If you were in the middle of somewhere (pick a place) and you had 4 towers all around you. (GSM thanks Greg)
What goes into why your phone chooses one tower over the other? If I made several calls in a row, would I get the same tower each time even if I don't move from the same footprint I was standing in when I made the call (maybe the answer to the first question answers the second part, sorry.)
If you have the hankering to help out one more time, applying the same situation as above, would/could the calls hit different faces of the tower if you aren't moving?
Thanks D
My thoughts I don't believe there would be a situation where you are "precisely" in the middle of several towers, but ultimately the phone should connect to the tower with the strongest signal - so whichever tower is closer to you would be the. If signals overlap and the towers are from the same carrier, it may hop between the towers if you're moving about, or if an obstruction that is not visible to you is cleared increasing the signal from one tower over another. Of course if it's a tower from another carrier, you'll hit the roaming…
As I understand the system, this is a very loaded question, but I'll give it a go….
If you were in the middle of somewhere (pick a place) and you had 4 towers all around you. (GSM thanks Greg)
What goes into why your phone chooses one tower over the other?The phone doesn't pick the tower, the system feeds the call to the tower of choice.
If I made several calls in a row, would I get the same tower each time even if I don't move from the same footprint I was standing in when I made the call (maybe the answer to the first question answers the second part, sorry.)
The tower could change for many reasons. One being system load, if there are a large number of calls in one tower yours could be shifted off to another to level the system load.If you have the hankering to help out one more time, applying the same situation as above, would/could the calls hit different faces of the tower if you aren't moving?
You could potentially hit more than one face of a tower if there is overlap at the edges of the field of view for the antenna.Thanks D
First, you should seperate the concepts of distance and signal strength. You could be in the balance point of signal strength between towers, theoretically…but not in practice. You could also be physically equidistant from the towers, which has little or nothing to do with the tower that would be used. Remember that your phone will pick the strongest signal, not the closest tower. This happens in reality all the time…because of line of sight (primarily).
Second, as to your question directly, tower traffic (load) and signal strength are the two primary determinants on which tower is selected. Additionally, latentcy (hanging on to last tower piloted) has signfiicance in determining which tower a call may be connected on. There are also differences in tower selection by the equipment on each tower (sectorized v. omnidirectional) and the amount of arrays dedicated to each sector per tower (traffic load capacity).
Equidistance and equal power, however, can only be theoretical, because traffic load and signal strength may change with the same physical location of the handset…so there will be changes to the tower connections based upon those, and other, factors even though the handset has not moved.
To keep the OP's question live, a few examples of typical cause values why a mobile network might use another mast is used, where a particular mast which appears to be ideally placed with respect to the location/positioning of a mobile phone is not used.
The assumption being made for these cause values is that the MS is seeking to obtain a service for speech calls
│7 6 5│ 4 3 2 1│ │
│
│0 0 0│0 0 0 0│ │Radio interface message failure │
│
│0 0 0│0 0 0 1│ │Radio interface failure │
│
│0 0 0│0 0 1 0│ │Uplink quality │
│
│0 0 0│0 0 1 1│ │Uplink strength │
│
│0 0 0│0 1 0 0│ │Downlink quality │
│
│0 0 0│0 1 0 1│ │Downlink strength │
│
│0 0 0│0 1 1 0│ │Distance │
│
│0 0 0│0 1 1 1│ │O and M intervention │
│
│0 0 0│1 0 0 0│ │Response to MSC invocation │
│
│0 0 0│1 0 0 1│ │Call control │
│
│0 0 0│1 0 1 0│ │Radio interface failure, reversion to old channel │
│
│0 0 0│1 0 1 1│ ││
│
│0 0 0│1 1 0 0│ │Better Cell │
│
│0 0 0│1 1 0 1│ │Directed Retry │
│
│0 0 0│1 1 1 0│ ││
│
│0 0 0│1 1 1 1│ │Traffic
Greg,
I'm unclear on 7654321. Is that the dialed number? Now with all the other fields is there a specific numerical value that the MS station is looking for to obtain a connection or is the MS just searching for the greatest value among available antennas? With all the fields you provided and the values in the columns could you look at two different fields/reports or whatever you actually call the example you provided and pick which antenna the MS will communicate with?
Hi Ed
In a bit of a hurry this morning, so maybe I haven't answered everything you are asking.
OK, I see what you mean, but no its not a dialled number.
It is a 'byte' (as in bits, nibbles and bytes). The cause values you see above show only the bit representations that are used to to define a cause (a reason for doing something).
These cause values would be used by the network, they are set out in the GSM standards - because this is a GSM question. These types of messages would run between the base station subsystem and the mobile services switching centre. The cause values are a reason why intra- or inter- cell 'handover' may take place.
Those reasons are not translated in every case as to the external set of events that caused a handover to take place relevant to an MS call e.g. a reason for traffic overload, merely an action necessitated moving the off-air-call-set-up, call establishment or a call in progress to another cell.
Placed in context with the OP's question, knowing cause values can help the person conducting cell site analysis identify possibilities why a network may use one mast for a call rather than another.
The MS wouldn't see the cause values above as it does not command which mast will be used.
The MS has three 'functions'
* Requests
* Action commands
* Receive responses
Equally the MS deals with the 'functions' across three strata
* (MM) Mobility Management
* (CC) Call Control (Connection Management)
* (RR) Radio Resources
Any of the strata may present to the MS a reason why something happens or doesn't happen. The reason detected by the MS can be presented on the MS screen either in alphabet (words), hexidecimal notation or decimal-digit notation. These may be displayed to the user either in normal mode or in engineering mode.
Cause values received by the MS are recorded in the GSM standards.
For the purposes of the OP's question, it allows the reader a broad range of options to consider about the use of a mast, which can be separated into two MS states
- Idle Mode
- Dedicated mode
However, there is a small area where the MS can have a direct affect or at least an impact when a mast/cell be used or not
- terminating a call request (end button)
- turn off the MS
- removing the battery from the handset
And also, a less well considered matter is where an MS has a call already in progress on one mast and another call is received at the same time from another mast. Usually that would be resource wasteful and doesn't often occur and should happen if a TCH is already in use and the operator should use a Stealing Flag to alert the user on the same TCH, but it can depend upon network conditions at the material time. In this case terminating the first call to answer the second may generating a receiving call record from a mast that might just put the MS in another area than expected.
Adding some more relevant to this thread
GSM Cell Selection Process - http//
CSA - GSM Timing advance and absolute delay - http//
CSA - R&TTE Directive - http//
Updated
Use of GSM Logical Channels for CSA - http//
Directed Retry
A mistake that experts and investigators could make would be to ignore the existence of Directed Retry and, even more problematical, not to have asked the question was Directed Retry active at cell/BSC level at the material time of the calls, apart from any intervention within the network.
http//
Jamie, we need a "thumbs up" or something widget on posts! I do not want to constantly exclaim in one-liners -
Thanks Greg, this is very good! mrgreen