IEEE802.11p ahead of  
LTE-V2V for safety applications
Alessio Filippi, Kees Moerman, Vincent Martinez and Andrew Turley – NXP Semiconductors.
Onn Haran and Ron Toledano – Autotalks.
Executive Summary
Car-to-car communication is attracting significant attention as it promises to drastically reduce road fatalities, im-
prove mobility and enable a high-level of vehicle automation. Supporting safety critical applications is at the core 
of car-to-car communication, and for years, the technology of choice for V2X has been IEEE802.11p. Recently, 
a new standard addressing V2X applications has started evolving under the umbrella of 3GPP, whose focus is 
mobile broadband standardization. Because the safety of millions of road users will depend on the performance 
of these technologies, it is important to compare them.
There are several relevant facts important to highlight when comparing IEEE802.11p to LTE-V2X: 
•  IEEE802.11p is ready now, LTE-V2X is not [4]. Today, IEEE802.11p-based products are available on the market  
 
from multiple silicon vendors. Some Tier1s have complete solutions available. In contrast, there is no LTE-V2X  
  product available in the market today, and it will most likely take several years before a complete solution will be  
 
ready and tested. The promised 5G version of V2X will have an even longer time horizon; 
•  IEEE802.11p is already installed in cars on the road. An end-user can buy a vehicle (e.g. GM Cadillac1 ) equipped  
  with IEEE802.11p technology today; 
•  The V2V NPRM has been published [1]. It clearly indicates that the US Government apparently has the intention  
 
to deploy IEEE802.11p as a technology thoroughly tested, validated and available for safety critical applications; 
•  IEEE802.11p mass deployment could begin soon. Volkswagen, one of the largest car manufacturers worldwide,  
	 publicly	announced	that	from	2019	onwards,	they	will	equip	their	first	model	series	with	IEEE802.11p	technology2.  
 
The cellular community is advocating that V2X implementations should wait for cellular technology to be ready and 
tested,	and	disregard	the	investments	and	field	tests	done	to	validate	IEEE802.11p	for	safety	critical	applications.	
More concretely, the cellular community claims that LTE-V2X offers: 
•  a strong cellular eco-system which leverages years of experience in providing paid-services and a mature techno- 
 
 
 
logy available worldwide. This is a valid argument, but it refers to entertainment services in a cellular-based  
technology. The communication between a device and a base-station is fundamentally different from the device- 
to-device communication in a dynamic environment; 
•  twofold better performance [6]. However, it is IEEE802.11p which outperforms LTE-V2X in important V2V use  
 
cases as we show in this article; 
•  minor added cost. This is questionable as the support of safety critical applications strongly indicates the need to  
 
 
separate those from the entertainment SW and HW. Therefore, LTE-V2X will likely be physically separated from  
the cellular modem; 
1 
2 
 
 
http://media.cadillac.com/media/us/en/cadillac/news.detail.html/content/Pages/news/us/en/2017/mar/0309-v2v.html 
https://www.volkswagen-media-services.com/en/detailpage/-/detail/With-the-aim-of-increasing-safety-in-road-traffic-Volkswagen-will- 
enable-vehicles-to-communicate-with-each-other-as-from-2019/view/5234247/6e1e015af7bda8f2a4b42b43d2dcc9b5?p_p_ 
auth=oyU0Lqiz
 
2
•  a roadmap of evolution and future proof technology due to the continuous effort in improving the technology  
 
via the well-tested mechanism of the 3GPP meetings. While this might be true, introducing an updated standard  
  every 12 to 15 months does not guarantee that older vehicles will be able to communicate with newer ones. This  
is in contrast with the need of creating a stable and universal international standard to enable the success of V2X  
 
 
technology. 
The proposed LTE-V2X technology is a derivative of the cellular uplink technology that maintains similarity with the 
current LTE systems: frame structure, sub-carrier spacing, clock accuracy requirements and the concept of a resource 
block, to mention a few. These properties were not made to fit the vehicular use cases, but rather are inherited from 
existing cellular technology. Consequently, LTE-V2X struggles to meet the specific application requirements of car-
to-car communications. 
Technically, LTE-V2X suffers when there is no network to support the communications. It has stringent synchronizati-
on requirements (section 2.1), it cannot properly receive messages from nearby and closed-by transmitters (section 
2.3) and it’s limited in its maximum range (section 2.4). Furthermore, it proposes a resource allocation scheme that 
does not properly handle messages with variable size (section 2.5) and a multiple user access mechanism that is not 
well suited for broadcasting messages (section 2.6) or for handling collisions of messages (section 2.9). The heavy-
weight design of LTE-V2X translates into a higher overhead (sections 2.7 and 2.8). 
Commercially, LTE-V2X cannot leverage the presence of the standard LTE modem in the car. Different safety require-
ments (section 2.10) and technology needs (section 4.1) strongly suggest that the safety critical domain of LTE-V2X 
will be separated from the entertainment domain of the standard LTE modem. The stringent synchronization require-
ments (section 3.2) could significantly increase the costs in the LTE-V2X hardware.
Strategically, LTE-V2X might not be the best technology for safety critical applications as its fast development cycle 
does not match the automotive development cycle (section 4.1). The 3GPP community has already started working 
on a new version of LTE-V2X while the current version has not been tested in the field yet. The next generation of 
IEEE802.11p is also being considered (section 4.2.1) to capitalize on the experience of multiple large-scale field 
trials to test safety critical applications. 
Our conclusion is that IEEE802.11p technology is ideal for safety critical applications that must be supported in ab-
sence of a network. If the cellular infrastructure is available, LTE-V2X is a valid alternative and offers a more mature 
eco-system for entertainment services. The win-win situation would be to focus on the strongest points of each tech-
nology and work together to provide the best car-to-car communication solution, continue deploying IEEE802.11p 
for safety critical applications and ensure that the upcoming LTE-V2X technology can coexist.
 
3
1 Introduction 
 
Since its introduction 10 years ago, the technology of choice for V2X has been IEEE802.11p3 , which has been 
standardized, implemented and thoroughly tested. Recently, a new standard addressing V2X applications has 
started evolving under the umbrella of 3GPP, whose focus is mobile broadband standardization. The safety of 
millions of road users will depend on the performance of these technologies; therefore, it is of outmost importance 
for policy makers, vehicle manufacturers and the wider automotive ecosystem to compare them.
1.1  V2X targeted functionality 
 
Working together and sharing information to make transportation safer, greener, and more enjoyable, is truly 
compelling. The technologies associated with this concept, collectively known as Cooperative Intelligent 
Transportation	Systems	(C-ITS),	promise	to	reduce	traffic	congestion,	lessen	the	environmental	impact	of	
transportation,	and	significantly	reduce	the	number	of	lethal	traffic	accidents. 
 
A key enabling technology of C-ITS is wireless communication, covering vehicle-to-vehicle (V2V) communication, 
vehicle-to-motorcycle (V2M) communication, vehicle-to-infrastructure (V2I) communication, and infrastructure-to-
vehicle (I2V) communication. Collectively, these wireless transactions are referred to as vehicle-to-everything, or 
V2X, communication. 
 
V2X technology will support many safety-related and possibly the non-safety-related use-cases of C-ITS systems. 
It needs to operate robustly in a very dynamic environment with high relative speeds between transmitters and 
receivers, and support the extremely low latency of the safety-related applications in fast highways, crowded urban 
intersections and tunnels.  
1.2  IEEE802.11p 
 
IEEE802.11p was designed to meet every V2X application requirement with the most stringent performance 
specifications.	In	1999,	the	U.S.	Federal	Communications	Commission	(FCC)	set	aside	75	MHz	of	spectrum,	in	the	
5.9 GHz region, for V2X. The IEEE802.11p standard operates within this range. 
 
IEEE802.11p is an extension of IEEE802.11a (WiFi), operating in an ad-hoc network mode without the need of a 
BSS (Basic Service Set, the WiFi ‘base station’). It is optimized for mobile conditions in presence of obstructions, 
handling	fast-changing	multi-path	reflections	and	Doppler	shifts	generated	by	relative	speeds	as	high	as	500	km/h.	
The typical Line-Of-Sight (LOS) range is 1 km, but the main purpose of IEEE802.11p is to ‘see around corners’ 
(NLOS, Non Line Of Sight) as no other sensor in the car is able to do. It has been shown that with state-of-the-
art technology, currently available as commercial off-the-shelf products, larger ranges of even several km are 
routinely achievable. IEEE802.11p multiple access mechanism (the Carrier Sense Multiple Access protocol with 
Collision	Avoidance,	CSMA-CA	protocol)	efficiently	handles	high	density	use	cases	when	combined	with	Distributed	
Congestion	Control	(DCC)	[7].
3 
Formally named IEEE802.11 operating outside the context of a BSS, and also known in the USA as DSRC
 
4
 
The	standardization	work	started	more	than	10	years	ago,	a	final	draft	was	approved	in	2009,	and	has	been	
extensively	tested	and	validated	since	that	approval.		The	first	large-scale	field	test,	the	simTD	project	[8],	
began	in	2009	and	included	over	one	hundred	vehicles.	Dozens	of	additional	field	trials	with	commercial	
IEEE802.11p products have been completed since then, while many are still on-going.  To mention a few, 
see	[8-13].	One	of	the	biggest	running	pilots	for	IEEE802.11p	is	funded	by	USDOT	(in	Wyoming,	Tampa	and	
New York city) including over ten thousand vehicles implementing diverse applications and an investment of 
more than $45 million [13].  Large investments are being made to guarantee the quality and reliability of this 
technology.  
Several	semiconductor	companies	have	designed	and	tested	automotive	qualified	IEEE802.11p-compliant	
products. A large number of hardware and software products are available from multiple suppliers, comprising 
a rich ecosystem. There are several car models on the market with IEEE802.11p technology, while others are 
planned to be launched soon, for example: 
•  GM’s Cadillac CTS is equipped with IEEE802.11p4 ;
•  Toyota has close to 100,000 cars in Japan equipped with IEEE802.11p;
•  Volkswagen selected IEEE802.11p technology to support V2X applications5. 
The	USDOT	has	declared,	based	on	collected	evidence,	that	IEEE802.11p	technology	can	significantly	reduce	
the	number	of	collisions	on	the	road.	Experts	expect	that	the	USDOT	is	in	the	process	of	mandating	the	use	of	
IEEE802.11p in all new light vehicles for safety-related use-cases [14].
1.3  LTE-V2X  
LTE-V2X	is	a	relatively	new	technology	(first	discussions	took	place	in	2015),	and	is	an	extension	of	3GPP	Rel-
12	Device-to-Device	(D2D)	functionality,	which	itself	is	based	on	using	the	LTE	uplink	transmission	and	uplink	
spectrum resources for direct communication between devices. Basic safety V2V functionality made its debut in 
LTE	Rel-14	specification.	 
LTE-V2X was designed with multiple deployment scenarios in mind, leading to the following requirements: 
1.  Operation with or without eNB (‘base station’) coverage. LTE-V2X in Rel-14 is based on the PC5 interface 
that allows users to directly broadcast messages to each other, with or without network coverage. 
Operation	under	cell-coverage	is	leveraging	all	the	benefits	of	a	synchronous	network,	where	central	
coordination, scheduling and management is realized by a series of base-stations. However, it should 
be noted that many scenarios exist where this setup cannot be operated, e.g. in rural areas with poor 
coverage, and highways and fast-speed users with many handovers. Reliable operations without coverage 
must be addressed by LTE-V2X technology; 
4 
5 
 
 
http://media.cadillac.com/media/us/en/cadillac/news.detail.html/content/Pages/news/us/en/2017/mar/0309-v2v.html
https://www.volkswagen-media-services.com/en/detailpage/-/detail/With-the-aim-of-increasing-safety-in-road-traffic-Volkswagen-will- 
enable-vehicles-to-communicate-with-each-other-as-from-2019/view/5234247/6e1e015af7bda8f2a4b42b43d2dcc9b5?p_p_ 
auth=oyU0Lqiz
 
5
2.  Standalone operation on a dedicated unlicensed carrier or under licensed spectrum; 
3.  Enhanced	D2D	air-interface	functionality	for	supporting	low-latency,	high-density	and	high	speed. 
To address the enhanced requirements, Rel-14 LTE-V2X introduced new Sidelink transmission modes (Transmission 
Modes	3	&	4),	see	Table	1.	These	differ	from	Rel-12	D2D	modes	(TM	1	&	2)	by	introducing	low-latency	
transmissions, improved support for higher speed and new distributed channel access mechanism [15].
Mode 1
Mode 2
Mode 3
Mode 4
Scheduling method Channel access
Use case
Release
eNB
eNB-controlled
Public safety VoIP
LTE Rel-12
Distributed
Random, with blind 
re-transmissions
eNB
eNB-controlled
Distributed
Sensing, with semi-
persistent trans- 
mission
Public safety VoIP
LTE Rel-12
V2X
V2X
LTE Rel-14
LTE Rel-14
Table 1: available operation modes in LTE-Sidelink Communications
Despite	the	recent	contributions	and	standardization	efforts,	the	LTE-V2X	standard	has	not	reached	maturity,	and	
many	technical	topics	are	still	being	discussed,	leading	to	some	significant	standard	changes	agreed	upon	during	
the last RAN meetings. The number of maintenance Change Requests (CR’s) related to V2X is large and makes 
it challenging for chip makers to settle on a set of functionalities, reach interoperability testing stage, freeze 
the hardware and software architecture and go to production. Automakers might also question the effective 
performance and support of the safety-critical use cases. At this point in time, the real-life performance of the LTE 
Rel-14 standard is practically unknown.  
The most relevant and challenging LTE-V2X operation mode for the safety-critical applications is Sidelink Transmission 
Mode 4, which can be seen as an ad-hoc mode. The comparison with IEEE802.11p technology will focus on this mode.
2 Comparison of LTE-V2V Mode 4 versus IEEE802.11p
Both	IEEE802.11p	and	LTE-V2X	use	the	well-known	Orthogonal	Frequency	Division	Multiplexing	(OFDM)	as	a	modu-
lation technique, in which a block of data is transmitted on equidistant subcarriers. However, 
IEEE802.11p
LTE-V2X
Multi-user allocation
single user per symbol
multiple users share the same symbol
Synchronization requirements
asynchronous
tight synchronization 
OFDM parameters
short symbol duration 
very long symbol duration
Channel access mechanism 
CSMA-CA
sensing based SPS transmission
Table 2 Difference between IEEE802.11p and LTE-V2x design parameters
 
6
as reported in Table 2, they choose very different parameters. LTE-V2X has inherited much of LTE mechanism which 
is suitable for centralized (i.e. non-ad-hoc) and synchronized network, with power control, synchronization adjust-
ments and which operates with low to moderate speed. As we show in the following sub-sections, it is less suitable 
for ad-hoc communication mode and fail in several important V2X use-cases. 
2.1  Synchronization
LTE-V2X is more sensitive to frequency errors and timing errors than IEEE802.11p. With inaccurate frequency 
synchronization,	the	residual	frequency	errors	lead	to	Inter-Carrier	Interference	(ICI).	In	LTE-V2X	the	OFDM	subcar-
riers	are	10	times	closer	than	in	IEEE802.11p	so	the	same	absolute	frequency	error	has	significantly	more	impact	in	
LTE-V2X than in IEEE802.11p. Consequently, LTE-V2X performance is limited, and the same absolute frequency error 
generates	100	times	larger	interference	power	[10].	This	is	quantified	in	the	time	and	frequency	accuracy	require-
ments of IEEE802.11p and LTE-V2X summarized in Table 3.  
Timing accuracy
Frequency accuracy
allowed error 
[µsec]
Reference
Allowed error 
[ppm]
Reference
IEEE802.11p *
LTE-V2X ** 
± 1000
± 0.39
Absolute (UTC)
± 20
sync source
±  0.1
Absolute 
sync source
Table 3 - transmit accuracy requirements. 
*		timing	accuracy	is	specified	in	IEEE	1609.4	for	channel	switching.	 
				IEEE802.11p	operation	has	no	timing	dependency;	frequency	accuracy	is	specified	in	IEEE802.11
**	timing	accuracy	is	specified	in	3GPP	TS	36.133;	frequency	accuracy	is	specified	in	3GPP	TS	36.101
Two main differences are apparent:  
 
 
  1.  LTE-V2X requirements are much more demanding; 
  2.  LTE-V2X requirements are relative to the user’s synchronization source. When users have different synchroni- 
 
zation sources, such as locking to different base stations, the requirements can no longer be maintained thus  
 
impacting performance when vehicles are communicating with one another.  
 
To address the synchronization requirements, LTE-V2X users rely on the Global Navigation Satellite System (GNSS) 
signal. However, this brings other challenges. For instance, it is a fact that the GNSS signal is not always available or 
not reliable enough in locations such as tunnels, underground parking lots and urban canyons. With no GNSS co-
verage, keeping synchronization within the required accuracy boundaries depends on the drift of the local oscillator 
of the user. The higher the accuracy, as required by the tight subcarrier spacing, the higher the costs. In the absence 
of reliable GNSS signal or no GNSS signal at all, a user will have to select an alternative source for synchronization 
which impacts reliable communications.  
IEEE802.11p operation does not depend on GNSS signal. IEEE1609.4 requires the GNSS signal as well, but simply 
to switch from one channel to another, i.e., with much lower time and frequency accuracy. 
 
7
2.2  High speed conditions
Transmissions	by	moving	vehicles	introduce	Doppler	frequency	shifts,	which	can	be	seen	as	additional	frequency	
errors	(in	addition	to	synchronization	errors).	Under	high	speed	conditions	these	Doppler	frequency	shifts	can	be	
two times or even four times larger than the synchronization errors (increasing with the vehicle relative velocities) 
and become dominant.
Figure	1	OFDM	symbol	of	
IEEE802.11p (top) and of 
LTE-V2X (bottom), at scale
As shown in Figure 1, in LTE-V2X the symbol duration is ten times longer than that of IEEE802.11p which puts a limit 
on	the	maximum	detectable	Doppler	frequency	shift,	and	therefore	maximum	limit	on	speed	(in	addition	of	tracking	
the fast-varying channel). In fact, this drawback was already observed internally in 3GPP simulation results where 
beyond	a	speed	of	140km/h,	messages	are	no	longer	detected	reliably	and	performance	is	quite	poor	[19].	The	
attempt of 3GPP to overcome the problem was by introducing complex processing methods which were found not 
to be robust enough [20] or by reducing the modulation and coding scheme (MCS) which did not solve the problem. 
Proposing to change the pilot symbols pattern or shorten the symbol duration [21] was not accepted and eventually 
LTE-V2X is strictly	limited	to	speeds	below	140km/h.	
IEEE802.11p	on	the	other	hand,	benefits	from	very	short	symbol	duration	and	selected	a	symbol	pilot	pattern	
such that does not impose any limit on performance in high-speed. And while LTE-V2X is limited to operate below 
140km/h,	IEEE802.11p	can	perform	well	even	at	speeds	of	250km/h	or	beyond.				
2.3  Near-far problem
LTE-V2X is sensitive to the scenario in which a user receives a signal from two or more transmitters with different 
power levels, i.e., the near-far problem, as illustrated in Figure 2. The power difference may occur even for two 
nearby	transmitters,	when	one	of	those	is	obstructed.	IEEE802.11p	allows	a	single	user	transmission	for	each	OFDM	
symbol, and the receiver sets its parameters, like the automatic gain controller (AGC), the time offset estimation and 
the frequency offset estimation, in the best possible way for each user independently, as symbols are not shared.
 
8