Vehicles are increasingly getting connected with 5G cellular connectivity, and are expected to be connected to 6G in the future. One of the key features of 5G and of 6G will be use of millimeter wave (mmW) bands. In this paper, we consider vehicles equipped with mmW connectivity. The mmW band requires beam tracking and management to achieve high data rate. This can be particularly challenging for vehicle mobility. We study how perception from cameras in a vehicle can be used to improve beam management for mmW in vehicles. We propose a reinforcement learning-based beam management technique to combine 3D maps generated from perception with current 5G beam management. Our results show that the perception can be used to significantly improve the tail of throughput. Additionally, our results show that perception can also be used to significantly reduce power needed for beam management.

Vehicles are increasingly getting connected with 5thGeneration (SG) cellular connectivity, and are expected to be connected to 6thGeneration (6G) in the future. One of the key challenges of enabling cellular connectivity in a vehicle is radio frequency (RF) cabling from the sharkfin antennas on the top of the vehicle to modem placed inside the vehicle. Such placement of antennas is needed to negate the impact of out-to-in (O2I) losses introduced by a car body. Cost of the cabling can particularly be a concern as cellular systems utilize higher bands. In this paper we study using optically transparent and transmissive meta-surface(s) on windows of a vehicle to eliminate/mitigate the O2I loss. This can eliminate need for RF cable. To do so, we determine the 3-dimensional O2I loss of a Volkswagen Atlas sports utility vehicle (SUV) through measurements in an anechoic chamber. Next, based on this measured O2I loss and the proposed theoretical link model for a transmissive meta-surface we perform link-level analysis of the in-car antenna system with meta-surfaces installed on various glass surfaces. We analyze actively controlled and fixed design meta-surfaces. Our results show that actively controlled meta-surfaces (controlled with full channel information) of desired dimension and placed on the driver and passenger side window can recover the O2I losses in almost all cases. We additionally show that an in-vehicle antenna with fixed design meta-surface (i.e., with fixed refraction angle) can negate the O2I loss to be within 3 dB 90% of the time.

Integrated sensing and communication (ISAC) is being discussed in 5G Release 19 and will likely be a critical component of 6G. In this paper we study ISAC for automotive applications. Specifically, we focus on adding sensing capability to a vehicle modem. Such additional capability will increase the value proposition of adding higher bands connectivity to vehicles. We perform a comprehensive study of challenges needed to enable automotive grade sensing in a vehicle modem, and propose solutions to those challenges. We show how cyclic prefix-orthogonal frequency division multiplexing (CP-OFDM), which is used for communication, can be combined with novel receive processing to meet (and even exceed) the automotive sensing requirements. Furthermore, we propose a two-stage resource allocation scheme to minimize the overhead needed for sensing. This simplifies the tradeoff between communication and sensing. We also perform system level simulations to quantify the impact of sensing interference on cellular communication. Finally, we carry out measurements to identify antenna placement that can be used for both communications and sensing. Then, we prototype our proposals to demonstrate automotive level sensing using CP-OFDM.

In this paper, we study the cooperation between a connected vehicle user equipment (UE) and a handset UE within the connected vehicle. Specifically, we investigate how a vehicle UE can assist the performance of a handset UE inside the vehicle. We propose to use the vehicle UE as an alternative link for the handset UE inside the vehicle, thus taking advantage of the vehicle UE's externally mounted antenna to eliminate vehicle penetration loss to the handset UE. We demonstrate dynamic switching between the vehicle UE link and the handset UE link is the simplest solution, and the solution achieving significant performance gain. A switching algorithm is designed to improve handset UE coverage when inside the vehicle, while placing no additional requirements on the cellular communication architecture or specifications. The performance of the switching algorithm is prototyped and verified through extensive test drives.

One of the key features of 5G is the NR V2X (vehicle to everything) sidelink that has been first standardized during Release 16. V2X is designed to operate in a distributed manner, enabling operation independent of wireless network coverage. This differs fundamentally from traditional cellular networks where resources are centrally managed by gNodeBs. For distributed operation, 3GPP has defined a set of pre-configuration parameters in the Radio Resource Control (RRC) specification which can be set by the network, or pre-provisioned in the UE for OOC operation. The standard does not assign values for the pre-configuration parameters, but rather defines value ranges to provide flexibility and accommodate a rich span of applications and environments. To enable efficient and inter-operable operation these parameters need to be set carefully. In this paper, we evaluate pre-configuration parameter selection for several key aspects of NR Release-16 sidelink operations. In particular, we focus on distance-based quality of service, congestion control and resource reservation. Our results suggest parameter values suitable for various V2X applications, traffic types and operating conditions.

Fast and reliable communication between vehicles is a key factor in improving the transportation safety. To enable such communication, the two main competing technologies are Cellular V2X (C-V2X or LTE-V2X) and IEEE 802.11p-based DSRC. This paper presents a comparison between the performance of C-V2X and DSRC in dense highway scenarios. C-V2X and DSRC implementations considered here are compliant to SAE J3161/1 and SAE J2945/1 standards, respectively. The results show that with a similar bandwidth of 10 MHz, in terms of both inter packet gap (IPG) and reliability, C-V2X performs better than DSRC at ranges greater than 100m, while both technologies exhibit good performance at shorter distances. Additionally, with SAE J3161/1-based one-shot transmissions for critical events, C-V2X performance at shorter distances is further enhanced. Moreover, with the larger bandwidth of 20 MHz, C-V2X further improves in performance.

Autonomous vehicles (AVs) have the potential to transform road transportation by improving safety and increasing traffic/fuel efficiency. Currently, AVs rely primarily on line-of-sight sensing technologies to gather information about their surroundings. Some of the surrounding objects (e.g., vehicles, infrastructure, and pedestrians), however, can potentially communicate with the AV to help it create a better digital map of the world around it. In this work, we quantify the gains vehicle-to-vehicle (V2V) communication can provide for AVs. Specifically, we focus on the safety and maneuver planning of the AVs. Our preliminary findings indicate that V2V can reduce a significant fraction of AV collisions. Further, V2V can considerably reduce the maneuver completion time.

The theoretical analysis of C-V2X based basic safety related ITS spectrum needs is presented. The spectrum needs are calculated by two methodologies: traffic load mapping and queuing theory. The assumptions for transport traffic modelling were based on China transport status. According to the calculation results, at least 20MHz should be allocated to support V2V/VI basic safety application and total 30MHz should be allocated for V2V/V2I/V2P. Based on the technical study, China government allocated 20MHz for LTE-V2X trial in the end of 2016, which is mainly for V2V/V2I. The methodologies presented in this paper, enables the spectrum requirement for other regions and countries to be evaluated with the adjusted assumptions.

Ad hoc wireless networks without central authority have attracted much attention in both academic society and industry. A key aspect of communications in such networks is the large dynamic range of interference. OFDMA (multiplexing simultaneous transmissions in different frequency tones) is an effective way of dealing with interference. On the other hand, there are many known distortions in transmit and receive chains of the communication devices, e.g. non linearity of the power amplifier, which can compromise the OFDMA orthogonality among the resources assigned to different transmissions. These distortions, together with the large dynamic range one would expect in ad hoc communications, can pose a threat to the effectiveness of communications in OFDMA ad hoc networks. In this paper, we propose and evaluate different signaling choices for communications in ad hoc networks, via both analysis and simulations. In particular, we observe that OFDMA based single tone signaling, i.e. each user transmits only on a single frequency tone, outperforms other signaling choices in many aspects.

We consider a cellular network or a wireless local area network (WLAN), deployed with attachment points (APs) capable of transmission and reception in multiple radio- frequency (RF) carriers. A major factor contributing to the efficiency and stability of the network is the mechanism determining the connection of terminals to the appropriate AP and the RF carrier. This paper describes a practical, distributed, network-assisted and terminal-driven mechanism to determine the connections for load-balancing among the available RF carriers. The mechanism is based on a metric that we term as the service level indicating metric (SLIM). We extend the notion of SLIM to cases where Quality-of-Service (QoS) parameters are specified and propose a load-balancing mechanism for such cases. We demonstrate the near-optimality of the proposed mechanisms through 3GPP based simulations.

We study the performance of an opportunistic scheduling schememaximum quantile scheduling, i.e., scheduling a user whose current rate is in the highest quantile relative to its current rate distribution, in a wireless system. In a practical scenario it is unlikely that users' rate distributions are known at the scheduler, and have to be estimated via measurement. Under the assumption of fast fading, we prove a bound on the relative penalty associated with such estimates, showing that number of independent samples need only grow linearly with thenumber of active users. This is a fairly limited cost, suggesting one could track distributional changes in users' channels. By contrast other opportunistic scheduling schemes require estimating or setting weights/thresholds that implicitly depend not only on the number of users, but alsotheir rate distributions, andpossibly their traffic characteristics. In other words the penalty associated with tuning weights for other schemes can be higher than that associated with estimating users' rate distributions for maximum quantile scheduling. This statement is supported by our simulation results. Furthermore we prove that if rates are bounded and number of users is high, maximum quantile scheduling is sum average throughput maximizing subject to temporal fairness.

We study reducing feedback overhead of users' channel state information required for opportunistic scheduling at a base station while minimizing the throughput penalty incurred due to reduced feedback. We first propose a simple contention based scheme known as 'static splitting' for best effort traffic. The idea is to divide users into static groups, with users that belong to a group, and have their currently supported rate above a threshold, contending to send their current feedback to the base station. This is combined with maximum quantile scheduling -scheduling a user whose current rate is high relative to its distribution, to obtain thresholds that are independent of users' rate distributions even when they are heterogeneous, allowing off-line optimization of thresholds. Next we develop the insight that for a traffic mixture of best effort and real-time traffic one has to combine contention and polling to reduce feedback while providing quality of service. We propose a scheme based on this insight. Under this scheme we prove a lower bound on the service seen by a real-time when users' channel capacities are fast fading. Furthermore, we propose a heuristic modification that is able to exploit a larger fraction of opportunism. Simulation results illustrate the performance advantage of the proposed schemes.