The Limitations of Cellular Data

SONY DSCIt’s hard these days to find anybody that is satisfied with the quality of data received over cellphones. A research report published by Aptelligent late last year showed that the US placed 10th in the world in overall cellular network performance, measured by the combination of reliability and speed. We all know that sometimes cellphone data is adequate, but can suddenly deteriorate to where you can’t load simple web pages. There are a number of factors baked into the cellular architecture that contribute to data performance. Following are a few of the key factors:

Data Power Drop-off. Cellular networks, by design, assume a massive drop-off of data speeds with distance. I don’t think most people understand how drastic the power curve is. Cellular companies show us bars to indicate the power of our connections – but these bars are not telling us a true story. The cellular architecture has a 100:1 data rate ratio from cell tower to the edge of the delivery area (generally a few miles). To provide an example, this means that if a cell site if designed to deliver 10 Mbps at the cell tower, that it will deliver only 1 Mbps at the mid-point of the cell tower range and only 0.1 Mbps at the edge.

Shape of the Cellular Footprint. It’s easy to think that there are nice concentric circles of cellphone signals propagating around cell towers. But nothing could be farther from the truth. If you go around any cell site and measure and plot the strength of signals you will find that the footprint of a cell tower looks a lot more like an amoeba, with the signal in some directions traveling a relatively short distance while in others it might travel much farther. If these footprints were static then engineers could design around the vagaries at a given cell site. But the footprint can change quite dramatically according to temperature, humidity and even the number of users concentrated in one portion of the footprint. This is why the delivery of broadcast wireless services is always going to more an art than a science, because the delivery footprint is constantly shifting, in many cases dramatically.

Proliferation of Antennas. Modern cellular networks have improved performance by significantly increasing the number of transmitting antennas on a cell tower (and also more receiving antennas in cell phones). This use of MIMO (multiple input, multiple-output) has produced a significant improvement for customers who are able to gain simultaneous signal from more than one transmitter. But there are two consequences of MIMO that actually decrease performance for some users. First, MIMO largely benefits those that are closest to the cell tower, and that means there are fewer quality connections available for those farther away from the cell tower. Second, MIMO has a secondary characteristic in that MIMO works best using cellular channels that are not-adjacent. And during time of heavy cellular usage this has the result of improving the signal strength in the MIMO channels but decreasing the strength of the other channels, again decreasing quality for customers that grab the weaker channels.

Impaired Hand Offs. Mobility is enabled in a cellular network when a customer is handed off from one cell site to the next while traveling. MIMO and other techniques that increase the signal to a given customer then make it more difficult for that customer to be handed to the next cell site. Hand offs were relatively error free when customers received a one channel signal from one transmitter, but now the quality of hand offs from one cell site to another can vary dramatically, resulting in more disconnects or drastic swings in the strength of connections.

Small-Cell Issues. All of the above issues will be compounded by the introduction of small-cells into the cellular network. In today’s cellular architecture a customer can only be handled by one cell tower at a time. Cellular networks don’t automatically assign the strongest connection to a customer, but rather the nearest available one. While small-cells will increase the opportunity to get a signal in a crowded environment, it also increases the chance of getting a poor connection, or of running into hand off issues for mobile customers.

2D Signal Propagation. Cell tower antennas are largely aimed to transmit close to the ground and do not waste signals by sending signals upwards in a 3D pattern. Anybody who has traveled to a big city and received poor signal on an upper floor of a tall hotel is familiar with this issue. The cellular signals are focused towards street level and not towards anybody higher. That’s not to say that you can’t get a cellular connection at the top of a highrise, or even in an airplane, but the vast majority of the connections (and the strongest connections) are aimed downward.

Crisis Propagation. Cell towers are arranged as an interconnected mesh. When something drastic happens to a given cell tower, such as losing power or being swamped with calls during an emergency, this not only shuts down the tower with a problem, but the problem cascades to nearby towers, often taking them out of service as well. This is similar to a rolling blackout in an electric grid. Carriers have been working on load balancing techniques to try to tamp down this problem, but it’s still relatively easy for a given cell tower to get overwhelmed and start a neighborhood and even regional cascade.

These issues all outline how complicated it is to design a great cellphone network. The above issues are worsened by the fact that in the US our cell sites were largely placed years ago to accommodate voice traffic and thus are not situated to instead provide optimum data traffic. But even a brand new cellular network designed to optimize data traffic would run into these same or different issues. It’s nearly impossible to design a cellular network that can handle all of the issues encountered in the real world. This makes me glad I’m not a cellular engineer.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s