It was about a decade back that I was first introduced to IP Multimedia Subsystem (IMS). We had some lab implementation and experimented with a wide variety of session based applications using Java based SIP servers and a Diameter Server – from voice services to interesting video applications. And, the technology was coming… coming.. and, coming!
I had almost got into a superstitious belief that any acronym that starts with “I” was doomed to fail in the telecom world. ISDN was still born, IN belied all promises and became a 3-application wonder (prepaid, toll-free and portability) and IPTV making us all scurry around was not one would call a success!
So, it was with great satisfaction that I saw the return of this old friend with VoLTE. However, having day-dreamed about a slew of applications around this platform, I find it unsatisfactory to see it employed in a narrow sense as VoLTE. I am sure that the range of session oriented capability for quad-play would expand in days to come.
Along with a friend who is an RF expert, I sat down to look at one more possibility and what it could mean to an operator – VoWiFi taken up as a WiFi offload service. We did a few planning models and found that when the IMS Core is used to offload about 30% of traffic, it can free up radio spectrum roughly to the tune of about 25% under certain assumed circumstances!
Imagine the improvement in data service experience that one could achieve if not redesign the number of carriers used.
Is 30% offload a feasible option? Honestly, we don’t know.
However, given that the modulation schemes progressively become less efficient as the users moves away from the eNodeB, we also found that the possibility of gaining from offload improves if the hot-spots are created farther away from the eNodeB sites.
And, based on some sample network data (geolocation snapshots) that I had looked at, I found quite a few sites to have a large user population (>= 50%) that would potentially be at distances where they use the least efficient modulation scheme. Unfortunately, this was only a sample data and a snapshot in time – so not generic enough to make the abovementioned savings in spectrum as The Rule.
I also realise that with a WLC on the cloud, it is possible to bring in a few more interesting scenario – SDN/NFV cases!
One last point, unexplored yet – the quality of the voice services over WiFi! Whatever that be – this resurrection would certainly go a long way in breaking my superstition about the”I” series acronyms! 🙂