**1. Introduction**

With the growth of wireless technology and intelligent transportation systems, vehicular ad hoc networks (VANETs) are attracting significant attention. Current goals are to make the ad-hoc network more efficient, secure and provide comfort to passenger on the road [1]. The main concern is to provide information regarding traffic congestion, collision notification, emergency, location services, weather conditions, and so on. VANETs can improve road safety and relief of vehicle drivers on the road. Traffic related information is analyzed and shared by vehicles in the network. VANETs is a subclass of mobile ad-hoc networks, which provide communication facilities to nearby vehicles in the road environment, which makes it different from others due to characteristics such as dynamic road topology, communication, sensing capabilities and transmission power for vehicles' function [2].

**Citation:** Ullah, I.; Shah M.A.; Khan, A.; Maples, C.; Waheed, A.; Jeon, G. A Distributed Mix-Context-Based Method for Location Privacy in Road Networks. *Sustainability* **2021**, *13*, 12513. https://doi.org/10.3390/su132212513

Academic Editor: Tan Yigitcanlar

Received: 1 October 2021 Accepted: 5 November 2021 Published: 12 November 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

The basic architecture of vehicular networks consists of Road Side Units (RSU), On-Board Units (OBU), Event Data Recorders (EDR), various sensors, and navigation systems (such as GPS) [3]. RSUs are a road infrastructure that increase the communication connectivity to vehicles. The OBU is fixed in the vehicle with a temper proof device that protects the cryptographic credentials of vehicles. This is used for wireless communication among vehicles and with infrastructure [4]. An EDR archives various events of vehicles communication during a trip on the road. GPS can be used to provide geographical location, vehicle speed, movement direction, and acceleration for a specific time interval [5]. Obstacles on the road are detected with the aid of radar and sensors. In-vehicle, an omnidirectional antenna is fixed for wireless channel access in the network.

The deployment of onboard units permits communication among nearby vehicles and fixed road infrastructure, which make possible three communication models, i.e., Vehicle to Vehicle (V2V), Vehicle to Infrastructure (V2I), and hybrid communication model [3,6]. In the V2V model, there is no support of infrastructure and vehicles are communicated directly. For data and information gathering, vehicles communicate with road side infrastructure through the V2I model. In the hybrid model, vehicles do not communicate with infrastructure directly but communicate in a single or multihop manner, depending on the transmission range of vehicles. This enables long-distance communication between vehicles in the network. Various wireless technologies are suggested for communication in vehicular networks, such Dedicated Short Range Communication (DSRC), Cellular network, WiMax, WiFi, and VeMAC protocol [6,7]. Among the existing technologies, DSRC has the potential for use in wide range variety of applications, including collision avoidance which can save thousands of lives and billions of dollars annually [8].

The mobile node (vehicle) in a network can broadcast Basic Safety Messages (BSMs), Cooperative Aware Messages (CAM) or beacons to disseminate road environment information. The beacon message's contents consist of vehicle identity, velocity, position, and other information [9]. The vehicle broadcasts beacon messages in plaintext format, and so other entities in the network are able to learn the actual identity and location of vehicles by analyzing these beacon messages. Similarly, an adversary can obtain the personal information of a vehicle driver by collecting beacon messages and tracking the various locations visited, thus coming to know the behavior and activities of the vehicle driver. This has the potential to pose several types of threat to the vehicle driver, such as damage to social reputation, physical harassment, blackmailing, and property loss [10]. To protect the privacy of the vehicle, pseudonyms can be used in place of the real identity in the message, and this is a commonly accepted solution. The pseudonym is an alias or randomized identity of a vehicle inserted in the beacon message. However, the use of fixed pseudo-identity is not suitable for protecting the privacy of a vehicle, and it must be changed over time to guard against the linking the pseudonym of a vehicle over time.

For the protection of vehicle location privacy, various pseudonym-changing strategies have been proposed in the literature. Some techniques use the concept of a mix zone [11–15] to hide the vehicle identities in a zone created by the vehicle to change pseudonyms cooperatively. However, the performance of the mix zone is degraded in conditions of lower traffic density [16]. Techniques based on group signatures are introduced [17–20] to protect location privacy of vehicles in which the broadcast beacons are signed with a key assigned to a group to protect the identity of a vehicle in a group. However, the management of signatures in the group administratively burdensome [16]; large groups have difficulty with managing signatures while small group size impacts privacy protection. Schemes based on a silent period [21–23] can hide the identity of vehicles. However, these schemes have a detrimental effect for road safety applications. To overcome the onward limitations of existing schemes, we propose a novel scheme, the Crowd-based Mix Context (CMC) method that efficiently provides location privacy protection under diverse vehicle traffic conditions.

The existing pseudonym changing scheme, Ref. [24] addresses the problem of location privacy in mix zones under lower traffic density. It uses a concept of creating fake

pseudonyms to anonymize the vehicle in the concerned region. However, generating fake pseudonyms in large quantities impact VANET applications and create computational overheads for the vehicle OBU. If a single-vehicle creates fake pseudonyms for anonymity purposes, the attacker may be able to find similarities in beacon messages, to ascertain that only one vehicle is on the road. Furthermore, fake pseudonyms create a liability issue in the network. The work undertaken in [25] is based on both mix zones and silent periods, where a large number of vehicles gather and anonymize identities in silent mode. However, while this works well in an urban scenario that consists of a higher number of vehicles, it is not consistent in the case of lower congested areas such as highways in which vehicles may rarely change pseudonyms [26].

Therefore, to preserve the location privacy of a vehicle, there is a need for an efficient approach that works under various traffic density conditions and also provides privacy outside mix zone areas. In this paper, we propose a distributed scheme CMC that offers privacy protection to vehicles in VANETs. In this paper, the terms network, road network, vehicular network, vehicular communication network and VANETs denote interchangeably and they all refer to vehicular ad-hoc networks. Our contributions in this paper are given below.


The rest of the paper is organized as follows: Section 2 contains details of the existing literature on location privacy schemes. System models and goals are discussed in Section 3. The proposed solution is explained in Section 4. Formal modeling and specification of the proposed model are given in Section 5. Section 6 provides details of the experimental setup and evaluation criteria for location privacy. In Sections 7 and 8, we discuss the performance analysis and comparison of the proposed scheme, respectively. Finally, the paper is concluded in Section 10.
