Q: In what ways are manufacturers preparing for 5G?
By Michael Chinn, Deputy General Manager ICT Group, Electronic Components Sales & Marketing Group, Senior Vice President, TDK
Referred to as the enabler of the “Fourth Industrial Revolution,” 5G facilitates wireless solutions for applications across many industries by reliably connecting massive numbers of devices, providing ultra-low communication link latency and gigabit speeds.
5G can wirelessly connect millions of Internet of Things (IoT) devices in a dense urban area, for city-wide remote sensing or beacon networks. Connectivity modules are effective deployment solutions, allowing pre-certified 5G radios to easily deploy, and integrate in massive IoT networks. To minimize size, modules rely on advanced manufacturing techniques such as high-density System in Package (SiP) assembly, including the embedding of active and passive elements in the module substrate, and use of highest performance passive components to complement the advanced ICs in use. Novel power storage technologies are also useful, including solid-state low-voltage batteries or energy harvesting from solar cells or other sources, increasing the autonomy of devices.
5G will eliminate cellular network data bottlenecks, enabling simultaneous ultra-high-speed wireless transmission (even in crowded environments), by opening new radio spectrum like millimeter wave (mmWave) bands over 20 GHz. At first, 5G mmWave will likely be used by wireless carriers to provide Gbit Fixed Wireless Access to residential and SOHO customers. The next step will be using 5G mmWave to deliver Gbit mobile, such as live streaming player’s eye 4k video to smart devices of a crowd at a big game. This demands new antennas and RF filters, based on novel ceramics and other materials. These advanced mmWave components are necessary for handheld devices and network infrastructure base stations, where massive MIMO antenna arrays will be used for dynamic beam steering to maximize network capacity.
5G will transform businesses and daily lives. High performance electronics, including the latest passive component technologies, are essential to enable the hardware needed to make 5G a reality.
By Benny Solomon, AOI and AOS Marketing Director, Orbotech
To effectively address the question of how manufacturers can prepare for the 5G wireless networking juggernaut, one can look ahead to the impact 5G will have on electronic end devices targeted to support 10X faster data rates and 1000X more data traffic. 5G will facilitate powerful connectivity throughout all industries, and manufacturing innovation must advance to support the scale and quality necessary for reliable implementation. The impact of 5G on PCB production provides a good specific illustration of the changes needed.
Essential to high-frequency 5G network connectivity are the ubiquitous PCBs at each node that will enable the projected increases in 5G data rates and bandwidths. Faster download speeds will drive seamless, real-time capabilities for new applications from augmented reality and virtual reality (AR/VR) to life-critical autonomous vehicle sensing. For these 5G-driven applications, the allowance for error is virtually nil, and with current IoT device growth forecasts in the billions, the network’s vulnerability to failure is exponential. These latency-sensitive applications will drive PCB reliability standards to new levels, directly influencing manufacturing practices to ensure an equally high level of quality production and inspection.
PCB manufacturing faces unique challenges due to the 5G network’s higher frequencies and shrinking form factors. Increased I/O in ever-smaller designs commands high-density interconnects (HDIs) with inherently thinner board traces. These ultra-thin lines can introduce possible signal performance degradation. For example, if the line’s physical characteristics, such as top and bottom width, vary from the original design, RF signal transmission can be delayed, negatively impacting downstream data flow. By necessity, manufacturers are challenged with deploying myriad new innovations, such as modified semi-additive processes (mSAP) to ensure designs are executed well in production. The next challenge in PCB production is tackling testing accuracy and reliability.
Automated Optical Inspection (AOI) tools have been successfully used to inspect PCBs for defects, but until now, AOI systems primarily inspected CAM designs to ensure the original design was faithfully produced and adhering to design rules. For 5G-enabled PCB boards, additional capabilities are needed for the physical measurement of trapezoidal and/or rectangular-shaped cross sections. This requires an AOI system that can measure both the top and the bottom of the PCB, as well as inspect different potential defects including laser vias and patterning with minimal handling. A few AOI tools have some measurement capabilities, but still only measure the trace width at the top of the conductor without measuring width at the bottom. Surprisingly, until now, bottom measurements have only been possible by taking samples and inspecting them manually by microscope, an unsustainable practice in light of the scale and yields required for future 5G deployment.
Innovation in AOI technology is showing advancement as demonstrated by PCB manufacturers’ ability to leverage 2D Metrology technology to automatically inspect and measure PCBs’ top and bottom trace conductors. This innovative testing capability can be performed at high throughput rates on a high sampling rate, ensuring better overall yield for manufacturers. This is an important step toward achieving the PCB quality levels needed for cost-effective 5G adoption.
There is much heralding of 5G’s imminent arrival, but its adoption will require change and continuous innovation across a spectrum of technologies. Manufacturers, as the true facilitators of this massive transformation, need to be at the forefront of this innovation to ensure 5G network quality and reliability. Just one small representative example of this is in the necessary evolution of AOI technologies to support high-frequency, low-latency 5G systems by enabling faster, higher-precision PCB inspection and verification.
By James Wilson, Senior Marketing Director, Timing Products, Silicon Labs
Radio access networks are going through a significant transformation in preparation for 5G. Legacy point-to-point Common Public Radio Interface (CPRI) networks between remote radio heads and baseband units are being replaced by Ethernet-based eCPRI fronthaul solutions. These solutions provide a more flexible, scalable way to support the higher bandwidth requirements that 5G will demand. This transition is also spurring the design and deployment of specialized pre-5G radio access equipment that increases network capacity and coverage. New designs for small cells, distributed antenna systems, massive MIMO, and other pre-5G radios face a new development challenge because they must simultaneously support LTE and Ethernet connectivity. This puts unique requirements on timing solutions to support low phase noise LTE clocking, low-jitter Ethernet clocking, and system clocking.
Another key innovation is now underway as metro and edge networks upgrade to higher bandwidths to supporting ramping demand for video streaming and mobile data. A tremendous amount of innovation is happening at the physical layer right now, as Ethernet switches/PHYs, FPGAs, and ASICs are migrating from 28 Gbps non-return-to-zero (NRZ) SerDes to higher-speed 56 Gbps and 58 Gbps PAM4 phase-amplitude modulation SerDes. PAM4 packs more bits into the same amount of time on a serial channel by squeezing in four states per cycle. The resulting signal-to-noise ratio (SNR) has to be much better to ensure the link’s bit-error-rate does not degrade. This, in turn, is driving the need for lower jitter clocks and crystal oscillators (XOs) to provide reference timing for 56G PAM4 SerDes.
By David Ryan, Senior Business Development and Strategic Marketing Manager, MACOM
The exciting evolution of 5G has been set in motion with the recently announced 3GPP standards; carriers are already announcing first deployments within 2018. From a radio perspective, OEMs and operators are treating these first deployments as the natural evolution of existing 4G networks and current architecture. Their initial steps will be to evolve this architecture in the traditional sense by adding more MIMO, splitting up the antennae into smaller pieces and putting a transceiver behind each one, similar to the 4G/LTE-A Pro Massive-MIMO strategy, which is already validated in both China and Japan. This solution, while not unlocking the potential of a full active antenna array, uses less complex hardware and fewer transceiver paths, thereby keeping initial costs down, and many manufacturers are leaning toward this natural step-by-step evolution.
In effect, these solutions put a lower power transceiver behind every antenna subsection—a typical 192 element antenna array, consisting of 12 rows, eight columns, and two polarizations, will be driven by 64 transceivers. Typically, such an implementation will be approximately 0.8 m tall and 0.4 m wide, fitting comfortably in the footprint for an existing Macro cell antenna.
Since the increased number of transceivers, combined with wider bandwidths, will generate a huge amount of raw data, the availability of cost-effective, high-speed, front-haul solutions as well as fiber capacity are also driving equipment vendors to reconsider solution partitioning. The new eCPRI standard effectively reduces the bandwidth requirement for fronthaul networks by integrating CPRI processing function in RRU. However, emerging low-cost 100G optics offer carriers the option to adopt CPRI for future-proof RRU implementation.
The 5G standards may have just been set, but it’s safe to say manufacturers around the world are already busy!