Portability Is Profitability In Radio Software

The Association of Public Safety Communications Officials (APCO) International Project 25 is intended to unify U.S. public safety bodies, whether their role is homeland security or responding to day-to-day emergencies, but progress has been slow. Etherstack’s protocol stack technology could enable legacy analog equipment to work with the new digital APCO standard.

Wireless air interface protocol stack technology created by an Australian firm is receiving development funding from In-Q-Tel, an independent strategic investment group launched by the Central Intelligence Agency.  This funding aims to bring new technologies to the U.S. intelligence community.

Etherstack has been writing portable protocol stacks that are independent of the underlying hardware and operation systems in different radios since the company formed in 1995. The company is the first from outside North America to receive funding from In-Q-Tel, largely because its capabilities cannot be readily found within the United States. In October 2007, Etherstack opened a software-defined radio research and development facility in New York, staffed by U.S. engineers to fulfill multiple contracts with a number of customers in the United States.

The rationale behind the formation of Etherstack grew out of the cellular market in 1995, explains David Deacon, its founder and chief executive officer. At that time, global system for mobile (GSM) protocol stacks were becoming a commodity. “You don’t buy a Samsung, Nokia or Motorola mobile phone based on the quality of its GSM stack,” Deacon elaborates.

Speaking about the public safety and homeland security markets with the emerging standards of Tetra and Association of Public Safety Communications Officials (APCO) International Project 25 (P25), Deacon says, “I believed that manufacturers would, over time, view protocol stacks less as intellectual core property and more as a transistor, battery or another component that they    purchased.” Etherstack’s success depended on this change happening and on its engineering accomplishments in portability to meet the resulting demand.

Deacon uses the terms protocol stack and waveform interchangeably. He explains that both describe all software on the hardware platform from the bottom of Layer 1 (modulation/demodulation) all the way through to Layer 2 (link layer) and Layer 3 (network layer) and potentially any application-specific software required for the use of that protocol.

Companies started to realize that they could procure the wireless protocol stack from an outside source, which would free up as many as 20 engineers over a period of up to two years, Deacon says. Etherstack’s work on the abstraction of the protocol stack or waveform from the underlying hardware was completed in the late 1990s. Among the first customers for its software was Thales (Racal at that time), for work on the Vector covert Tetra radio used by special forces and others and delivered in 2001. “That is exactly the same protocol stack that we have just delivered to the Swedish defense ministry, six years later, with only minor modifications, which were all about porting that Tetra protocol stack to the software communications architecture (SCA),” he adds.

In developing hardware-agnostic, software-defined portable waveforms,  Deacon states that details matter. He cites the different endianness of microprocessors, which is the direction bit orders are electronically sent. “These are the sorts of things that we make as independent as possible in our software so that we don’t need to worry what hardware platform we are running on.” He continues, “What matters is that enough processing power exists for the software.”

Endianness is one of more than 100 different items that Etherstack engineers considered in developing the abstraction in its software and the underlying platform. “We came up with our own hardware abstraction layer (HAL), which we instead called the core services. That is exactly the same as the HAL in the SCA as a concept, although they are quite different in how they work. With it, we can do ports to other platforms, measured in handfuls of weeks compared to many months if not years,” Deacon says.

It is not the SCA that makes the waveforms portable, Deacon argues. “We try not to confuse waveform portability and the SCA, although the SCA is relatively a good model and a useful tool for industry to start to discuss the issues associated with waveform portability. [The SCA] is great as a concept to get people thinking in the right direction about waveform portability, but you can write portable waveforms without the SCA because that is what we’ve been doing for 13 years now.”

Programmers still can write an SCA-compliant waveform that is not portable to someone else’s platform if they haven’t considered other waveform portability issues, Deacon allows. In recent years, Etherstack has made a number of connections in the U.S. software-defined radio markets in both the defense and homeland security arenas, including, Harris and Raytheon on APCO P25 as well work on Tetra as part of a jointly funded program between the U.S. Defense Department Joint Tactical Radio System (JTRS) community and the Swedish defense ministry.

In the U.S. homeland security market, Etherstack teamed with Raytheon JPS to offer the P25 channel controller. “That product is capable of converting any analog FM repeater into a digital APCO 25 repeater, simply by cabling to the back of it.” Currently, 90 percent to 95 percent of all deployed public safety infrastructure in the United States is legacy analog. “With one simple component, the Channel Converter transforms a legacy analog FM base station into digital P25 base station by cabling it to the back. Then, all of a sudden you are talking digital P25 over the air, and all Internet protocol (IP) out the back, which can connect into IP software switching products for connection into larger public safety radio networks.”

A recent trial with the Wisconsin State Highway Patrol demonstrated conversion of five different legacy manufacturers’ repeaters using this device. These were used in a multi-site trunked radio network, with handheld terminals from six different manufacturers including EF Johnson, Kenwood, Icom, M/A-COM and Tait. Deacon says, “Estimates of the total cost of replacing the public safety network infrastructure in the United States during the next 10 years are approximating $45 billion. The Channel Controller device, however, brings a typical network operator’s costs down into the region of a spare parts budget.”

Etherstack is providing Icom waveform or protocol stack technology across a range of products and markets to support digital private mobile radio, or PMR. The company hopes to become a single source for military specific protocols and waveforms for the defense community.  For example, Etherstack’s APCO P25 waveform on the Harris Communications Falcon III radio enables U.S. defense forces to interoperate with civilian emergency services such as the police and the fire brigade in the event of natural disasters such as Hurricane Katrina.

“The technology skill set is identical between the two markets,” Deacon says, pointing out the company’s expansion from homeland security and public safety to the defense market. “In many respects, the public safety or network controlled protocols are often far more complicated than the military link-based protocols. If you are a Tetra or an APCO P25 mobile, you are under the control of a network. You have to signal for permission to use channel or make a public switched telephone network call. Most military protocols are link based. You key up a link between one radio and another. Each country has security requirements regarding access to those waveforms, usually considered on a case-by-case basis,” Deacon explains.

Deacon cites Etherstack’s U.S.-Swedish work on the Tetra waveform as evidence of expertise in complex waveforms. “That Tetra waveform is the first complex waveform that has been ported to the SCA because all the more military waveforms are fundamentally link-oriented waveforms as opposed to network centric … you don’t get a second job if you don’t get the first one right,” emphasizes Deacon, referring to a recent follow-on contract from the Swedish defense ministry.

Limited work by Etherstack on undisclosed military-specific waveforms already has taken place, Deacon elaborates. “We currently have under development both Universal Mobile Telecommunications System (UMTS) and Worldwide Interoperability for Microwave Access (WiMAX) waveforms for lead SDR customer use in both commercial and military applications. The Layer 1 [modulation/ demodulation] for both those protocols is very similar to the Orthogonal Frequency-Division Multiplexing required in the Wideband Networking Waveform,” he adds.