The development of the smart city (SC) paradigm relies on the need for more interconnected public Intelligent Transportation Systems (ITSs). In fact, nowadays, smart cyber physical systems in the transportation domain are expected to play an important role in the ambition to develop passenger-centric services. Like other utilities, transport infrastructures are slowly moving forward to more intelligent, connected, user-centric and collaborative systems. This movement is partly supported by the increasing availability of low-cost smart objects with wireless interconnection capabilities and wireless indoor positioning systems. A significant revolution has been envisaged in the transportation domain by the introduction of these low-cost elements with their wireless interconnection capabilities mostly in the near-field environment. In fact, cars, trains, buses, bicycles and road infrastructures are becoming increasingly equipped with sensors, RFID tags and NFC devices, sending critical information to the traffic control centers to better route traffic and to provide users with real-time relevant transportation information.
New research, engineering methods, tools and simulation studies for this cyber-physical—and near-field—scenario in the transportation domain have to be developed. This chapter is outlined in the context of modeling techniques to explore this challenging Big Data or Internet of Things (IoT) scenario in the ITS domain. Simulation modeling is a necessary and crucial step in the design, development, test and performance evaluation of any communication network implementation or strategy. To provide a global performance study of the IoT environment in a specific transportation scenario, a system level simulation approach is necessary. Traditionally, system level simulators receive results from link level simulation studies as an input. However, well-known discrete event system level network simulators have so far not included near-field wireless communication behavior despite the fact that these low-cost smart devices are widespread and are relevant actors who impact and stress the communication infrastructures.
Our goal, in this chapter, is to provide guidelines on modeling these smart low-cost near-field wireless objects and on how to integrate their behavior in traditional network Discrete Event Simulation (DES) tools. The ultimate aim is to provide an insight into the available tools in order to study—with an overall perspective—their behavior and their impact on the access and core communication infrastructures of ITS.
This chapter is structured as follows. Section 2.1 provides an overview of near-field wireless technologies, introducing the near-field and far-field concepts together with a taxonomy of the near-field wireless technologies which can be found in the transportation domain. Section 2.2 introduces the two main existent techniques for the characterization of the near-field communication link. This link level characterization traditionally serves as an input to system level simulation techniques based on DES tools. Section 2.3 presents the state of the art of these DES frameworks and their different approaches to near-field modeling for performance evaluation purposes. The last section covers the main conclusions and further research opportunities.
This section deals with the definition of near-field compared to far-field in communications with the aim of providing the reader with a better technical understanding of these concepts. Then, the most relevant near-field wireless technologies that are being used today for a number of transport applications are detailed.
Near-field and far-field concepts are related to the generation of an electromagnetic field in the area surrounding an antenna, where an alternate current flowing through a conductor loop mainly generates a magnetic field (H), and an alternate current flowing through a conductor dipole mainly generates an electric field (E). As these fields (H or E) propagate, an electromagnetic field is created (a field composed of electric and magnetic fields). The interaction between these fields creates an electromagnetic wave able to travel into space.
Depending on the distance from the source, the field that surrounds an antenna can be broken up into two segments: near-field and far-field. Typically, near-field is defined as the field around the antenna up to λ/2π away, where λ is the wavelength. After this point, the electromagnetic wave begins to separate from the antenna and therefore, the ability to interact by inductive or capacitive coupling is lost, reaching the far-field zone after a transition zone. Figure 2.1 shows the different zones and their names.
Near-field and far-field have different energies so they typically require a corresponding antenna type because the near-field applications primarily employ the magnetic field, while the far-field has both electric and magnetic components. When moving from near-field to far-field, the wave impedance varies, being constant at a value of 377Ω in the far-field region as shown in Figure 2.2, together with the different zones.
The closest region to the antenna is known as near-field reactive. In this region, the most important characteristic is the presence of a dominant magnitude. Depending on the physical characteristics of the antenna, one field (electric or magnetic) will prevail over the other. More specifically, in the case of two rectangular loops, the coupling may be characterized taking just the magnetic field into account.
The region beyond two wavelengths is called the far-field, and in this case the electric (E) and magnetic (H) fields support and regenerate one another as their strength decreases inversely as the square of the distance. Table 2.1 shows the distance of EM field separation (λ/2π) for several frequencies associated with communication systems.
Table 2.1. Value of λ/2π (cm) for different wavelengths
Working in near-field or in far-field differs as summarized as follows:
- – the coupling between two antennas in the near-field (reactive) is similar to the AC transformer, whereas the far-field coupling is commonly described as RF communication;
- – in near-field communications, the received signal mainly depends on the characteristics of the source. In far-field links, the most relevant parameter is the communication channel;
- – inductive coupling is mainly determined by the relative distances of transmitter and receiver parts, whereas radiation is usually more closely related to differences in propagation time and/or path.
As mentioned, wireless communications (via the antenna) occur using a process known as electromagnetic coupling. There are two types of coupling:
- – inductive: a near-field antenna uses inductive coupling which means that it uses a magnetic field (H) or an electric field (E). A magnetic (H) or electric (E) field is created in the near-field region that allows the antenna to transmit the signal;
- – capacitive: a far-field antenna uses capacitive coupling (or propagation coupling). Capacitive coupling occurs when the signal sent by the antenna propagates and an electromagnetic signal is available.
In contrast to far-field antennas which transmit a propagating electromagnetic field, a near-field antenna generates a local magnetic (H) or electric (E) field suitable for short read-range applications. Near-field antennas are not very commonly used and therefore there are limited options available in the market. On the other hand, far-field antennas present a wide variety of shapes and sizes offering a larger coverage than the near-field antennas. Many options are available regarding far-field antennas such as linear or circular polarization, varying gain, indoor or outdoor use, and multi-band; many wireless technologies are based on far-field communications, such as Zigbee and Bluetooth.
Specific near-field wireless communication technologies have been used for a long time in transport systems such as trains, undergrounds and trams.
In this sense, balises are very common devices present in many Automatic Train Protection (ATP) systems. Balises’ role is to increase safety and avoid collisions or other kinds of accidents like derailments due to high speed. Balises are powerless devices located on the track and they are telepowered by the train when the train passes by. Once the balise is active, it transmits information—an inductive communication—to the Balise Transmission Module (BTM) of the train to warn the driver or even to stop the train if the driver is not acting as expected. Balises are not only a fundamental part of many national ATP systems such as the Spanish ASFA or the German PZB, but they are also part of the most recent and modern ATP systems such as the European Train Control System (ETCS)—where the balise is called Eurobalise [UNI 12]—or Communications-Based Train Control (CBTC).
Euroloop [UNI 08] is another example of inductive technology used in ETCS. It is quite similar to a balise. In fact, it is an extension of the Eurobalise over a particular distance to be able to continuously transmit data to the Loop Transmission Module (LTM) of the vehicle over cables emitting electromagnetic waves in a similar way to other national systems such as the German LZB or Thales’ Euroloop.
Apart from these traditional railway technologies, transportation systems are also embracing the SC and IoT paradigms. In fact, in [GUB 13], the authors introduce the concept of smart transportation and smart logistics as an area of applicability of IoT concepts. Some of the services pointed out in [ZAN 14] are related to these domains such as traffic management and smart parking. Furthermore, in [ATZ 10], the authors add the following services and applications: assisted driving, mobile ticketing, monitoring environmental parameters and augmented maps.
The number of communication technologies that can be used in the SC and IoT domains is huge and it depends on the physical characteristics of the devices and the specific use case [ZAN 14]. Some commonly used communication technologies are wired (e.g. Ethernet, Fiber Optic), whereas others are wireless (e.g. WiFi, UMTS, LTE). Only two of those wireless technologies can be considered inductive: RFID and NFC.
In fact, nowadays RFID and NFC are two of the most common near-field wireless technologies and, thus, they are detailed in the following subsections.
Radio Frequency IDentification (RFID) [WAN 06] is an automatic wireless data collection technology with a long history [LAN 05]. This technology is usually employed to identify items by means of radio waves. The basic composition of an RFID system covers a tag and a reader. The reader sends an interrogating signal to the tag, and the tag responds with its unique information. RFID tags are classified into Active, Semi-Passive or Passive:
- – Active RFID tags contain a battery and therefore they rely on their own power source. As a result, the active tag can be read by signals up to 100 meters. This long read range makes active RFID tags ideal for many industries where asset location and other improvements in logistics are important. Active tags may be either read-only or read/write, thus allowing data modification by the reader. Other benefits such as data storage and faster data rates make this kind of tag very appropriate for electronic toll collection. There is an endless variety of tags. This kind of tag is the most expensive one;
- – Semi-passive RFID tags are similar to passive tags in using the reader signal to provoke a response from the tag, and similar to active tags in containing a battery to power all the electronics of the tag itself. Usually, the semi-passive tag presents a longer operating life in terms of power supply compared to active tags, but on the other hand these tags have some of the limitations of the passive tag in terms of slow read speeds and short read distances. The price of semi-passive tags is lower than the active tags and higher than the passive tags;
- – Passive RFID tags do not contain any power source. Instead, the energy employed to power these tags is the energy of the electromagnetic signal sent by the RFID reader. Therefore, passive tags depend on the reader’s RF signal to respond. Usually, passive RFID tags have a read range from near contact and up to 25 meters. Currently, the most employed type of RFID tag is the passive one. In general, their design is simpler and does not contain a battery. The tag may be used in many applications thanks to the different forms it can take, ranging from identification cards for public transportation to tags embedded in license plates for car identification. This type of tag is the cheapest one. In a passive RFID system, the reader transmits a modulated RF signal to the tag consisting of an antenna and an integrated circuit chip. The chip receives power from the antenna and responds by varying its input impedance and thus modulating the backscattered signal. There were functional passive RFID systems already being reported in the early 1970s [KOE 75]. Since then, RFID has advanced [FIN 04, KAR 03, GLI 04, DEV 05] and has experienced tremendous growth.
RFID tags primarily operate at three frequency ranges. These frequency ranges also set the type of communication that can be employed:
- – Low Frequency (LF) 125–134 kHz;
- – High Frequency (HF) 13.56 MHz;
- – Ultra-High Frequency (UHF) 856–960 MHz.
Low-frequency (LF, 125–134 KHz) and high-frequency (HF, 13.56 MHz) [EPC 13] RFID systems are short-range systems based on inductive coupling between the reader and the tag antennas through a magnetic field (near-field).
Ultra-high frequency (UHF, 860–960 MHz) [EPC 13] and microwave (2.4 GHz and 5.8 GHz) RFID systems are long-range systems which use electromagnetic waves propagating between reader and tag antennas (far-field). EPC provides a specification for this kind of RFID.
There are many RFID standards depending on the application RFID is intended to be used for. DIN, ISO and VDE are some normalization bodies offering these standards.
Near-Field Communication (NFC) is a specific subset of High-Frequency (HF) RFID. NFC allows for the secure exchange of data. Moreover, an NFC device acts as an NFC reader and NFC tag. Therefore, NFC devices are able to communicate peer-to-peer.
NFC devices operate at the same 13.56 MHz frequency as HF RFID readers and tags. The standards and protocols of the NFC format deal with the use of RFID in proximity cards and are based on RFID standards [ISO 16, ISO 13].
As a finely honed version of HF RFID, near-field communication devices have taken advantage of the short read-range limitations of its radio frequency. Because NFC devices must be in close proximity to each other, usually no more than a few centimeters, it has become a popular choice for secure communication between consumer devices such as smartphones.
Peer-to-peer communication is a feature that sets NFC apart from typical RFID devices. An NFC device is able to act both as a reader and as a tag. This unique ability has made NFC a popular choice for contactless payment, a key driver in the decision by influential players in the mobile industry to include NFC in newer smartphones. Also, NFC smartphones pass along information from one smartphone to the other by tapping the two devices together, which turns sharing data such as contact information or photographs into a simple task. Recently, you may have seen advertising campaigns that use smart posters to pass information to the consumers.
Also, NFC devices can read passive NFC tags, and some NFC devices are able to read passive HF RFID tags that are compliant with ISO 15693. The data on these tags can contain commands for the device such as opening a specific mobile application. You may start seeing HF RFID tags and NFC tags more frequently in advertisements, posters and signs as it is an efficient method to pass along information to consumers.
In effect, NFC builds upon the standards of HF RFID and turns the limitations of its operating frequency into a unique feature of near-field communication.
Currently two different approaches can be distinguished regarding the characterization of near-field communications, namely theoretical calculations based on electrical models and electromagnetic simulators. Although the latter also rely on the electrical models, the way of working with both of them differs. For the electrical models, mathematical equations are derived, and therefore a huge mathematical background is required. On the other hand, simulators allow us to graphically model the elements to be characterized and the computation is done by the simulator itself, in the time domain or in the frequency domain.
As a result of the characterization, a transfer function of the near-field communication is obtained. This transfer function can be employed in the DES framework as the model of the link level near-field zone between the communication devices.
2.2.1. Electrical models
Figure 2.3 shows the basic electric model used for inductive loops [DOB 12]. It represents a simplified model of a transformer, incorporating the particular elements found in this field.
In this figure, L1 and L2 are the inductances of each coil, UQ2 is the voltage induced in the coil 2, R1 and R2 are the ohmic losses of each coil, Cp is the parasitic capacitance of loop 2 and C2 is the tuning capacitor. C2 is used to provide the appropriated resonant frequency.
One of the key points of the model is the analysis of RL. In [CHA 10] the RL of an UHF RFID passive tag is presented, showing the differences in near- and far-fields. In the case of many commercially available devices, when the tag is located in the near-field the received antenna power increases, therefore the system includes some kind of nonlinear control circuit to avoid potential damages. Another interesting model is shown in [JAN 11], including a simple description of each building block.
In any case, the performance of the complete system is conditioned by the ability of the emitting antenna to induce voltage into the passive loop. The next point presents the analysis of one of the most frequent cases used in RFID communications.
Taking into account the Biot-Savart law [2.1], the magnetic field created by any segment of the loop at any point of the space can be calculated as follows:
where represents a differential element of the conductor carrying a current I, the unit vector in the direction of the straight line between the differential element and the point of interest, and r is the distance between these items.
[2.1] allows us to compute the magnetic field created by a rectangular loop at any point of the space. In the case of rectangular loops, it is very important to know the z-component of the field.
The differential field created by is given by:
The integral of expression [2.5] over the EF segment gives the total field created:
The field created by the rest of the elements of the loop can be calculated in the same way. Finally, expression [2.7] will enable the calculation of the vertical component of the magnetic field at any point.
In order to characterize the interaction between two rectangular loops, the situation shown in Figure 2.5 can be considered. It is considered a distance Δx between the centers of both loops, and a perfect alignment between their longitudinal axes.
In order to calculate the voltage induced in loop ABCD, we must express the magnetic flux created by loop EFGH through loop ABCD as a function of time. We can achieve this goal combining expressions [2.8].
The magnetic flux will be given by [2.9]:
The final step is to express the time-variant magnitudes of [2.9] as a function of time. In the particular case of loops for railway applications, there are two sources of time variations:
- – the current of loop EFGH, usually given by a sine function;
- – the center distance Δx, which may be expressed as a function of train speed.
In any case, operator [2.10] will give the final expression:
The most powerful tool available today for electromagnetic analysis is the computer simulator. There are several commercial packages capable of performing accurate 3D simulations, taking into account different boundary conditions. These platforms can replicate virtually any physical system. However, there are two main drawbacks: on the one hand, the use of computational resources may be high even for simple models, and on the other hand, it may be difficult to relate the performance obtained with the basic design parameters.
It is also possible to combine the computer’s computation capacity with the equations mentioned in the previous section. For example, BTM antennas are defined as rectangular loops, and therefore, the analysis of this particular system may be carried out by applying the basic electromagnetic equations to the system as shown in section 2.2.2. With this approach a greater insight into the physical problem is obtained, and as a result, the conclusions obtained can be used to optimize the system with quantitative and qualitative rules. In order to simplify the expressions, a symbolic mathematical solver can be used. This methodology requires less computational resources than the conventional electromagnetic simulator but it shows some constraints due to the simplifications required in the mathematical analysis.
Currently, electromagnetic simulation software is a tool that supports communication designers, obtaining accurate predictions for more complex structures than two-faced rectangular loops. There are a number of different electromagnetic analysis programs that differ from each other in a number of different underlying technologies. Each simulation technology offers particular benefits, and therefore solving a specific problem type requires the use of one particular type of electromagnetic simulator that best suits the problem.
Computer-Aided Engineering (CAE) software has only been used for around 25 years, although currently it is one of the key parts of the design process. Nowadays, efficient and powerful personal computers are capable of running highly computationally demanding CAE programs in a reasonable time. CAE tool developers have taken advantage of this computer performance improvement which has resulted in the availability of unprecedented levels of simulation capability. This is a significant advantage in the field of electromagnetic simulators since the problem sizes associated with solving Maxwell’s equations can be quite large. Nevertheless, CAE simulation tools’ performance constraints are generally not the speed of the simulation engine, but the accuracy or availability of the models employed for the simulation. Usually, designs can be classified into active devices represented by nonlinear models or passive devices that are represented by linear models. But since even passive components such as cables and connectors exhibit nonlinear behavior, complex models are often needed for them. Moreover, passive components can be classified into discrete or lumped components (resistors, capacitors, and inductors) and distributed components, such as those formed of microstrip transmission lines [SWA 03].
Electromagnetic simulators solve different circuit problems based on Maxwell’s equations. Currently, most of the electromagnetic simulators rely on three key technologies: Method of Moment (MoM), Finite Element Method (FEM) and Finite Difference Time Domain (FDTD) methods. These simulation methods tend to use a similar approach to solve the problems:
- – first, a physical model is created. This usually consists of layout geometries, material properties, etc.;
- – then, the simulator is set up with the boundary conditions, the extent of the simulation and the assignment of ports and other specific simulation options;
- – once the model is defined and the simulator is set up, the simulation is performed. The simulation involves the use of mesh cells to transform the physical model into discrete elements. The simulator makes us of local functions to approximate the field/current across the mesh cells;
- – finally, the local function coefficients are adjusted until the boundary conditions of the simulation are fulfilled. Design information such as S-parameters, field level and/or radiation patterns can be calculated during post-processing.
This process is similar for simulators based on MoM, FEM and FDTD [HES 09]. However, the differences among them make each one best suited for particular applications:
MoM: this technique only requires that the metal interconnects in the structure to be simulated are meshed. Therefore, simulations are speed up compared to the other technologies because a “planar” MoM mesh is simpler and smaller than the equivalent “3D volume” mesh required for an FEM or FDTD simulation. MoM algorithms solve Maxwell’s equations implicitly by solving a matrix;
FEM: this simulation method is a true 3D field solver that allows arbitrarily shaped 3D structures to be analyzed. The advantage over MoM simulation is that it can be used for any type of 3D structure and is not confined to a layered stack up. FEM simulation requires that objects being simulated are placed into a truncated space. This volume of the simulation domain is converted into discrete elements, usually tetrahedral mesh cells with a denser mesh being created around the geometric model being simulated. FEM algorithms solve Maxwell’s equations implicitly by solving a matrix;
FDTD: this simulation method is a true 3D field solver which can analyze arbitrary shaped 3D structures like FEM. FDTD algorithms solve Maxwell’s equations in a fully explicit way. FDTD employs a time-stepping algorithm that updates the field values across the mesh cell time-step by time-step, thereby explicitly following the EM waves as they propagate through the structure modeled.
In order to select the most suitable EM simulators based on MoM, FEM and FDTD analysis methods for a given application, the geometry of the design and the circuit response type are the first parameters to consider:
- – MoM-based simulators offer the most efficient simulation method for truly planar structures. For that reason, an MoM-based simulator would be recommended for analysis of on-chip passive elements and components on a PCB and planar antennas. However, it is not the best method for communication between two antennas or between a passive tag and a reader as in the case of RFID. Either FEM- or FDTD-based EM simulators are usually more appropriate for true 3D structures [HES 09];
- – both MoM and FEM methods solve natively in the frequency domain, which makes them more appropriate than FDTD for the analysis of circuits with a high quality factor (high Q), such as filters, cavities, resonators and oscillators. In contrast, the FDTD method solves natively in the time domain, making them useful for connector interfaces and transitions.
Wireless communications have attracted considerable interest in the research community, and many wireless networks are evaluated using Discrete Event Simulation (DES) tools. This section examines different simulation tools and makes a comparison between them, taking into account their capabilities to simulate near-field wireless communications. In fact, this section is focused on four widely used network DES software: Riverbed Modeler, OMNeT++, ns-2 and ns-3.
2.3.1. Riverbed Modeler
Riverbed Modeler1—previously known as Opnet Modeler or OPNET—is a commercial network discrete event simulator. It provides a wide range of libraries to model packet networks and it makes use of most of the technologies (e.g. WiFi, WiMAX, ADSL) and protocols (e.g. IP, TCP). Furthermore, it is one of the most popular network simulators and, thus, the third-party support and availability of additional libraries is high.
Riverbed Modeler allows us to exhaustively model the physical characteristics of the wireless communications. Packets are the data chunks processed through the transceiver pipeline of Riverbed Modeler in order to simulate wireless transmissions. The term pipeline is used to outline that Riverbed Modeler processes every wireless packet in a sequence of 14 stages in which the physical characteristics of the wireless link are split. Six of these stages are related to the transmission of the packet: Receiver Group, Transmission Delay, Link Closure, Channel Match, Tx Antenna Gain and Propagation Model. In contrast, the remaining eight stages are related to the reception: Rx Antenna Gain, Received Power, Interference Noise, Background Noise, Signal-to-Noise Ratio, Bit Error Rate, Error Allocation and Error Correction. The stages are performed sequentially and each stage has available the results of the calculations performed in previous stages. Thus, the Signal-to-Noise Ratio stage, for example, has the information about the received signal and interference power estimated in previous stages available (Received Power, Interference Noise and Background Noise) to calculate the Signal-to-Noise Ratio (SNR) and make it available to following stages. In general, the objective of this pipeline workflow consists of calculating the SNR of the packet (Signal-to-Noise Ratio stage), then obtaining the Bit Error Rate (BER) of the packet based on the SNR and the modulation used (Bit Error Rate stage), estimating the number of wrong bits in the packet (Error Allocation stage) and, finally, considering the number of wrong bits to decide if the packet is received correctly or not (Error Correction stage).
The simulator comes with several sets of pipeline stages predefined to simulate the most common wireless technologies nowadays such as WiFi or LTE. It also provides a basic set of stages for generic wireless communications such as generic Time Division Multiple Access (TDMA) wireless communication, which could be valid as a basis to model TDMA wireless technologies like GSM.
There are proposals to model RFID in Riverbed Modeler [YAN 09, MAR 13]. The authors of [YAN 09] provide an improved channel model for RFID communications by performing a correct parameterization of the wireless channel and by improving the BER pipeline stage with a new BER-SNR curve that takes the fading effect into account. However, no other pipeline stages are modified according to the paper. Instead, the authors of [MAR 13] do not modify any pipeline stage. They study only the link layer protocol of RFID and the wireless coverage is limited with the use of antennas with different diagram patterns.
OMNeT++2 is a powerful and modular DES software. It consists of modules that can be simple or compound, depending on whether they are atomic or consist of inner modules respectively. The most common way of interaction among modules is by sending and receiving messages via gates and connections. First, gates can be used for sending (output gates) or receiving (input gates) messages. Second, connections can be assigned with transmission properties such as transmission delay or data rate.
There are a huge number of models implemented by individuals and groups made publicly available with open-source licenses. Some of the most important models are provided together in packets known as simulation frameworks. One of the most important is the INET Framework3 which provides support for the IP family of protocols, wired and wireless link layer protocols, and other popular technologies and protocols. The framework for modeling wireless networks is MiXiM4. MiXiM joins and extends several existing simulation frameworks developed for wireless and mobile simulations in OMNeT++. It provides detailed models of the wireless channel (fading, etc.), wireless connectivity, mobility models, models for obstacles and many communication protocols especially at the Medium Access Control (MAC) level. However, MiXiM is now supposed to be deprecated and some of its functionality has been included in INET framework since version 3.0. Finally, it is also worth pointing out the Castalia5 framework, which is focused on the simulation of Wireless Sensor Networks (WSNs), Body Area Networks (BANs) and networks of low-powered devices. This framework uses the lognormal shadowing model to calculate the average path loss between nodes whose distance is between a couple of meters and hundreds of meters. Moreover, it also provides the alternative of a specific path loss map, e.g. based on real measures, which could be used for BAN and near-field communications.
In [FER 15], the authors introduce a novel simulator to test RFID anti-collision proposals based on OMNeT++ and the Castalia simulation framework. The Propagation module of the simulator is responsible for calculating the propagation loss and delay in addition to providing the mechanism to detect RFID collisions.
GreenCastalia6 [BEN 13] is an extension to the Castalia framework to allow the simulation of protocols and devices that should cope with the energy harvesting typically required in WSNs.
ns-27 is a widely used tool to simulate the behavior of wired and wireless networks. It is an open-source object-oriented DES software organized according to the Open Systems Interconnection (OSI) model. Simulations are based on a combination of C++ and OTcl. In general, C++ is used for implementing protocols and extending the library, and OTcl is used to create and control the simulation environment itself, including the selection of output data. Simulation is run at the packet level, allowing for detailed results.
The MannaSim Framework8 extends ns-2 to cope with WSNs. In this sense, it introduces new modules for design, development and analysis of different WSN applications. This framework allows for the selection of different types of wireless channels, radio propagation models and antenna models as well as many other physical characteristics of wireless communication.
SensorSim9 [PAR 00] is a simulation framework developed on top of ns-2 by the US Naval Research Laboratory in order to ease the simulation of sensor networks. It supports different sensor channels which simultaneously support multiple propagation models. Apart from the wireless channel, this framework also focused on other critical aspects of WSNs such as the power model or the energy consumption.
ns-310 is a new simulator, not compatible with ns-2 and built from scratch to replace it. It is entirely built in C++, and OTcl programming language is not used. ns-3 is primarily targeted for research and academic purposes. The large majority of its users focus on wireless/IP simulations that involve models for WiFi, WiMAX or LTE for layers 1 and 2.
Each release of ns-3 is provided with a well documented model library. This model library has support for wireless communication technologies, low-powered wireless communications and up to 15 propagation models that can be extended with other modules.
There are multiple surveys that compare these simulators qualitatively [SIN 08, XIA 08, KUM 12], by focusing on the characteristics of each simulator, and quantitatively [KHA 14], by focusing on the performance. From the point of view of modeling near-field wireless communication, none of them provide models to at least simulate the most common near-field communications such as RFID—the near-field version of the RFID specification—and NFC. In fact, there are not even any third-party frameworks that extend the simulators to provide this functionality. Thus, there are two main approaches to integrate the near-field wireless communications in these DES frameworks.
The first approach would be to integrate the near-field wireless communication inside the DES tool by adapting the wireless channel to be able to model near-field wireless communications. Although the physical characteristics of the near-field induction and the far-field propagation are quite different, the features provided to model the far-field propagation could be used to model the near-field induction. In this sense, the four analyzed DES tools present similar capacities. First, Riverbed Modeler has a pipeline of stages to model the wireless channels that can be customized by implementing new stages in C code. OMNeT++ with the Castalia framework and ns-2 with Mannasim or Sensorsim frameworks are focused on Wireless Sensor Networks, but they are also good frameworks to include new channel models. Even ns-3, which is a much more recent simulator, can be extended with new propagation models.
The authors of [ROD 16] present the design of new five pipeline stages of Riverbed Modeler to model the near-field communication between the Eurobalise on the track and the BTM inside the train. Among the modified pipeline stages, the Closure stage ensures that the communication is only performed when devices are located very close and the Power stage provides an equivalent received signal power related to the produced induction that is obtained from an equivalent received signal matrix whose values has been decided according to real measures. The design values and validation of the model are carried out with real measures performed in a laboratory with real equipment.
The second approach would be to use simulation-in-the-loop or co-simulation features of the DES software. Both solutions imply the removal of near-field computation from the DES tool and reliance on measures to be performed in an external device. The simulation-in-the-loop is usually used to connect the simulator to a real device or a real network, whereas the co-simulation is used when the simulator is connected to another simulator. For near-field wireless communications, it would be more appropriate to use the co-simulation option in order to link the DES tool with the electromagnetic simulator. One example of this approach is the COSMO network simulator [ZHA 10] built on top of OMNeT++ and MATLAB to provide an improved indoor wireless simulator. This example could be followed to build a DES simulator with near-field wireless communication capabilities included.
Nowadays, the increasing availability of low-cost smart objects with wireless near-field interconnection capability represents a growing opportunity to develop smarter and enriched public ITSs. Consequently, a significant research community effort is dedicated to build the necessary tools to evaluate the performance and potential of this massive near-field deployment and their impact on the existent communication infrastructures. System level simulations, such as DES frameworks play a crucial role within this tooling. For complexity reasons, system level frameworks rely on link simulation models accurate enough to capture the essential behavior.
Traditional characterization of near-field communication at the link level relies on two approaches: a theoretical approach normally based on mathematical calculations and another one based on electromagnetic simulators. The last approach has benefited from recent increases in computational power and reached unprecedented levels of simulation capability. With modeling of this low level, the system level simulators benefit from precise communication links and accurate response to the events.
Although most of the commonly used system level simulation frameworks—Riverbed Modeler, OMNeT++, ns-2 and ns-3—target modeling the wireless communication link to measure global end-to-end quality of service performance indicators, their scope is normally far-field technologies. Nevertheless, in recent years, a few initiatives worthy of mention have appeared and they are identified in this chapter.
[ATZ 10] ATZORI L., IERA A., MORABITO G., “The internet of things: a survey”, Computer Networks, vol. 54, no. 15, pp. 2787–2805, 2010.
[BEN 13] BENEDETTI D., PETRIOLI C., SPENZA D., “GreenCastalia: an energy-harvesting-enabled framework for the Castalia simulator”, ENSSys’13 Proceedings of the 1st International Workshop on Energy Neutral Sensing Systems, ACM, New York, pp. 7:1–7:6, 2013.
[CHA 10] CHAKRA S., FARRUKH U., GARCIA B., “Electrical model simulation for a UHF RFID system in near and far fields”, International Journal of Simulation: Systems, Science and Technology, vol. 11, no. 1, pp. 14–20, 2010.
[DEV 05] DE VITA G., IANNACCONE G., “Design criteria for the RF section of UHF and microwave passive RFID transponders”, IEEE Transactions on Microwave Theory and Techniques, vol. 53, no. 9, pp. 2978–2990, 2005.
[DOB 12] DOBKIN D.M., The RF in RFID, UHF RFID in Practice, 2nd ed., Newnes, 2012.
[EPC 13] EPCGLOBAL, EPC Radio-Frequency Identity Protocols Generation-2 UHF RFID; Specification for RFID Air Interface Protocol for Communications at 860 MHz–960 MHz, EPCglobal Inc., November 2013.
[FER 15] FERRERO R., GANDINO F., MONTRUCCHIO B. et al., “A novel simulator for RFID reader-to-reader anti-collision protocols”, 2015 International EURASIP Workshop on RFID Technology (EURFID), pp. 59–64, October 2015.
[FIN 04] FINKENZELLER K., RFID Handbook: Radio-frequency Identification Fundamentals and Applications, Wiley, 2004.
[GLI 04] GLIDDEN R., BOCKORICK C., COOPER S. et al., “Design of ultra-low-cost UHF RFID tags for supply chain applications”, IEEE Communications Magazine, vol. 42, no. 8, pp. 140–151, 2004.
[GUB 13] GUBBI J., BUYYA R., MARUSIC S. et al., “Internet of things (IoT): a vision, architectural elements, and future directions”, Future Generation Computer Systems, vol. 29, no. 7, pp. 1645–1660, 2013.
[HES 09] HESE J.V., SERCU J., PISSOORT D. et al., State of the Art in EM Software for Microwave Engineers, Agilent Technologies Application Note 5990-3225EN, February 2009.
[ISO 13] ISO, IEC, Information Technology – Telecommunications and Information Exchange between Systems – Near Field Communication – Interface and Protocol (NFCIP-1), ISO/IEC, no. 18092, http://cp.literature.agilent.com/litweb/pdf/5990-3225EN.pdf, 2013.
[ISO 16] ISO, IEC, Identification Cards – Contactless Integrated Circuit Cards – Proximity Cards, ISO/IEC, no. 14443, 2016.
[JAN 11] JANKOWSKI-MIHULOWICZ P., KALITA W., “Application of Monte Carlo method for determining the interrogation zone in anticollision radio frequency identification systems”, in TURCU C. (ed.), Current Trends and Challenges in RFID, InTech, July 2011.
[KAR 03] KARTHAUS U., FISCHER M., “Fully integrated passive UHF RFID transponder IC with 16.7-uW minimum RF input power”, IEEE Journal of Solid-State Circuits, vol. 38, no. 10, pp. 1602–1608, 2003.
[KHA 14] KHAN M.A., HASBULLAH H., NAZIR B., “Recent open source wireless sensor network supporting simulators: a performance comparison”, 2014 International Conference on Computer, Communications, and Control Technology (I4CT), pp. 324–328, September 2014.
[KOE 75] KOELLE A., DEPP S., FREYMAN R., “Short-range radio-telemetry for electronic identification, using modulated RF backscatter”, Proceedings of the IEEE; (United States), August 1975.
[KUM 12] KUMAR A., KAUSHIK S.K., SHARMA R. et al., “Simulators for wireless networks: a comparative study”, 2012 International Conference on Computing Sciences (ICCS), pp. 338–342, September 2012.
[LAN 05] LANDT J., “The history of RFID”, IEEE Potentials, vol. 24, no. 4, pp. 8–11, 2005.
[MAR 13] MARINO F., MASSEI G., PAURA L., “Modeling and performance simulation of EPC Gen2 RFID on OPNET”, 2013 IEEE International Workshop on Measurements and Networking Proceedings (M N), pp. 83–88, October 2013.
[PAR 00] PARK S., SAVVIDES A., SRIVASTAVA M.B., “SensorSim: a simulation framework for sensor networks”, Proceedings of the 3rd ACM International Workshop on Modeling, Analysis and Simulation of Wireless and Mobile Systems, MSWIM’00, ACM, New York, pp. 104–111, 2000.
[ROD 16] RODRIGUEZ L., PINEDO C., LOPEZ I. et al., “Eurobalise-Train communication modelling to assess interferences in railway control signalling systems”, Network Protocols and Algorithms, vol. 8, no. 1, pp. 58–72, 2016.
[SIN 08] SINGH C.P., VYAS O.P., TIWARI M.K., “A survey of simulation in sensor networks”, 2008 International Conference on Computational Intelligence for Modelling Control Automation, pp. 867–872, December 2008.
[SWA 03] SWANSON D.G., HOEFER W.J., Microwave Circuit Modeling Using Electromagnetic Field Simulation, Artech House, London, 2003.
[UNI 08] UNISIG, FFFIS For Euroloop v2.3.0, ERTMS/ETCS Class 1, SUBSET, no. 044, February 2008, available at: http://www.era.europa.eu/Document-Register/Pages/Set-2-and-3-FFFIS-for-Euroloop.aspx.
[UNI 12] UNISIG, FFFIS For Eurobalise v3.0.0, ERTMS/ETCS Class 1, SUBSET, no. 036, February 2012, available at: http://www.era.europa.eu/Document-Register/Pages/Set-2-and-3-FFFIS-for-Euroloop.aspx.
[WAN 06] WANT R., “RFID explained: a primer on radio frequency identification technologies”, Synthesis Lectures on Mobile and Pervasive Computing, vol. 1, pp. 1–94, January 2006.
[XIA 08] XIAN X., SHI W., HUANG H., “Comparison of OMNET++ and other simulator for WSN simulation”, 2008 3rd IEEE Conference on Industrial Electronics and Applications, pp. 1439–1443, June 2008.
[YAN 09] YANG D., LIU W., “The wireless channel modeling for RFID system with OPNET”, 2009 5th International Conference on Wireless Communications, Networking and Mobile Computing, pp. 1–3, September 2009.
[ZAN 14] ZANELLA A., BUI N., CASTELLANI A. et al., “Internet of things for smart cities”, IEEE Internet of Things Journal, vol. 1, no. 1, pp. 22–32, 2014.
[ZHA 10] ZHANG Z., LU Z., CHEN Q. et al., “COSMO: co-simulation with MATLAB and OMNeT++ for indoor wireless networks”, 2010 IEEE Global Telecommunications Conference (GLOBECOM 2010), pp. 1–6, December 2010.
Chapter written by Christian PINEDO, Marina AGUADO, Lara RODRIGUEZ, Iñigo ADIN, Jaizki MENDIZABAL and Guillermo BISTUÉ.