Introduction

Abstract

Within 20 years of the introduction of electronic computers, the need to have a connection between them emerged. The early foundations of the modern day computer network dates back to the 1960s when large universities and research labs wanted to share information between their computers.

Within 20 years of the introduction of electronic computers, the need to have a connection between them emerged. The early foundations of the modern day computer network dates back to the 1960s when large universities and research labs wanted to share information between their computers. Consequently, Ethernet was developed in the 1970s to interconnect local computers via cables and wires. Ethernet, standardized as IEEE 802.3, provides a framework for wiring, and protocols for signaling between computers that are not geographically far off. A network comprising computers connected with wires is known as the Local Area Network (LAN). As of today, LANs not only provide local information sharing, but can also be connected to a router (or hub) to access the external networks. The Internet itself is an interconnection of LANs that allows information sharing over a global scale (Bodden 2008).

Soon after the popularity and success of wired LANs, interest in getting wireless services began to mount. The overwhelming interest in wireless network services led to the introduction of many wireless technologies. The main reason for the popularity of wireless services was not high data rates, instead, the provision of mobility has been the primary cause of the popularity of wireless networks and technologies. The wireless links can provide coverage in areas where it is difficult to lay cables and wires (Stallings 2008). Due to the increasing demands of mobility and freedom from wires, wireless products and services rapidly became popular in both domestic and commercial sectors. The wireless technologies penetrated into the market through two main directions. The first is with the evolution of cellular systems which primarily supported voice services but are now providing data services as well. On the other hand, Wireless Local Area Networks (WLANs) were introduced as a wire-free version of the conventional LANs. The cellular networks were meant to provide services in both indoor and outdoor environments while WLANs were meant for data services in the indoor applications only (Hassan et al. 2009b).

Over the last few years, wireless networks have seen massive development with introduction of innovative technologies. Various wireless networks and technologies have been launched each of which targets a specific application. In terms of coverage, wireless networks can be classified as those supporting long range communication and those that suffice for communication needs over a shorter domain. In the long range category, cellular networks became popular and different underlying technologies such as Global System for Mobile communication (GSM), Universal Mobile Telecommunications System (UMTS), Enhanced Data for Global Evolution (EDGE), etc., gained popularity (Kwok and Lau 2007). WiMAX has also been introduced as a communication service for far-fetched areas where laying cables and wires is difficult. Like long range communications, communication over shorter range also witnessed innovation with the advent of technologies such as Bluetooth, Zigbee, and 802.11 Wireless LANs, etc (Garroppo et al. 2011). IEEE standardized WLANs as 802.11a/b/g networks. The first in the 802.11 series was IEEE 802.11b that was introduced in the late 1990s. After its success, IEEE standardized a range of 802.11 variations that cater for different purposes. While 802.11 networks inherently suffice for short range indoor communications, this book explores its use in highly mobile outdoor environments.

1.1 802.11 Wireless LANs

WLANs and LANs serve the same purpose, that is, information sharing, except that the former offers network services without interconnecting the devices with cables and wires. The devices in WLAN can communicate with each other wirelessly and hence are not fixed at one particular location. The legacy 802.11 standards operate on 2.4 GHz (802.11b/g) and 5 GHz (802.11a) frequency bands, and can support different data rates ranging from 6 to 54 Mbps. The advantage of using 802.11a networks over 802.11b/g is that the frequency spectrum they operate on is rarely used. On the other hand, the spectrum used by 802.11b/g is already in use by several other devices (Zhu et al. 2004). With regard to the transmission technologies, 802.11a and 802.11g use OFDM (orthogonal frequency division multiplexing) while 802.11b uses DSSS (direct sequence spread spectrum). 802.11g networks theoretically offer 54 Mbps data rates and are backward compatible with the 802.11b networks. A comparatively newer standard, IEEE 802.11n, offers data rates as high as 600 Mbps. A more detailed account of this standard has been given in Sect. 2.3.3.

Fig. 1.1

Extended service set in 802.11 infrastructure networks

The 802.11 networks have two modes of operation. One is the ad hoc mode, in which the mobile nodes in close vicinity connect with each other wirelessly and employ routing protocols to communicate with each other. The second mode is called the infrastructure mode (shown in Fig. 1.1), in which the mobile nodes communicate with each other via a central station called the Access Point (AP). The 802.11 AP together with the mobile nodes within its footprint constitute a Basic Service Set (BSS). A set of BSSes connected to the Internet is called the Extended Service Set (ESS) (Bhola 2002). Note that the end-users connect wirelessly to the AP, which itself has wired connections to the external network. In order to connect to WLAN, mobile nodes must have its Service Set Identifier (SSID). SSIDs are periodically transmitted by the WLAN AP in the beacon messages. A mobile node can gain network access by listening to the beacon signals and extracting SSID information therefrom. Some vendors also give an option of disabling SSID broadcast to ensure security (Oppenheimer 2004). However, with the advent of new and more robust authentication mechanisms, SSID-based authentication is rarely employed.

A single WLAN AP deployed inside a building can allow multiple users to connect to the Internet without having to stay at a fixed location at all times. Such wireless access to the Internet and other network services increases the work force productivity by 35 %, allowing easy communications among the co-workers, efficient maintenance of schedules, and quick access to emails (Reinward 2007). Ever since their introduction in the late 1990s, 802.11 networks have seen massive deployment across the cities throughout most of the developed world. Their popularity and growth was so rapid that by the summer of 2002, the number of 802.11 networks ranged between 15 and 18 million (Kanellos and Charny 2002). Schmidt and Townsend (2002) have pointed out that by the end of 2002, Wi-Fi¹ connectivity would be available in most of the universities and large corporations. A recent study has suggested that the number of domestic WLAN APs is more than 14 million in the US alone (Bychkovsky et al. 2006). Hull et al. (2006) report encountering 32000 WLAN APs during their drive tests that lasted for 290 drive hours.

As of today, a WLAN AP can be traced in most of the businesses, offices, restaurants, airports, shopping malls, university campuses, and houses. WLANs have now become so ubiquitous that their use as a replacement of cellular service is being considered. Several cellular companies have started offering paid Wi-Fi services before WLAN emerges as a wide area technology (Drucker and Angwin 2002).

1.2 Expanding the Mobility Domain of WLANs

The feature of mobility, combined with the provision of data rates that are much higher than the cellular systems, 802.11 WLANs, became extremely popular in the late 1990s and early 2000s. However, 802.11 WLANs are designed to support “restricted” mobility applications and therefore support network services only inside a building. Unlike the cellular systems, which provide network services over a larger geographical expanse by virtue of their planned base station deployment, WLANs were initially meant to cover smaller coverage regions. Therefore, WLANs inherently possess a limited mobility domain. This limitation of WLANs is now being addressed in response to the increased demand of 802.11 access over a larger area.

Presently, when a WLAN mobile node moves out of an indoor environment (home, office, etc), it has to switch over to another technology for the continued use of the network services. Instead of changing the wireless service, the idea being investigated is to use WLANs in the challenged environments. Recently, the concept of using WLANs from outdoors in high mobility vehicular environments has come under consideration. Since WLANs are capable of providing high data rates and because they are already available in large numbers, they can offer a cost-effective solution to allow communications between vehicles and between vehicles and the roadside infrastructure. Vehicular communication is an emerging research area in the Information and Communications Technology (ICT), that allows the use of safety application on roads and highways (Chisalita and Shahmehri 2004). In the following, an introduction to vehicular communication is given, which is followed by a brief discussion on the candidate wireless technologies for vehicular communications.

1.2.1 Vehicular Communications

Vehicular communication is concerned with enabling communication between vehicles, and between vehicles and roadside infrastructure to improve on-road safety. Thousands of fatalities and serious casualties in road accidents are reported every year across the world. Such accidents involve vehicles as well as pedestrians (David and Flach 2010). World Health Organization (WHO) and the World Bank predict that traffic injuries shall become the third biggest contributor to the burden of disease if necessary steps are not taken (Strom et al. 2010). It is expected that vehicular communications can play an effective role in reducing the traffic casualties consequently improving the transportation safety. This idea is not completely new because it existed in the form of “telematics” previously. However, the recent innovations in the low-cost communication technologies have substantially increased the research interest in this field (Bilchev et al. 2004). Housing communication devices within vehicles has now become commercially and technically viable as these devices become increasingly portable. In today’s research world, vehicular communication is a hot issue that is being explored from different perspectives in several ongoing projects, such as the Intelligent Transportation Systems (ITS) (Joseph 2006a).

ITS is concerned with using information and communication technologies from vehicles for various purposes. Despite its advantages and widespread applications, ITS faces a variety of challenges in different countries of the world. Ezell (2010) has classified ITS applications into two types to better understand its associated challenges. The first set of applications include those ITS solutions that can be deployed independently. For example, different communities and councils can deploy roadside cameras on independent basis. The second set of applications involve those which depend on other systems. For example, roadside communication stations may be deployed independently; however, their presence shall be useless unless communication devices are installed in all vehicles of a city. If vehicles are not equipped with communication devices, they will not be able to get useful information from the roadside communication units. Lack of funding is another problem faced by large-scale ITS deployment. Government transportation departments usually do not have enough funds to take ITS-related initiatives. On the technical side, ITS currently does not have sufficient standards which makes it difficult to integrate multiple ITS applications in a single system (Ezell 2010). Despite these challenges, various countries have successfully deployed ITS and are reaping its benefits in everyday life.

1.2.2 V2V and R2V Communications

The success of ITS applications depend significantly on the communication between vehicles. Like the classification of 802.11 as ad hoc and infrastructure networks, vehicular communication is also classified into two types. One is the Vehicle-to-Vehicle (V2V) communication that exploits the characteristics of the ad hoc networks. The vehicular nodes communicate with other nodes that are within their transmission range without requiring the services of a central entity. This is often referred to as the Vehicular Ad hoc Network (VANET) (Toor and Muhlethaler 2008), in which communication between the vehicular nodes is regulated by different routing algorithms and protocols (Wilke et al. 2009). One sample application of the V2V communication may be the transmission of a signal from a vehicle to others when it is about to change lanes on a freeway. This shall allow the neighboring vehicles to anticipate the lane change even in blind spots to reduce the risks of casualties.

Using cooperative communications between vehicles in a V2V scenario is getting increasingly popular. Vehicles traveling in a group are employing cooperative communication to improve road safety. Bauza et al. (2010) use cooperation among vehicles to detect traffic congestion on roads. Every vehicle traveling in a group estimates traffic conditions separately. These estimations are joined together to collectively decide whether the roads are congested. Similarly, neighboring vehicles in a V2V network can avoid traffic casualties by exchanging collision warnings in a cooperative manner (Yang et al. 2004).

In 2009, General Motors (GM) introduced innovative V2V solutions that have the potential to make roads safer for commuters. The developed technology uses V2V transponder placed on a vehicle’s roof, which can determine its own location and that of other vehicles in close vicinity. The device is capable of transmitting audible and visual warning messages in harsh driving conditions. The GM V2V device can detect sudden application of brakes by the vehicle ahead in order to warn the driver to slow down or change lane.

Fig. 1.2

R2V and V2V communications between RSUs and OBUs

The other type of vehicular communication is the Roadside-to-Vehicle (R2V) communication which conforms to the principles of infrastructure networks. R2V communication is also often referred to as Vehicle-to-Infrastructure (V2I) communication. In R2V communications, the vehicles communicate with the roadside infrastructure, e.g., base stations and AP, to send (or receive) information. Figure 1.2 shows the V2V and R2V communication between OBUs (On Board Unit, the vehicle) and RSUs (Roadside Unit, roadside base station or AP). The sample applications of R2V communication may include regular advertisements from a gas station, indication of a free parking space in an airport, etc (Sichitiu and Kihl 2008). R2V communications may also be used to upload and download traffic information from a central server and can also support Internet services on the move. For the rest of this book, the primary focus remains on the R2V communications.

Miller (2008) has proposed an architecture that combines V2V and R2V communication in a single network. The so-called Vehicle-to-Vehicle-Infrastructure (V2V2I) network allows ad hoc communication between vehicles like in V2V network, and uses a ‘Super Vehicle’ to communicate with the roadside infrastructure. All vehicles in a network transmit data to the super vehicle, which in turns communicates the same to the roadside base station. Super vehicles in one V2V network can also communicate with the super vehicles of other networks. It has been shown in Miller (2008) that V2V2I architecture can reduce the bandwidth requirement of the roadside base station by a factor proportional to the number of vehicles in a network.

Note that V2V and R2V communications augment each other and are not separate networking paradigms. However, in order to provide a more focused account, this book deals with R2V communication scenarios and their associated issues.

1.2.3 Wireless Technologies for Vehicular Communications

In order to facilitate vehicular communications, the use of a suitable wireless technology is a must. Among the various wireless technologies, cellular, 802.16 WiMAX, and 802.11 WLANs have emerged as the three major candidates for use in vehicular communications. These technologies originated at different times with different aims and objectives. The cellular networks first emerged in 1981 in the form of Nordic Mobile Telephone (NMT) systems in Scandinavia. NMT was followed by the release of Advanced Mobile Phone Services (AMPS) in 1983 in the US. These cellular networks accommodated a large number of users with their planned base station deployment and frequency reuse mechanism. They were initially meant to provide voice services with mobility over a large geographical domain. The cellular networks face signal degradation problems when the outdoor base stations are accessed from the indoor environments. The cellular signals have to penetrate through the walls to reach an indoor mobile node. To tackle this problem, femtocells are recently introduced to improve system capacity by avoiding the penetration of signals through walls and buildings (Hasan et al. 2009b). On the other hand, Wireless Interoperability for Microwave Access (WiMAX) was introduced as IEEE 802.16 in 2001. WiMAX provides broadband network access with range as long as 30 miles. While cellular and WiMAX offered network services in the outdoor setups, WLANs were introduced to provide broadband network services in the Small Office/Home Office (SOHO) setups. WiMAX, WLAN, and cellular networks are all being considered for vehicular communications because of their unique features and advantages. However, they also possess various drawbacks because they are not designed to meet the requirements of vehicular communications. The suitability of these technologies in the vehicular context and their brief comparison in terms of data rates, cost, and deployment issues have been given in the following.

1.2.3.1 Cellular Networks

The cellular networks may be a reasonable choice for use in the vehicular context because the cellular base stations are already massively deployed across the cities and are already offering network services on the move. Santa et al. (2008) have shown the feasibility of using the cellular infrastructure in vehicular communications. The reported delay analysis suggests that HSPA technology over the European UMTS will further enhance the suitability of cellular systems for both V2V and R2V communications. Nevertheless, the cellular systems suffer from low data rates. For example, GSM EDGE and UMTS HSPDA theoretically offer 1 and 7.5 Mbps data rates, respectively. Second, the cellular systems operate on licensed frequency spectrum which is purchased (or rented) by the cellular companies for dedicated use. The communication cost over the cellular frequencies may be comparatively higher than WLANs because they account for charges associated with using the dedicated spectrum. In summary, the cellular network, despite being almost ubiquitous, provides a low speed network connection with comparatively high communication costs.

1.2.3.2 802.16 Networks

Worldwide Interoperability for Microwave Access (WiMAX) is another candidate technology that is considered to support vehicular communications. Although it provides a larger coverage area than the WLANs, it is not as well deployed as WLANs and cellular networks. Therefore, dedicated WiMAX base stations must be deployed across the areas of the interest for enabling WiMAX-based vehicular communication. This certainly incurs significant deployment and labor cost. These heavy investments have also been recognized in the report published by the Center for European Policy Studies (CEPS) (Renda et al. 2009). Therefore, enabling vehicular communications by deploying WiMAX base stations shall have considerable economical constraints. On the other hand, WLANs (and cellular networks) incur no deployment costs because they are already available in large numbers in most of the developed world.

In addition to providing larger coverage area, WiMAX theoretically promises high data rates (75 Mbps for 802.16 e) for fixed wireless communications but provides much lower data rates under mobile conditions. For instance, the data rates up to 10 Mbps for 10 km Line-of-Sight (LoS) conditions have been reported by Ahmed and Habibi (2008). Intuitively, the data rates will further decrease with mobile non-LoS communication such as that expected in vehicular environments. Chou et al. (2009) have compared the achievable throughput from 802.11 WLAN and 802.16 WiMAX networks in the vehicular environments. The experiments which compared 802.11g with 802.16d reveal that the throughput from the former is much higher. Therefore, WLANs outplay WiMAX in terms for data rates.

1.2.3.3 802.11 Networks

There are two main reasons for preferring 802.11 WLANs in vehicular communications over the cellular and WiMAX networks. First, the WLAN APs are massively deployed across most of the developed cities of the world and hence provide reasonable infrastructure support. The already available WLAN infrastructure eliminates the need of heavy investments required for deploying the roadside infrastructure. Additionally, since they operate on free and unlicensed Industrial, Scientific, and Medical (ISM) frequency band, they do not incur additional cost of dedicated spectrum as is the case with the cellular systems. Second, WLANs support data rates that are much higher than WiMAX and cellular networks. WLANs can support fast exchange of information even at vehicular speeds (Tufail et al. 2008). With increasing interest in exploring 802.11 networks in vehicular environments, IEEE has standardized 802.11p WAVE to support information exchange among vehicles, and between vehicles and roadside infrastructure. 802.11p has been discussed in detail in Sect. 2.3.1.

Despite the apparent advantages of 802.11-based vehicular communications, there are some outstanding issues that must be addressed before vehicular communication can be realized using WLANs. The main research challenges addressed in this book are introduced in the following section.

1.3 Challenges in 802.11-Based Vehicular Communications

802.11 networks have several issues associated with them such as the rate adaptation, fair carrier access techniques, QoS provisions, interference and security, etc. The fact that WLANs are not meant to support outdoor communications further increases the challenges associated with the 802.11-based vehicular communications. V2V and R2V communication scenarios may have separate issues and challenges. However, as mentioned earlier, this book is concerned with R2V communication only; therefore, it explores the challenges in R2V communication environments. More specifically, the focus of this book is on addressing two important challenges that are pertinent to R2V communications, namely disruption and handover latency.

Disruption in wireless networks is used to mean interruption in communication services. Various sources of disruption exist in wireless networks most of which relate to the fading characteristics of the wireless channel. Fluctuations in the wireless channel and different sources of refraction and interference result in disrupted wireless services. Using 802.11 networks over a larger mobility domain introduces another kind of disruption that is specific to WLAN-based vehicular communications. This kind of disruption is different from the general interruption due to the changes in environment and wireless channel, and is primarily due to the unplanned deployment of WLAN APs. The unplanned deployment of APs leaves areas with no coverage in between two APs. Figure 1.3 shows a typical scenario in which a vehicle faces a disruption period between two connectivity periods. A vehicle faces periods of connectivity and dis-connectivity as it tries to access WLAN APs on the move. This phenomenon is termed as disruption.

One of the main challenges with WLAN-based vehicular access is reducing or tolerating this disruption. In fact, reducing irregularity in network services is a specialized area of research in wireless networking which is referred to as Disruption Tolerant Networking (DTN) (Farrell et al. 2006). The concept of DTN was initially introduced as delay tolerant networking with its main application in the deep space communications. The idea was to tolerate the elongated delays in the long distance communications (Fall and Farrell 2008). The term disruption is also being used in the context of vehicular communication. Disruption in vehicular communication is the irregularity in network services received by a vehicular node due to unplanned deployment of roadside infrastructure (Eriksson et al. 2008).

Fig. 1.3

Disruption in WLAN-based vehicular communications due to the unplanned deployment of 802.11 APs

In order to make the effective use of WLAN infrastructure on the move, the mobile node must be able to establish quick connections with the roadside APs. In 802.11-based vehicular communications, the mobile node leaves and enters the footprints of the APs very frequently. The process of connecting to a new AP after moving out of the footprint of the previously associated AP is called handover. The current handover procedure between a mobile node and an AP takes a considerable amount of time, which is often larger than the time a vehicle spends within the footprint of an AP. Table 1.1 shows the observations on the time spent by a vehicle within the footprint of an AP (Hasan et al. 2009a). The observations are drawn from the drive tests that are discussed in detail later in this book. It can be seen from Table 1.1 that the vehicle has to perform a handover after every 15 s. The delay in a process as frequent as this must be small. The latency associated with the increased number of handovers can also affect the QoS offered by the WLAN (Kwak et al. 2009). Therefore, the handover delay in the WLANs must be reduced for uninterrupted use of 802.11 services from vehicles.

Table 1.1 Encounter duration between a vehicle and an AP as observed in two different areas

The WLAN access for complete mobility over a large geographical expanse also requires consideration of interoperator handovers, i.e., handovers to APs that belong to the unsubscribed (foreign) Internet Service Providers (ISPs). The ISPs, or simply operators, are the network service providers that connect the end-user AP to the Internet. Different APs deployed across the cities may belong to different ISPs. Generally, an end-user is subscribed to one single ISP, and may not access the APs that belong to other ISPs. For enabling WLANs to provide complete mobility, some sort of universal roaming must be enabled among different ISPs. If handovers to the foreign APs are prohibited, the overall connectivity of the vehicle shall become limited. In addition to reducing the handover latency, the use of inter-ISP handovers is important for 802.11-based vehicular communications.

1.4 Summary

WLAN APs have been massively deployed by the end-users ever since their introduction in the late 1990s. WLAN APs can now be traced in various commercial entities, for example shopping malls, restaurants, businesses, airports, etc, as well as in domestic buildings such as houses. Their heavy presence across most of the developed cities and their ability to support high data rates have motivated the researchers to analyze the performance of WLANs in the vehicular environments. Vehicular communication can play a vital role in ensuring passenger safety on roads and highways by facilitating various applications such as traffic congestion monitoring, exchanging lane changing messages, warning about possible traffic hazards in advance, etc. Instead of deploying dedicated roadside infrastructure, already available WLAN APs can be used for vehicular communication. This book explores different techniques and challenges in applying these WLAN APs in R2V communications. The use of 802.11 networks from vehicles has two major limiting factors, namely disruption and handover latency, both of which have been discussed in the rest of this book.