Foundation Forum

Brigadier General Howard J. Mitchell
Director, Space Systems
Office of the Secretary of the Air Force
Streamlined Acquistion Case Study
AFA Symposium
Colorado Springs, Colo.
May 24, 1996

It gives me great pleasure to be here today. Colorado Springs was my home for four years while I was at the Air Force Academy.

I have a second hat that I wear in Washington. I am the Director of Communications, Acquisitions and Operations for the National Reconnaissance Office. If this meeting was a year ago, you probably wouldn t have an NRO presentation and probably wouldn t have seen the NRO trying to build consensus on anything. We are trying to build consensuses and do things with the rest of the space community. We are very dedicated to making sure we are accomplishing our missions within this new framework of working with General Dickman and the Space Architect office and Mr. Bob Davis in the DUSD(Space) organization as well as cementing our ties with the users, U.S. Space Command as well as Air Force Space Command. I am going to give you today a little bit of a case study on the Global Broadcast Service system.

Let me tell you what the need was, how we went from that need to the Dissemination Studies, to building an Integrated Product Team and then to where we stand today on this program. We started out with a set of studies in November 1994, and 17 months later, we have gone through the process of demonstrations, have a Mission Needs Statement approved by the JROC [Joint Requirements Oversight Council], and have had an expanded Defense Resources Board actually put money into the budget for the Air Force s 1998 POM, which is pretty good work. What was the need? For communications, we have the need for High Bandwidth Terrestrial Communications to Fixed Sites. We are in pretty good shape there with fiber optics and commercial leased lines. We don t have a real big problem in that area.

On the other end of the spectrum, for the Assured and Two-way Satellite Communications for the mobile users, we have a robust system to do that at a low-data rate. We have the anti-jam capabilities and all those things you are familiar with.

In the middle, we have a need for High Bandwidth Satellite Broadcast Communications to service mobile needs as well as to get theater information from the CONUS out to the theater. That is what we studied to see how we can best solve that high bandwidth requirement? There were several studies that had been done. Most were centric within the intelligence community. We recognized we had a lot of requirements because we push a lot of big data files and a lot of bytes. We had our Plans and Analysis organization in the National Reconnaissance Office look at a direct downlink concept which fundamentally said we could just go from our satellite assets on orbit straight into the theater and let them process the information in theater versus a dissemination architect.

Then we looked at how we did our Integrating S-Band Tactical Network and Tactical and Related Applications, day-to-day dissemination and how to upgrade those. We had a study look at future technology options. Our imagery folks were looking at Imagery Intelligence Dissemination. We ad an Intelligence Systems Secretariat at the community level looking at common dissemination system formats. We had a lot of activities going on that were disjointed so we brought those together into one study that we called the Data Dissemination Architectures Study. It was run from the Directorate of Communications organization by Colonel Kirk Washington who works for me. We brought together our Plans and Analysis folks, folks from the Imagery Intelligence Directorate, Communications Directorate, the Defense Space Project Office, Operational Support Office, our Signals Intelligent Directorate, the Technology Office, and our Office of Systems Applications. We then brought in the Central Imagery Office, Defense Intelligence Agency and the Defense Dissemination Program Office in the intelligence community to form the team that looked at this integrated architecture.

At this point, you can see we still had not brought in the services and everything else. This is business as usual for the NRO, or the way it has done business in the past. I am going to show you how we transitioned from this into the community.

We came out with a sizing of this Broadcast Service for the intelligence products and the system solutions to try to meet those requirements. The Tactical and Related Applications is a broadcast that currently shares channels in some instances and is a 2.4 kilobit-per-second broadcast. It is how we get the "producers" information out from the signals intelligence [SIGINT] side of thing as well as ELINT. We need to be worldwide above 60 degree South latitude.

The Primary Imagery for the Major Regional Conflicts [MRC], is 140 megabytes maximum data rates, clearly not something that we can handle over existing communication systems. We look at what earth coverage should be and the sizing was based on seven spots to receive elements for the DDS, Data Dissemination System. In any footprint of a satellite, you could expect to serve seven of those sites with a 24 megabit pipe. In addition to that, you have the Navy who is going to be in transit and have forces in transit, and you must have some kind of capability to take care of them no matter where they are. Instead of a spot beam onto them, you may put a broad area beam at a lower data rate to get the information out to them. Measurement and Signature Intelligence was included in the imagery needs. We also looked at using the current existing two-way Comm for the Tasking/User pull systems.

This objective system was the system from which we could frame the debate. It could be changed or modified, but at least we ve looked at a lot of the issues, we ve looked at a lot of the questions, we have a lot of the answers, and we have a lot of the data. We had a target architecture that could then be discussed and modified.

SLIDE-Mil GBS Satellite Coverage

It was a three-ball system with seven spot beams of greater than 20 megabits each and one broad area coverage beam of six megabits. That related to what came out of the study. We are looking for seven moveable spot beams to service any of the customers at a hot spot and then the broad area takes care of anybody at sea and in transit.

These are the kinds of antennas in the field today. You are all familiar with the 18 to 24-inch dishes we are looking at for the GBS terminal. In one of our demonstrations, we used a one-meter dish at Roving Sands. We were trying to go from big dishes to smaller dishes to really small dishes.

We weren't just looking at intelligence data. As we started to brief this study around and look at how we were going to go about doing this, rather than taking the traditional approach closed our doors, gone off in the back room, put money in the NRO budget and done this we took the opposite approach and said we are going to go out into the community and see what other needs could this system serve. There are clearly other very large bandwidth users other than just intelligence products. There are logistics information, weather information, imagery, and telemedicine things that drive information not just down to the Joint Task Force, but down to the land, air and naval components.

With the recognition this entirely changes some of the doctrine that the Joint Chiefs of Staff has in place, it makes them think differently. All of a sudden there isn t a limitation on how much information you can get to the Air Component guy and he may be able to get information that the Joint Task Force doesn t think he really needs to have, but yet he wants to have and he ll have access to it. There are doctrinal issues that go into the Concept of Operations development for these things.

Why do we want to do GBS? For each requirement, GBS provides substantial capability. We looked at the Air Tasking Order, Tomahawk Mission Data Updates, Imagery and the TFIDL [Tactical Forces Integrated Deployment Logistics] for the logistics guys. The ATO tends to be about a megabit in size, based on Desert Storm experience. Using a PC modem at 2.4 kbps requires 1.02 hours while the GBS operating at 23 mbps takes only .38 seconds to get this information out. For the TFIDL, you go from 9.6 days to 1.45 minutes. There is a substantial ability to move data forward to where it needs to be in the time frame that it needs to be there in order to support the warfighters.

How do we build a consensus? We started in the traditional NRO way as this study actually started before the NRO became unclassified in April 1995. From that internal team, we went out to ASD C3I, DISA [Defense Information Services Agency], DARPA [Defense Advanced Research Projects Agency] and other organizations, the services, and the Joint Staff with this concept saying, "We think there is great value here for the military and let s try to jointly put together a program and take a look at what could happen." That spawned an ASD C3I-led Integrated Product Team and a General Officer Steering Group. This was subsequently changed with the stand up of DUSD(Space) who took over the leadership for that IPT. We brought in the Space Architect as they were being formed up after this initial IPT started. We added members as time went on to keep the team together and keep the information flow going.

J-6 [Directorate for Command, Control, Communications and Computers] was the big driver in this. VAdm. Arthur Cebrowski and his troops were really pushing this "bandwidth on demand" concept and this idea fit perfect with where they were trying to take the JCS. We had a natural match that helped from the user side and opened a lot of doors for us.

There are many decision-makers in the process. We briefed the Joint Requirements Oversight Council in August of 1995 for the Mission Needs Statement. Probably for the first time in history, an NRO guy stood up and briefed the Mission Needs Statement for a system that had a military application as well as intelligence application. I don t think it has ever been done before.

It was the right way to go. We came up with an idea that the military could use and would go through the JROC and ORD [Operational Requirements Document] process. Also at that same time the GBS requirements flowed into the CRD [Capstone Requirements Document] which was exactly what needed to happen. We now had a system that was going to move a lot of intelligence data, but it also showed up in the overarching requirements document that fed into the Space Architect s regime. The JORD [Joint Operational Requirements Document] for the GBS is being done by the Army and the CONOPS is being done by the Unified Space Command. We built this team and got the concept into the decision makers and went with a full-court press in briefing them. We purposely kept this entire effort at the unclassified or Secret-collateral level.

There was no SCI [Special Compartmental Information] level information contained in any of the briefings. We did that on purpose to open up the doors and the lines of communication. Some of the data that flows over this may be SCI level, but the program itself did not require SCI access. That was a big stride foward for the NRO. Some of my folks had a really tough time doing that, but it was the right thing to do.

How do we build that consensus? We briefed, and we briefed, and we briefed at all levels of the government up to and including the Assistant to the Speaker of the House of Representatives. It included all of the committees, directors of operations and intelligence and the whole gamut of everybody you could imagine from the intelligence side as well as on the military side.

It is one thing to build consensus by vu-graph, it is another to get the user involved early. There are a couple reasons to do demos. One of those reasons is to build user support for the program. That is exactly what we did here. We had a step-wise process to demonstrate incremental capabilities and get the warfighter and the users on board by actually seeing it demonstrated. In Radiant Storm, we demonstrated military feasibility with a straight, commercial off-the-shelf transponder, leased a DBS transponder that was on orbit and showed we could do T-1 data rates. As you know, DBS, or direct broadcast, is geared for video. They are expanding that now. They are in the third generation box, but, at that point, it was still unique to put data over it.

We then built on Radiant Storm with Special Project 95 where we brought in both data and video on multiple-receive locations and passed primary imagery for the first time. With the very minimal capability we had for that initial broadcast capability, we could do our own broadcast versus relying on somebody else to do it for us. The Air Force C4 Agency, which is at St. Louis, then took the next step by showing a full, 23-megabit-per-second service, asynchronous transfer mode [ATM] over the system and multi-media service. JWID 95 [Joint Warrior Interoperability Demonstration] last summer was the Broadcast Management Information Services. We programmed ATM encryption, showed we were interoperable with the DISN [Defense Information Service Network] and the Defense Information Structure. We took a bite at a time as we marched up this chain. This summer, we are associated with JWID - 96.

What are we doing today? Two things that are up and running. We are using commercial satellite transponders in the Ku-Band, which is not a military frequency. Eventually, we ll probably go to Ka-Band.

The Global Broadcast Service testbed was funded by the NRO as a concept development tool. It supported all of those exercises and demonstrations. We made it available for U.S. Space Command to task in support of exercises. They tell us where they want receivers, and what they want on the broadcast, where they want it and when they want it to support exercises. They are the clearing house. That has worked very well. We ve provided them the technical means, and they ve provided the user input.

We are supporting military exercises in the CONUS and Hawaii. We must drop Hawaii shortly. I m unhappy about that, however, the people in Hawaii would rather have the Olympics on the commercial transponders than have us do GBS. We will reinstitute it after the Olympics.

When you are doing commercial stuff, you ve got to consider the other buyers competing for that service.

The second thing we are doing is the Joint Broadcast Service. The Bosnia Command and Control Augmentation, BC2A, is done by a joint DARPA/DISA program office in Northern Virginia. It is a very large scale information effort for the warfighter in support of EUCOM in Bosnia. JBS makes it all work. It takes the information from the servers here in the CONUS and broadcasts it out. That was the original concept. We got into the program this fall when they asked us to do Predator data. We are today taking Predator data down to a ground site, up through a VSAT, which is a very small aperture terminal, down to a T-3 line back to the CONUS into our uplink site, which is in the Naval Research Laboratory, [Washington, DC] and broadcast it back out across the entire theater. There is almost no perceptible loss in the resolution of the imagery from the Predator even though we are making all those hops and broadcasting it back out throughout th theater.

It takes the Predator data, and almost in real time, broadcasts it across the whole theater rather than just back down to a single site or the analysis center. In fact, we are standing up a total of 29 sites in support of EUCOM, including on Navy ships that are supporting the effort.

We are learning big things, and the JCS is learning big things, and the users are learning big things as this ACTD is stood up and provides information. We are providing both video, intelligence data and a common operating picture. Folks are getting more comfortable with the system.

All of this is prototype stuff the GBS testbed and the associated boxes were about a $2-3 million effort. The current effort is a much larger mainly because of the receive site suites. We are now deploying suites that cost in the $80,000 to $100,000 range. That is not because of the boxes, but because of the cryptological equipment that we have to put with it. We are also putting out SPARC-20s [computer] and other kinds of tools that the users didn t have for this demo. We are trying to drive down that price to about $2,000 by having the encryption built into the box. NSA is working that part of the program.

Now, the Joint Space Management Board gave the Acquisition Executive Agency lead to the Air Force. It was like an auction. Everybody came in and told how they would do this program. The NRO, DISA and the services participated. The JSMB took that information and Dr. Kaminski made a recommendation that the Air Force be selected as the lead agent for the program.

As part of that effort, the current budget contains about $386 million for this program. There was a reprogramming package in FY96 that brought the total up to $459 million. That reprogramming was geared to support a commercial transponder lease and putting four Ka-Band transponders on UFO [Ultra High Frequency Follow-on] Satellites 8, 9 and 10 a Navy initiative. That will actually give a capability in 1998 that will be owned by the military and will be in the Ka-Band, whereas these are currently the Ku-Band.

The first two phases of the program are at the entry level. The third phase is awaiting the MILSATCOM Architecture Development Team s effort, and it is integrated through the CRD and ORD processes. We think this is a perfect initiative and a demonstration of how we can quickly field capabilities and quickly get out prototypes to answer a lot of questions for the users and participate from intelligence community standpoint in both supporting requirements and coming up with the ideas. We are going to transition the GBS and JBS testbed operations over to the Air Force this fall because they are the executive agents. Then the NRO will be out of the business except from a user perspective.

What are the keys to this success? We got the Joint Staff and the user involved early in this process. The J-6 helped drive things and the services came on board with the right folks.

Senior DoD and intelligence community spin-ups through our briefings really got the message across that there were new capabilities in the commercial world and that we ought leverage off of these systems.

We gained credibility because we ve done the analysis and initiative comparisons up front in the concept phase, rather than just saying, "We ve got this great satellite thing." We said, "You ve got to worry about the information side on one end and about the distribution side on the other end."

This also allowed us to separate contractor marketing from engineering facts. Our team comprised of the services and the intelligence community was able to separate what was true from pure vu-graph pitces, and that helped keep everybody on track.

We also built "end-to-end" concepts to frame the debate. Again, we initially came up with an option for three on-orbit satellites, 3,000 boxes, three uplink sites, and $1.2 billion to deliver it by 1998. We knew that wasn t going to get funded in the POM, but it allowed us to frame the debate and allow trades to be made. It was a very good approach to have a solution that everybody could modulate against.

We had the resource flexibility to pursue good ideas. The NRO has that capability. We are able to fund these kinds of demonstrations and prototypes from approved sources. I did not have this flexibility when I was the director of developmental planning at SMC, Space and Missile Systems Center. We were always trying to fight for money to do these type of prototype things. I fully support having that money in the budget.

We should leverage off the commercial investments to do early prototyping. Most of what we did was based on commercial off-the-shelf products. The box that we wound up selecting for Bosnia was right out of the catalogue. We did not do any RDT&E. It was one of three options by the three contractors. The one we picked matched the schedule

Also, obtain bright people and guide them but don t micro manage them. Let them go explore stuff.

Finally, anticipate, anticipate, and anticipate. We sat down and had a planning session on marketing this thing and building a consensus. We stuck pretty well to it. We identified the folks we needed to brief and whose calendars to get on. We set the demos and the information we were going to pull out of those demos and pass to folks.

This is a success story when we went from a study to a funded program in 17 months. We ve got good consensus and it has been incorporated into the normal mainstream activities of the Department of Defense. It is being looked at by the Space Architect and in the trade space for the MILSATCOM architectures. We have been participating with the Space Architect in doing both that ADT as well as the other two ADTs that he has ongoing.

That concludes my presentation, and I ll be happy to take any questions.

MS. LANTZY: In the case of the Space and Missile Tracking System, SMTS, how do we build consensus on the mission of SMTS through IPTs when there is an MOU [Memorandum of Understanding] between the Air Force and BMDO [Ballistic Missile Defense Organization] that seems to give BMDO a degree of veto power over AF Acquisition decisions or plans?

GEN. MITCHELL: I struggled with that one when this program was called Brilliant Eyes and I was the Ballistic Missile Defense program director. Any time you have that kind of situation where the service, the intelligence community and another department have a stake in this, it is very difficult to make those things work as smoothly as you can if they are internal to the service. I think there is a big challenge there, and BMDO has been the big pusher for the requirement for SMTS as they were for its predecessor, Brilliant Eyes. It is going to be a continuing dialogue that has to take place at senior levels. I think you can build the Integrated Product Teams. You just have to make sure that everybody understands what everybody s equity is and what their concerns are and then address them. Building consensus is difficult to do sometimes, but you have to make sure that you do it. You must have willing partners. It is going to be difficult, particularly because BMDO programs tend to be very political. BMDO often ends up doing things they didn t want to do but are at the direction of Congress.

MS. LANTZY: Would you please explain "Predator data?"

GEN. MITCHELL: Predator is the UAV [Unmanned Aerial Vehicle] that is flying in Bosnia. It has sensors on board, both visual and IR and as it flies, it collects the data. Actually, it collects the data, sends it up via satellite link back to the command and control station on the ground, which is then VSATed back over another satellite link to a ground station which is then sent back over to us in the CONUS for us to broadcast back over. It is not a national asset, but is a theater asset.

MS. LANTZY: What impact will the protest of the Phase 2 Award to Hughes have on the overall program?

GEN. MITCHELL: I d have to defer that question. I am not the GPS or UFO guy. The UFO is a Navy procurement. I don t know where that stands now. As the decision was made for the Air Force to be the acquisition lead the GBS program, we have backed away from those things and not gotten into the acquisition strategy for either UFO, which the Navy is doing, or the Follow-on, which will come out of the architecture General Dickman is doing. We are supporting, but the Joint Program Office, as part of the MILSATCOM program, officed out of Los Angeles, has those responsibilities. So I wouldn t want to speak for the Navy since they are the acquisition agent for that.

MS. LANTZY: Where will the GBS program office be located?

GEN. MITCHELL:It is within the Space and Missile Center organization in Los Angeles. They will probably have a presence in the Washington area over the next couple years as they are trying to get through the budgeting and early parts of the program. But the program was given to the MILSATCOM program office in Los Angeles.

MS. LANTZY: What does the move of the JBS uplink site from NRO mean?

GEN. MITCHELL: We hope it doesn t mean anything. We hope it will be transparent to the user, which is our goal. All we are changing over is the responsibility for the operation and maintenance of that facility from the National Reconnaissance Office to the Joint Program Office or to DISA, whomever is their agent for it. We are in the process right now of working through what needs to transfer over to them, how they are going to do it and to make sure we have uninterrupted service because the Predator data is critical to the effort over there. We will continue operating at the NRL site until we are sure there is another site up and running and have successfully transitioned before we break down the site at NRL. Part of the problem with that is there is no prime contractor for the site at NRL. We built this uplink site with a bunch of smart engineers who can make it work. Fortunately, we are collocated. The GBS testbed and the JBS testbed, our uplink site, are two strings of equipment in the same building. If one of them goes down, we can put the other stuff on line and just bring the testbed over and swap it to the operational site. We have a nice back up there, and we need to make sure that is maintained in this transition as well. I don't want to give people the impression that the NRO is going to walk away from this. We transitioning things in accordance with the decisions that were made by the JSMB that the Air Force is the executive agent for this program. We are going to take the things we ve been doing, transfer the lessons learned, transfer the hardware and then support them in any way we can after that.

MS. LANTZY: Thank you, General Mitchell.

Return to the Colorado Springs '96 Foundation Forum