Many, if not most, of my posts recently have involved seemingly non-wireless related topics. I tend to focus primarily on computer and networking related topics. Part of the reason is due to networks being what I work with on a day in day out basis but the other reason is much larger in the scope of wireless communications. Over the last 10 years most major radio systems have migrated from circuit-switched technologies that were commonplace in telephone systems such as AT&T’s Bell System to IP switched and IP routed systems that resemble the networks operated by thousands of enterprises all over the world. Channel banks, analog telco switches and T-carrier based microwave radios are giving way to Ethernet switches, routers and newer IP native technologies. Leased analog pairs are being dropped from support by multiple carriers around the world and becoming ever more impossible to order and are having to be replaced by Voice/Radio over IP systems (VoIP/RoIP) just in order to keep the systems online.
Why IP?
There are many reasons to go with IP. For one, many of the guys with analog switching knowledge are reaching retirement and most communications engineers these days are ingrained with IP technology in school. IP can also be much more efficient.
As an example, take a circuit switched multi-site trunking system. In that type of system, any voice traffic destined for another site would be sent down a T1/E1 trunk. Since I’m based in North America, I’m more familiar with the T-carrier system so for my sanity we will the example system off a T1. An on premises trunking controller would “route” any wide-area (meaning multi-site) talkgroup traffic down the T1. Since a T1 is made up of 24 Time Slots with analog 8 bit audio sampled at an 8,000 Hz rate (that’s 64 kbps per slot), each active call would get a slot dedicated to it. With one slot used for command and control for the trunking controllers with the rest of the system, that means there was a physical limitation of 23 active simultaneous calls which the T1 (and thus the site) could support. Fast forward to the first generation of digital radio systems where voice calls are processed by the system as raw data. The switching hasn’t changed, each talk group active gets a time slot on the T1 but as stated earlier a T1 time slot has 64 kbps of capacity but the data streams being dealt with are 9600 bps which appears to be wildly inefficient as that is but a fraction of the capability of the time slot. Bring in the IP switched and routed systems where we can aggregate the data handling capability of the the T1’s time slots to 1.544 Mbps and just send the necessary data as IP packets. All the sudden a system that was using the fully 1.544 Mbps T1 in a circuit switched configuration is now only using 512 kbps as IP…which means we can expand the capacity and capabilities of the system without having to upgrade the old T1 to a multi-T1 bonded setup or even a T3.
IP Doesn’t Mean Internet Required
One of the common misconceptions that is often heard, often times from amateur radio operators, is that services of an IP based system will go down when the internet goes down. This isn’t at all the case as IP doesn’t mean the internet is involved. In fact, many modern radio systems are 100% air gapped from the internet and operate in their own standalone networks relying on leased point to point circuits from telco providers, microwave radio systems and other methods of extending ethernet from point A to point B. Add in, that one of the beautiful things aspects of IP based systems is that it is far easier and more cost effective to build in additional redundancy. This allows the creation of networks which are dynamic, self healing and are capable of uptimes measured in years without suffering any major outages. IP hardware is even lower in cost with a $20 Mikrotik router having enough “oomph” to handle the networking needs at the average P25 site but this should also be a great time to mention, just because a specific product can handle the needs of something doesn’t mean it’s the best product for the job and I am in no way bashing on Mikrotik as I actually run several DMR trunking sites off Mikrotik routers for friends as they provide the best capability for the price compared to some of the more mainstream Cisco, Juniper and Arista offerings.
One of the great benefits to IP is just how scalable it can be. Take the above diagram of a typical site. In the event of a point to point path failing, the other path takes over. In the event of a router failing, the other router takes over. In the event a switch failure, the site may lose half of it’s capacity…but it won’t be completely down. That’s the beauty of the redundancy offered with IP networking and it can be as redundant as needed (for example, remove a router and switch and you can remove the redundancy).
Concluding
The industry is changing and IP is certainly the direction everything is headed. Very often these new radio systems are completely air gapped and often have system updates tightly regulated. Understanding how IP correlates with modern Land Mobile Radio systems is critical however it is not the only part of the systems that need to be known. Many system controllers and devices run various services and can range from embedded Linux, to Windows, to some high end virtualization and it will all typically communicate in some form over IP. There are various vendors out there that provide network equipment. What is chosen often depends on the manufacturer and sometimes the customer so it can vary. One manufacturer may choose to design their network appliances around Cisco, another may choose Juniper, a third may choose VyOS on TAA x86 machines but that is why being familiar with how these networks are built has become critical in the industry and will continue to be a highly desirable skill for those wishing to pursue careers or further their existing careers or even maintain an amateur radio network such as the Rocky Mountain Ham Radio club’s system.
