Digital Transformations, Success and Challenge Stories

The 3rd industrial revolution, aka 3rd Wave refers to advance Mobile/Wireless Handheld Communications to its highest level and accessibility of massive amount of data/information instantly to mass populations.  4G/LTE has started the 3rd wave and is gaining more momentum with 5G.

Companies have to aim for bigger projects which require multi disciplines professionals and execute in much faster pace, if they want to have any chance of selling their products in the fast-changing market with ROI.

Recruiting professional in many disciplines is not only time consuming but also expensive. Even when a company successfully do that, they don’t have enough work for these highly skilled yet specialized professionals throughout the year.

Nowadays, many corporations and startups rely on using subcontractors.  These companies plan a big project and don’t necessarily have all the required skills to complete it.  Then, they hire subcontractors to deliver specific part of that big project.

This model works well when the HW/FW/SW interfaces are defined well and can be validated.  Here’s handful of corporations using subcontractors, Qualcomm, Apple, Amazon, Meta, and Google.  Their success level is not the same, precisely due to the interface definitions.

Qualcomm was one of the first adaptor of this model.  They started it back in ~2010 before anyone else in the wireless market.  They have successfully executed many projects since then by using subcontractors to lower the cost and execute faster and bringing the product to the market before their competitors.

Currently, Kuiper project is relying in this model.  The user terminal front end is mostly in house design and development, yet the baseband modem and processor are not.  The challenge is the integrations of the interfaces which have yet to materialize as a successful SATCOM UT with the promised datarate.

To be successful, adequate advance project concept and plannings are the essential.

This planning should comprise of business goals and proper decomposition of these goals into technical requirements which consists of HW/FW/SW.  Then, each partition can be done independently whether it is an internal team or subcontractor executing them.  For instance, BT IP, Frac N PLL IP, AGC algorithm, or Beamforming Algorithm could be subcontracted.

This model referred to as digital transformations.

Partner with ORTENGA for your products planning and executions.

 

Qualcomm Declining Mobile Market Share

Many are wondering why Qualcomm mobile market share is shrinking year over year.

There are some external developments which are related to the competitions, such as; Samsung, Media Tek, Apple, etc.  These are outside of Qualcomm control and it cannot directly do anything about them.

There are also some key internal business developments plan and decision which not only are directly controlled by Qualcomm, but also can easily be changed.

Qualcomm is not investing enough in new vertical technologies and is behind the curve.  The business development team’s expectation is that once the market changes direction, so would Q.

That passive business mindset is not what the original founders were thinking when they conceived the company when no one else in the market believed on CDMA technology.  That was a significant undertaking and calculated risk to invest in something that nobody else has tried and believed it can pay off.

The Qualcomm founders choose outside of the box technology and business development plan which fitted their time and market.  In order to prove the CDMA technology, they invested from A to Z.  Then showed the rest of the world and market the spectral efficiency superiority of CDMA vs. FDMA/TDMA.  Once that was accomplished, they sold all the peripheral businesses and capitalized on the ROI.  At that point, they monetized the licenses and the CDMA product portfolio.

It requires new business plan and vision to capture verticals that may not be apparent to everyone in the mobile market.

Partner with ORTENGA for new product concept, definition, and development.

 

What Cause a Technical Project Fail?

In technical world of business, typically the failures are due to having a business goal, yet could not decompose that to technical requirements which derives the execution.
By the time, the management becomes aware of that issue, if ever, significant time has elapsed (competition has gained ground) and resources are diminished.
The failure in the execution costs the company their IP and they end up selling it without ROI.

Partner with professionals who have appropriate experiences to decompose your business goal to the technical requirements and avoid betting and risking your IP.

Apple RFIC and SoC

Apple has long been working on its own silicon to support their products portfolio, iPhone, iPad, Laptop.

The processor, A1x is the legacy and established in all Apple products.  This is how Apple differentiated their products.

The wireless communication historically has been challenging for them but that significant milestones are made during the past decade.  It all started in and around 2013 when Apple business plan changed to reduce silicon cost by internally design and developing one silicon at the time.

After the A1x processor in Apple products, the MODEM (Modulator and Demodulator) is the heart of any radio.  It controls the incoming and outgoing radio signals through transmitter and receiver chain of RF signals.  Also, it is the most expensive silicon in the communications chain.  Therefore, naturally Apple wanted to own and reduce the overhead paying for it.

At the beginning, it appeared another silicon project.  Apple has successfully designed and developed the processor and this would be another piece of the silicon puzzle.  After handful of years of engineering and professional recruitment, they realized the technology know how is much deeper and requires much more specialized engineering to design and develop a MODEM.  In 2019, Apple acquired /bought Intel smartphone modem business for $1B.  That realization and the decision changed the trajectory of Apple upward.

In order to expedite the design and development process, Apple opened design center at the heart of communication technology center in San Diego, CA, couple of years before that.  That decision allowed them to recruit top notch industry experienced right from Qualcomm (existing MODEM supplier), also save operating expenses by investing in the design center as capital (a Win-Win for both Apple and Engineers in San Diego).

Meanwhile, Apple has also expanded the business goal to include RFIC which is the interface with antennas.

During the past 4 years, Apple is continued in the pursuit of their own MODEM which is expected to bear fruit by 2025.

Apple is still missing Antenna technology expertise and leadership for future generations of mobile communications.  6G will include significant changes to antenna which is a missing whole in Apple vision for years to come.

Partner with ORTENGA for Antenna Design and Development of your next generation product.

Antenna Design by AI

Frequency selective filters can be designed by CAD tools such as Genesys.

Antenna is spatial selective filter and similarly can be designed by CAD.

In facts, antenna array governing equations are similar to frequency selective filter and can be similarly modeled.

Design optimization tools will be key to useful simulator for the end users.

The remaining challenge is how to form factor the antenna and conform to required space.

Partner with ORTENGA for Antenna Design and Development to fit small spaces and form factor of the product.

Appropriate Message Format

With the advance of technology and receiving multiple messages on daily basis, the question arises about what should be the format of an appropriate message?

First, any message should have “introduction” of the sender.

“Hello, my name is ….”, “Blue Bank Financial …”, “Orange Motors Service Department…”

Second, the message should provide “authentication” of the sender.  This gets more specific and the level of authentication is typically determined/agreed between the sender and the receiver.

Once the authentication is established, the receiver chooses to either continue or terminate receiving the information.

This process actually occurs in radio communications as well as radar.

In radio communication, there is a protocol for each “frame”, referred to frame format. Each frame starts with “preamble”, which is fairly the first and second steps mentioned above, plus “synchronization” and “equalization” process with the receiver.

The sender sends a “pilot” for synchronization for the receiver, then authentication between the sender and receiver takes place.

Also, the sender may/could send “training” signal to allow the receiver to prepare appropriate “equalizer” to cope with the “channel” or the radio environment.

This is where the advance radio communication protocol allow handshake between “sender” and “receiver”.  It is like choosing the language and the speed before the actual dialog starts.  The sender transmits number of signals and the receiver acknowledges which of those received properly.  Then, the sender continues the dialog, actual information, with that particular waveform.

The waveforms are designed for various “channel” or environment conditions and “frequency dependent”.

In radar, similar validation occurs to authenticate the echo from the target was actually sent by the radar itself, before the radar process the echo signal.  That process embedded in “matched filter” signal processing.  Also, the waveforms are designed and down-selected depending on the target, frequency, and the environment.

These steps and process are also valid for personal messaging and/or face to face dialog and any authentic sender follows before delivering the message.

Partner with ORTENGA for your new product design and development.

MIMO Radar

MIMO aka Spatial Multiplexing is being used in 5G as well as WiFi technologies.

New Radars also utilize MIMO waveforms to optimize for position and velocity accuracy/resolution, simultaneously.

MIMO Radar also allows combination of Active and Passive illuminators, which can enhance the above resolutions.

MIMO relies and exploits multipath scatters and combines the received signals in a such way to optimize for SINR.

In radio communications application, there are handful of transmitted coding which are optimized per channel and predefined. When SRS is transmitted, the receiver replies back a number which reveals the best signal coding received for that channel.  Then, the transmitter starts transmitting data with the same coding until the channel conditions are changed.

Similar algorithm can be utilized in the MIMO Pulsed Radar to enhance SINR.

Partner with ORTENGA in design and development of MIMO Radar.

 

LiDAR Systems Performance vs. ADC ENOB

Does your LiDAR have required operation range? Have you considered various applications and its required LiDAR range into your product definitions?

The following plot illustrates the ADC impacts on the range and error for few applications.

The plot reveals as the range, consequently its error increase, required ENOB increases.
Is your product defined and specified ENOB adequately which can meet the end user needs and requirements?
The customer does not necessarily know about the technical impact and expect the supplier to provide a solution which meets their need.

How about the ADC clock rate or sampling rate/frequency? Have you considered what would be the optimum frequency? Do you have a frequency planning for the ASIC/SoC?
When your product gets selected to interface with other system (smart phone, automotive, etc.), that frequency planning could cause make or break the deal.
Have you considered the margin/tolerances for each parameter in a such way that operationally withstand part to part (process), voltage, and temperature variations, i.e., collectively known as P.V.T. corners for that application?
There is a 1 x 10 x 100 rule in design and development of any product.
Every dollar that is assumed to be saved during the product definitions and design, could cost you 10x more if there is one missing specification, once the product enters development phase.
If by any chance, that product makes it to production phase, that one missing specification would cost 100x to be fixed. Going back to drawing board and find a “kludge” for implementing and/or beefing up one specification that end user cannot live without is costly and could break the deal.
Re-spinning an ASIC is way more costly than HW. Also, every re-spin drives more HW/FW/SW changes within your organization. That implies one change in silicon has a rippling effect in HW/FW/SW changes for the customer integration platform. The end user may also have to go through HW/FW/SW changes to accommodate and benefit that one silicon change.
Leverage experienced product definer and technology SME to join your team.
Partner with ORTENGA to define and spec out your new product.

 

Good Business Decision Story of Facebook

Any business plan which has clear definitions of scope, schedule, and budget could be successfully executed.  Then, if the expected ROI pan out and become reality, then the benefit would be harvested.

Visionary product provides revenue streams for many years into the future, e.g., iPad or iPhone.

Good commercial products provide revenue for couple of years at the most, before incrementally enhanced product replace the legacy.

These business plans are well thought off and vetted by both technical and business professionals and moved forward to executions and produce revenue streams.

In technical industry, there are times that a business plan is launched and only critically vetted by business professionals.  The technical challenges are unknown to the business community, therefore are not properly scoped, and the cost of doing the project increases significantly without a clear path for execution and harvesting any ROI.

The sooner the burning cash is realized, the sooner the cost saving begins, and increases the opportunity for investment in the area of expertise which could yield ROI.

Sometime in 2013, Facebook business executives, decided to launch their SoC design and development for XR applications.  The idea was to start design and developing SoC and associated HW/FW which would support Facebook XR business goals.

Facebook went to hiring hundreds number of Engineering from semiconductor industries with many years of successful SoC experiences from bottoms up, e.g., engineers and some managers.  The business plan was in full motion.  The business plan has not been vetted before launch by technical exports of the semiconductor industry.  The technical scopes and responsibilities were loosely defined and there were not metrics to identify technical challenges and risk.

Fast forward 9 years later in the Summer of 2022, couple of Billion dollars investments later, Facebook realized that SoC design and developments requires many engineering disciplines that Facebook does not have and cannot necessarily recruit in time to complete the business plan concept of 2013.

Facebook laid off thousands SoC/HW/FW/SW engineers which supported that XR business plan.

The new business decision was to partner with SoC experts to harvest this new XR verticals.

Qualcomm’s bread and butter business is design and developments of SoC and has resources, network, proven track record, and understand the cost model of working within semiconductor industry.

The Summer of 2022 Facebook decision was an excellent decision made by Facebook executives to partner with Qualcomm.

The 2022 business decision will bear fruit, much more profitable for both companies which alone could not have realized.

Partner with ORTENGA for vetting that business plan before execution.  ORTENGA is comprised of SME and appropriate network in the industry.

 

How to promote innovations?

Many starts up and corporations lack innovative algorithm, process, and/or technology to survive and breed to the new wave of product lines and market.

They even hire the brightest and intelligent skilled workers in the market, yet they cannot solve some of the problems which was originally seen and assumed to be resolvable.

These organizations go through multiple layers of skilled workers to address the lack of innovation in their line of thinking to address technical problems and end up empty handed after years of investments.

They presume that lack of innovation is due to not having right people working on their projects and are missing the fundamental aspect of any innovations.

Innovations requires challenging the status quo.

If you have skilled workers which have multiple technical degrees and/or engineering disciplines and still come short with innovative results, it is not them, it is the working environment and atmosphere.

Here is why.

Anyone who has advance technical degrees and gone through higher educations and in particular engineering graduate school, has learned to critically think.

Critical thinking drives innovations, if one challenges the status quo.

Foster an atmosphere of critically thinking, allow everyone to challenge anything and everyone within the organization, then watch and harvest innovations.

Do you have radio front end HW which cannot be SW driven to perform?

Do you have radio front end that cannot be used by your customers?

Do you have baseband MODEM which needs to be interfaced with an appropriate radio front end to perform?

Are you designing an ASIC which does not have any design specifications which can be validated with simulations and/or tests?

ORTENGA fosters entrepreneurship mindset and critical thinkers of SME which can help your organizations.

 

Data Analysis for Machine Learning, ML

Any technical data has three layers.

By technical, it implies anything that there is science which reveals cause and effect for various events.

Engineering, medicine, physics, chemistry, biology, and astronomy are among many scientific fields.

To have any chance of predicting any future event which is not necessarily an experiment in controlled environment, the science and theory of that fields have to be known.  That knowledge comes from studying that science, and not only observing prior events.

There are 3 layers of data.

  1. Analytical information which are typically derived from theory.
  2. Statistical information which are typically extracted from experiment or occurring current events.
  3. Correlation of information which are connecting the dots between items 1 and 2.

Successful Machine Learning algorithms have adequate theoretical models which are validated during data collection of current events, and correlate the collected information to predict the impact or high probability of next event to occur in future.

Standalone item 3 above is data analytics.  Without the knowledge of item 1, it misses many obvious scientific cause and effect relation.  And Without item 2, it lacks the science of large populations of events and misses the randomness of inputs or parameters that impact and scientific event.

If the prediction is correct, Machine has learned, otherwise the Machine does not add any value.

Not every ML algorithm has all or adequate level of the above components.  Skipping item 1, would not produce ML, it is book keeping which may rely only on statistics to predict future event and is as good as betting in the horse race.

Partner with ORTENGA for ML algorithm development for your product and/or technology.

 

Machine Learning Algorithms

There are many design problems which do not have known theoretical or analytical solutions, yet they have practical applications, therefore demand in the technical market space.

M2M communications require some level autonomous in learning and decisions making to ensure, no human interactions.

If you have any of these problems, ORTENGA has subject matter expertise to help you define the scope and then provide appropriate ML algorithm to address that.

 

Layoffs Aftermaths and Lessons Learned

Any company or individual gone through a round of layoff knows that is unpleasant experience and have lasting feelings and lessons learned for all parties involved.

The impacted employees learn that never assume any job is “permanent” and “no job” or company is safe.

The companies learn that the easiest part of the ordeal is laying off employees.  Their issues kick in as soon as layoff is executed.  Many loyal employees who remain are also emotionally impacted and the shockwave helps them to look at their options elsewhere more seriously.  The blind trust is gone.

Good companies pay anywhere from a few to several months of severance package and treat their employees with full respect and offer impacted employees training during and post layoffs to land another job.

The companies learn that productivity diminishes as time goes by and recruiting “fresh blood” becomes more challenging than before.  The employees’ referrals diminish significantly.  These impact the business bottom line, which was supposed to be improved after the “layoff”.

The companies either start changing their brand and/or even name or sometime being taken over by bigger corporations due to lackluster business outlook.  After all, layoff implies failure of the business plan which was set to motion by the business executives and leadership.

For those of you who were impacted, there is light at the end of the tunnel.  There are better opportunities waiting for you outside of your previous company.

ORTENGA was found after layoffs of late 2015 by network of colleagues and subject matter exports in Antenna, HW, FW, SW, and IC designs for radio communication and radar applications.  And most of us, if not all are better off and live happier and healthier life.