Smoothening the GSM-R co-existence with UMTS900
It seems not easy for the regulator to find a fair solution satisfying both camps, but in fact it turned out that the difficulties are commonly exaggerated.
One could think that frequency regulation is rather simple, and that the complexity of allocating frequencies is commonly exaggerated. Isn’t the role of the regulator basically just to follow standardisation and sell the expensive frequency licences for mobile operators? This of course is not true, which I came across again while studying the co-existence issues of GSM-R and UMTS900 networks.
Being constantly involved in mobile broadband consulting engagements, LTE and HSPA+ rollouts, smartphone market studies and other bells and whistles of the mobile industry, I need to admit that GSM-R system has not been so close to my heart lately. From mobile industry perspective it is a marginal technology as it is designed only for non-commercial railway communications. The co-existence of the systems was however a far more interesting topic than I could have imagined.
In principle the GSM-R system is basic “good old GSM”, enhanced with some services and features tailored for railway use. It is operated right below the commercial 900-band for GSM and UMTS without protecting guard bands. UMTS900 on the other hand has been a true success story. The operators are able to boost the mobile broadband capacity and coverage in the rural areas and cost-efficiently supply the increasing demand. UMTS900 deployment has also woken up the GSM-R operators who are now worried about the increased interference levels. As an independent party following the mobile industry, we wanted to study if there is a real reason for the worry of the GSM-R camp. Furthermore, we wanted to build an understanding on the best practises for GSM-R and UMTS900 deployment in the same geographical area.
Commercial UMTS900 operators are often located next to each other in frequency domain without guard bands, thus one could think that the co-existence of GSM-R and UMTS900 wouldn’t be more complicated. But we found that the differing design principles of the networks and a bunch of other factors cause GSM-R networks suffering from the co-existence of the systems. Though GSM-R is the victim, it’s actually easy to sympathise any player of the game. Why should the commercial UMTS/GSM operators satisfy with limited usage of the scarce and expensive frequency assets while they are using standard equipment and following regulations. On the other hand the GSM-R operators have been allocated a spectrum, which was thought to be practically free of interference.
It seems not easy for the regulator to find a fair solution satisfying both camps, but in fact it turned out that the difficulties are commonly exaggerated. We were able to identify a set of deployment rules allowing a rather smooth co-existence of the systems, without excessive rollout limitations. In my view, with collaboration and by stretching just a little bit further than one considers the maximum (gentleman players), the conflicting parties could share the living room in harmony. The regulator has a critical role as well: A peace enforcement without shared responsibility can easily become a “Treaty of Versailles” – the underdog will have scores to settle, which kills the living conditions of collaborative attitude (objective referee). And do-nothing is a true time bomb (good rules).
The co-existence of GSM-R and UMTS900 is treated with greater detail in a related White Paper, which you can find here.
Although GSM-R is not so sexy in the era of the current mobile broadband hype, to me the topic was very interesting and it opened my eyes for the complexity of our industry and regulation. For anyone being curious to learn something new about telecoms regulation problem zones, I highly recommend reading the paper.