sensors Article
Use of Energy Efficient Sensor Networks to Enhance Dynamic Data Gathering Systems: A Comparative Study between Bluetooth and ZigBee Razvan Andrei Gheorghiu *
ID
and Valentin Iordache
Transports Faculty, Politehnica University of Bucharest, Bucharest 060042, Romania;
[email protected] * Correspondence:
[email protected] Received: 29 April 2018; Accepted: 1 June 2018; Published: 3 June 2018
Abstract: As road traffic conditions worsen due to the constantly increasing number of cars, traffic management systems are struggling to provide a suitable environment, by gathering all the relevant information from the road network. However, in most cases these are obtained via traffic detectors placed near road junctions, thus providing no information on the conditions in between. A large-scale sensor network using detectors on the majority of vehicles would certainly be capable of providing useful data, but has two major impediments: the equipment installed on the vehicles should be cheap enough (assuming the willingness of private car owners to be a part of the network) and be capable of transferring the required amount of data in due time, as the vehicle passes by the road side unit that acts as interface with the traffic management system. These restrictions reduce the number of technologies that can be used. In this article a series of comprehensive tests have been performed to evaluate the Bluetooth and ZigBee protocols for this purpose from many points of view: handshake time, static and dynamic data transfer (in laboratory conditions and in real traffic conditions). An assessment of the environmental conditions (during tests and probable to be encountered in real conditions) was also provided. Keywords: sensor networks; vehicular communications; ZigBee; Bluetooth; Wi-Fi; data acquisition system; sensors
1. Introduction The Internet of Things is a modern concept that allows the connection to the Internet and, hence, remote access and control for many devices that were initially meant to be used locally. This concept, that basically represents the possibility of having a network of different types of devices that are able to work together in an integrated system, may be transferred to the road transport field, supporting the development of new applications, such as dynamic information gathering, or even an integrated concept of Internet of Vehicles. Communication is the key element in the progress of future applications. Even if it is about the data exchange between a vehicle and a road side unit (RSU) with the purpose of informing the driver about the conditions on the road ahead, the data collected by the vehicle along the path that is downloaded into the RSU, or even collaborative vehicle driving, the communication system is the backbone of each development. However, communication in the vehicular environment presents many challenges: signal propagation issues, environmental conditions, Doppler shifts depending on the speed of the receiver relative to the transmitter (or vice versa), on how crowded the communication channels are and not least, all the possible interferences. Each of these represent an issue that may have a negative influence on the communication. Sensors 2018, 18, 1801; doi:10.3390/s18061801
www.mdpi.com/journal/sensors
Sensors 2018, 18, 1801
2 of 17
There are many studies that have tried to develop a new communication system that will address all (or at least most of) these issues using dedicated communication technologies, such as Dedicated Short Range Communication (DSRC). A useful data collection system in a road environment should be able to collect information from many sources via a multitude of sensors, thus being similar to a large sensor network. The main disadvantage of special technologies is the final cost of the system, being based on expensive solutions, especially for large-scale networks. In addition, the operating cost of the system is relatively high, as the main concern is the signal propagation aspect, and not the energy efficiency. The advantages are not able, in many situations, to offset the disadvantages. The approach in this paper starts from the disadvantages presented above, thus, in the development of a large sensor network able to collect meaningful data to be used in traffic management, the focus should be on the costs, but having also in mind the communication reliability. If it is expected to extend such system beyond the usage of the public vehicles (that can be included in the system more easily) to private cars, the solution must be inexpensive and provide enough facilities to drivers to convince them to be a part of the data collection network. The foreseen benefits will be for all the parties involved, as the general traffic management system may obtain more information from the whole network (as opposite of having data collection equipment only in fixed points, like road intersections), but also the drivers may acquire useful information for their trip: incident/accident locations, congested areas, travel speeds, alternative routes, road works, etc. Modern vehicles already have many installed sensors: rain, light, speed, distance to objects in front or in the rear, etc. Therefore, the components that can acquire information are, in many cases, already installed. The key element is the communication system that must be installed to allow data exchange. The main features on the system, from a private vehicle owner point of view, are purchase costs, operation costs, and benefits obtained. In this article, the focus will not be on the applications that may be implemented, or the types of sensors that may be used, as there are many studies for these (like [1,2]), but the technology that may be used (and in what conditions will it function) to support such systems. In Section 2 are presented some brief details about the technologies that were considered. The authors have performed a literature review, trying to find similar analysis for Bluetooth, ZigBee, or even a comparison of the two technologies. In the end is emphasized the main topic of this paper and the original contributions. Section 3 presents all the details about the tests performed: environmental assessment, handshake time, static data transfers and data exchange in motion. In Section 4 the conclusions of these studies are drawn. 2. Background and Related Work There are several wireless technologies that can be used to transmit data from vehicles, with reduced costs, all of them working in the Industrial, Scientific and Medical (ISM) frequency band. Among them we can mention: -
-
Wi-Fi is the most common wireless communication. Its main disadvantage is the considerable number of users (transmission crowdedness) and interference with other communications. This communication is used in almost all office buildings, but it can be activated on the smartphones inside vehicles. In addition, in some modern vehicles Wi-Fi access is provided, multiplying the devices of this kind that can be detected along the road. Bluetooth: is the most common connection method between two portable devices. It is usually found in vehicles, connecting the phone with the audio system, but it’s also used for headset connections. Therefore, a significant number of Bluetooth communications are likely to be found near the road network. Bluetooth’s advantage is the usage of frequency hopping, that continuously search for free channels to be used in data exchange. This ensures the successful data exchange, but the communication time depends heavily on the frequency band congestion.
Sensors 2018, 18, 1801
-
3 of 17
ZigBee: is an uncommon technology to be used for vehicle communications, as it was designed for smart home device networks, being capable of fast data transfers (at low data rates) between a substantial number of devices. ZigBee may use fixed (pre-selected) channels, that may be chosen far enough from usual interferences in 2.4 GHz frequency band. ZigBee uses an interference mitigation technique (frequency agility mechanism [3]), that can be divided into three phases: interference detection, channel evaluation and interference mitigation [4]. The main feature of this technology is its very efficient use of energy [3]. As we will present in the following sections, it is proven to be a very reliable communication, more reliable in fact than Wi-Fi or Bluetooth. Adding the layered network topology and the reduced price of the equipment, we consider it to be the most suitable technology to be used in wireless sensor networks.
Considering all these aspects, a comparison between solutions needs to be performed. Although there are several papers dedicated to Bluetooth analysis (such as [5–7]), or ZigBee evaluation (such as [8,9]), very few compare these technologies (one probable cause may be that they were developed in the beginning for different purposes). ZigBee is an unusual choice for a vehicular communication protocol, or even communication between moving devices, but has great advantages from the power consumption point of view. In addition, none of the studies found that focused on Bluetooth vs. ZigBee that seemed to be representative for the purposes of this article [10–13], included comparative evaluation of the communication aspects involving these two technologies, focusing instead on power consumption or just presenting technical specifications from datasheets. Therefore, an in-depth analysis of all the aspects related to communication between two devices appeared as a requirement for this study. This is the main topic of this paper and the main contribution of the authors to this field. Real-life tests have been performed (not just data taken from datasheets) and the tests made in similar conditions create a complete and real image of the two technologies and possibility to implement them in real traffic applications. There are several types of influences for these technologies, being based on radio signals, but the most influence on the communication are the limitations of the technology used (in terms of distance, data rate, etc.), and interference with other communications in the same frequency band. As the most known (and, therefore, used) solutions are Wi-Fi, Bluetooth, and ZigBee, is seems the analysis should study the influence of each technology on the others. However, previous work [14,15] has revealed minimum influence on Bluetooth over ZigBee and vice-versa, therefore the tests performed had the goal to evaluate these technologies in a Wi-Fi environment. 3. Assessment Tests Performed and Results For the evaluation of possible usage of Bluetooth and ZigBee technologies, we have tested the handshake time (device connection time) and data transfer time for specific amounts of data. Both are important in a dynamic environment, in which vehicles are moving at high speeds and, therefore the devices must be able to connect and transmit the required information very fast. The tests were performed both in a laboratory environment and on a road network. 3.1. Hardware Used Test equipment is based on Arduino Uno v3 development boards, that allow communication with and control of Bluetooth and ZigBee communication modules which were chosen as following due to their low price and high availability on the market. Bluetooth version 4.0 modules HM-10 with a CC2541 chipset were used (Figure 1), having the following relevant specifications [16–19]: -
Data rate: up to 2 Mbps. RF power: up to 6 dBm (5 mW). Electric current consumption: in active mode—8.5 mA, and in sleep mode—400 µA~1.5 mA.
Sensors 2018, 18, 1801
4 of 17
Sensors 2018, 18, x
4 of 17
-
Power supply: 3.3 V DC, 50 mA. Wireless range: up to 100 m. Wireless range: up to 100 m. According to the Bluetooth network specifications, one of the modules was set as Master and According to the Bluetooth network specifications, one of the modules was set as Master and the the other one as Slave, with search for a pair and connection establishment being set to automatically. other one as Slave, with search for a pair and connection establishment being set to automatically. It is to be underlined that Bluetooth v4 was used in these tests due to the fact that it was the state It is to be underlined that Bluetooth v4 was used in these tests due to the fact that it was the state of the art Bluetooth technology available at the beginning of the research project. Although during of the art Bluetooth technology available at the beginning of the research project. Although during the the meantime the Bluetooth v5.0 specifications were made available for the market, test modules were meantime the Bluetooth v5.0 specifications were made available for the market, test modules were not not available until recently. Some aspects regarding Bluetooth v5 were tested in [20], such as available until recently. Some aspects regarding Bluetooth v5 were tested in [20], such as throughput throughput against distance or influence of obstacles, or in [21] for RSSI against distance. However, against distance or influence of obstacles, or in [21] for RSSI against distance. However, it must it must be mentioned that the tests performed revealed only static behaviour (but not in Wi-Fi be mentioned that the tests performed revealed only static behaviour (but not in Wi-Fi interference interference conditions), and future assessments have to be performed to evaluate the improvements conditions), and future assessments have to be performed to evaluate the improvements of Bluetooth of Bluetooth v5 against older versions in sensor networks, using similar conditions to the ones v5 against older versions in sensor networks, using similar conditions to the ones defined in this paper. defined in this paper. For the ZigBee technology, XBee Series 2 with Wire Antenna modules were used (Figure 1), having For the ZigBee technology, XBee Series 2 with Wire Antenna modules were used (Figure 1), the following relevant specifications [3,22]: having the following relevant specifications [3,22]: Data rate: up to 250 kbps. Data rate: up to 250 kbps. -RF power: power: up up to to 33 dBm dBm (2 (2 mW). mW). RF -Electric current current consumption: consumption: in in active active mode—40/45 mode—40/45mA, mA,and andininsleep sleepmode—1 mode—1μA. µA. Electric -Power supply: supply: 3.3 3.3 V V DC. DC. Power -Wireless range: up to 120 m. Wireless up to 120 According to the ZigBee network specifications, one of the modules was set as Coordinator and the other one as End-Device. An XBee Shield was used as an interface between the module and the Arduino board board (Figure (Figure 1). 1).
Figure 1. Hardware components.
To create create interference interference in in the the 2.4 2.4 GHz GHz spectrum spectrum on on specific channels the To specific channels the authors authors inserted inserted between between the test modules a Wi-Fi router and a computer, and transferred large files with speeds up to 70 the test modules a Wi-Fi router and a computer, and transferred large files with speeds up to 70 Mbps, Mbps, the maximum maximum data datarate ratethat thatcould couldbebeachieved achievedwith with the equipment used. The router manually the the equipment used. The router waswas manually set set to the desired channel and bandwidth. to the desired channel and bandwidth.
3.2. Wireless Channels Assessment In this section, all wireless technologies used are analyzed in terms of interference between each channels will be used in the tests. To other to determine which channels To test the hypothesis hypothesis about the congested Wi-Fi environment environment we we have have performed several tests in the Bucharest road network (Bucharest being Wi-Fi the capital city of Romania). The city is ranked in the 5th position among the most crowded cities in worldtop top according to Tom-Tom [23], making it one of thetobest places to test field the world according to Tom-Tom [23], making it one of the best places test field communications communications in a crowded environment least, considering all the mentioned communication issues in a crowded environment (at least, considering(at all the communication issues in Section 2). mentioned in Section several 2). In this environment, several junctions and public transport routes haveonbeen In this environment, junctions and public transport routes have been selected, based the selected, based on the probability fortoother communications to be or present (Wi-Fi, Bluetooth or probability for other communications be present (Wi-Fi, Bluetooth ZigBee). The test sites were ZigBee). The test sites were chosen near blocks of flats, or office buildings to maximize the data that could be collected. The aim of this test was to evaluate the communication density and to assess the
Sensors 2018, 18, 1801
5 of 17
Sensors Sensors2018, 2018,18, 18,xx
55of of17 17
chosen near blocks of flats, or office buildings to maximize the data that could be collected. The aim signal for Wi-Fi channel. The presented Figure 22 signal and in [24] were of this power test was toeach evaluate the communication density and toin the power for signal power for each Wi-Fi channel. The results results presented inassess Figure and published published ineach [24]Wi-Fi were meant to set the reference for the following interference tests between Wi-Fi and other channel. Theset results Figure and published in [24] were meant to set the reference the meant to the presented reference infor the 2 following interference tests between Wi-Fi and for other communication technologies. following interference tests between Wi-Fi and other communication technologies. communication technologies.
Figure Figure 2. Wi-Fi channel density [24]. Figure2. 2.Wi-Fi Wi-Fichannel channeldensity density [24]. [24].
From on the road network, both for junctions and public transport routes, itit Fromthe the tests testsperformed performed routes, From the tests performed on on the theroad roadnetwork, network,both bothfor forjunctions junctionsand andpublic publictransport transport routes, resulted that, overall, the most used Wi-Fi channels are 1, 11 is practice due that, overall, thethe most used Wi-Fi channels areare 1, 6, 6,1,and and 11 which which is aa common common practice due itresulted resulted that, overall, most used Wi-Fi channels 6, and 11 which is a common practice to the fact they are not overlapping (as detailed in [25,26] and seen in Figure 3) and interference from to the fact they are not overlapping (as detailed in [25,26] and seen in Figure 3) and interference from due to the fact they are not overlapping (as detailed in [25,26] and seen in Figure 3) and interference nearby devices can avoided, but the of is increasing, and nearby devices can be becan avoided, but as asbut the number number of devices devices is continuously continuously increasing, and routers routers from nearby devices be avoided, as the number of devices is continuously increasing, and implement modern algorithms to switch from one channel to another in case of congested implement modernmodern algorithms to switch fromfrom one one channel to to another congested routers implement algorithms to switch channel anotherinincase case of of congested communications, so changes may appear this pattern and Wi-Fi communications tend to communications,so sochanges changesmay mayappear appearinin in this pattern and Wi-Fi communications to exceed exceed communications, this pattern and Wi-Fi communications tendtend to exceed the the typical spectrum usage and extend to other, less congested, frequencies, although they are the typical spectrum usage and extend to other, less congested, frequencies, although they are typical spectrum usage and extend to other, less congested, frequencies, although they are overlapping overlapping each other. overlapping each other. each other.
Figure Figure3. 3.Wi-Fi Wi-Fichannel channelallocation. allocation. Figure 3. Wi-Fi channel allocation.
The Themain maininfluence influenceof ofthis thisWi-Fi Wi-Fispectrum spectrumis ison onBluetooth Bluetoothtechnology, technology,as asit ituses usesthree threeadvertising advertising The main influence of this Wi-Fi spectrum is on Bluetooth technology, as it uses three advertising channels and two of them are overlapping Wi-Fi channels 1 and 6 (as detailed in [27] and channels and and two two of of them them are are overlapping overlapping Wi-Fi Wi-Fi channels channels 11 and and 66 (as (as detailed detailed in in [27] [27] and and seen seen in in channels seen in Figure 4). Figure 4). Figure 4). The great advantage of this technology is that it uses Adaptive Frequency Hopping, a technique that automatically selects the communication channel according to the detected degree of interference. ZigBee, as opposed to Bluetooth, may communicate on a specific channel, making it (at least in theory) sensitive to other communications in the same frequency band. For the evaluation of this interference, we considered the ZigBee channels 12, 17 and 22, which are closest to the center frequency of the Wi-Fi channels 1, 6 and 11, respectively, and clearly overlap with them (as detailed in [28] and seen in Figure 5). In addition to these, we have tested ZigBee channels 25 (0X19) and 26 (0x1A), to verify that they have the least probability of being influenced by any of the selected Wi-Fi channels.
Figure 3. Wi-Fi channel allocation.
The main influence of this Wi-Fi spectrum is on Bluetooth technology, as it uses three advertising channels Sensors 2018, 18,and 1801two of them are overlapping Wi-Fi channels 1 and 6 (as detailed in [27] and seen in 6 of 17 Figure 4). Sensors 2018, 18, x
6 of 17
Figure 4. Bluetooth vs. Wi-Fi channel allocation.
The great advantage of this technology is that it uses Adaptive Frequency Hopping, a technique that automatically selects the communication channel according to the detected degree of interference. ZigBee, as opposed to Bluetooth, may communicate on a specific channel, making it (at least in theory) sensitive to other communications in the same frequency band. For the evaluation of this interference, we considered the ZigBee channels 12, 17 and 22, which are closest to the center frequency of the Wi-Fi channels 1, 6 and 11, respectively, and clearly overlap with them (as detailed in [28] and seen in Figure 5). In addition to these, we have tested ZigBee channels 25 (0X19) and 26 (0x1A), to verify that they have the least probability of being influenced by any of the selected Wi-Fi Figure 4. Bluetooth vs. Wi-Fi channel allocation. channels.
Figure5.5.ZigBee ZigBee vs. vs. Wi-Fi Figure Wi-Fi channel channelallocation. allocation.
3.3. Handshake Time Tests
3.3. Handshake Time Tests
The first tests that have been performed for the evaluation of each technology employed The first tests that have been performed for the evaluation of each technology employed measuring measuring the handshake time. Connection time evaluation was performed under laboratory the handshake time. Connection time evaluation was performed under laboratory conditions, inside conditions, inside a building, in an open space, with direct line of sight between the emitter and a building, an open space, direct line of sight between emitter and receiver. This allowed receiver. in This allowed thewith monitoring and control of the the radio environment, especially the thecommunication monitoring and control of the radio environment, especially the communication that provide that provide potential interferences. potential interferences. The environment was assessed using Wi-Fi Analyzer application on a smartphone (Figure 6). The environment assessed using Wi-Fi Analyzer application on a smartphone (Figure 6). Typical environmentalwas noise was considered as one with Wi-Fi communications evenly distributed Typical noise was considered with Wi-Fi communications distributed on theenvironmental specific channels, covering the whole as 2.4one GHz spectrum. This seemed to beevenly an appropriate onsimulation the specific covering theinwhole GHz spectrum. This seemed to beinan appropriate of channels, the real road network, which2.4 unknown communications may occur every point, simulation of the real(from road network, whichpoint unknown communications may occur in every point, without any control the sensor in network of view) of the channels used. An example of this environment is presented in Figure in whichpoint the channel numbers are represented horizontal without any control (from the sensor6network of view) of the channels used. on Anthe example of this axis and theissignal strength, measured in dBm,the onchannel the vertical one. are represented on the horizontal environment presented in Figure 6 in which numbers The tests have been performed in the following scenarios: axis and the signal strength, measured in dBm, on the vertical one.
tests have been performed in the following scenarios: - The Typical, uncontrolled, environmental noise. Heavy traffic on Wi-Fi channels 1, 6 and respectively 11. The router was set to each channel, with Typical, uncontrolled, environmental noise. a 40 MHz bandwidth, to consider the worst-case scenario that can occur. A 40 MHz wide channel Heavy traffic on Wi-Fi channels 1, 6 and respectively 11. The router was set to each channel, will double the use of the available Wi-Fi spectrum, leaving fewer non-overlapping channels with a 40 MHz bandwidth, to consider the worst-case scenario that can occur. A 40 MHz wide available for Bluetooth or ZigBee and worsening the interference problem. channel will double the use of the available Wi-Fi spectrum, leaving fewer non-overlapping channels available for Bluetooth or ZigBee and worsening the interference problem.
Sensors 2018, 18, 1801
7 of 17
Sensors 2018, 18, x
7 of 17
Sensors 2018, 18, x
7 of 17
Figure 6. An example of evenly distributed Wi-Fi communications.
Figure 6. An example of evenly distributed Wi-Fi communications. Figure 6. An example of evenly distributed Wi-Fi communications.
The evaluation started with the test devices put to sleep. An additional wired connection between emitter and receiver to signalput both to additional wake up. The connection was done TheThe evaluation started withwas the used test devices tomodules sleep. An wired connection between evaluation started with the test devices put to sleep. An additional wired connection as shown in Figure 7. used to signal both modules to wake up. The connection was done as shown emitter and receiver was between emitter and receiver was used to signal both modules to wake up. The connection was done in Figure 7. in Figure 7. as shown Wake-Up Button Wake-Up Button
Arduino Development Arduino Board Development Board
Bluetooth / ZigBee Module Bluetooth / ZigBee Module
Communication Cable Communication Cable Wireless Router Wireless Router
Wireless Communication Wireless Communication
Arduino Development Arduino Board Development Board
Bluetooth / ZigBee Module Bluetooth / ZigBee Module
Computer Computer
Figure 7. Handshake time measurement test bed. Figure7.7.Handshake Handshake time time measurement Figure measurementtest testbed. bed.
Sensors 2018, 18, 1801
8 of 17
Two moments were considered, the difference between them representing the handshake time: -
-
The moment when the button is pressed, considered as reference. This will power up the modules that will connect with each other and the Master/Coordinator module will start transmitting time stamp messages to the Slave/End-device module. The moment when the first message is received and displayed by the Slave/End-device module. Sensors 2018, 18, x 8 of 17
Both communication modules require a wake-up timethem which should not be considered Two moments were considered, the difference between representing the handshake time: as a part of the handshake and needs to be subtracted from the obtained measurements. According to their The moment when the button is pressed, considered as reference. This will power up the modules specifications, that the will value is 504 µseach forother the Bluetooth module [16] and 13.2will msstart fortransmitting the ZigBeetime module [29]. connect with and the Master/Coordinator module messages to the Slave/End-device Therefore,stamp the handshake time will be: module. -
The moment when the first message is received and displayed by the Slave/End-device module.
th_Bluetooth = atSlave − ttime − 504 µs not be considered as a part Both communication modules require wake-up which should Master of the handshake and needs to be subtracted from the obtained measurements. According to their specifications, the value is 504 µ s= fort the Bluetooth module [16] and 13.2 ms for the ZigBee module th_ZigBee End− Device − tCoordinator − 13.2 ms [29]. Therefore, the handshake time will be: the Bluetooth technology, handshake time was determined for each considered
(1) (2)
For scenario, the distance between the modules being modified and the 𝑡ℎ_𝐵𝑙𝑢𝑒𝑡𝑜𝑜𝑡ℎ = 𝑡𝑆𝑙𝑎𝑣𝑒 between − 𝑡𝑀𝑎𝑠𝑡𝑒𝑟 − 0 504 μs 5 m, with 1-m step. Increasing (1) distance over 5 m led to an increase in handshake time that made this communication technology 𝑡ℎ_𝑍𝑖𝑔𝐵𝑒𝑒 = 𝑡𝐸𝑛𝑑−𝐷𝑒𝑣𝑖𝑐𝑒 − 𝑡𝐶𝑜𝑜𝑟𝑑𝑖𝑛𝑎𝑡𝑜𝑟 − 13.2 ms (2) unsuitable forFor data exchange between moving devices, in the event of strong interference from the the Bluetooth technology, handshake time was determined for each considered scenario, the considereddistance Wi-Fi between channels. Average measurement values are presented in Table 1 and Figure 8. the modules being modified between 0 and 5 m, with 1-m step. Increasing the distance over 5 m led to an increase in handshake time that made this communication technology unsuitable forTable data exchange between moving time devices, in the values event of(milliseconds). strong interference from the 1. Bluetooth handshake average considered Wi-Fi channels. Average measurement values are presented in Table 1 and Figure 8.
Distance (m) 0 Distance (m) 1 2 3 4 5
0 1 2 3 4 5
Environmental Wi-Fi Channel 1 Wi-Fi Channel 6 Wi-Fi Channel 11 Table 1. Bluetooth handshake time average values (milliseconds). Noise Interference Interference Interference Environmental 284.8856 Noise 294.484 284.8856 280.2852 294.484 277.894 280.2852 284.582 277.894 284.582 325.2228 325.2228
Wi-Fi Channel 1 268.8632 Interference 619.2584 268.8632 470.3724 619.2584 628.2776 470.3724 3840.1712 628.2776 3840.1712 8558.9332 8558.9332
Wi-Fi Channel 6 270.4752 Interference 657.5296 270.4752 277.1484 657.5296 297.224 277.1484 5840.5988 297.224 5840.5988 16,323.9836 16,323.9836
Wi-Fi Channel 11 279.6896 Interference 282.4408 279.6896 295.7724 282.4408 270.3096 295.7724 277.8568 270.3096 277.8568 8938.8316 8938.8316
Figure 8. Bluetooth handshake time.
Figure 8. Bluetooth handshake time. With environmental interference present, which usually can be described as low Wi-Fi traffic, the handshake time has an average value of about 290 milliseconds. From Figure 8 it can be observed With environmental interference present, which usually can be described as low Wi-Fi traffic, that handshake time is not significantly influenced by Wi-Fi traffic on any of the tested channels, if the handshake time has an average value of about milliseconds. Figure 8 it can be observed the distance between the modules is small, up to 3290 m, the average value From being around 380 milliseconds. Forsignificantly greater distances, Wi-Fi channel has thetraffic higheston influence handshake that handshake time is not influenced by 6Wi-Fi any ofover the the tested channels, if the
distance between the modules is small, up to 3 m, the average value being around 380 milliseconds. For greater distances, Wi-Fi channel 6 has the highest influence over the handshake time, but the other
Sensors 2018, 18, 1801
9 of 17
channels will affect it too, determining average values which will make this technology difficult to implement if the considered applications will need a free communication environment at all times. The influence of specific Wi-Fi channels (1, 6, 11) occurs due to the initial connection for Bluetooth protocol, that used three advertising channels (presented in Figure 4) that are close to, or even overlap, Wi-Fi channels 1, 6, and 11, respectively. When heavy traffic is induced on these channels, the initial connection between Bluetooth devices is delayed. However, for communications disturbed only by environmental noise (as it would be likely the case in real life), the handshake time remains about the Sensors 2018, 18, x 9 of 17 same for all tests. time, buttime the other will affect it too, determining average values which willBluetooth, make this with the Handshake was channels determined for the same considered scenarios as for technology difficult to implement if the considered applications will need a free communication exception that the modules were only placed at 0 m from each other. Few tests were made for distances environment at all times. The influence of specific Wi-Fi channels (1, 6, 11) occurs due to the initial up to 5 m and no significant differences between measurements were observed. connection for Bluetooth protocol, that used three advertising channels (presented inAverage Figure 4) measurement that close to, or in even overlap, channels 1, 6, in andTable 11, respectively. When heavy traffic is12, induced values areare presented Table 2. Wi-Fi It can be seen 2 that ZigBee channels 17, 22 and 25, on these the (Figure initial connection between Bluetooth devices is delayed. However, interference. for that overlap Wi-Fichannels, channels 5), are easily influenced by Wi-Fi environmental communications disturbed only by environmental noise (as it would be likely the case in real life), This behaviour is accentuated if heavy Wi-Fi traffic is generated. the handshake time remains about the same for all tests. Handshake time was determined for the same considered scenarios as for Bluetooth, with the 2. ZigBee handshake time (milliseconds). exception that Table the modules were only placed at 0average m fromvalues each other. Few tests were made for distances up to 5 m and no significant differences between measurements were observed. Average Wi-Fi 1 inWi-Fi 6 Wi-Fi Channel measurement values Environmental are presented in Table 2. ItChannel can be seen TableChannel 2 that ZigBee channels 12, 17,11 22 ZigBee Channel Noise Interference and 25, that overlap Wi-Fi channels (Figure 5), are easily Interference influenced by Wi-FiInterference environmental interference. is accentuated if16,059.7 heavy Wi-Fi traffic is generated. 12 This behaviour 649.7
17 22 25 ZigBee 26 Channel
7477.6
39,063.4
Table 2. ZigBee handshake time average values (milliseconds). 5010.6
4721 Environmental 23.2 Noise
5431.11 Wi-Fi Channel 22.8 Interference 16,059.7
Wi-Fi5630.4 Channel 6 22.6 Interference
14,349.3 12,565.711 Wi-Fi Channel 22.7 Interference
12 649.7 17 7477.6 39,063.4 5010.6 that ZigBee is also influenced by the nearest Wi-Fi 14,349.3 From Table222 it is obvious channels: in normal 25 4721 5431.1 5630.4 12,565.7 operating conditions there are among Wi-Fi channels (considering 26 23.2 evenly distributed 22.8 communications 22.6 22.7
that auto-configurable routers properly select non-congested channels to improve performance, From Table 2 it is obvious that ZigBee is also influenced by the nearest Wi-Fi channels: in normal resulting in a relatively even communication distribution among the whole spectrum). But in special operating conditions there are evenly distributed communications among Wi-Fi channels interference conditions, the routers wereproperly configured induce thechannels most influence, (considering that when auto-configurable routers select to non-congested to improvethe values have significantly increased. The only exception is channeldistribution 26, which is thethemost performance, resulting in a relatively even communication among wholeprotected, spectrum). being the Butin inthe special interference conditions,and when the routers to induce the influence, furthest one frequency spectrum, which seemwere notconfigured to be influenced by most Wi-Fi at all. Therefore, the values have significantly increased. The only exception is channel 26, which is the most protected, it is obvious that for relevant results, this is the one that should be used in future tests. being the furthest one in the frequency spectrum, and which seem not to be influenced by Wi-Fi at To compare Bluetooth and ZigBee in the results, same conditions, have been all. Therefore, it is obvious that for relevant this is the one handshake that should betests used in future tests.performed for ZigBee channel 25, in environmental noise. The behaviour is relatively as the values To compare Bluetooth and ZigBee in the same conditions, handshake tests haveconstant, been performed for ZigBee channel 25, 23.83 in environmental noise. The behaviour is relatively constant, as the varies between 23.20 ms and ms. This is presented in Figure 9, where the red linevalues represents the varies between 23.20 ms and 23.83 ms. This is presented in Figure 9, where the red line represents the average value. average value.
Figure 9. ZigBee handshake time.
Figure 9. ZigBee handshake time.
Sensors 2018, 18, 1801
10 of 17
Sensors 2018, 18, x
10 of 17
A A comparison comparison between between Bluetooth Bluetooth and and ZigBee ZigBee handshake handshake time time measurements measurements is is presented presented in in Figure 10. It can be seen that Bluetooth handshake is faster than the ZigBee’s one for all Figure 10. It can be seen that Bluetooth handshake is faster than the ZigBee’s one for all scenarios scenarios with exception, ZigBee ZigBeechannel channel2626which whichisisfurthest furthest from Wi-Fi channels. channel, with one one exception, from Wi-Fi channels. ForFor thisthis channel, the the handshake time is ten times smaller than Bluetooth lowest one, and it seems to be the right solution handshake time is ten times smaller than Bluetooth lowest one, and it seems to be the right solution for for implementing implementing ZigBee ZigBee communications. communications.
Figure 10. Bluetooth and ZigBee handshake time. Figure 10. Bluetooth and ZigBee handshake time.
3.4. Data Trasfer Time 3.4. Data Trasfer Time Data transfer tests represented the second stage in technology evaluation. The connection time Data transfer tests represented the second stage in technology evaluation. The connection time is is important, especially when taking into consideration moving devices, due to short interval the important, especially when taking into consideration moving devices, due to short interval the emitter emitter and receiver are in each other’s communication range. But after the devices are connected and receiver are in each other’s communication range. But after the devices are connected data transfer data transfer is the main parameter that can reveal the quantity of data that can be exchanged, with is the main parameter that can reveal the quantity of data that can be exchanged, with a fair probability a fair probability of error-free communication. Data transfer tests were performed in two cases: static of error-free communication. Data transfer tests were performed in two cases: static tests, that had tests, that had the goal to determine the maximum communication distance, and dynamic tests, that the goal to determine the maximum communication distance, and dynamic tests, that intended to intended to evaluate the capability of devices to exchange a certain amount of data when moving at evaluate the capability of devices to exchange a certain amount of data when moving at different different speeds against each-other. speeds against each-other. 3.4.1. Static Static Tests Tests 3.4.1. Static tests tests were were performed performed in in laboratory laboratory conditions, conditions, inside inside aa building, building, in in open open space, space, with with direct direct Static line of sight between the emitter and receiver modules. Similar to connection time assessment, the line of sight between the emitter and receiver modules. Similar to connection time assessment, the data data transfer tests considered the following scenarios: transfer tests considered the following scenarios:
---
Environmental across the the 2.4 2.4 GHz GHz spectrum). spectrum). Environmental background background noise noise (communications (communications spread spread across Heavy traffic on Wi-Fi channels 1, 6, and 11. Heavy traffic on Wi-Fi channels 1, 6, and 11.
For each scenario, three situations have been addressed regarding the length of the message: For each scenario, three situations have been addressed regarding the length of the message: 256, 512 and 1024 bits, considering the message length analysis from [17] from which it resulted that 256, 512 and 1024 bits, considering the message length analysis from [10] from which it resulted that these values may be used in vehicle communications, being suitable for data exchange, depending these values may be used in vehicle communications, being suitable for data exchange, depending on on the application. the application. The evaluation took into consideration 100 consecutive message exchanges. For each of them, The evaluation took into consideration 100 consecutive message exchanges. For each of them, the emitter will send a message, of a specific length, and enter a listen mode. The receiver will the emitter will send a message, of a specific length, and enter a listen mode. The receiver will evaluate evaluate the message and, if it will be correct, it will return the same message. If it will be wrong, it the message and, if it will be correct, it will return the same message. If it will be wrong, it will send will send a different message back, to trigger a data resend. The emitter, being in listen mode, will a different message back, to trigger a data resend. The emitter, being in listen mode, will wait for the wait for the response. If a resend will be necessary, it will send the initial message again. If the
Sensors 2018, 18, 1801
11 of 17
Sensors 2018, 18, x
11 of 17
response. If a resend will be necessary, it will send the initial message again. If the response will be correct, timebeelapsed the first message to this was calculated, as a two-way responsethewill correct,since the sending time elapsed since sending themoment first message to this moment was correct communication. calculated, as a two-way correct communication. The tests tests have havebeen beenperformed performedwith with the modules already connected to each other. The main The the modules already connected to each other. The main goal goal of this evaluation was to test the maximum distance for which the communication of this evaluation was to test the maximum distance for which the communication can can stillstill be be performed. Although from previous data seemstotobebeirrelevant irrelevanttototest test Bluetooth Bluetooth for for distances performed. Although from previous data it it seems distances greater handshake time may cause impossibility for data transfer), therethere are some greater that that55mm(as (asincreased increased handshake time may cause impossibility for data transfer), are applications that does notdoes require instead distance is key parameter as downloading some applications that not speed, require speed,theinstead the distance is key(such parameter (such as data from busses at from the end of theatroute). As of thethe goal of thisAs research was testresearch all possible downloading data busses the end route). the goal of to this wasconditions, to test all we included these tests in the evaluation. possible conditions, we included these tests in the evaluation. The distance distance between the modules modules was modified between for the the Bluetooth Bluetooth ones ones and and The between the was modified between 00 and and 15 15 m m for between 0 and 50 m for the ZigBee ones, with 5-m step. Placing the Bluetooth modules at a distance between 0 and 50 m for the ZigBee ones, with 5-m step. Placing the Bluetooth modules at a distance greater has made it almost impossible to greater than than 15 15 m, m, and andZigBee ZigBeemodules modulesatatover over50 50mmfrom fromeach eachother, other, has made it almost impossible transfer messages. Test bedbed setup is presented in Figure 11.11. to transfer messages. Test setup is presented in Figure
Arduino Development Board
Wireless Router
Arduino Development Board
Wireless Communications Bluetooth / ZigBee Module
Bluetooth / ZigBee Module
Computer
Figure test bed. bed. Figure 11. 11. Static Static data data transfer transfer time time measurement measurement test
As resulted from the handshake time tests, ZigBee channel 26 was used for the corresponding As resulted from the handshake time tests, ZigBee channel 26 was used for the corresponding message transfer time measurement. message transfer time measurement. Test results are presented in the Figures 12 and 13. It should be mentioned that, since no Test results are presented in the Figures 12 and 13. It should be mentioned that, since no significant significant interference was determined between ZigBee channel 26 and considered Wi-Fi channels, interference was determined between ZigBee channel 26 and considered Wi-Fi channels, the authors the authors did not test these scenarios for distances greater than 25 m and use as reference the values did not test these scenarios for distances greater than 25 m and use as reference the values obtained in obtained in presence of the environmental background noise. presence of the environmental background noise. In Figures 12 and 13 it can be noticed that the message transfer time is relatively constant for In Figures 12 and 13 it can be noticed that the message transfer time is relatively constant for ZigBee modules with values around 60 milliseconds (this was to be expected due to lack of ZigBee modules with values around 60 milliseconds (this was to be expected due to lack of overlapping overlapping between channels) and increasing with the distance for the Bluetooth ones to over 1 s, between channels) and increasing with the distance for the Bluetooth ones to over 1 s, with channel 1 with channel 1 having the greatest influence. However, even considering this increase, Bluetooth having the greatest influence. However, even considering this increase, Bluetooth allowed messages to allowed messages to be transferred successfully. be transferred successfully.
Sensors 2018, 18, 1801
12 of 17
Sensors 2018, 18, x Sensors 2018, 18, x
12 of 17 12 of 17
Figure 12. Bluetooth and ZigBee message transfer time (256 bits message/Environmental and Wi-Fi Figure 12. Bluetooth and ZigBee message transfer time (256 bits message/Environmental and Wi-Fi Figure Bluetooth and ZigBee message transfer time (256 bits message/Environmental and Wi-Fi channel12. 1 interference). channel 1 interference). channel 1 interference).
Figure 13. Bluetooth and ZigBee message transfer time (256 bits message/Wi-Fi channel 6 and 11 Figure 13. Bluetooth and ZigBee message transfer time (256 bits message/Wi-Fi channel 6 and 11 interference). Figure 13. Bluetooth and ZigBee message transfer time (256 bits message/Wi-Fi channel 6 and 11 interference). interference).
In Figures 14 and 15 it can be noticed that the message transfer time for ZigBee modules In Figures 14 and 15constant it can be noticed the 150 message transferBluetooth time for message ZigBee modules continues to be relatively with valuesthat around milliseconds. transfer In Figures 14 and 15 it can be noticed that the message transfer time for ZigBee modules continues continues to be relatively constant with values around 150 milliseconds. Bluetooth time has the same behavior as for previous tested message length, reaching values message of almosttransfer 3 s. For to be has relatively constant withasvalues around tested 150 milliseconds. Bluetooth message transfer time 3has the time the same behavior for previous message length, reaching values of almost s. For Bluetooth, it becomes difficult to successfully transfer considered messages as the distance between same behavior as for previous tested messagetransfer length, considered reaching values of almost s. For Bluetooth, Bluetooth, becomes difficult to successfully messages as the3distance between modules isitincreasing. it becomes difficult to successfully transfer considered messages as the distance between modules modules is increasing. is increasing.
Sensors 2018, 18, 1801
13 of 17
Sensors 2018, 18, x Sensors 2018, 18, x
13 of 17 13 of 17
Figure 14. Bluetooth and ZigBee message transfer time (512 bits message/Environmental and Wi-Fi Figure 14. Bluetooth and ZigBee message transfer time (512 bits message/Environmental and Wi-Fi Figure Bluetooth and ZigBee message transfer time (512 bits message/Environmental and Wi-Fi channel14. 1 interference). channel 1 interference). channel 1 interference).
Figure 15. Bluetooth and ZigBee message transfer time (512 bits message/Wi-Fi channel 6 and 11 Figure 15. Bluetooth and ZigBee message transfer time (512 bits message/Wi-Fi channel 6 and 11 interference). Figure 15. Bluetooth and ZigBee message transfer time (512 bits message/Wi-Fi channel 6 and 11 interference). interference).
In Figures 16 and 17 it can be noticed that the message transfer time for ZigBee modules In Figures 16 and 17constant it can be noticed the 190 message transfer but time ZigBee modules continues to be relatively with valuesthat around milliseconds, thefor maximum distance In Figures 16 and 17 it can be noticed that the message transfer time for ZigBee modules continues continues constant with values around 190 milliseconds, maximum distance decreased to tobe 45relatively m. Bluetooth message transfer time continues to havebut thethe same behavior as for to be relatively constant with values around 190 milliseconds, buttothe maximum distance decreased decreased to 45 m. Bluetooth message transfer time continues have the same behavior asmfor previous tested message lengths, reaching values of more than 5 s. For distances greater than 10 it to 45 m. Bluetooth message transfer time continues to have the same behavior as for previous tested previous tested message reaching values ofdeliver more than 5 s. For distances greater than 10 m it is a very difficult task forlengths, Bluetooth to successfully the message under heavy interference. message values of more than 5 s. For distances greater than 10heavy m it isinterference. a very difficult is a very lengths, difficult reaching task for Bluetooth to successfully deliver the message under task for Bluetooth to successfully deliver the message under heavy interference.
Sensors 2018, 18, 1801
14 of 17
Sensors 2018, 18, x
14 of 17
Sensors 2018, 18, x
14 of 17
Figure 16. Bluetooth and ZigBee message transfer time (1024 bits message/Environmental and Wi-Fi Figure 16. Bluetooth and ZigBee message transfer time (1024 bits message/Environmental and Wi-Fi channel16. 1 interference). Figure Bluetooth and ZigBee message transfer time (1024 bits message/Environmental and Wi-Fi channel 1 interference). channel 1 interference).
Figure 17. Bluetooth and ZigBee message transfer time (1024 bits message/Wi-Fi channel 6 and 11 interference). Figure 17. Bluetooth and ZigBee message transfer time (1024 bits message/Wi-Fi channel 6 and 11 Figure 17. Bluetooth and ZigBee message transfer time (1024 bits message/Wi-Fi channel 6 and 11 interference). interference). In conclusion, Bluetooth will allow for exchange of messages but only for low moving speeds or
In conclusion, Bluetooth will allow forofexchange ofthe messages but onlyequipment. for low moving speeds or for vehicles that will spend short periods time near infrastructure ZigBee, on the Invehicles conclusion, Bluetooth will allow forofexchange messages but only for low moving for that achieve will spend short periods time near the infrastructure equipment. ZigBee,speeds on the or other hand, will successful communication at of higher moving speeds, smaller messages being hand, willwill achieve successful communication at higher moving speeds, smaller messages recommended and for non-critical applications. forother vehicles that spend short periods of time near the infrastructure equipment. ZigBee,being on the recommended and for non-critical applications. other hand, will achieve successful communication at higher moving speeds, smaller messages being 3.4.2. Dynamic Tests recommended and for non-critical applications. 3.4.2. Dynamic Tests The tests in motion were performed on a side road with very low traffic flows, in an industrial 3.4.2. Dynamic Tests motion were performed on test a side road with low was traffic flows, in industrial area,The andtests within clear line of sight between the modules. Onevery of them deployed inan a fixed point area, and with clear line of sight between the test modules. One of them was deployed in a fixed point and the other one was installed on a car. Urban environment was assessed, with car speeds between The tests in motion were performed on a side road with very low traffic flows, in an industrial and the50 other one was installed on a the car.data Urban environment was assessed, with car speeds 10 and km/h. For each exchange wasOne tested vehicle coming and out area, and with clear line of evaluation sight between the test modules. ofwith themthe was deployed in ainbetween fixed point 10 andother 50 km/h. For evaluation the data exchange tested with the vehicle coming in and out ofthe the road side communication range and the was results were published by the in [30]. and one module waseach installed on a car. Urban environment was assessed, with carauthors speeds between the50 road sideZigBee module communication range andand thedifferent results were published bywere thecoming authors inand [30]. The same channel 26 was considered message lengths tested. in The useout 10 of and km/h. For each evaluation the data exchange was tested with the vehicle The same ZigBee channel 26 was considered and different message lengths were tested. The use of 256 bit messages in this mobile environment led to acceptable message transfer times (Figure 18). of the road side module communication range and the results were published by the authors in [30]. of 256 bit messages in this mobile environment led to acceptable message transfer times (Figure 18).
Sensors 2018, 18, 1801
15 of 17
The same ZigBee channel 26 was considered and different message lengths were tested. The use of 256Sensors bit messages 2018, 18, x in this mobile environment led to acceptable message transfer times (Figure 15 of 17 18). Sensors 2018, 18, x 15 of 17 Regarding greater length messages (512 and 1024 bits), as seen from previous results they led to Regarding greater length amessages (512data andtransfer, 1024 bits), as seen from previous results led to and, difficulties in accomplishing successful especially for Bluetooth, and tothey increased Regarding in greater length messages (512data andtransfer, 1024 bits), as seenfor from previous results they led to difficulties accomplishing successful Bluetooth, and toworsened increased and, in most of the cases, unsuitableamessage transfer time. especially Obviously, these problems when the difficulties in accomplishing a successful data transfer, especially for Bluetooth, and to increased and, in most of the cases, unsuitable message transfer time. Obviously, these problems worsened when data transfer was attempted between two modules moving towardsthese or away fromworsened each other, so these in most the cases, message two transfer time.moving Obviously, when the data of transfer was unsuitable attempted between modules towards orproblems away from each other, so two types of messages were no longer considered because very few or none of them were successfully the data was attempted between two modules moving towards away from of each other, so these twotransfer types of messages were no longer considered because very or few or none them were transmitted at first tests. At last, to increase the success rate of the message transfer, a smaller length these two types of messages were no longer considered because very few or none of them were successfully transmitted at first tests. At last, to increase the success rate of the message transfer, a was considered (128 bits message, Figure 19) keeping in mind that in many situation smaller amounts successfully transmitted at first tests. At last, to increase the success rate of the message transfer, smaller length was considered (128 bits message, Figure 19) keeping in mind that in many situationa smaller length was bits message, Figure 19)shorter keeping in mindshorter that inmessages. many situation of data needamounts to be transferred orto it(128 is feasible to split into smaller of considered data need be transferred or it isitfeasible to splitmessages. it into smaller amounts of data need to be transferred or it is feasible to split it into shorter messages.
Figure Averagemessage messagetransfer transfer time messages [30]. Figure 18.18. Average timefor for256 256bitbit messages [30]. Figure 18. Average message transfer time for 256 bit messages [30].
Figure 19. Average message transfer time for 128 bit messages source: [30], with extended data Figure 19. Average message transfer time for 128 bit messages source: [30], with extended data included. Figure 19. Average message transfer time for 128 bit messages source: [30], with extended included.
data included. As in previous tests, ZigBee provided more constant results, with average message transfer in previous tests, more constant results, with average transfer timesAs below Bluetooth, andZigBee with a provided communication range significantly larger. The message final conclusions times below Bluetooth, and with a communication range significantly larger. The final conclusions are presented in the next section. As in previous tests, ZigBee provided more constant results, with average message transfer times are presented in the next section.
below Bluetooth, and with a communication range significantly larger. The final conclusions are 4. Conclusions presented in the next section. 4. Conclusions Handshake time measuring was the first step in communication evaluation. The capability of
Handshake time measuring firststarting step in from communication evaluation. capability of 4. Conclusions the devices to quickly connect to was eachthe other, a sleep mode (that willThe increase energy the devicesisto quicklyinconnect to each other, starting from a sleep mode (thatperformed will increase energy efficiency) essential a dynamic environment like road traffic. From the tests it resulted
Handshake time measuring was environment the first steplike in communication evaluation. The capability efficiency) essential dynamic road traffic. From tests performed it resultedof the that ZigBeeischannel 26inisaundisturbed by other communications, while the Bluetooth, with its frequency devices to quickly connect to each other, starting from a sleep mode (that will increase energy efficiency) that ZigBee channel 26 is undisturbed other communications, while Bluetooth, frequency hopping approach, might encounter by congested conditions, thus delaying the with first its step of the is essential in aapproach, dynamic might environment likecongested road traffic. From thethus testsdelaying performed resulted ZigBee hopping encounter conditions, theitfirst step that of the
Sensors 2018, 18, 1801
16 of 17
channel 26 is undisturbed by other communications, while Bluetooth, with its frequency hopping approach, might encounter congested conditions, thus delaying the first step of the communication process. This result influenced the following test conditions, as from that moment, for ZigBee only channel 26 was used. Bluetooth, due to its specific protocol, cannot avoid interference by default. Static and dynamic data transfer tests have proved again that ZigBee has the following advantages against Bluetooth v4: -
Higher communication range Lower times for data transfer A more uniform set of values.
Considering the recent implementation of Bluetooth v5, the necessity to re-run these tests to properly evaluate the improvement of the new technology in similar conditions must be emphasized. All environmental conditions for the field tests were assessed using specific applications (as described earlier) to ensure the validity of these results. From all these we may conclude that ZigBee technology may provide a valuable support for large-scale energy efficient sensor networks. The next steps will be to evaluate critical messages transfer parameters to extend the possible usage of this technology in vehicular environment. Author Contributions: All the authors contributed equally to this work. Conflicts of Interest: The authors declare no conflicts of interest.
References 1. 2.
3. 4.
5. 6. 7.
8. 9.
10.
11.
12.
Płacze, B. Selective data collection in vehicular networks for traffic control applications. Transp. Res. Part C Emerg. Technol. 2012, 23, 14–28. [CrossRef] Lopes, J.; Bento, J.; Huang, E.; Antoniou, C.; Ben-Akiva, M. Traffic and Mobility Data Collection for RealTime Applications. In Proceedings of the 2010 13th International IEEE Annual Conference on Intelligent Transportation Systems, Madeira Island, Portugal, 19–22 September 2010. ZigBee Alliance. ZigBee Specifications. Available online: http://www.zigbee.org (accessed on 20 May 2018). Mohd, A.S.; Mohd, S.A.; Lo, A.; Rozeha, A.R. Experimental Studies of the ZigBee Frequency Agility Mechanism in Home Area Networks. In Proceedings of the IEEE 39th Conference on Local Computer Networks Workshops (LCN Workshops), Edmonton, AB, Canada, 8–11 September 2014. [CrossRef] Phan, R.C.W.; Mingard, P. Analyzing the Secure Simple Pairing in Bluetooth v4.0. Wirel. Personal Commun. 2012, 64, 719–737. [CrossRef] Ferrigno, L.; Paciello, V.; Pietrosanto, A. A Bluetooth-Based Proposal of Instrument Wireless Interface. IEEE Trans. Instrum. Meas. 2005, 54, 163–170. [CrossRef] Mikhaylov, K.; Tervonen, J. Multihop data transfer service for Bluetooth Low Energy. In Proceedings of the 3th International Conference on ITS Telecommunications (ITST), Tampere, Finland, 5–7 November 2013. [CrossRef] Soijoyo, S.; Ashari, A. Analysis of Zigbee Data Transmission on Wireless Sensor Network Topology. Int. J. Adv. Comput. Sci. Appl. 2017, 8, 145–151. [CrossRef] Zhang, F.; Wang, F.; Dai, B.; Li, Y. Performance evaluation of IEEE 802.15.4 beacon-enabled association process. In Proceedings of the 22nd International Conference on Advanced Information Networking and Applications, Okinawa, Japan, 25–28 March 2008; pp. 541–546. Rajeesh, K.; Bhuvana, C.; Anushya, S. Comparison of Zigbee and Bluetooth Wireless Technologies Survey. In Proceedings of the 2017 International Conference on Information, Communication & Embedded Systems (ICICES 2017), Chennai, India, 23–24 February 2017. Idris, Y.; Muhammad, N.A. A Comparative Study of Wireless Communication Protocols: Zigbee vs. Bluetooth. In Proceedings of the 33rd Annual Conference of the IEEE Industrial Electronics Society (IECON 2007), Taipei, Taiwan, 5–8 November 2007. [CrossRef] Ekström, M.C.; Bergblomma, M.; Lindén, M.; Björkman, M.; Ekström, M. Comparison Study of ZigBee and Bluetooth with Regards to Power Consumption, Packet-Error-Rate and Distance. Available online: https://www.diva-portal.org/smash/get/diva2:574502/FULLTEXT01.pdf (accessed on 20 May 2018).
Sensors 2018, 18, 1801
13.
14. 15.
16. 17.
18. 19. 20.
21. 22.
23. 24.
25. 26.
27. 28. 29. 30.
17 of 17
Lee, J.S.; Su, Y.W.; Shen, C.C. A comparative study of wireless protocols: Bluetooth, UWB, ZigBee, and Wi-Fi. In Proceedings of the 33rd Annual Conference of the IEEE Industrial Electronics Society (IECON), Taipei, Taiwan, 5–8 November 2007. NXP Laboratories. Co-Existence of IEEE 802.15.4 at 2.4 GHz. Application Note. 2013. Available online: http://www.nxp.com/documents/application_note/JN-AN-1079.pdf (accessed on 20 May 2018). Sikora, A.; Groza, V.F. Coexistence of IEEE802.15.4 with other Systems in the 2.4 GHz-ISM-Band. In Proceedings of the 2005 IEEE Instrumentation and Measurement Technology Conference, Ottawa, ON, Canada, 16–19 May 2005; pp. 1786–1791. Texas Instruments Incorporated. 2.4-GHz Bluetooth™ Low Energy and Proprietary System-on-Chip; CC2541 Datasheet, SWRS110D; Texas Instruments Incorporated: Dallas, TX, USA, January 2012; Revised June 2013. Gheorghiu, R.A.; Iordache, V.; Cormos, C.A. Cooperative Communication Network for Adaptive Truck Platooning. In Proceedings of the 3rd International Conference on Vehicle Technology and Intelligent Transport Systems, VEHITS 2017, Porto, Portugal, 22–24 April 2017; Volume 1, pp. 228–235. [CrossRef] JNHuaMao Technology Company. Bluetooth 4.0 BLE Module Datasheet; JNHuaMao Technology Company: Jinan, China, 2014. Tinysine Electronics. Tinysine Bluetooth 4.0 BLE Module, version 1.1.; User Manual; Tinysine Electronics: Hefei, China, 2015. Yaakop, M.B.; Malik, I.A.A.; Bin Suboh, Z.; Ramli, A.F.; Abu, M.A. Bluetooth 5.0 throughput comparison for Internet of Thing usability. A survey. In Proceedings of the International Conference on Engineering Technology and Technopreneurship (ICE2T), Kuala Lumpur, Malaysia, 18–20 September 2017. [CrossRef] Pau, G.; Collotta, M.; Maniscalco, V. Bluetooth 5 Energy Management through a Fuzzy-PSO Solution for Mobile Devices of Internet of Things. Energies 2017, 10, 992. [CrossRef] Hou, J.; Chang, B.; Cho, D.K.; Gerla, M. Minimizing 802.11 Interference on Zigbee Medical Sensors. In Proceedings of the Fourth International Conference on Body Area Networks, Los Angeles, CA, USA, 1–3 April 2009; Article No. 5. Available online: http://corporate.tomtom.com/releasedetail.cfm?releaseid=1012517 (accessed on 20 May 2018). Gheorghiu, R.A.; Iordache, V.; Badescu, I. Analysis of Possible Wi-Fi Interferences of Wireless Communications Implemented in Vehicular Environments. In Proceedings of the 2017 13th International Conference on Advanced Technologies, Systems and Services in Telecommunications (TELSIKS), Nis, Serbia, 18–20 October 2017; pp. 244–247. Yang, D.; Xu, Y.; Gidlund, M. Wireless coexistence between IEEE 802.11- and IEEE 802.15.4-based networks: A survey. Int. J. Distrib. Sens. Netw. 2011, 7. [CrossRef] Liang, C.J.M.; Priyantha, N.B.; Liu, J.; Terzis, A. Surviving Wi-Fi interference in low power ZigBee networks. In Proceedings of the 8th ACM Conference on Embedded Networked Sensor Systems (SenSys’10), Zurich, Switzerland, 3–5 November 2010; ACM: New York, NY, USA, 2010; pp. 309–322. Microchip Technology Inc. Wireless Communications. 2018. Available online: http://microchipdeveloper. com/wireless:start (accessed on 20 May 2018). Sahoo, P.K.; Pattanaik, S.R.; Wu, S.-L. A reliable data transmission model for IEEE 802.15. 4e enabled wireless sensor network under WiFi interference. Sensors 2017, 17, 1320. [CrossRef] [PubMed] Digi International Inc. XBee® /XBee-PRO® ZB. Product Datasheet, 2008–2011. Available online: https: //cdn.sparkfun.com/datasheets/Wireless/Zigbee/ds_xbeezbmodules.pdf (accessed on 20 May 2018). Iordache, V.; Gheorghiu, R.A.; Minea, M.; Cormos, A.C. Field Testing of Bluetooth and ZigBee Technologies for Vehicle-to-Infrastructure Applications. In Proceedings of the 2017 13th International Conference on Advanced Technologies, Systems and Services in Telecommunications (TELSIKS), Nis, Serbia, 18–20 October 2017; pp. 248–251. © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
