Technology is driving a new “interconnected world.” In this dynamic, ultra-connected environment, memory has become a critical subsystem for the storage, processing, and delivery of data 24/7, particularly between individual embedded systems and the cloud. Factors contributing to the growing importance of memory include:
- Sophisticated, advanced, multicore processors that enable more software and data to be processed in real time
- Always-on, intelligent connectivity that requires high-performance data transfers to and from the device
- High-level operating systems and tablet-like user interfaces that require higher performance and higher onboard capacities
- Big data generation and analysis that drive the need for additional capacity and increased real-time processing
The automotive sector is just one of many industries undergoing a significant transformation—one that requires more diversified memory subsystems. Today’s automobile has become more like a second home/officethan simply a means of getting from point A to point B. To support this new capability, electronics has continued to grow as a percentage of a vehicle’s total cost.
Traditionally, automotive electronics has been used in three areas: powertrain, advanced driver assistance systems (ADAS), and infotainment. Powertrain represents the most basic use of memory for engine and transmission control units. Memory must meet strict requirements to operate across broad temperature ranges and adhere to strict quality and product lifecycle standards. ADAS and infotainment applications are more complex.
ADAS are individual systems that help increase safety and efficiency. They include applications such as lane tracking, collision avoidance, adaptive cruise control, night/through-fog vision, parking assistance, and rearview camera displays. Some of these systems are becoming mandated by regulations. For example, the Kids Transportation Safety Act (KTSA), aimed at preventing backover accidents, would require rear camera displays in all new cars sold in the U.S., beginning in 2015.
Automotive infotainment systems are experiencing significant growth. Navigation systems are being installed in mid-range automobiles and are trending toward high-resolution 3D images. In-car telematics displays are becoming more sophisticated, featuring intelligent displays of information that match the situational context. In-car entertainment is featuring more robust content and Internet connectivity. Some of these systems are adopting the tablet-like user interface that consumers have grown to love.
Another important trend is the evolution of these formerly single systems into a networked intelligent system that can handle complex scenarios. The best illustration is the quest for a self-driving car. Google is the most visible company working on these types of vehicles, but similar projects are under way at other organizations, including California Institute of Technology. Auto makers, including Mercedes Benz, Audi, and BMW, are beginning to integrate autonomous operations into their own cars and have developed working prototypes.
These cars use a combination of technologies, including radar sensors on the front, video cameras aimed at the surrounding area, and various other sensors and artificial-intelligence software to help steer. Google’s self-driving car uses a roof-mounted laser to take a 3D map of the car’s environment and compare it against high-resolution maps to help it drive safely and in accordance with the law.
The emergence of these intelligent systems is driving a more diverse set of requirements for embedded applications. The requirements encompass performance, volatility, capacity, cost, reliability, operating environments, power, and supportability and, not surprisingly, require tradeoffs.
For the most sophisticated systems, multiple memory technologies are needed—from DRAM and NOR/NAND Flash, to managed memory solutions like e·MMC and SSDs, to next-generation technologies. The memory must also be available in multiple packaging options, including known good die (KGD) for specialized environments and multichip packages.
Types of memory used in automobile applications include:
- Serial NOR: Typically used for booting applications and maintaining data related to road signs, preferred languages, etc.
- NAND Flash: Increasingly important for in-vehicle infotainment (IVI) systems where the application code is increasing in complexity.
- DRAM (DDR2/DDR3/LPDRAM): Used as main memory in sophisticated operating systems and used to buffer audio and video data.
- eMMC: Used mainly for storage and management of big data for navigation (3D maps), music, vehicle information, and advanced safety systems.
The automotive segment is clearly an area where intelligent systems are a source of exploding memory requirements. From automotive systems to mobile environments to medical systems—memory is becoming a more critical aspect when it comes to storing, accessing, protecting, and delivering important information.