Transitioning from Agency Owned to Commercial Wireless Data Networks Kitsap County Sheriff's Office Total Run Time: 27:41:00 [intro music] Dan Hawkins: The following is another in a series of recorded audio interviews on lessons learned and best practices from projects funded through COPS technology grants. These podcasts are presented by SEARCH, the National Consortium for Justice Information and Statistics through funding from the U. S. Department of Justice Office of Community Oriented Policing Services, COPS, under Cooperative Agreement 2007-CK-WX-K002. Today's topic is Transitioning from Agency-Owned to Commercial Wireless Data Networks. Our guest is Lieutenant John Sprague, with the Kitsap County Sheriff's Office in Washington. I'm Dan Hawkins, the Public Safety Technology Specialist for SEARCH and Dan of this podcast. Kitsap County is a key participant in a project funded by a COPS 2007 technology grant to connect agencies on the western side of the Puget Sound to shared data systems. Bremerton Police Department received the grant and partnership of Kitsap County, other municipal jurisdictions, and two Indian nations. At the center of the project has been replacement of an agency-owned wireless data network with commercial services. John, welcome. Can you tell us a bit about the project with Bremerton and share your background with it? John Sprague: Sure. I've been with the county here in the sheriff's office for about 23 years. And since about 1996 I've been involved in the planning for the mobile data system, and it's gone through a couple of variations. But we found that our old UHF 9600 Baud network just wasn't getting us where we wanted to be as far as providing applications out to the field. And so that was the crux of our project was to bring in a new network and bring new applications out to the field so that the officers could have better tools rather than having to come back to the office or rely on support staff to do a number of their tasks. Dan: Let's talk a little bit about your previous mobile data network. It was similar to others around the country based on land, mobile, or radio technologies. What was it put in originally for? John: The original thrust was to reduce the congestion on our voice channel. We used to be isolated to just two voice dispatch channels for all the agencies in the county. So that would be four police departments, the two travel police agencies, and the sheriff's office. In clear back in mid-nineties they commissioned a report, and basically what the report came back as saying was that we would have to add voice channels and dispatch staff. But one of the ways to mitigate that was to bring in mobile data so the officers could be doing more of the work in the field and to get them to the point where they could actually see the CAD [Computer Aided Dispatch], or the computer data printouts, right out in the field without having to be on the voice radio quite so much. So that was the start of it. And when it first came in, pretty soon they found that there was going to be some interface work needed to get the CAD system online. And so they started out taking a little slower step by bringing it in with just running messages between units and having querying ability for looking at licensing wants and warrants of that sort. So it took a few years before all the designing and engineering could get to the place where we could bring CAD out to the cars. Dan: Did you eventually see some reduction in voice traffic after it was installed? John: Right. Initially there was some reduction, and that's when we were just doing the wants and warrants licensing checks on the cars. But when we brought on the ability to actually see the dispatcher's computer data, that's when we saw a real significant change in what we heard on the air. For example, in the past you might have somebody calling back to the dispatcher to verify what the address was, or asking for a phone number for the caller, asking for driving directions...a number of things that were available right in what the call receiver had put into the CAD system. But in the past, you could never get that information without asking for it over the air and, in effect, needlessly tying up the voice channel. Dan: How was coverage of the system? John: With the old system we started off with just two sites we could get licensed. Obtaining licensing was a little bit of a trick due to the congestion. We're just across the water from the Seattle, Tacoma, and Everett area where they have larger systems. And so there's always this fight for available spectrum in channels. At that point, we could cover about 75 percent of the county. Then over about the next five years they were able to get some licensing and as other people moved off to trunk systems and such. And so we ended up with having seven sites available in the county. We're about 37 to 38 miles long and about 10 miles wide on average on a peninsula. And the terrain was just such that even with those seven sites and running six channels, we could only get up to about 80, maybe 85 percent coverage on the county on the old system. Dan: Interesting. You had a number of agencies that were using the systems. Did you have any problems keeping equipment that was in use by those agencies synchronized across the agencies? John: The problems weren't so much because of the network. We were having more problems because of separate agency funding abilities. And that's something that we set out to change when we got the technology grant, was to get us all on one synchronized path for bringing updates into the system and starting a replacement fund. So the grant really helped us have that breather to where we could use that more for the capital expenses while we were able to offset some money to start a replacement fund and get everybody on a common platform. The other part of the story is on the old network it was hard to keep the updates of software and the patching done. The old network really did not support much throughput, and it was on a proprietary protocol where we didn't have full IP [Internet Protocol] connectivity. And we couldn't push software updates or patches to operating systems, anti-virus, any of that. Of course, we weren't out on the network as far as the Internet goes, so we didn't have the need quite as much. But we found as we wanted to bring in more applications that providing updates for those applications was going to be nearly impossible on the old network. Dan: Was that the primary drive to switch to the commercial services? John: The primary drive, well, it's a couple of things, but the primary drive was to get more applications out to the field. And that's what this new network's allowed us to do, is really get something out there where we can connect both to the client-server type applications and the web-based applications, which we just couldn't consider on our previous network. Dan: I see capacity, whether you're looking at it from throughput or simultaneous usage is obviously a big difference. What sort of impact do you believe that will have on the applications you use? John: Well we are already finding...we've had just the new network in place for basically less than a year, and we're already finding that people are finding new web-based applications that they want to use in the field. Before, we were pretty much limited to accessing our CAD system in making some queries. We really couldn't get back to our own records management system or the system that we share with the agencies around us. We have both a county shared records management, and then there's a federally sponsored one that covers the Puget Sound region and beyond. And we couldn't access those. Now we can get to those with a truly IP-based network. And it seems like every few months we hear of a new application. We've been able to bring in electronic ticketing and electronic collision reporting, for instance. And in the past if we'd done that we would have had to bring the laptops out of the car every day to synchronize onto our wired network at the offices, and now that can be done live in the field. We've had some experimenting we're doing lately with license plate readers in the field. That's also something that can be carried on the network in school mapping, Critical Government Facilities Mapping. It just seems like every few months we're back to thinking about a new application that we can provide out to the field. Dan: With the availability of common Internet protocols like TCP/IP and others like that, it seems like, any application generally that's available we would expect over our wired data networks might be available to you in the car too at this point. What sort of decisions have you had to make about the type of applications that you use? Are there any that are off limits or just not practical that you ran across so far? John: Well, it comes to technical practicality or technical ability. We did run into a small challenge with our records management system because we found that the vendor did provide us a mobile version, a field version, of that. But they really didn't give us all the functionality that we wanted the deputies and officers to have. For instance, they were going to have a little trouble doing some research or even seeing if we had a mug shot in our photo imagery system. So, we went out and bought another technology to basically allow us to run almost a thin client operation where we can pretty much get the screen shots from the office and from our client there. But from operational perspective what we have to limit our people to, are things basically... if an application can make an officer's job more efficient or save them time, we're trying to provide it out to the field. However, if there was an employee that thought that they needed to be able to check on their personal email from the field, well then that's something we look at and scrutinize as to whether or not that would be necessary or wanted. And in most cases we filter something that we don't deem as necessary out of the system. Dan: What kind of risk do you see in using commercial services? John: Well, I suppose you do have the risk that you could have a vendor selected that then goes out of business or changes pricing on you. But what we've found is with most of the carriers, and we concentrated on cellular carriers, that the cellular carriers by and large were not companies that we were worried about them disappearing. And they were tied into government purchasing and rate contracts where we didn't believe that we'd have unexpected business consequences there. Beyond that then we were looking at the technical capabilities and did we have a risk with turning over network infrastructure to a private company, risk with survivability and then down to the disaster, earthquake, large storm, things of that that sort. So that's what we had to look at next. Dan: What did you look at, at survivability and how did that affect your choice of who you went with as a carrier? John: When we started speaking to the providers in the area that was a key question is... what do you have in place for survivability? Do you have emergency power out at your sites? And then of course, what are your coverage areas? And as luck would have it, perhaps not luck on their part but it was fortunate for us ... the carrier that had the most sites providing coverage to our area was the same carrier that provided backup generator and battery power for almost all of their sites. There were a couple of non-critical sites where they didn't have it. But we thought that was a good thing and it did play into our selection, I should say. Dan: Have you had any opportunity to consider what the impact is, or any concern you have about priority access for public safety services or lack thereof that's guaranteed to you? John: At this point it really hasn't been that big a concern for us. Access to the system on the data side of the house has been sufficient. I think at some point, especially as more and more civilian and general business applications get put onto these networks, that we'll see that becoming a larger concern. So I would certainly hope that we see some type of priority access system in the future, but at this point it hasn't been a very large concern for us. Dan: Has your coverage changed much with this? Do you see improved coverage? John: We do. We found that with the cellular providers, we had probably more than double the number of sites and towers within the county as well as some that were across the water, across Puget Sound from us, that also looked into our shoreline. And so we've gone now to where I believe we're in excess of 95 percent coverage in the county for our mobile units. Now saying that I just have to caveat that what we did in some of our outlying areas is we realized we had to put in the cellular bi-directional amplifier and a gain antenna on the vehicle to give us coverage in some of the tougher spots to get the RF signals in. But by and large, we're seeing a lot better coverage. Dan: Can you talk a little bit about the difference in the equipment? I know in the city of Bremerton you're using somewhat different equipment than what you're using out about the county. Can you just describe that for us? John: What they found in the city areas is...I'll step back a moment here. [laughs] But in the city areas there's more cell site density, and even some places where the cell sites overlap the coverage. And so you can actually get sufficient coverage in the cities, at least in this area, for the most part using the antenna right on the computer or on the AirCard itself. Now, when you start to get out away from the business areas and the more populated areas, then we found that we really needed to add something extra to that and get the external antenna on. Because the city police will come out into the county to assist us or investigating cases and things of that sort, they went ahead and put the antennas on their vehicle. They didn't have to put amplifiers. And at times, of course, we'll take a laptop out of a car and just run an internal antenna. And when you're inside the city and business areas, that really works quite well. Dan: Let's talk a little bit about costs. Some agencies around the country that are contemplating switching from their legacy mobile data networks to commercial wireless service struggle with the thought they're going to have monthly recurring costs. For example, grants won't cover for ongoing service outside of the grant period. How did you evaluate costs of the switch overall, and ongoing costs and how to put your grant to use on this? John: I don't have all the finance figures right in front of me but I can say in general that it seems that it'll be about $50 a month, at least we found in our area, for our AirCard fees, and that we have to compare that with our infrastructure and upkeep on the old system. With the old system we had base stations, and at some of the tower sites there, we were paying a lease fee. And we had to have technicians out to those sites. We had controllers and interface software that we were paying annual licenses on. And then we had at least $1,500 worth of data capable radio and modem installation on each of those cars. Well if you looked at it as a five-year lifespan of those installations, that's $25 a month. So we were about half the cost of what we pay in the cellular fees just to maintain our own equipment, and probably a little bit larger percentage. Whereas we found with the cellular providers, in some cases they could provide, inclusive of our monthly cost they would agree to provide the modem for us and then we were picking up the cost of amplifier and antenna. So on a monthly basis I suspect it is slightly more expensive. But if you look at the big picture for what it's costing you to run a private network, it's substantial. And you have to weigh the ability of, do you want to spend that substantial money to maintain a system that won't provide you all of the ability for the applications and it won't provide you all the throughput, or do you spend a bit more each month and actually get a tool that does a lot more of what you need it to do. And there's a real significant difference incidentally when you look at the throughput on these systems. Even if we had decided that we were going to go out and update our proprietary...the government owned network, there would have been a lot of infrastructure costs. And we would also be facing the narrowbanding that's coming at us as our old channels at 25 kilohertz channels in UHF. And so we were going to have to get equipment that could give us a higher throughput to get licensed and maybe combine some of those channels therefore reducing the number of sites we could run. Or we were going to have to live with a slower baud rate. Even at that, probably not be able to maintain the number of sites we wanted to keep our licensing. Dan: Sounds like you did avoid some expenses that would have occurred had you even stayed with the initial system. Were there any other specific costs that were offset by making the switch? John: We had a few other things that played into the financial picture. For instance, we had the integrated biometric identification system, we call IBIS. Basically a handheld fingerprint system in the field. We have about 60 units fielded for that. Well, each of those units had its own AirCard and data connection to get back to the server. We've been able to work with the vendor on that project to change it over so that they can actually use a short range wireless connection from the handheld device back to a laptop that's in the patrol car and then piggyback onto our mobile data network, onto the cellular network to get back into the servers. So, for those 60 units, they no longer have that monthly cost, that additional AirCard connection that we had in the past. Dan: I see. Do any of the agencies bear a different cost for participating or is it pretty well leveled across agencies? John: It's pretty well leveled across the agencies, it's... per each laptop that's fielded. Now, we have a few of the city agencies that are running their patrol cars or their laptops for two shifts per day. So they have fewer laptops out in the field and therefore their monthly costs are a little less. Whereas out in the county and with some of the tribal and city agencies, you have a program where the officers and deputies are assigned to a particular patrol car and that patrol car is also taken to their residence and they're available for call out and things of that sort. And of course, on the fire side of the house, it's a little bit different type of fleet. There are some differences there in cost. But if you look at ...if an agency comes in and says that it wants 20 units for its staffing and the agency next to it comes in and says they'd another 20 units, they're both going to pay the same cost per unit. Dan: Economic times have been pretty tough around the country here in the last year to two years. Have there been any economic impacts in the region that have led to more or less support for using commercial services? John: Well, you might think that, with budget cuts, they'd be looking to cut back on the commercial network. But the tack that's been taken at this point is, that since we're working with fewer deputies, fewer officers, fewer firefighters than we've had in the past that we need to enable all of the field personnel to do more out there. So we need to leverage the technology. So the support up to this point has been very good for giving the tools to the guys in the field and letting all of them work in the field much more autonomously and not rely so much on the support staff that we would have required in the past. Another issue that's come to light is, if we get to the point in our budgeting where we end up having to close some of our remote offices where we're renting office space and paying for a wired network out to those offices, it's been discussed that, now that we've provided a better wireless ability, that we can pull that off and the deputies and officers can actually work from their laptops without having to be so dependent upon some of our remote rented office space. Dan: You've had a project here to improve your communications generally and information sharing. Has the project improved interagency cooperation in other ways? John: They have technology-wise, when we brought in automatic vehicle location, where we've seen this for multiple agencies. And so we went ahead and enabled all of the agencies to see each other on the mapping screen. So we can get better field coordination. If you're requesting mutual aid from another agency, they're tied into the same system. What you're getting out of the CAD system is the same. What you're seeing on the map is the same. It's just the coordination works better there. And the messaging are the same formats. But, on a less formal basis, just bringing the leaders of the agencies together to talk about how to format the system has had a better team building perspective. It's just been really nice to see how everybody pulled together to make this system work in a cooperative manner. Yeah, informally you also get the conversations that just spring out of... if you're talking about technology, then you end up having to talk about...well, what's the practical use on the field and how do we perform our jobs better. Dan: John, in closing, what are your takeaway messages for agencies similarly switching from agency owned systems to commercial network services? John: Well, I think the first one is to look at the future. If you look at technology in the last 10 years and probably into the future 10 years, wireless technology has been evolving. We hooked up our system at 9600 Baud and we thought oh, this is great. It's going to carry us for quite a while. Then 19.2 became available. But today we're looking at networks that are carrying easily 100 kilobytes per second, 200 and then we're going to have 4G coming down the pike. And these will be coming fairly rapidly. So when you're comparing the systems, think about what the future's going to bring. If you want to go out and upgrade your network, say in five years your technology's changed quite a bit, carriers can provide it to you without a great deal of cost. How do you pull that off internally? Do you have the infrastructure, if you had a private network, could you do that? And I think that's a key issue to look at is getting the applications out there but providing a ability to upgrade in the future. The other thing about this is work with the neighbors. We've seen a lot of benefit. We can share some of the infrastructure. In fact, most all of the infrastructure we'll be able to share with the neighbors. There's things we can do as a group with the police departments, the fire districts, the sheriff's office that we never could have made happen if it wasn't for working as a team and working cooperatively, not only on the operations into the house but with our information services, information technology, our dispatch center technicians, and administration. So the biggest issue with all this I think to make it affordable and manageable, is work with your neighbors. Dan: John, it sounds like you've had a successful project there and your work with Bremerton Police Department has been very progressive. Other agencies around the country are contemplating similar shifts and your sharing your experiences here is very much appreciated. John: Oh, you're welcome. I hope some people can get some value out of it. Dan: You've been listening to SEARCH podcast on Transitioning from Agency-Owned to Commercial Wireless Data Networks. We'd like to thank our guest, Lieutenant John Sprague with the Kitsap County Sheriff's Office in Washington. This project was supported by Cooperative Agreement 2007-CK-WX-K002 by the U.S. Department of Justice Office of Community Oriented Policing Services. Points of view or opinions expressed in this podcast are those of the guest and moderator; do not necessarily represent the official position or policies of the U.S. Department of Justice. [music] Transcription by CastingWords p.