MCI planned to enter the wireless market by building a nationwide GSM network, which would have been the largest in the world. Instead, Burt Roberts, MCI CEO, was persuaded to purchase a share of Nextel, which at the time was a startup, using a then-unknown Motorola proprietary air interface protocol called iDEN. The proposal was to buy 2.5% of the company for $1.3B, implicitly valuing the company at $52B.
I was tasked to evaluate the proposed investment.
Performed a review of the iDEN technology, including excessive voice channel delay, poor voice quality, capacity, and coverage. Met with Motorola’s iDEN team to discuss technical issues. Performed drive testing on Motorola’s demonstration network in Los Angeles. Prepared business case scenarios comparing iDEN and GSM.
My team and I concluded that the iDEN technology had serious, potentially fatal, flaws; that Motorola was not interested in fixing these flaws; that the technical and business risks were unacceptable; and that the valuation of Nextel implied by the proposed acquisition was greatly inflated.
After reporting to senior management on our findings, MCI cancelled its plans to invest in Nextel, and instead entered the market as an MVNO – mobile virtual network operator – reselling re-branded services offered by existing cellular carriers.
Motorola’s intransigence regarding fixing the iDEN technologies’ fatal flaws was finally broken by Craig McCaw, who invested in Nextel with the understanding that the flaws would be fixed. They were fixed, and Nextel went on to become a part of Sprint Nextel, contributing about $10B in revenues to the overall $41 B company.
2. The First CDMA Cellphone System
AT&T, then a monopoly in most areas, announced that it would enter the cellular mobile telephone market using an inefficient analog technology called AMPS – Advanced [sic] Mobile Phone System. I was tasked to compare analog vs. FDMA. TDMA and CDMA to determine which technology made the most efficient use of the available spectrum.
Performed mathematical analysis and constructed computer-simulated networks using each technology.
Concluded that CDMA was the most efficient technology provided that it was equipped with effective power control.
Awarded two US Patents on CDMA cellular systems. Published numerous papers that turned out to be highly controversial. Eventually sold rights to the patents for a confidential sum.
The concept was picked up by Drs. Irwin Jacobs and Andrew Viterbi, who founded a company called Qualcomm. In the mid 90s the service was launched and has become a global industry worth $billions.
When the GSM standards body deliberated on third-generation digital phones, it also chose CDMA as its technology base. By the end of the decade more than 85% of the world’s cellphones will be CDMA.
A Westminster, Colorado startup company called TensorComm was developing an interference-cancellation technology for cellular and other mobile systems. TensorComm was working with venture capital firms Sequel Venture Partners (Boulder, CO) and iSherpa Capital (Greenwood Village, CO) to secure its next round of funding. The venture firms wanted to know if TensorComm’s intellectual property was well-protected by its portfolio of US and foreign patents.
The venture firms knew that as an expert witness, I was familiar with the security and validity of patent claims, and how potential infringers try to invalidate claims and/or prove non-infringement. They asked me to look at TensorComm’s IP portfolio.
The quality of a patent is determined primarily by three factors:
- Novelty: The patent must disclose inventions that have not existed before (prior art)
- Non-Obviousness: The patent’s claims should not be obvious to one skilled in the art, for example by combining two existing inventions.
- Enablement: The patent must describe the invention in sufficient detail so that one skilled in the art can create it without excessive research and development.
The entire portfolio must be examined for each quality factor.
- Novelty: Searched US and foreign patent databases as well as the technical literature for prior art that could invalidate one or more claims.
- on-Obviousness: Searched the same sources for inventions that when combined produce one or more of the patent claims.
- Enablement: Read each patent specification (the main descriptive body of the document) to determine if each claim can be implemented as described.
- Conclusion: Each and every claim in each portfolio patent was novel, was not obvious and was adequately enabled by the specification.
I reported these findings to the VCs.
TensorComm received $7m in financing from Centennial Ventures, Access Venture Partners, Crawley Ventures, and iSherpa Capital among others.
TensorComm’s portfolio now exceeds 75 patents.
In the press, FierceWireless named TensorComm as one of its “Fierce 15” wireless companies of 2007
Formus Communications, Inc was formed in Denver with a view to acquiring LMDS (i.e. 28GHz) spectrum in the US and overseas. The US spectrum was acquired at auction and sold shortly thereafter with a profit of $400m. Formus also wished to acquire spectrum outside the US.
Depending on the country, spectrum can be licensed on a first-come, first-served basis; for a fee; as a result of a competitive bid process; or at auction.
Formus used all of the above techniques to acquire spectrum in various countries. I participated in “road show” meetings to raise $450m capital; delivered technical seminars to the staffs of several communications ministries; managed the creation of bid documents in the case of competitive bidding; and planned and installed networks in several of the countries.
Formus acquired spectrum in 14 European countries with population coverage of about 200m. Before the “Telecom bubble” burst the company had been valued at around $1B, without breaking even.
5. Network Design in Beijing
Motorola had sold an iDEN network to an operator in China. Generally RF coverage design either is done by the vendor, or by a third party RF planning company such as MSI or LCC. In this case the task was taken on by the Beijing Railway University.
Motorola was concerned that the University design, having been done by an inexperienced team without computer assistance, would be flawed. The problem: how to effect a re-design while allowing the University people to save face.
I was engaged as a “diplomatic third party engineer” to travel to China, meet with all parties and propose an acceptable solution.
Without critiquing the University design, I demonstrated the use of the MSI RF computer planning tool and suggested they check their “pencil and paper” design against it.
I also persuaded Motorola to pay for a copy of the software and provide it free to the University.
Motorola made its $200m sale and deployed a properly designed system to the satisfaction of the local operator.
Pfizer Animal Health desired to enter the nascent market of pet tracking – a service that would allow an owner to find a “missing” pet. Having no wireless expertise on staff, Pfizer hired me to do a preliminary design using off-the-shelf components, and to estimate the cost and physical size of the product.
Conducted extensive research into available products: cellphone board, GPS and Bluetooth components, batteries and packaging.
Selected candidate components based on size, battery drain, performance, and cost. Put together a sample package (on paper) to demonstrate the size and form factor; estimated battery life; contracted to create sample moldings of the packages to show Pfizer.
Interviewed several industrial design firms at Pfizer headquarters (NY) and made recommendations on the best company to do the design.
Rather than go through the pain and risk of developing a completely unfamiliar product, I counseled Pfizer to acquire GlobalPetFinder, a startup company that had already fielded a similar product and that had a strong patent portfolio in the area. The acquisition was accomplished in a few months for an undisclosed sum.
In the early 90s the FCC was proposing to award spectrum licenses in the 1.9GHz region for “personal communications services” (PCS) However, in a break from past practice, the FCC declined to specify what service should be offered and what technology should be used to deliver it.
As a result dozens of technology contenders clamored for market attention by claiming the superiority of their approach. Each technology had been demonstrated, but in different locales and under different circumstances.
As Director of Wireless R&D at U S WEST Advanced Technologies, I was assigned the task of finding a way to compare these candidates on an apples-to-apples basis.
The “apples-to-apples” requirement suggested the use of the same geographical area and cell layout for each technology. This suggested the use of a “test bed” in which each base station consisted of an empty box furnished with wired interconnect and power, into which each vendor could install its candidate equipment.
The network would then be subjected to a standard battery of tests to determine its performance, and by agreement these results would be published for all to see.
The conclusion was that CDMA was the obvious choice, and U S WEST’s New Vector division – operator of U S WEST’s cellphone service – chose to use CDMA for its digital network. Eventually there were deployments of CDMA cellular technology throughout the US and in many other countries.
The service generates $billions worldwide and with the adoption of CDMA for third generation systems globally, some 85% of the estimated $600B market will be CDMA by the end of the decade.
Among radio engineers it is well known that the earth’s atmosphere is a strong absorber of radio energy in the region of 60GHz. This is due to the resonant frequency of the oxygen molecule.
The US Air Force wished to use this frequency for battlegroup air-to-air digital voice communication. The thinking was that due to the strong atmospheric absorption the close-range link could be neither intercepted nor jammed from the ground or from distant aircraft.
The main problem was that due to the short range and the high velocity of the aircraft, Doppler shift would move the signal out of its passband. It was therefore imperative to track the signal as its frequency changed, and to do so fast enough to avoid losing track.
My approach was to use a Fast Fourier Transform chip to track the signal incrementally. The same device could also be used to demodulate a frequency-shift-keyed signal.
The air force awarded Litton a $200m contract to deliver hundreds of the modems for use in fighter aircraft.
A Golden, Colorado startup, Superconducting Core Technologies, had been sustaining itself with small Government contracts and wished to move into a market with a more expansive scale. They hired me as a consultant to suggest target markets for commercialization.
After considering several alternatives I suggested that SCT target the cellular base station market. Because base stations usually have antennas separated from receivers by at least the height of the tower, the receivers are notoriously noisy and insensitive. By installing a superconducting filter and amplifier at the masthead, the uplink system performance could potentially be improved between twofold and tenfold.
SCT designed and built several prototype units and demonstrated them at trade shows and in actual commercial networks.
SCT was acquired by ISCO International, a $10M “boutique” supplier of superconducting cellular products.
Ball Aerospace spun out a new division, Ball Wireless, to commercialize its considerable expertise in antenna design. However, the new division lacked a strategic direction for its products. They invited me to spend a day brainstorming possible alternatives.
I gave a brief presentation to Ball Wireless personnel including top management, and then we spent the rest of the day brainstorming how Ball would differentiate itself from the large number of antenna suppliers already in the market.
The general agreement by the end of the meeting was that low-margin “me too” products would get the new division nowhere. What was needed was a set of innovative products that addressed specific needs that arise in the field; for example, antennas with dynamically steerable main lobes and nulls to improve signal to interference ratio. Ball had the expertise to develop such products quickly and cheaply.
Ball Wireless decided to ignore my “high road” recommendation of innovative products and to take the “low road” of me-too offerings. The parent company closed the division’s doors after a few months.