Analyzing the role of RF-SOI technologies for smart mobility.

The transportation industry is experiencing fundamental changes as it embraces the concept of sustainable and smart mobility by increasing its focus on consumers and society’s needs, habits and preferences [Figure 1].

New technologies have enabled a first wave of mobility services, among which ubiquitous wireless connectivity has played a critical role. Nowadays, popular services such as carpooling and ride hailing are commonplace thanks to the widespread adoption of smartphones and other mobile internet connected devices.

Newly adopted 5G, Wi-Fi 6E and V2X, among other connectivity systems, are helping further extending the range of mobility services, for example, by facilitating vehicle wireless interactions with road infrastructure to enhance driving safety, with other vehicles to optimize traffic conditions. Vehicle wireless connectivity can also help make high quality video and audio entertainment content easily accessible in-vehicle for passengers’ comfort.

In [1] and [2], we discussed how RF CMOS technologies and RF-SOI substrates have been evolving together to make high performance RF and mmW Front Ends (RFFE) available that have the degree of reliability and robustness required for new emerging applications. This article aims to provide a more detailed analysis of the role of RF-SOI technologies for smart mobility.

We have put a particular emphasis on light passenger cars since we consider that connectivity concepts could then be extrapolated to commercial and utilitarian vehicles. 

Section 2 provides general concepts of vehicle connectivity and the systems and network protocols on which it relies. In sections 3 to 6, a particular emphasis is made on the different RF technology blocks needed to secure sustainable, reliable and robust vehicle connectivity. Section 7 provides an overview of one of the most rapidly adopted vehicle hands-free systems: UWB. Section 8 provides conclusions and summarizes how vehicle connectivity relies on a multitude of RF technologies and techniques, and how most of them leverage on the rich know-how gathered over many years of CMOS on RF-SOI Front End (RFFE) evolution. 

As presented in [3] and [4], and depicted in Figure 2, connected vehicles rely on multiple systems and networks that provide:

In terms of hardware, several antennas and radiofrequency front ends (RFFE) are required to ensure full vehicle wireless connectivity. In passenger cars, the most common placement for such antennas has been the shark-fin located on the car’s rooftop but as car aesthetics evolve, antennas integrated in conformal panels are also being widely adopted. In both cases, the antennas are placed near the surface of the vehicles to avoid the Faraday cage effect of the car metallic bodies. Furthermore, to minimize losses and interferers that could jeopardize the sent/received information integrity, various Original Equipment Manufacturers (OEMs) are making the choice to place the RFFEs as near as possible to the antennas, as shown in [5].

The following points discuss in further details the specificities of each connectivity system and the particularities of each one of their RFFEs.

As all wireless devices, connected vehicles depend on several RF ICs and RF modules for reliable wireless connectivity. Most of such elements are contained in a “box,” commonly referred as T-BOX for telematics box, also known as TCU for Telematics Control Unit. As shown in Figure 4, the TCU contains function blocks for sensing, positioning and data storage, processing and transfer. Among all these blocks, the NAD (Network Access Device) is composed of all the circuitry required to secure a reliable and robust cellular network (4G LTE or 5G) communication – including the RFFE depicted in Figure 5. It is the main focus of this segment.

Different car models and their optional features are designed to be commercialized in specific world regions. The car’s NAD needs to comply with the rules and regulations for cellular connectivity in each of these regions. Figure 6 shows how mid-band frequencies (bands around 3.5GHz) are among the most used cellular spectrum around the world; it is therefore not a surprise that many NAD providers and users choose C-band as the privileged band for the RFFE operation.

C-band offers a good trade-off between coverage and bandwidth, and consequently data rate [6]. It is important that NADs using such bands coexist with minimum interference among themselves but also with other connectivity systems, such as Wi-Fi and C-V2X; RFFE linearity is therefore a key element of design.

As discussed in [1] and [2], interferers could arise at any point of the RFFE passive (transmission lines, inductors, etc) or active circuitry (transistors, diodes, etc). By using trap-rich RF-SOI substrates, over which much of the RFFE circuitry is built, such unwanted signals are minimized wherever they take place. Figure 7 shows how the different frequency bands used in the TCU are located very near one another and how leakage from one frequency band could affect the neighboring one; RF-SOI helps minimize such leakage.

Figure 8 shows how an RF signal’s second and third order harmonics non-linearities are minimized by using trap-rich RF-SOI substrates. The improvement is clear when compared to non-trap rich RF-SOI (HR-SOI) substrates. A Coplanar Waveguide (CPW) is used to perform representative RF characterization of the non-linearities.

In order to provide the required flexibility to address different regional markets, modularity is another key element of the NAD design. RFFE components that integrate the transmit, receive and filter functions in a few modules are preferable as they could be rapidly swapped, depending on the region in which the vehicle is commercialized and to comply with regional regulations and with local customer preferences.

As shown in Figure 5, RF-SOI provides an unmatched level of integration flexibility for RF Front End Modules (RF FEM), facilitating the integration of high-performance Low Noise Amplifiers (LNA), switches, and Power Amplifiers (PA) in monolithic dies or in high value-added multi-technology modules associated with filters and other support functions.

A common public misconception is that vehicles – given their capacity to integrate a large battery – can accommodate electric power-hungry systems. This is very far from the truth; modern vehicles’ constraints on power consumption efficiency are as high as those of flagship smartphones or any other modern portable wireless device. Modern vehicles need to accommodate an increasing number of sensors, MCUs and other electrical systems whose power consumption must be strongly optimized.

NAD designers should closely watch the RFFE current consumption, the power dissipation and the overall power efficiency. Minimizing RFFE insertion loss and the overall RF link budget are a must. Furthermore, a power efficient NAD’s RFFE dissipates reduced amounts of power as heat, greatly helping to ensure a highly reliable operation of the whole NAD.

To minimize the losses associated with a lengthy connector, the NAD – i.e. the TCU it is part of – is typically located not far from the antennas and therefore exposed to the wide temperature variations of the vehicle body. Furthermore, it is well known that modules integrating PAs could see their operating temperatures rise to more than 85°C. The combination of both poses a serious challenge for reliable and robust automotive RFFE operations and should be carefully considered from the early stages of the design.

Under wide temperature variations [7], an RF-SOI substrate that provides a stable linearity performance provides a definitive advantage. Helping to ensure that a rise in temperature won’t considerably increase RFFE non-linearities, susceptible to jeopardize the NAD functionality and/or interfere with neighboring wireless systems, is a major differentiator for any automotive RFFE.

Low linearity drift over temperature – RFeSIxT – is a new feature added to Soitec’s RFeSI family of products [8]. As shown in Figure 9(b), it confers stable linearity performance at temperatures beyond 85°C while keeping all other RFeSI substrates benefits [1]. As reference, Figure 9(a) shows Soitec’s RFeSI (without the RFeSIxT feature), intended for consumer grade products, linearity behavior over temperature.

Providing safety to the car driver and passengers is crucial role for all automotive systems, including telematics. The RFICs that are part of the car emergency call system are discussed more in detail in point 3.1.

One of the key components of the wireless emergency system of a car is the one that provides critical information to emergency response teams in the case of an incident – ambulance, firefighters, etc. Critical information includes car positioning (e.g. GNSS coordinates), timing of the incident, car’s and passengers’ conditions (cause of the emergency call) and car identification, plus any other information that could help save crucial time when providing first aid support [Figure 10].

Different world regions and mobility players have different strategies to provide emergency connectivity. Such service is provided by private players [9][10] or, in the case of the European Union (EU), through a public car emergency call (eCall) response service, accessible in all EU countries [11]. The eCall is a mandatory system for all new cars sold in the EU since 31 March 2018. The EU specifies that the eCall systems in new vehicles must be able to:

In case of a vehicle accident, ensuring that the data to be transmitted can find a path to a working TCU antenna, regardless of the state of the vehicle involve, and reach the emergency response team, is not an easy task and demands a very robust RF switch. Emergency call RF switches should be capable of hot switching at powers compatible with cellular networks (up to several hundred milliwatts) and be certified ASIL A [12]. RF-SOI technology is the most commonly used technology to design switches [13] complying with both requirements.

If the NAD has an integrated eCall system, the NAD itself should be certified ASIL. Some manufacturers would therefore opt to have the eCall system located outside the NAD, which will remain to reach a certified automotive consumer grade, reducing design complexity and costs associated associated to the certification. Once again, a modular TCU design, facilitated by the use of CMOS technologies on RF-SOI substrates, helps with a flexible TCU + eCall design as shown in Figure 11(a). RF-SOI, as mentioned before, is the technology of choice for different topologies of eCall switches, Figure 11(b).

In order to provide not only emergency services but also regular ones (e.g. firmware updates), in some cases car OEMs partner with MNOs [14], leaving car owners with little to no choice on their cellular operator. To provide car owners more flexibility with regards to their own cellular service provider, some car OEMs implement a Dual SIM Dual Active (DSDA) [15] system approach, which is discussed in more detail in section 3.2.

The Subscriber Identification Module, or SIM card, is intended to store the international mobile subscriber identity (IMSI) number and other data unique to any cellular network user. This card is an essential element to link a device to a cellular network.

A DSDA configuration relies on, as its name suggests, two SIM cards and requires two separate transceivers, and the RFFE transmitters (Tx) and receivers (Rx) associated with them, in order to provide connectivity to two active operators. By using a DSDA system, car OEMs could continue to rely on their partner operator while providing vehicle owners with the flexibility to concurrently use their own preferred operator, and thus benefit from personalized services such as family data plans and others.

While using two RF pathways provides a certain advantage, it also increases the NAD power consumption and the RFFE complexity. Linearity and the associated power consumption efficiency are therefore a key design consideration for a DSDA system. As previously discussed, RF-SOI provides a clear linearity advantage and helps optimize the power consumption of the NAD’s RFFE. Furthermore, thanks to the modularity of the NAD, it is possible to envisage alternative approaches such as having a 2×2 MIMO diversity, instead of the 4×4 one previously depicted in Figure 5, to optionally reduce power consumption and complexity of a Cellular (4G LTE or 5G) DSDA NAD, as shown in Figure 12.

Luis Andia is senior manager RF business development at Soitec.

Yvan Morandini is RF business development and product marketing manager at Soitec.

Jean-Marc Lemeil is general manager of RF-SOI Business Unit at Soitec.

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.