IDEA

The idea comes from the fact that "mobility" which is the integral feature of the mobile networks does not discriminate against users, as well as system nodes. So, installing radio nodes onto moving vehicles and actually moving them in space will not change network functionality as long as backbone and air interface are functioning without disruptions. Therefore, it is possible to use UAV or other type of aircraft as Radio Base Stations, directing antenna towards earth and using satellite backbone or even Ground stations for backbone where possible (closer to the shore or in continental remote areas, within the reach of ground stations)

This way users in remote areas will be able to use their own, existing mobile devices, setting initial cost for the user essentially equal to $0!

There are challenges that needs to be analyzed and the main ones are described in the following sections.

ISSUES DISCUSSED DURING PROJECT

1. User Range: The first obvious question is weather airplanes are too high, and therefore to far for any kind of decent reception? Today's LTE network, depending on the network settings, has coverage up to 100 km (Check Aircom LTE Training on YouTube especially PRACH settings section). That is of course, if the radio conditions are favorable to allow for such propagation of the signal. Having in mind that we are trying to cover users from the air, most often having line of sight between airplane and the user, we are actually facing much better coverage conditions than in existing networks in urban and dense urban areas. Taking into considerations today's usual commercial airplane speed of about 900 kmph and the height of about 10 km, there is roughly up to a 15 min window of opportunity for the communication with over-passing plane. As one can observe from the one of the available web sources that are dealing with commercial air traffic (i.e. flightaware.com) some areas of the ocean are pretty busy with traffic, with many opportunities for communication, like north Atlantic ocean between Europe and North America. Others are unfortunately not that covered. For the areas that are not covered with commercial traffic, fleet of specialized UAV may be used. Additional UAVs may also be used for well covered areas of the ocean to decrease disruption times and improve network capacity and performance. UAVs also have the possibility to fly lower and slower, therefore, increasing time for connection opportunity. Usage of so called Low Bands (i.e frequencies in the 600, 700, 850 and 900 MHz bands) which is very common in today's network, now that GSM is phased out and that part of the spectrum is becoming available for new technologies.
2. Airplane Speed: Today LTE network does not have strictly defined moving speed limitations, and all moving speed issues are related to rapid changes in the environment and excessive handovers that are reducing connection quality. In the ocean or other remote areas, where environment will not be changing rapidly due to high speed, airplane speed should not be an issue. Also, as previously discussed, 15 min window is more than enough not to cause frequent handovers, even if there are other airplanes with connection capabilities in the area.
   
3. Handover: Each Radio Base Station has a list of neighboring Base stations where connection may be handed over due to mostly user mobility. In an environment where Radio Base Stations are moving as well, keeping the list of neighbors does not makes sense. Fortunately, starting with LTE system ANR feature is used for automatic neighbor list creation (Automated Neighbor Relations). This feature is generally set in such a way that the first time the new neighbor appears in the UE measurement report with enough power, neighbor relation is automatically created. Similarly, relation is automatically deleted if not used for certain amount of time. ANR feature is essential in the case of moving base stations.
   
4. Code planning: In order to distinguish between all available Radio Base Stations in the area, which are all using same frequency in the same time, code planning is needed. PCI (Primary Cell Identifier) is embedded into the signal of each base station and UE (user equipment) uses known protocols to look for them, synchronize and decide about which station is using for communication. Code planning is relatively simple procedure and due to fact that there was no need so far, it is performed outside of the network and planned codes are later fed to each cell. Pretty good code planning can be performed using desktop tools like Microsoft Access and few simple queries with known locations for base stations, and author of this text was performing it multiple times. Moving base station complicates this procedure a bit, but the problem is far from being unsolvable, especially with small number of base stations (airplanes) and high number of available codes that is 512 in LTE network.
   
5. Antennas: Antennas that are used in the mobile networks today are mostly directional High Gain antennas with 65 deg horizontal beam-width and about 10 deg vertical beam-width. This is highly desirable on ground based systems but totally useless when used on airplane with the intention to cover towards ocean/ground. In this case we need some dish type antennas with much wider coverage areas, possibly about 90 deg beam-width in all directions. It does look very clumsy, ugly and quite possibly, installing Dish antenna on the bottom of the airplane will affect aerodynamics of the airplane. But fortunately, there are other solutions and beam-forming panel antennas are becoming increasingly popular in LTE, and integral part of 5G system. Beam-forming antenna is dynamically changing it's coverage pattern so it can focus on each or group of users at the time, increasing range of coverage as well as the quality of the connection. This type of Dual Mobile system (where both, users and system nodes are moving) is an excellent use case for the new generation of beam-forming antennas. This kind of panel antenna will have minimal to no effect on aircraft performance, in the same time providing better user perception. More on beam-forming here (https://www.rfglobalnet.com/doc/a-comprehensive-guide-to-active-antennas-or-beamforming-0001)
   
6. Satellite Backbone: Satellite communication is a common solution for backbone on ground base stations located in the remote areas today, so it shouldn't be an issue for system based on flying base stations. Capacity and performance provided by satellite compared to microwave or optical backbone is certainly lower, but high level of traffic and high number of users is not expected in this case. Satellite coverage and capacity is increasing daily and will provide for the possibility for the growth of such kind of mobile network.
   
7. Network disruptions and delays: As mentioned above, this network will have a lot of coverage gaps, both in space and time domain. It is essential to provide the network protocols that will ensure connection is not entirely interrupted and all data lost in the case of sudden or planned coverage loss. Users in the middle of the ocean or in remote areas can not afford to wait for another connection opportunity to start from scratch. Application must not fail and data must be preserved for reasonable amount of time, until the next connection is available. Based on the description of the NASA DTN technology, this system is essential for the function of such, lets name it Dual Mobile system (DTN Reference page). It contains most of the necessary features to enable disruption and delay tolerance. Possibly handover supporting feature needs to be added so the system is compatible with mobile network. This feature should predict next possible base station to be connected to end user in case if network interruption and data that was stored in previously used base station needs to be resent to the new one. Users will not be able to use all existing software on their mobile devices. Phone software needs to be updated with DTN support and possibly browsers and other applications with DTN support need to be installed.
   

NETWORK CARRIERS

It is hard to imagine one business taking over such complex task of providing network coverage across multiple oceans and continents. Possibly strong international organisations and/or cooperations like the one that is running ISS will be needed to lead this endeavor. It may be possible for multiple entities to be connected into global network each one profiting depending on the amount of traffic generated by base stations they own, like airline carriers or even each private jet for its own owner, existing carriers that are looking to extend their coverage. Also users may be treated as the part of the coop, where users by paying fees, are becoming partial owners of the coop.

Existing carriers may use the idea to cover remote areas quickly.

OTHER USES

Africa and Asia have areas that cannot be called remote because of high population numbers, but are certainly lacking infrastructure that may delay or even stop classic ground mobile network penetration. It may be good idea to start using UAV fleet to cover this areas quickly and efficiently.

NB IoT sensors that do not need constant connection to the network. (Narrow Band Internet of Things)