analyzed in order to display results in an Internet Website.
Based also on the PLC scheme, a home energy management system (HEMS) is proposed in Ref. [83]. The proposed system does not require any additional electric construction as power lines are available in cities and houses. In addition to the energy saving in cities and home appliances, the proposed system takes into account the energy usage of the HEMS itself, in order to decrease the total system cost. The architecture of the proposed HEMS is presented in Fig. 2.13.
Sensors, called smart power trips nodes, perform measurements on the power consumption each 10 ms in order to obtain accurate measurements. Note that the accuracy of an electrical energy consumption monitoring system can be improved using a technique proposed in Ref. [84]. This technique is based on a network of Hall-effect wireless sensors attached to the wire in the electrical distribution panel. In order to mitigate the gain errors due to the distance between the sensor and the monitored wire, a single high-precision current transformer sensor has to be added at the main electrical input.
To reduce the energy consumption, the sensors send the average measurements obtained each 1 s. Moreover, they use a low-power wireless communication technology (IEEE 802.15.4) for transmitting results to the sink node. Note that the energy consumption can be enhanced when reducing the amount of the transmission of messages Clear to Send (CTS) [85]. When receiving data from the sensors, the sink node forwards it to a data acquisition component that includes a Data Stream Management System (DSMS) for data processing and data writing in the database. Finally, the results are displayed using web interfaces.
A similar architecture composed of three layers is proposed in Ref. [86] for the purpose of designing a zero-energy home using WSNs. The first layer is the sensors layer where it is essential to investigate the possible avenues of harvesting energy in homes and other buildings and so to efficiently localize the energy harvesting. The second layer is the communication and computation layer where the energy generation is analyzed in order to manage and control the energy system using Information and Communication Technologies (ICT). In order to reduce the energy consumption, sensors should have the ability to decide the actions to take in real time [87]. For example, a smart heating control system was adapted to the available sensors opportunistically in order to reduce the energy consumption while providing the comfort temperature [88]. Note that heating represents a great source of energy consumption. For example, in Switzerland, heating represents 65% of the total consumption in residential buildings [89]. The third layer is the storage layer using the small-scale energy storage devices such as batteries stationed in various locations in houses and cities and therefore the storage devices are distributed.
Finally, in order to evaluate the consumption of energy of the sensor itself, a sophisticated energy model has to be defined. This model has to take into account many energy sources such as:
• the radio transmission and reception energy representing the energy consumed for exchanged information;
• the control access energy representing the energy consumed when waiting the liberation of the physical channel and when exchanging control packets such as Request-To-Send (RTS) and CTS packets;
• the sensor sensing energy enabling the sensor to connect to the physical world for monitoring;
• the sensor logging energy representing the energy consumed for writing information in a database;
• the aggregation energy for processing in order to reduce the quantity of information exchanged and to eliminate redundancy;
• the transient energy representing the dissipated energy when the sensor changes from states (reception, transmission, idle, and sleep states [90]). Note that in the sleep state, the sensor consumes the lowest energy. In the reception and transmission states, the sensor is receiving and transmitting data, respectively. In the idle state, the sensor is awake but it is neither transmitting nor receiving its bursts.
Many energy models were proposed in the literature such as in Refs. [91–95]. The energy sources considerations of the energy models are presented in Table 2.3. Evidently, we recommend using the energy model proposed in Ref. [94] in order to use a realistic model.
Table 2.3
Energy Sources Considerations of Energy Models
| Energy Sources | Model in Ref. [91] | Model in Ref. [92] | Model in Ref. [93] | Model in Refs. [94,102] |
| Radio transmission and reception energy | Yes | Yes | Yes | Yes |
| Control access energy | No | No | No | Yes |
| Sensor sensing energy | No | Yes | No | Yes |
| Sensor logging energy | No | No | No | Yes |
| Aggregation energy | Yes | Yes | Yes | Yes |
| Transient energy | No | No | Yes | Yes |
6. Security
Security in applications for smart homes and cities is almost similar to WSN application environment and therefore the security and privacy concerns are also similar. Security issues in WSNs are presented in Ref. [95] detailing specific and physical attacks as well as security protocols and requirements.
However, the communications in sensor networks applications in smart homes and cities is exposed to more serious threats especially when concerning privacy in homes and vital risks on health of persons. Moreover, in access technologies, many wireless networks can be used simultaneously and thus security in these technologies is a major challenge because of the different characteristics of security architectures used within each wireless network; see Refs. [95–97].
Among the attacks that can be harmful to WSN applications deployed in smart homes and cities, we cite the following:
• Illegal access: A user that does not have authorization may access the system when there is no consistent authentication.
• Data alteration: An attacker may modify the information transmitted in the network. Moreover, it can access the database and alter stocked information.
• Fake data injection: An attacker may insert erroneous information disrupting the system functioning and even causing damages.
• Denial-of-Service (DoS) attacks: An attacker may make the network or equipment unavailable by inserting a great amount of information and/or requests.
• Replay attacks: An attacker may replay the requests sent with false answers.
Attacks can be classified into passive or active. Passive attacks aim to obtain information without performing actions such as alteration and dropping and therefore it is too difficult to detect this kind of attacks. Active attacks are more harmful than passive attacks as they steal and alter information in addition to causing drops and even blocking of system functionality. For example, active attacks on hospital and healthcare WSN applications may conduce to life-threatening situations [98,103].
Note that any efficient and useful application has to consider the security. For example, a general architecture taking into account the security in the medical environment is proposed in Ref. [99]. Furthermore, the security is considered in all healthcare projects. In the CodeBlue project, in addition to the event delivery, filtering, aggregation, and handoff modules, a flexible naming and discovery scheme as well as authentication and encryption procedures is used (see Fig. 2.14 [26]).
Figure 2.14 Main Modules in CodeBlue Project Architecture
In the WSN4QoL project, the middleware layer contains the security block (see Fig. 2.15 [27]). This block monitors the acknowledgment packets exchanged at the network layer in order to identify threats or equipment malfunctioning and instruct the MAC layer to encrypt data.
Figure 2.15 Protocol Stack Used in the WSN4QoL Project Architecture
Therefore, in order to design robust applications, the following security requirements have to be considered:
• Confidentiality: As data is highly confidential especially in smart home applications, it is crucial to ensure the secrecy of data.
• Integrity: It is essential to guarantee that data received was not modified by attackers. This can be insured by hashing and digital signature. Error resilience as well as communication and data reliability are vital for medical applications [23], especially under chirurgical operations and emergency situations.
• Availability: Applications have to offer availability and therefore absolutely avoid system crash due to DoS or Distributed Denial-of-Service (DDoS) attacks. In fact, a lack of availability leads to several problems for saving lives in healthcare applications [100] or fatal accidents when attacking driving management applications in cities [95,101,103].
• Secure localization: Things and persons have to be exactly located for efficient applications without unveiling the localizations for the safety and privacy of users.
• Secure routing information: Routes have to be safe and information has to cross only trusted equipment.
• Intrusion protection and detection: Any network is susceptible to intrusion and therefore prevention and detection techniques have to be deployed. The system has to be well protected and the source of attack has to be determined for further actions [102].
7. Summary
In this chapter, we have presented different wireless sensor networks applications in smart homes and cities. We have focused on applications for energy saving, noise and atmospheric monitoring, and healthcare monitoring. The energy saving is primordial as energy resources have become more and more scarce while the energy use significantly increases. The monitoring of noise, atmospheric, and healthcare aspects certainly makes lives more healthy and comfortable.
To perform these applications, we have presented three access technologies differing in networks and equipment used and therefore presenting various costs and efficiencies. Finally, in order to enhance applications’ performance, we have discussed routing strategies, power-saving methods, and security requirements.
In order to enhance WSN applications for smart cities and homes, it is interesting to consider other challenges and techniques such as the use of IPv6 over Low-power Personal Area Network (6LoWPAN), the determination of efficient services that not only connect sensors to the web but also build a service, and design of software architecture.
..................Content has been hidden....................
You can't read the all page of ebook, please click here login for view all page.