X

Connect (X)

Tag Archives: edge computing

EdgeMicro Launches No-Cost Testing Environment for Edge Implementations

EdgeMicro is launching a Proof-of-Concept Program that will enable companies to accelerate their timeline for proving the viability, scalability and performance of their edge computing pilot projects. Companies seeking to test low-latency applications that take advantage of edge-located data cache and compute services can apply to utilize a no-cost testing environment in an EdgeMicro data center that is dedicated to these pilot projects.

“All companies seeking success at the edge recognize near-zero latency can only be achieved by placing their IT hardware and services closer to the consumer. These companies include content providers, wireless companies, IoT and other service providers. While each company delivers unique value, they all have a common challenge: the hurdles of cost and speed to market are too high,” said Mike Hagan, CEO of EdgeMicro.

The testing environment is in a secure, fully-redundant 48 kW EdgeMicro data center in Denver and is supported by key partners including Megaport, Fiber Mountain and BitBox. Customer testing will commence November 1, 2018 with a target end date of February 2019.

“This program gives companies much more than a testing environment,” said Josh Snowhorn, chief strategy officer. “Selected participants will be among the first to leverage EdgeMicro’s ETX (Edge Traffic Exchange), a distributed micro internet exchange program designed for mobile interconnectivity.”

 

The Tower is the New Data Center

By Cole Crawford

Today’s Internet is broken; and the tower can help fix it.

For the last decade, the Internet has been dominated by very large, centralized, data centers built in remote locations by the likes of Amazon, Microsoft, Google, and Facebook.

It is hard to understate how much of today’s Internet depends on, these and other companies, servers housed in the world’s largest data centers – often in the middle of nowhere. Nearly everything we do on our connected devices – from checking the weather on an Alexa device to sending emails from an iPhone – requires dozens of round trips to these massive warehouse-scale cloud computers in remote locations.

However, there is an emerging class of applications – including IoT, mixed reality, immersive mobile gaming, and autonomous driving – where the existing model of remote, centralized data centers, with best-effort Internet routing, becomes insufficient. These applications need optimized connections to the last mile network and require compute and storage to be located more closely to the device or application. The round trip back to a centralized data center takes too long, and the amount of data, that needs to be transferred, is too large.

We call this edge computing, and its reinventing the Internet and the role of the tower in the process.

Re-architecting the Last Mile

When we talk about the “last mile,” we usually mean that stretch of cable or spectrum that connects the service provider to the device.

In order to survive, the Internet must redistribute the power of its current data center footprint. Centralized data centers will not disappear, but they will be augmented by thousands of micro data centers, with increasingly larger amounts of computing power and storage, at the edge of the last mile.

Edge computing will transform the Internet into a geographically dispersed fabric of computing. Tens of thousands of micro data centers will extend the cloud to the edge of the network, making it possible to run applications and services, in the precise locations in the network where they are needed. Whether it is to support virtualization of the wireless carrier networks, cloud-assisted autonomous vehicles, or mobile mixed reality applications, all will benefit.

Leveraging the Tower Footprint

Building out our next generation Internet demands facilities that are, by definition, near the edge of the last mile network. In large metropolitan areas, where the demand for edge computing will be most dramatic, real estate is expensive and not readily available. The prospect of assembling thousands of small urban parcels, securing building permits, power supplies, enclosures, HVAC, and fiber connectivity would be a daunting, if not near impossible, task.

With hundreds of thousands of cell towers, the wireless industry has an extensive portfolio of locations. These are ideal for micro data centers at the edge of the last mile network. These locations are numerous ,and distributed for geographic coverage, allowing for a wide range of networking topologies and data center placement options.

Tower sites are largely built out and, often, have existing structures with sufficient floor space, security, power, HVAC, fiber connectivity. These, and other preconditions, readily support the addition of micro data centers at a marginal incremental cost. Moreover, because they are literally part of the cellular infrastructure, peering into wireless networks in an edge meet-me room, makes it possible to deploy IT infrastructure that is one hop from the RAN.

The wireless industry is also on an aggressive path to virtualize, densify, and upgrade their networks to 5G. As the wireless industry makes these investments, they will look for ways to reduce their costs or leverage their returns. This creates opportunities for all stakeholders to collaborate around joint interests.

For example, wireless carriers are virtualizing their radio access networks and looking to concentrate the software functionality on “white box” servers running in edge locations. This creates an opportunity for shared cost-reduction, by co-locating infrastructure with other server operators. These include public cloud providers and presents the opportunity for more complex relationships, such as the opportunity for cloud providers, and carriers, to collaborate on building the platform on which to evolve the network.

Finally, the tower industry has a unique set of financial tools to help accelerate this transformation. Most relevantly, a substantial portion of the real estate, tower and equipment-housing capital expense are held in REITs like Crown Castle, which are tax advantaged and have access to vast amounts of low cost capital.

Summary

As wireless carriers and infrastructure providers gear up to offer new services to subscribers (and building out their newer 5G networks), we will see an insatiable demand for edge computing.

Since the turn of the century, we have been building large, centralized data centers in far off locations. To date, this has served us well; land and power was cheap, and the round trip from edge to core was “fast enough.”

However, this is no longer sufficient. We must augment today’s centralized data centers with micro data centers at the edge, and the tower will often be the location of choice.


Cole Crawford is founder & CEO of Vapor IO, the leading provider of colocation, interconnection and SD-WAN at the edge of the wireless network. For more information, visit vapor.io. This article will be featured in the Fall issue of Applied Wireless Technology.

Cell Towers Will Serve as Micro Data Centers in Low-latency Networks

By Don Bishop, Exec. Editor, Assoc. Publisher, AGL Magazine

EdgeMicro is rolling out dozens of micro data centers at cell tower sites across the country to move the most popular cloud content and services closer to end users. The use of micro data centers at cell tower sites greatly reduces latency and makes cell towers the front lines of how web-based services get delivered to consumers.

EdgeMicro’s CEO, Mike Hagan, spoke with AGL eDigest at the Connectivity Expo, Connect (x), conducted by the Wireless Infrastructure Association in Charlotte, North Carolina.

AGL eDigest: How does the placement of micro data centers at cell tower sites reduce latency?
Hagan:I’m so glad you’re asking about latency because it is a huge pain point for everyone involved in delivering mobile access to the internet, including both wireless companies and content providers. In fact, the problem is getting worse as consumers and corporate end users look to do more and more via mobile devices that utilize LTE networks.

To answer your question, let’s look at how content and computing services are delivered to wireless devices via cell networks today. Let’s say you are a consumer who wants to watch the latest episode of Game of Thrones on your phone. Today, that episode lives in a centralized data center that may be hundreds or thousands of miles away from the end user. That would be fine if data truly travelled at the speed of light, but the reality is that delivering the packets to a viewer from that central data center involves sending traffic over a variety of networks, across multiple hops, through a long line of switches and gateways (many of which are outdated and which therefore create bottlenecks) until the content finally arrives at the cell tower.

EdgeMicro’s model is to place our micro data centers close to end users, including at the base of the cell towers at thousands of sites in the U.S. and across the globe. That eliminates the latency-killing hops and stops over hundreds or thousands of miles. We are also deeply engaged with the biggest-name highest-traffic content providers in the industry, which will populate our data centers with the content users demand most. And all of that traffic is coordinated by our patent-pending Tower Traffic Exchange IP translation solution in each data center. That allows direct connection from the cell radio to the content, which enables traffic to get to end users even faster. The end result is that we effectively deliver zero latency at the mobile edge, which allows all mobile connected devices to become faster.

AGL eDigest: What is driving the need for data centers at the edge?
Hagan: It’s a perfect storm of several factors that are compounding one another. One of the biggest is that consumers are using their smartphones as a primary device for entertainment, whether that is streamed video, streamed music, live sports and more. That is an enormous amount of data that is being delivered from centralized data centers out to consumers via LTE networks. The trend is only going to grow as the MNOs deploy 5G over the coming years. The consumer’s expectations for speed is elevating. They will demand instant, no-latency access to that regardless of what device they utilize and what kind of network it is delivered by. The demands and pressures MNOs face to deliver are significant, and EdgeMicro helps wireless companies serve their customers more efficiently in a very cost-effective way.

One important thing to keep in mind is that the need to deliver at the edge is urgent now. It’s not just a problem down the road. It’s already something MNOs and content providers need to be solving. And it will intensify over time, accelerating with the massive implementation of the internet of things and upgrades to 5G. These trends are not going away. They are only going to grow as consumers and companies rely more and more on wirelessly-connected devices. It’s important to point out that the use of large centralized data centers will not go away. But they cannot serve the consumer at the edge efficiently. Moving content closer to the consumer is the core solution for improving latency.

However, it is not all about latency. We must look at backhaul too, because that is the other problem that MNOs and content providers are struggling with. The current backhaul infrastructure was not designed to support this kind of volume, which is why moving content to the edge is so important. That is the best way to lower or delay the cost of new fiber installations, and it is the only way to reduce the skyrocketing backhaul costs for companies delivering content and services to end users. If we get this right, MNOs may not have to spend as many billions of dollars as planned on backhaul infrastructure. Moving content to the base or the tower and connecting to consumers locally is a wining solution for everyone.

AGL eDigest: Who are the customers for the micro data center equipment? Wireless carriers, tower owners, others or a combination?
Hagan: EdgeMicro’s approach is unique because we are not simply selling a box to companies that want to go to the edge. To us, we think the idea of MNOs and content providers building out their own edge micro data centers doesn’t make technical sense or business sense. We believe the capital expense is too high and the speed of deployment will be too slow.

We also believe that going individual companies going it alone creates risks they just don’t have to take. And then there are the practical issues like: How do you fit multiple edge data centers at a cell tower if every wireless provider and content provider has their own facility? Do individual companies want to manage the contrition process and operations and maintenance requirements of thousands of remote data centers? Every company has the same goal in mind: to get to the edge as quickly as possible. I believe working collectively and solving problems together is the key for everyone’s success and, most importantly, making the consumers experience at the mobile edge the best it can be.

EdgeMicro’s model is to bring all of the major players together in a common environment that doesn’t require them to build on their own. We keep it simple by using a highly proven model called colocation. We provide space, power, cooling and use the same interconnection model was that was the foundation of land based broadband boom of the last two decades. Our vision for the industry is to get everyone to the edge in fast, cost effectively while minimizing risk.

AGL eDigest: How are target cell sites selected for the micro data centers?
Hagan: I will avoid too much detail here for competitive reasons. The simple answer is we are working with all key stake holders in the edge ecosystem to deliver our colocation and interconnection services to the most target rich environments in the United States. We have the best team to get this right. This is where we believe experience and relationships really count.

AGL eDigest: How well do micro data centers fit at cell sites?
Hagan: Space is definitely an issue at many cell tower sites, so we have factored that into our design so that it works not only at cell towers but in many other locations where space is at a premium. Our micro data centers are 20 feet long by 8.5 feet wide and are designed in a way that works with anywhere: towers, fiber junction points, outside central offices, parking lots and rooftops. We can deploy anywhere.
The biggest issue that must be addressed is available power. We spent a significant amount of time studying the space to get this right. The industry average per rack in the colocation space is 7 kilowatts, so that is what we designed to. Edge sites are not only size-restricted but can be power-restricted, so getting the power aspect is important. Upgrading utility to serve a microdata center is costly and time consuming, so our approach works with what is typically available at a tower site, allowing us to move as quickly as our customers want us to.

AGL eDigest: With all the talk about small cells, how much does the use of cell tower sites for micro data centers increase the importance of cell tower sites and perhaps increase the likelihood that their usefulness will be extended?
Hagan: We are very excited to see the investment in DAS, small cells, 5G additional towers and the fiber network to make the edge work and the consumer happy. We are in a great space, and a new and powerful ecosystem is being created by all of the companies making these investments to better serve customers. When it comes to mobile access to the internet, the key principle that all the parties need to rally around is simple: moving content as close as possible to end users is the best way to reduce latency and serve their needs. It’s all about reducing the distance that data has to travel in order to get from where it lives to where end users want to consume it. Everyone working at the edge has a stake in accomplishing that, and it benefits everyone in the ecosystem because all of that real estate becomes more sticky and more valuable to the owner in the process.


Don Bishop
Executive Editor and Associate Publisher
Don Bishop joined AGL Media Group in 2004. He helped to launch and was the founding editor of AGL Magazine, the AGL Bulletinemail newsletter (now AGL eDigest) and DAS and Small Cells magazine (now AGL Small Cell Magazine). He served as host for AGL Conferences from 2010 to 2012, appearing at 12 conferences. Bishop writes and otherwise obtains editorial content published in AGL Magazine, AGL eDigest and the AGL Media Group website. Bishop also photographs and films conferences and conventions. Many of his photographs have appeared on the cover, in articles and in the “AGL Tower of the Month” center spread photo feature in AGL Magazine. During his time with Wiesner Publishing, Primedia Business Information and AGL Media Group, he helped to launch several magazines and edited or managed editorial departments for a dozen magazines and their associated websites, newsletters and live event coverage. He is a former property manager, radio station owner and CEO of a broadcast engineering consulting firm. He was elected a Fellow of the Radio Club of America in 1988, received its Presidents Award in 1993, and served on its board of directors for nine years. Don Bishop may be contacted at: [email protected]

Connect (x) 2018 – Technology at the Edge

By Ernest Worthman, AWT Exec. Editor, IEEE Sr. Member

This year’s Connectivity Expo, Connect (x), broadened the traditional infrastructure with a wider reach. One of those areas was a number of discussions around Edge networks. Below are some of what was presented in these cutting-edge sessions.

LTE on The Edge
One of the more interesting issues with upcoming 5G is the role 4G, specifically LTE, will play in it. Of late, there has been an uptick in the noise of this and there were some very interesting discussions about it here at Connect (x).

Many say that 4G will be the initial enabler of 5G, with its expanded capability to handle the demands of next-generation networks. It appears, at least from the data presented here, that this definitely has legs. While it is some fog as to all the possibilities of how this will shake out, there is one area where interest is keen – the CBRS band.

One of the promising enablers of licensed LTE, which is attractive to a lot of players, from carriers to web engines such as Amazon and Google, is the CBRS band. There is a lot of excitement here. For one, the barriers to entry are, relatively, low. Another enabling factor is the push for spectrum sharing.
This is extremely important since the CBRS band is being eyed by many factions for many use cases. And, there is broad support, from a technological perspective, from vendors to support the interested stakeholders.

In fact, one of the speakers in the session I attended around this, believes that CBRS deployments, at least in the early stages, will occur by the end of this year. To quote Mike Hart, CTO, Vivint Internet “CBRS is inevitable, now. It is not if, but when, and the when is down to months. This isn’t the end of CBRS, it is only the beginning.”

There also seems to be agreement about that from the FCC. In fact, Commissioner Michael O’Reilly, in his keynote earlier today was confident that progress in the CBRS band, as far as regulations go, will go quickly. As well, there is confidence that other spectrum regulatory agencies, around the world, are on board with making the 3.5 GHz band a common global platform.

It is, actually, quite amazing how fast progress in the band has occurred. This band has the potential to become a model scenario for deployments of next-generation technologies and platforms, while working with incumbents, as well as, employing edge-of-the-envelope spectrum-sharing metrics.
In closing, Hart noted that, “the CBRS band offers a great opportunity for disrupting the traditional models. Over time, we are very bullish about the neutral-host model as well, for a host of platforms, such as in-building.”

There is lot of investment beginning to go into CBRS. It will, definitely, be one of the more exciting spaces to watch.

Collocation on The Edge
Edge computing is definitely a hot topic here at Connect X. One message that came from the sessions is that The Edge is presently a confusing space. Some say it is here, others say not yet.  Connect (x) sessions had extensive discussions on connectivity on The Edge and brought some interesting opinions and perspectives on what is going on at The Edge.

The most interesting session was the discussion on the next wave in digital collocation, and it talked about the issues around the convergence at The Edge of the network. Panelists included, Zachary Smith CEO, Packet, Cole Crawford, CEO, Vapor.IO,  Josh Wolff Senior Vice President Lumos Networks and Josh Snowhorn, founder and chief strategy officer, EdgeMicro.

It was an interesting discussion because the panelists came from four very distinct and different perspectives, which made sense considering the diversity of the group. They covered just about all the angles. It was a group of very knowledgeable individuals with a pulse on the industry.

One of the more revealing statements came from Crawford when the moderator presented the question about when The Edge will arrive. According to Snowhorn, we are already too late. He wasn’t the only one. There was agreement among the panelists, in general, that The Edge cannot come too soon.
The reasoning for that is because the amount of data being consumed is already nearing capacity in the more congested locations. Without deploying Edge technology, bottlenecks are just waiting to happen. The biggest problem is latency. Typically, networks run around 10 to 20 milliseconds, which is just about the same time the various pathways around the human sensory information highway takes. That metric must be matched by networks otherwise our sensory capabilities are waiting for more input.

That is an interesting angle, yet very on target. Future networks must become transparent and lag times are a primary metric in this. This is of tantamount concern and a primary driver for deploying Edge data center networks as soon as possible.

Another interesting angle is the definition of exactly what constitutes The Edge. This is not a redundant question. The Edge is defined by applications such as autonomous vehicles, drone applications, and medical apps, which are much more sensitive to speed than apps such as video games, enterprise data movement and, in many cases the Internet of Anything/Everything (IoX).

The Edge is really the place where collocated infrastructures exist. The driving metric for these is performance and cost. Therefore, The Edge can really be towers, DAS deployments, small cells and other nodes on the networks, defined by application and data load requirements.

As well, The Edge must be a neutral host to be effective. It doesn’t matter if it manages carrier data, if it is a tower, or a hyperscale company installation. As Crawford puts it, “it is simply a meeting room at The Edge and QoS is the driving metric.

Overall, the message about The Edge, in general, is that we are still in the petri dish stage and we need to grow some results quickly so the industry can begin to get some clarity and begin to deploy it.


Ernest Worthman
Executive Editor/Applied Wireless Technology
His 20-plus years of editorial experience includes being the Editorial Director of Wireless Design and Development and Fiber Optic Technology, the Editor of RF Design, the Technical Editor of Communications Magazine, Cellular Business, Global Communications and a Contributing Technical Editor to Mobile Radio Technology, Satellite Communications, as well as computer-related periodicals such as Windows NT. His technical writing practice client list includes RF Industries, GLOBALFOUNDRIES, Agilent Technologies, Advanced Linear Devices, Ceitec, SA, and others. Before becoming exclusive to publishing, he was a computer consultant and regularly taught courses and seminars in applications software, hardware technology, operating systems, and electronics. Ernest’s client list has included Lucent Technologies, Jones Intercable, Qwest, City and County of Denver, TCI, Sandia National Labs, Goldman Sachs, and other businesses.  His credentials include a BS, Electronic Engineering Technology; A.A.S, Electronic Digital Technology. He has held a Colorado Post-Secondary/Adult teaching credential, member of IBM’s Software Developers Assistance Program and Independent Vendor League, a Microsoft Solutions Provider Partner, and a life member of the IEEE. He has been certified as an IBM Certified OS2 consultant and trainer; WordPerfect Corporation Developer/Consultant and Lotus Development Corporation Developer/Consultant. He was also a first-class FCC technician in the early days of radio. Ernest Worthman may be contacted at: [email protected]

Edge Computing Edges into the Mainstream

By Ernest Worthman, IEEE Senior Member

While movement in edge technologies and platforms has been visible for a couple of years now, much of it has been lofty. There have been some deployments but they have been limited in both scope and functionality. That is primarily due to the fact that there is not much of a common platform for edge computing. The word is that is it good for off-loading data and crunching data that does not need to move in or out of the core, as well as multimedia. Those are pretty broad boundaries and putting that into actionable items has been elusive.

Much of that is because, so far, there is not a compelling case for edge computing. 4G is handling the load relatively effectively and 5G is not yet in existence. That is going to change starting in 2018 but how quickly depends upon how fast other technologies and platforms gain traction.

As peripheral platforms move forward (the IoX and 5G particularly) the case for edge computing becomes more pressing. Edge computing has a lot of application to the IoX. That is because there will be scads of IoX devices and networks that will not require a lot of communication with a core network, (WLANs such as smart “X”, Wi-Fi, small cells, mmWave, and others). And when they do, they use wideband, dedicated backhaul channels to the core (IP, fiber, cable, or microwave).

For example, some experts believe at least 40 percent of IoT-created data will be stored, processed, analyzed and acted upon close to, or at, the edge of the network by the end of 2018.

Edge computing will lead to something called “computing everywhere.” That is where computing is done where it lives – be it the edge, the data center, the cloud, wherever. However, the trend will be to do as much as possible at the edge.

2018 will see the development of edge computing become a priority. The movement will be to increase funding, research and develop new edge computing capabilities as the data tsunami continues to roll. This is necessary because of the proliferation of low-power, wide-area (LPWA) networks that consist largely of resourced-constrained IoX edge devices. The current practice of using on-premises data centers or cloud resources will begin to shift to the edge in 2018, if edge technology moves far enough.

Much of this will be driven by the acceleration of the industrial IoT (IIoT), aka the next evolution of machine-to-machine (M2M).  This is an area where use cases and RoI exist and edge computing is a natural evolution to improve data analysis and reduce costs. As well, IIoT vendors are moving quickly to develop and manufacture devices capable of working at the edge.

Furthermore, edge computing will provide a cornucopia of analytical data in other segments (such as retail, transportation, utilities, remote monitoring, etc.) that will not have to rely on constrained bandwidths with limited data I/O capabilities. Analytics become almost instant with edge models offering real-time decision capabilities.

2018 will also see more attempts to make sense out of the “big data” that has been promised as the solution to everything and anything. Movement will be seen in actually turning these mountains of data into leverageable information. Edge technologies will pick up momentum this year in proving business value. Therefore, we can expect the adoption rate to rise, possibly exponentially, as the demand of IoX devices and applications emerges.

There will be an acceleration in edge products and applications as well as more serious talk about standards. There will be an uptick in alliances, partnerships and M&As as companies begin to position for the ramp up in edge advancements in all vectors.

Dr. Spincer Koh of Gorilla Technology makes a good point when he says, “edge computing is coming and it’s going to be big.” 2018 will give us a better look at how all this will shake out in the wireless, as well as other segments.

Spoilers

There are a few possible disruptions to edge computing. Number one is security. However, the advantage here is the containment of the network. That makes it a bit easier to secure them. On the other hand, there will be so many of them, that keeping all of them secure will be a monumental undertaking.

There are solutions on the drawing board. One is to reposition existing products to address edge network security concerns. Another is to develop targeted security solutions that are specific to a layer in the stack, or a particular vertical. However, edge deployments are still in the early stages; so is the security thinking. Much of the security mentality is of the wait-and-see kind.

There is validity to that. Investing in security before the industry really has traction is difficult to sell, regardless of lessons already learned.

A second dilemma is managing edge devices. From the IoX to smartphones, tablets, laptops, sensor networks, peer-to-peer and ad hoc networks.  Add to that the next generation of roving data centers (smart transportation elements, remotely piloted vehicles) and the management of edge networks and devices become even more complex. AI will be required to wrap one’s arms around all of this. That will start to emerge in 2018 and we will see edge networks melding with AI.

Another issue is the diversity in edge models. It will be challenging to manage distributed applications across a wide girth of players, platforms, interfaces and inter-edge connectivity. Some vendors are working on solutions to help manage this disparity, which will start to emerge in 2018. However, it will be well past 2018 before any clear directions in edge computing emerge.

Overall, there is little doubt edge computing will catch on. There are just a lot of issues that need to be resolved. This year will see progress, but more in the development of solutions for the issues versus actual deployments.


Ernest Worthman
Executive Editor/Applied Wireless Technology
His 20-plus years of editorial experience includes being the Editorial Director of Wireless Design and Development and Fiber Optic Technology, the Editor of RF Design, the Technical Editor of Communications Magazine, Cellular Business, Global Communications and a Contributing Technical Editor to Mobile Radio Technology, Satellite Communications, as well as computer-related periodicals such as Windows NT. His technical writing practice client list includes RF Industries, GLOBALFOUNDRIES, Agilent Technologies, Advanced Linear Devices, Ceitec, SA, and others. Before becoming exclusive to publishing, he was a computer consultant and regularly taught courses and seminars in applications software, hardware technology, operating systems, and electronics. Ernest’s client list has included Lucent Technologies, Jones Intercable, Qwest, City and County of Denver, TCI, Sandia National Labs, Goldman Sachs, and other businesses. His credentials include a BS, Electronic Engineering Technology; A.A.S, Electronic Digital Technology. He has held a Colorado Post-Secondary/Adult teaching credential, member of IBM’s Software Developers Assistance Program and Independent Vendor League, a Microsoft Solutions Provider Partner, and a life member of the IEEE. He has been certified as an IBM Certified OS2 consultant and trainer; WordPerfect Corporation Developer/Consultant and Lotus Development Corporation Developer/Consultant. He was also a first-class FCC technician in the early days of radio. Ernest Worthman may be contacted at: [email protected]